The
Collected Works
of
Edward Sapir
VI
The Collected Works of Edward Sapir
Editorial Board
Philip Sapir
Editor-in- Chief
William Bright
Regna Darnell
Victor Golla
Eric P. Hamp
Richard Handler
Judith Irvine
The
Collected Works
of
Edward Sapir
VI
American Indian Languages
2
Volume Editor
Victor Golla
1991
Mouton deGruyter
Berlin • New York
Mouton de Gruyter (formerly Moulon. The Hague)
is a Division of Waller de Gruyter & Co., Berlin.
?
537
@ Printed on acid-free paper which falls within the guidelines of the
ANSI to ensure permanence and durability.
Library of Congress Cataloging in Publication Data
(Revised for vol. 2)
Sapir, Edward, 1884-1939.
American Indian languages.
(The Collected works of Edward Sapir : 5 — 6)
Vol. 1 edited by William Bright.
Vol. 2 edited by Victor Golla.
Includes bibliographical references and index.
1. Indians of North America — Languages. I. Bright,
William. 1928- . II. Golla, Victor. III. Sapir, Edward,
1884-1939. Works. 1990 ; 5-6. IV. Title. V. Series.
PM108.S26 1990 497 89-13233
ISBN 0-89925-654-6 (v. 1)
ISBN 0-89925-713-5 (v. 2)
Deutsche Bibliothek Cataloging in Publication Data
Sapir, Edward:
(The collected works]
The collected works of Edward Sapir / ed. board Philip Sapir
ed. -in-chief. ... — Berlin ; New York : Mouton de Gruyter.
ISBN 3-11-010104-1 (Berlin)
ISBN 0-89925-138-2 (New York)
NE: Sapir. Edward: [Sammlung]
6. American Indian languages. — 2. Vol. ed. Victor Golla. —
1991
ISBN 3-11-012572-2
(f) Copyright 1991 by Walter de Gruyter & Co.. Berlin 30.
All rights reserved, including those of translation into foreign languages. No part of
this book may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopy, recording, or any information storage and
retrieval system, without permission in writing from the publisher.
Printing: Gerike GmbH, Berlin. Binding: Luderitz & Bauer, Berlin. Printed in Germany.
Edward Sapir, 1937
University of Michigan, Ann Arbor, Michigan,
taken by Kenneth Pike
( Courtesy of Sapir family )
Edward Sapir (1884-1939) has been referred to as "one of the most brilliant
scholars in linguistics and anthropology in our country" (Franz Boas) and as
"one of the greatest figures in American humanistic scholarship" (Franklin
Edgerton). His classic book, Language (1921), is still in use, and many of his
papers in general linguistics, such as "Sound Patterns in Language" and "The
Psychological Reality of Phonemes," stand also as classics. The development of
the American descriptive school of structural linguistics, including the adop-
tion of phonemic principles in the study of non-literary languages, was pri-
marily due to him.
The large body of work he carried out on Native American languages has
been called "ground-breaking" and "monumental" and includes descriptive,
historical, and comparative studies. They are of continuing importance and
relevance to today's scholars.
Not to be ignored are his studies in Indo-European, Semitic, and African
languages, which have been characterized as "masterpieces of brilliant associa-
tion" (Zellig Harris). Further, he is recognized as a forefather of ethnolinguistic
and sociolinguistic studies.
In anthropology Sapir contributed the classic statement on the theory and
methodology of the American school of Franz Boas in his monograph, "Time
Perspective in Aboriginal American Culture" (1916). His major contribution,
however, was as a pioneer and proponent for studies on the interrelation of
culture and personality, of society and the individual, providing the theoretical
basis for what is known today as humanistic anthropology.
He was, in addition, a poet, and contributed papers on aesthetics, literature,
music, and social criticism.
Note to the Reader
Throughout The Collected Works of Edward Sapir, those publications whose
typographic complexity would have made new typesetting and proofreading
difficult have been photographically reproduced. All other material has been
newly typeset. When possible, the editors have worked from Sapir's personal
copies of his published work, incorporating his corrections and additions into
the reset text. Such emendations are acknowledged in the endnotes. Where the
editors themselves have corrected an obvious typographical error, this is noted
by brackets around the corrected form.
The page numbers of the original publication are retained in the pho-
tographically reproduced material; in reset material, the original publication's
pagination appears as bracketed numbers within the text at the point where the
original page break occurred. To avoid confusion and to conform to the existing
literature, the page numbers cited in introductions and editorial notes are those
of the original publications.
Footnotes which appeared in the original publications appear here as
footnotes. Editorial notes appear as endnotes. The first endnote for each
work contains the citation of the original publication and, where appropriate,
an acknowledgment of permission to reprint the work here.
All citations of Sapir's works in the editorial matter throughout these vol-
umes conform to the master bibliography that appears in Volume XVI; since
not all works will be cited in any given volume, the letters following the dates
are discontinuous within a single volume's references. In volumes where
unpublished materials by Sapir have been cited, a list of the items cited and the
archives holding them is appended to the References.
Contents
Frontispiece: Edward Sapir 6
Preface 13
Introduction to Volumes V and VI 15
Section Six: Athabaskan and Na-Dene Languages
Introduction 21
Notes on Chasta Costa Phonology and Morphology (1914) 27
Corrigenda to Father Morice's "Chasta Costa and the
Dene Languages of the North" (1915) 95
The Na-dene Languages, a Preliminary Report (1915) 105
The Sino-Dene Hypothesis
[excerpts from a letter to A. L. Kroeber] (1921) 133
Athabaskan Tone (1922) 141
A Type of Athabaskan Relative (1923) 143
The Phonetics of Haida (1923) 151
Pitch Accent in Sarcee, an Athabaskan Language (1925) 169
The Similarity of Chinese and Indian Languages (1925) 191
Review of Berard Haile, Manual of Navaho Grammar (1926) 193
A Summary Report of Field Work among the Hupa,
Summer of 1927 (1928) 195
The Concept of Phonetic Law as Tested in Primitive Languages
by Leonard Bloomfield [excerpt] (1931) 199
Two Navaho Puns (1932) 203
Problems in Athapaskan Linguistics 205
1 Contents
Review of A. G. Morice, The Carrier Language (1935) 207
Internal Linguistic Evidence Suggestive of the
Northern Origin of the Navaho (1936) 209
Cornelius Osgood, The Distribution of the Northern Athapaskan Indians
[contribution by Sapir]: Linguistic Classification within the
Northern Athapaskan Area (1936) 221
Section Seven: Penutian Languages
Introduction 225
Preliminary Report on the Language and Mythology of the
Upper Chinook (1907) 231
Franz Boas, Chinook [contributions by Sapir] (1911):
Diminutive and Augmentative Consonantism in Wishram
Post-positions in Wishram
Wishram Text and Analysis
Modal Elements 243
A Characteristic Penutian Form of Stem (1921) 263
A Chinookan Phonetic Law (1926) 275
L. S. Freeland, The Relationship of Mixe to the Penutian Family
[with notes by Sapir] (1930) 283
(with Morris Swadesh) Coos-Takelma-Penutian Comparisons (1953) .... 291
Comparative Penutian Glosses 299
Section Eight: Wakashan and Salishan Languages
Introduction 319
The Rival Chiefs, a Kwakiutl Story Recorded
by George Hunt (1906) 323
Some Aspects of Nootka Language and Culture [excerpt] (1911) 353
Abnormal Types of Speech in Nootka (1915) 357
Noun Reduplication in Comox (1915) 381
The Rival Whalers, a Nitinat Story (1924) 435
Contents 1 1
Nootka Baby Words (1929) 465
Morris Swadesh (ed.), Salish-Wakashan Comparison (1949) 467
Section Nine: Other American Languages
A Tutelo Vocabulary (1913) 471
Review of B. Bibolotti, Moseteno Vocabulary and Treatises (1918) 475
Appendix
A. G. Morice, Review of Sapir, Notes on Chasta Costa
Phonology and Morphology (1915) 481
A. G. Morice, Chasta Costa and the
Dene Languages of the North (1915) 485
A. G. Morice, Misconceptions Concerning Dene Morphology:
Remarks on Dr. Sapir's Would-be Corrigenda (1917) 499
E. Sapir, Corrigenda and Addenda to Takelma Texts (1914) 513
Phonetic Key to Publications of Edward Sapir 515
References 525
Index to Volumes V and VI 543
Preface
Volumes V and VI of The Collected Works of Edward Sapir are devoted to
shorter works on American Indian languages (mainly of North America),
including some previously unpublished material. Volume V, edited by William
Bright, contains papers of a general nature on typology, classification, and
phonetic notation, followed by work on Hokan languages, on the Uto-Aztecan
family, and on the relationship of Algonkian, Wiyot, and Yurok. Volume VI,
edited by Victor Golla, contains articles on Athabaskan and Na-Dene lan-
guages, on Penutian, and on the Wakashan and Salishan families, plus two
short papers on languages of other groups. Appendices in both volumes con-
tain papers written by other authors which were discussed in papers by Sapir. A
combined index to Volumes V and VI appears in the latter.
The editors of these two volumes have worked together in planning the entire
sequence. Two possible ways of organizing the material were considered. One
would be purely chronological, without considering topic; the other, adopted
here, separates the articles into topical divisions and then arranges them chron-
ologically within each division. This has the advantage, we believe, of making it
easier for the reader to consult related papers in close proximity.
In addition to the articles contained in these two volumes, a number of arti-
cles which discuss one or more specific American Indian languages appear in
Volumes I through IV of The Collected Works. These are listed below, orga-
nized by language or language group. The volume in which a paper is to be
found is indicated by the appropriate roman numeral in brackets.
Athabaskan Languages: 1923c, A Note on Sarcee Pottery [IV]; 1924d, Per-
sonal Names among the Sarcee Indians [IV]; 1933c, La realite psychologique
des phonemes [I]; 1935b, A Navaho Sand Painting Blanket [IV]; 1936e, Hupa
Tattooing [IV]; 1936h, Kutchin Relationship Terms [IV]; 1930, A Note on
Navaho Pottery (with Albert G. Sandoval) [IV].
Comox: 1939e, SongsforaComox Dance Mask (edited by Leslie Spier) [IV].
Nootka: 1913b, A Girls' Puberty Ceremony among the Nootka Indians [IV];
1915h, The Social Organization of the West Coast Tribes [IV]; 1919e, A Flood
Legend of the Nootka Indians of Vancouver Island [IV]; 1933c, La realite psy-
chologique des phonemes [I].
Southern Paiute: 1910d, Song Recitative in Paiute Mythology [IV]; 1933c, La
realite psychologique des phonemes [I].
Takelma: 1907b, Notes on the Takelma Indians of Southwestern Oregon
[IV]; 1907d, Religious Ideas of the Takelma Indians of Southwestern Oregon
[IV].
14 VI American Indian Languages 2
Tsimshian: 1915g, A Sketch of the Social Organization of the Nass River
Indians [IVj; 1920c, Nass River Terms of Relationship [IV]; 1921c, A Haida
Kinship Term among the Tsimshian [IV].
Yana: 1908a, Luck-Stones among the Yana [IV]; 1916g, Terms of Rela-
tionship and the Levirate [IV]; 1918j, Yana Terms of Relationship [IV]; 1922d,
The Fundamental Elements of Northern Yana [IX]; 1923m, Text Analyses of
Three Yana Dialects [IX]; 1928j, The Unconscious Patterning of Behavior in
Society [III].
Volumes VII-XV, which contain Sapir's work of monographic scope on
American Indian languages and cultures, also include some shorter, closely
related articles containing lexical inventories and textual analyses. Note that
Sapir s Language: An Introduction to the Study of Speech (Volume II) cites
some thirty American Indian languages, and his 1916 monograph, Time Per-
spective in Aboriginal American Culture (Volume IV), one-third of which is
devoted to "evidence from linguistics," cites dozens of American Indian lan-
guages or language groups. It should also be noted that all references to specific
languages in each article are listed in the indices of each individual volume, as
well as in the comprehensive index in Volume XVI.
Preparation of this volume was supported in part by grants from the Phillips
Fund of the American Philosophical Society, the National Science Foundation
(grant no. BNS-8609411), and the Andrew W. Mellon Foundation.
The editor also acknowledges the contributions to the preparation of this
volume by Jane McGary and the help of Dr. Marie-Louise Liebe-Harkort,
editor in chief of Mouton de Gruyter.
Introduction to Volumes V and VI
It has often been said that Franz Boas is to be considered the father of anthro-
pological linguistics in North America, and in particular the initiator of serious
research on American Indian languages. But surely Edward Sapir, who began
his career as a student of Boas, became the most influential scholar of the twen-
tieth century in both these fields . Consider the diversity of the Native American
languages on which Sapir did original research — Chinook, Takelma, Yana,
Southern Paiute, Nootka, Sarcee, Navajo, and others; or the language families
in which he did ground-breaking comparative work — Hokan, Uto-Aztecan,
Algonkian, Athabaskan, and Penutian; or the types of studies he carried out —
descriptive, historical, comparative, ethnolinguistic, and what would now be
called sociolinguistic. Even before his untimely death, Sapir's achievements
were monumental; after 1939, his stature as an Americanist only grew, as many
of the materials he left in manuscript were edited and published by his students.
His stature grows yet more in subsequent volumes of these Collected Works,
with the publication of several major collections of texts (Sarcee, Kutchin, and
Hupa) and other important longer manuscripts, now edited by students of his
students.
It is possible to attempt some general comments about the overall course of
Sapir's work on North American Indian languages as it is reflected in the pres-
ent pair of volumes. Publications from the period 1906-1910 are primarily
descriptive, including the first results of field work on Wishram Chinook, Ta-
kelma, and Yana. In 1911, typological interest emerges in "The Problem of
Noun Incorporation in American Languages" (1911c) and is pursued most nota-
bly in the two reviews (1917k, 19171) of works by Uhlenbeck. Comparative lin-
guistic research, aimed at establishing relatively remote linguistic relationships
on the basis of both lexical and grammatical comparisons, comes to the fore in
1913 with "Southern Paiute and Nahuatl, a Study in Uto-Aztekan" (19131, 19151)
and "Wiyot and Yurok, Algonkian Languages of California" (1913h). During
the following half dozen years, Sapir's enthusiasm for tracing remoter rela-
tionships is manifest in such papers as "The Na-Dene Languages" (1915d), "The
Hokan and Coahuiltecan Languages" (1920b, written in 1915), and "A Charac-
teristic Penutian Form of S-tem" (1921b, written in 1918). This interest reached
its culmination in a drastic proposal to reduce 58 North American "stocks" (as
formulated by John Wesley Powell in 1891) to just six "great groups. " This classi-
fication, based on grammatical and typological rather than lexical corre-
spondences, was presented in a lecture at Chicago in 1920 (the notes for which
are published here in "Materials Relating to Sapir's Classification of North
American Indian Languages"). With little change, this formed the core for
Sapir's influential Encyclopaedia Britannica article on "Central and North
15 VI American Indian Languages 2
American Languages" (not published until 1929, 1929a). After the early 1920s,
Sapir s interest in these problems seems to have cooled; however, his last major
work in this genre, "The Hokan Affinity of Subtiaba in Nicaragua" (1925b),
argues for a Central American extension of the far-flung Hokan
(-Coahuiltecan) group, and presents what is perhaps Sapir's most detailed argu-
ment for the importance of "submerged" structural features in recognizing
remote linguistic relationship.
Sapir's sixfold classification and the methodology supporting it constituted,
during his lifetime, the most controversial part of his work on North Amer-
ican languages (it was never accepted, for instance, by his onetime teacher
Boas). It should be remarked, however, that what Campbell and Mithun
(1979: 26) have called the "reductionist zeal" of this classification was not
unique to Sapir. Large-scale genetic regrouping of North American languages
was initiated by Alfred L. Kroeber and Roland B. Dixon, who, in a series
of papers beginning in 1913, proposed assigning most of the Powellian
language families of California to one or the other of two new "stocks,"
Penutian and Hokan (Dixon and Kroeber 1913, 1919). Sapir joined in this
work only after the groundwork had been laid, and at Kroeber's urging
(GoUa 1986: 178). Sapir brought to the task a thorough famiharity with the
methods and data of Indo-European comparative philology, and — after a
brief period of skepticism — he became convinced that a rigorous application
of philological principles to American languages would yield important new
insights. He moved from one bold synthesis to another, and his comprehen-
sive classification of 1920 must be regarded as little more than a report on
work in progress. It is noteworthy, however, that Sapir did relatively little
after 1920 either to support or to revise that classification. His 1925 paper
on Subtiaba, while introducing some new structural arguments for Hokan,
is based on essentially the same group of cognate sets as in his earlier work,
and it refers only briefly to the larger Hokan-Siouan grouping introduced in
his 1920 lecture.
In contrast with the wide-ranging comparative work that had absorbed him
during the preceding decade, Sapir s research during much of the 1920s focused
narrowly and intensively on one group of languages: the "Na-Dene" stock of his
1920 classification (comprising Tlingit, Haida, and the widespread Athabaskan
family). As early as 1906 he had worked briefly, during his Takelma field work,
with a speaker of Chasta Costa, an Oregon Athabaskan language; in preparing
this material for publication (1914c), he saw Athabaskan as a family having the
diversity and the relatively good documentation to make it a match for his skills
as a comparativist. He was soon embroiled in controversy with older
Athabaskan scholars (e.g. , Father Morice, 1915c, see Volume VI and Appendix
to Volume VI); this was exacerbated by his 1915 proposal (1915d) of a genetic
relationship among Athabaskan, Tlingit, and Haida. Sapir concluded that only
through extensive field work of his own could he hope to accumulate the evi-
Introduction 17
dence necessary to convince his critics. His feeling about the necessity of such
work became even stronger when, around 1920, he came to suspect that an
intercontinental genetic connection between Na-Dene and Sino-Tibetan was a
distinct possibility.
Sapir's plan for Na-Dene field research was extraordinarily ambitious, and
it was never completed. Except for a foray into Haida phonetics (1923d), his
field work was entirely devoted to Athabaskan, involving four major inves-
tigafions: Sarcee, in 1922; Kutchin and Ingalik, in 1923; Hupa, in 1927; and
Navajo, principally in 1929. Only the Sarcee work is significantly represented
in Sapir's bibliography; even here the major published study was prepared
in collaboration with his student Li Fang-Kuei (Li 1930, see Volume XIII).
A good deal of the material collected by Sapir has been published post-
humously, but the definitive grammar of Navajo which Sapir planned (and
was working on even during his last illness) will never be written. Of his
comparative insights into Athabaskan, Na-Dene, and Sino-Dene, we have
only fragmentary notes.
Sapir's active research career extended from 1905 to 1938, or 33 years. During
the first two decades of this period — until his move from Ottawa to a teaching
post at the University of Chicago — he was engaged almost exclusively in Amer-
ican Indian research, the bulk of it descriptive linguistics. After 1925 his inter-
ests began to turn toward other types of study, particularly the psychology of
culture; and his linguistic field research virtually came to an end when he
moved from Chicago to Yale in 1931. He remained, nonetheless, a central figure
in American Indian linguistics, second only to Boas in status and pre-eminent
in intellectual influence. Nearly all his important students took up the study of
American Indian languages. It was left to them, and to their scholarly progeny
in turn, to continue the many facets of his research. We will do no more here
than mention the names of Harry Hoijer, Morris Swadesh, George Trager,
Stanley Newman, Li Fang-Kuei, Benjamin L. Whorf, Charles F. Voegelin, and
our own teacher, Mary Haas. All these scholars have transmitted to their own
students not only an enthusiasm for American Indian linguistics, but, even
more important, Sapir's commitment to the study of language within the broad-
est context of human understanding.
William Bright
Victor Golla
Section Six:
Athabaskan and Na-Dene Languages
Introduction
While Sapir began his involvement with Hokan and Penutian languages as a
field linguist, turning to comparative and classificatory studies only after com-
pleting several major descriptive works, with Athabaskan the reverse was the
case. Except for some unsystematic notes on Chasta Costa collected during his
Takelma field work in 1906 (published as 1914c; see below), Sapir had had no
first-hand experience with Athabaskan languages when he launched what came
to be the most absorbing of his historical hypotheses: the genetic relationship of
Athabaskan, Tlingit, and Haida, and the deep connection of this "Na-Dene"
family to the Sino-Tibetan stock.
Neither the Na-Dene hypothesis nor the possibility of connections with Asi-
atic languages were, strictly speaking, original to Sapir (see the discussion of
earlier speculation in Krauss 1973), but there is little doubt that Sapir was the
first to explore these questions in the light of modern comparative linguistics.
The evidence from Sapir's correspondence and manuscripts is that he probably
took up this work late in 1912 or early in 1913, shortly after completing a
detailed study of Uto-Aztecan (Sapir to Kroeber, 23 December 1912, in GoUa
1984: 71). In the ensuing months he apparently read through all the extant mate-
rial both on comparative Athabaskan and on Tlingit and Haida. In the spring of
1913 we find him complaining in a letter to Robert Lowie that Pliny Earle God-
dard, a leading student of comparative Athabaskan, was over-cautious:
He does not . . . seem to me to get very much beyond descriptive Athabascan sketches cast in
parallel grooves. The unifying reconstructive spirit, the elimination of secondary features and
the emphasis on essential ones, seem to be lacking, on the whole. It seems to me that with all
the experience that Goddard has had with Athabascan he would have felt irresistably drawn by
this time to a serious consideration of Haida and Tlingit as possibly genetically related, though
remotely, to Athabascan. This may not seem eventually as far-fetched as it does now. But I am
afraid that Goddard is rather timid in these matters. (Sapir to Lowie, March 1913.)
Three months later he wrote to A. L. Kroeber:
A propos of larger linguistic units, which seem to be somewhat in favor just now, I may say that
I have been occupying myself of late with Athabaskan, Tlingit, and Haida, and that I have
collected enough evidence to convince myself at least of the genetic relationship of these three .
(Sapir to Kroeber, 30 May 1913, in Golla 1984: 104)
It was during this period that he went back to his Chasta Costa notes and worked
them up into a short monograph with a distinctly comparative emphasis. Notes on
Chasta Costa Phonology and Morphology (1914c). The publication of this work
immediately brought him into the small circle of serious Athabaskanists, of
whom the most established and productive was the French-Canadian Oblate
priest, A. G. Morice. Father Morice reviewed Sapir s Chasta Costa monograph
in glowing terms (Morice 1915a), but followed his review with a more critical
22 V7 American Indian Languages 2
appraisal, "Chasta Costa and the Dene Languages of the North" (Morice
1915b). Sapir replied to Morice's strictures in "Corrigenda to Father Morice's
'Chasta Costa and the Dene Languagesof the North'" (1915c), to which Morice
gave rejoinder in "Misconceptions Concerning Dene Morphology: Remarks on
Dr. Sapir s Would-be Corrigenda" (Morice 1917). Morice s review and papers
are reprinted in the Appendix to this volume. In the assessment of at least one
scholar, "in the long run . . . the interaction between Morice and Sapir turned
out to be productive," with the two showing "grudging respect for each other"
(Krauss 1986: 153-154). The relationship, stormy or not, was essentially
between two generations of scholars, and between a man thoroughly familiar
with the concrete details of several Canadian Athabaskan dialects and a com-
parativist primarily interested in historical reconstruction. As the nature of his
Athabaskan work shifted in the 1920s and 1930s from comparative to descrip-
tive, Sapir s appreciation of Morice's extensive knowledge of Athabaskan struc-
ture grew, as is evidenced by his short but appreciative review (1935c) of Mor-
ice's massive grammar and dictionary of Carrier (1932).
Between 1913 and 1915 Sapir continued to devote much of his research time to
the Na-Dene project, combing the published documentation of Northern,
Pacific Coast, and Southwestern Athabaskan languages for material to be com-
pared with Swanton's descriptions of Haida and Tlingit. By 1915 he had
amassed about 300 lexical comparisons and had begun writing a "systematic
presentation" of the material (Sapir to Radin, 17 July 1918, quoted in Krauss
1986: 156; see Dallaire 1984: 169, letter no. 263). At this point, at Goddard's
request, he prepared the shorter paper reprinted here, "The Na-Dene Lan-
guages, a Preliminary Report" (1915d), iox pub\\cdi\\on'mi\\Q American Anthro-
pologist. The manuscript of the full study has unfortunately been lost, although
the ledgers in which Sapir entered his Na-Dene lexical comparisons have sur-
vived (manuscript 497.3 B63c Na20a.3, vols. 1, 3, and 4, American Philosoph-
ical Society Library).
Even after the publication of Sapir's evidence Boas and Goddard remained
skeptical about Na-Dene, and Boas challenged Sapir's methods in a heated
exchange at the annual meeting of the American Anthropological Association
the following December. The strength of Boas's opposition (which reached
print in 1920 in a scathing attack on the misuse of genetic classification),
together with the appearance of Boas's own descriptive study of Tlingit (Boas
1917), seems to have taken the wind out of Sapir's sails, at least temporarily. He
devoted hardly any time to Athabaskan or Na-Dene from 1916 through 1920,
which was in general a period during which he was more occupied with literary
and artistic matters than with linguistic research.
Late in 1920, as he was completing his general book Language (1921d), Sapir
experienced what he described to Kroeber as a "considerable recrudescence of
interest in linguistics" (Golla 1984: 347), particularly in classificatory work. On
the one hand, this led to Sapir's working out of a general classificatory scheme
for all North American languages, grouping most of them into six "great
Six: Athabaskan and Na-Dene Languages 23
Stocks" (1921a). But it also led him to reconsider Na-Dene. As he put it to
Kroeber:
I am just now interested in another big linguistic possibility. I tremble to speak of it, though I've
carried the germinal idea with me for years. I do not feel that Na-dene belongs to the other
American languages. I feel it as a great intrusive band that has perhaps ruptured an old
Eskimo-Wakashan-Algonkin continuity In short, do not think me an ass if I am seriously
entertaining the notion of an old Indo-Chinese offshoot into N.W. America. ... I have
already carefully gone over two Tibetan grammars (Jiischke and Foucaux) and find in
Tibetan pretty much the kind of base from which a generalized Na-dene could have devel-
oped, also some very tempting material points of resemblance. (Sapir to Kroeber, 4
October 1920, in GoUa 1984: 350; reprinted in volume V: 81-83).
In the ensuing months Sapir delved deeply into Chinese and Sino-Tibetan lin-
guistics, working to some extent under the guidance of the anthropologist and
orientalist Berthold Laufer. By the end of the summer of 1921 Sapir had developed
the outlines of his "Sino-Dene" hypothesis, and he discussed the matter at some
length in a letter to Kroeber (a copy of which was also sent to Laufer), dated
October 1, 1921, and printed here almost in its entirety as "The Sino-Dene Hypoth-
esis." The complete text of this and a short follow-up letter can be found in Golla
(1984: 374-384).
More pressing than the need to acquaint himself thoroughly with the Asiatic
side of the relationship, however, was the need Sapir felt for a more complete and
accurate documentation of Athabaskan, Tlingit, and Haida. Seeing them now
through a Sino-Dene lens, it was clear to Sapir that many important aspects of Na-
Dene phonology and grammar had been missed by previous investigators. Late in
1921, he drew together extensive comparative evidence on the morphophonology
of relativization in Athabaskan. This paper, "A Type of Athabaskan Relative"
(1923n), calls attention to the distinctive clause-like nature of Athabaskan poly-
synthetic structure and, as Sapir wrote to Kroeber, "insidiously prepares for far
bigger things than its ostensible theme" (Sapir to Kroeber, 24 November 1921, in
Golla 1984: 386).
Especially important in Sapir's eyes was accurate information on the presence of
tonal systems, which the Sino-Dene connection made him certain was fundamen-
tal throughout Na-Dene. A pitch accent had been reported for Tlingit by Boas
(1917), and Sapir himself, working briefly with a speaker of Haida in March 1920,
had noted tonal and other phenomena of potential historical importance in Haida,
later described in "The Phonetics of Haida" (1923d). Tone had not, however, been
definitely reported in Athabaskan, and Sapir was certain that this was an over-
sight. To his delight, the first Athabaskan language he had the opportunity to work
on, Sarcee, turned out to have a well developed system of pitch accent. In an
immediate announcement in the American Anthropologist, "Athabaskan Tone"
( 1922a), he flatly stated that in the light of his Sarcee observations "it is well nigh
inconceivable that [tone] should be absent in any other Athabaskan dialect."
This view was reiterated in Sapirs full analysis, "Pitch Accent in Sarcee, an
Athabaskan Language" (19251), which is less a descriptive study than a general
theory of Athabaskan tone illustrated with Sarcee data. The paper, moreover.
24 VI American Indian Languages 2
ends with a list of questions that Sapir felt could be answered satisfactorily only
from the standpoint of Na-Dene (1925f: 204-205).
His wife's declining health forced a postponement of Sapir s planned visit to
the Hupa in 1923, but he was able to spend much of the summer of that year
doing productive field work with two young Alaskan Athabaskans, speakers of
Ingalik (Anvik) and Kutchin, who were working at a camp in Pennsylvania. In
the turmoil following his wife's death early in 1924, and his subsequent move
from Ottawa to Chicago in 1925, Sapir carried out little Athabaskan work dur-
ing the next two years. That the Sino-Dene hypothesis still strongly attracted
him, however, is shown by an interview he gave to Science shortly after arriving
at the University of Chicago, printed under the title "The Similarity of Chinese
and Indian Languages" (1925o).
From the beginning of his teaching at Chicago until his death 14 years later,
Sapir's Sino-Dene research had to vie for time with his many other involve-
ments. The evidence of his manuscripts is that he did little further with the
larger historical questions, although he continued the serious study of Tibetan
and Chinese. Certainly his publications after 1925 show little direct concern
with the Sino-Dene relationship, or even with Na-Dene, except insofar as the
stock was represented in his general classification of Central and North Ameri-
can languages (1929a). The major area in which he continued his comparative
research was Athabaskan. "A Summary Report of Field Work among the
Hupa, Summer of 1927" (1928i) and the Athabaskan portions of "The Concept
of Phonetic Law as Tested in Primitive Languages by Leonard Bloomfield"
(1931b) give brief glimpses of Sapir's progress in working out the intricate
details of comparative Athabaskan phonology in the late 1920s. (In the latter
paper, written in 1928-29, Sapir equates his Athabaskan work with Bloomfield's
Algonquian in a methodological discussion.) Even here, however, his work
slowed considerably in the following decade. He offered a course in Com-
parative Athabaskan twice during his teaching at Yale (in 1931-32 and again in
1936), and from the students' notes that survive there is little evidence that
Sapir's views had evolved much after 1930. A short survey of "Problems in
Athapaskan Linguistics" found among Sapir's papers, apparently dating from
about 1932 and published here for the first time, contains little not found in his
earlier published work. The statement of "Linguistic Classification within the
Northern Athapaskan Area" (1936i), which Sapir provided Cornelius Osgood
for inclusion in his ethnographic survey of the Northern Athabaskan Indians, is
also unremarkable.
From a brief foray into Chasta Costa in 1906 through major field work on
Navajo beginning in 1929, Sapir collected a very large corpus of descriptive data
on several Athabaskan languages. It was the analysis of this material that came
more and more to occupy Sapir's attention after 1925, much of it done in collab-
oration with his students, Fang-Kuei Li and Harry Hoijer (also, more briefly,
Mary Haas), and with his Navajoist colleague. Father Berard Haile. Only frag-
ments of this work saw print during Sapir's lifetime, and some has remained
unpublished to the present day.
Six: Athabaskan and Na-Dene Languages 25
The materials from Sapir's Sarcee field work in the summer of 1922, his first
extensive synchronic study of an Athabaskan language, have been more fully
published than most later materials. Sapir himself, in addition to preparing a
largely comparative paper on the Sarcee tone system (1925f, see above), had
nearly completed a volume of Sarcee texts when he left Ottawa in 1925. This
manuscript is being published for the first time in Volume XIII of The Collected
Works. Also in that volume are reprinted two studies based on Sapir s mate-
rials: Fang-Kuei Li's "A Study of Sarcee Verb-Stems" (1930), originally written
as a masters thesis at the University of Chicago under Sapir's direction, and
Harry Hoijer and Janet Joel's "Sarsi Nouns" (1963). Sapir also wrote two short
papers on ethnographic aspects of his Sarcee work, "A Note on Sarcee Pottery"
(1923c) and "Personal Names among the Sarcee" (1924d); both are found in
Volume IV.
Sapir's Kutchin materials, collected from John Fredson at Camp Red Cloud,
Pennsylvania, during the summer of 1923, are much more poorly represented in
his published work. Sapir extracted the kinship terms for inclusion in Cornelius
Osgood's Ethnography of the Kutchin (Sapir 1936h, printed in Volume IV), but
otherwise published nothing of his Kutchin data. As with Sarcee, he had begun
preparing a volume of Kutchin texts while still at Ottawa, and during the 1930s
Mary Haas, as his research assistant, worked on a Kutchin stem list. In 1961-62
Victor Golla completed a preliminary stem list but did not publish it. The texts
and a stem list are being published in Volume XIII of The Collected Works.
The Anvik (Ingalik) notes that Sapir obtained from Thomas Reed, also at
Camp Red Cloud in 1923, are far less extensive than his Kutchin materials.
Essentially a wordlist, of no great descriptive or comparative interest, the
material was never utilized by Sapir and is not published in The Collected
Works.
Sapir collected extensive Hupa data during a northwestern California field
trip in the summer of 1927, during which he also worked more briefly on Yurok
and Chimariko. He was accompanied by his Chicago student, Fang-Kuei Li,
who carried out his own work on Mattole and Wailaki. (For Sapir's lively
description of this trip see Volume IV [1927b].) Other than a short "summary
report" of his linguistic findings (1928i) — containing, inter alia, the (to Sapir)
distressing information that Hupa lacks a tonal system — and the Hupa data
incorporated into papers on the comparative method in American Indian lin-
guistics (1931b) and on the northern origin of the Navajo (1936f, sec below),
Sapir published only one paper based on his Hupa work, a largely ethnographic
study of Hupa tattooing (1936e) written for A. L. Kroeber's Festschrift.
Sometime during the 1930s he began work on a volume of Hupa texts, with
extensive ethnographic notes, but it was far from complete at the time of his
death. The texts and notes, edited by Victor Golla, are being published in Vol-
ume XIV of The Collected Works, together with a lexical index to Sapir s data.
Sapir regarded his Navajo work, begun with a native speaker in Chicago in
1926 but largely carried out in the field in 1929 and in later collaboration with
Father Berard Haile, as "by far the most extensive and important linguistic
26 yf American Indian Languages 2
research" he ever accomplished (Sapir to Boas, 12 April 1938, quoted in Krauss
1986: 166). The data he collected, particularly lexical and paradigmatic mate-
rial, was extraordinarily rich, and his relationship with his principal consultant,
Albert G. (Chic) Sandoval, was especially close. Sapir prepared a large collec-
tion of texts for publication , and his correspondence indicates that he was plan-
ning to write a full Navajo grammar. In the years immediately preceding his
death, he was actively working with Father Haile, a Franciscan missionary and
scholar (see Sapir's review of Haile's earlier work [1926f]), in preparing literacy
and language teaching materials for Navajo (Krauss 1986: 164-166). Despite all
of this activity, at the time of his death in 1939 Sapir had in fact published hardly
anything based on his Navajo work. A short paper, "Two Navaho Puns" (1932d),
and a comparative-historical tour dc force, "Internal Linguistic Evidence Sug-
gestive of the Northern Origin of the Navaho" (1936f), nearly exhaust the list,
except for two brief ethnographic notes, "A Navaho Sand Painting Blanket"
(1935b) and "A Note on Navaho Pottery" (with A. G. Sandoval, 1930), and a
newspaper article describing the circumstances of the 1929 field trip (1929c). A
volume of Sapir s Navajo texts was seen through the press in 1942 by Harry
Hoijer, Sapir's principal Athabaskanist student, who much later published a
Navajo grammar based on Sapir's materials as The Phonology and Morphology
of the Navajo Language (Sapir and Hoijer 1967). These two publications are
reprinted in Volume XV of The Collected Works, together with extracts from
the voluminous correspondence on Navajo linguistics that Sapir and Father
Haile carried on between 1929 and 1938 (Berard Haile Collection, University of
Arizona Archives, Tucson). Besides work directly attributed to Sapir, the influ-
ence of Sapir's materials and interpretation is strong in at least two other pub-
lications: Father Haile's Learning Navaho (1941-48), the first volume of which
incorporates many of Sapir's insights into Navajo phonology; and Harry Hoi-
jer's A Navajo Lexicon (1974), which faithfully reproduces the organization of
Sapir's lexical files.
NOTES ON CHASTA COSTA PHONOLOGY
AND MORPHOLOGY.
INTRODUCTION.
In a large part of southwestern Oregon and contiguous
territory in northwestern CaUfornia were spoken a number
of apparently quite distinct Athabascan dialects. The terri-
tory covered by tribes or groups of villages speaking these
dialects embraced not only a considerable strip of Pacific coast^
but also much of the interior to the east (Upper Umpqua and
Upper Coquille rivers, lower Rogue river, Chetco creek and
Smith river) ; some of the tribes (such as Tolowa and Chetco)
were strictly coast people, others (such as Galice Creek and
Umpqua or J\kwa}) were confined to the interior. While
some of the Athabascan dialects spoken south of the Klamath
in California, particularly Hupa and Kato, have been made
well known to students of American linguistics, practically
nothing of linguistic interest has as yet been published on any
of the dialects of the Oregon- California branch of Pacific
Athabascan. It is hoped that the following imperfect and
fragmentar}'- notes on one of these dialects may prove of at
least some value in a preliminary way.^
' Outside of a few points in southern and southeastern Alaska (Cook Inlet, mouth
of Copper river, Portland Canal) this is the only region in which Athabascan tribes have
found their way to the Pacific.
* My ^ denotes nasalization.
* The material for these notes was secured in a very incidental manner. While the
writer was at work on Takelma in the latter part of the summer of 1906, he was living
with Mr. Wolverton Orton, a full-blood Chasta Costa Indian. At odd moments Mr.
Orton and the writer whiled away the time with Chasta Costa.
(273)
28 VI American Indian Languages 2
274
The Chasta Costa (or Cis/ta qlwAs/ta) Indians, now
gathered in Siletz Reservation in western Oregon, formerly
occupied part of lower Rogue river; between them and the
coast were other Athabascan tribes or villages of practically
identical speech, above them to the east were the unrelated
Takelma.'' Among these tribes of nearly or quite identical
speech were the YiV^/gwl or Euchre Creek people, the Tee' /me
dA/ne or "Joshuas" of the mouth of Rogue river, the Du/Vil
dA/nl, the Ml/klu/nu"* dA/nl, and the GwA/sd. All these formed
a linguistic unit as contrasted with the coast people {d/yds/ta
"lower tribes") or, as they are now commonly called by the
Indians of Siletz, "Sol Chuck" Indians, a Chinook Jargon
term meaning "salt water, coast" people; the dialect of these
coast tribes was probably identical to all intents and purposes
with Chetco. While Chasta Costa and Coast Athabascan
are thus more or less distinct, they seem to have been mutually
intelligible without very much difficulty, the coast dialect
sounding merely somewhat "strange" and "drawn out" to a
speaker of Chasta Costa. At least three other Athabascan
dialects of this region, however, seem to have differed so much
from Chasta Costa as to be but partly understood, if at all^
by speakers of the latter; these are Upper Umpqua, Upper
Coquille, and Galice Creek.
* It has already been pointed out (American Anthropologist, N. S., 9, p. 253, note 2)
that there is reason to believe that J. O. Dorsey was incorrect in assigning the Chasta
Costa villages above those of the Takelma (see his map in Journal of American Folk-Lore,
III, p. 228). On p. 234 Dorsey gives a list of Chasta Costa villages.
Six: Athabaskan and Na-Dene Languages 29
PHONOLOGY.
Vowels.
The vowels of Chasta Costa are a, a, e (open as in Eng-
lish met), e (long and open), o (close as in German Sohn),
0, u (apparently variant of o), u, i (generally open), I, and A
(like u of EngHsh hut)\ 6 (short and open as in German voll)
sometimes occurs after velars as variant of o {sxo/ld "five,"
cf. Hupa^ tcwo/la), a (as in English hat) occurs after velars
as variant of e {tsxci/xe "child," cf. Carrier^ cezkhehkhe "chil-
dren").
Vocalic quantity is of considerable importance in Chasta
Costa, not so much etymologically as phonetically. On the
whole, long and short vowels interchange on regular mechan-
ical principles; open syllables (that is, syllables ending in a
vowel) with long vowel regularly shorten this vowel when the
suffixing of one or more consonants to the vowel makes the
syllable closed. Examples of a thus varying with a are:
do/ydc/tla "I won't fly;" do/ydt/tla "we won't fly" (cf.
dd/yd/t!a "he won't fly")
dad /da "he is sitting down" (cf. da/ 6 Ad /da "I am sitting
down")
tcIdsL/se "he cries;" tcldch/se/Ve "I shall cry" (cf. tcla/Bil/se
"you cry")
tc!a/ydsL/se "they cry" (cf. tc!a/yd/6il/se "we cry")
nac/tlb "I swim" (cf. nd/tcll/tlo "you bathe")
' Hupa examples are taken from P. E. Goddar^l, "The Morphology of the Hupa
Language," Univ. of Cal. Publ. Amer. Arch, and Ethn., 3.
* Carrier examples are taken from Rev. A. G. Morice, "The D^n^ Languages," Trans-
actions of the Canadian Institute, I, pp. 170-212.
(275)
30 ^f American Indian Languages 2
276
An example of c shortened to e is:
7ies/ts!Al/l "I am seen" (cf. lie' /ts!Al/l "he is seen")
Original long vowels may lose their quantity even in an
open syllable, provided they are immediately followed or
preceded by a syllable with relatively strong accent. Such
are tela-, na-, and 7ie- in:
tc!a/ya/dil/se "we cry;" tc!a/ydsL/se "they cry"
ne/7id/ts!Al/l "we are seen"
Id na/dit/t!d "don't bathe;" {na/dit/t!d is phonetically
enclitic to strongly accented ld\ contrast 7id/dit/t!d/Ve
"you will bathe")
In general, however, stress accent cannot be said to be
particularly well marked in Chasta Costa." Each syllable
is a fairly well-defined phonetic unit tending to hold its own
against others, so that an approximately level accentual flow
with but few peaks results. Such writings as 7id/dit/t!d and
tc!d/dil/se, with apparent accent preceded by long vowels,
are doubtless but imperfect renderings of forms with level
stress on first and second syllables (they might perhaps better
be written nd/dU/t!d and tc!a/Bll/se with secondary accent
on second syllable). It does not seem that every vowel in an
open syllable is organically long; thus e in future -Ve and in
-de of Vwl/de "everything" is regularly short. Many such
cases are, however, probably only apparent, the short vowel
being followed by a glottal stop; thus plural ya- of ya/dAl/ni
"they make a sound" should doubtless be ya-.
Short a of closed syllables is regularly reduced from long
a; original short a becomes /i in a closed syllable. Examples
of A thus dulled from original a are:
VAc/yAc/Ve "I shall go" (cf. t't/dic/ya "I go;" -yAc =
Hupa -yauw)
' Weak stress accent seems characteristic of Athabascan generally. Father Moric6
goes so far as to say, "there is no accent in Den^" {op. cit., p. 173).
Six: Athabaskan and Na-Dene Languages 31
277
nd/xAn/do "eight, two less" {;nd/xA- = Kato'^ nqk/ka"
"two")
do/na/yAct/xwl "I do not vomit" (cf. na/yd/dAdt/xwf
"I vomit")
VAl/dAc "he runs" {-dAc= Hupa -dauw)
jAn/na/'Ac "he will bring" {-'ac = Hupa -aiiw)
Ve/A7i/yit/lAl "we are sinking" (cf. Ve/nit/lat' "we drown;"
Hupa -lat, -la "to float")
Not to be etymologically confused with this a is inorganic
A. Whenever a consonant is not followed by a definitely
determined vowel and yet, for some reason or other, is not
phonetically appended to the preceding syllable, it must begin
its own syllable and takes an inorganic, in other words ety-
mologically meaningless, ^ -vowel after it. This syllable may
either be .completed by a consonant of etymological value
(such as first person singular c, verb class signs I, t, I) never
followed by a definite vowel or, if it is immediately followed
by a syllable beginning with a consonant, this consonant is
borrowed to complete the inorganic syllable (-/ closes inorganic
syllables preceding d-, tl-, dj-, tc!-, ts!-, tdf-, tc'-, lI-), so that
a doubled consonant results of which the first half is of no ety-
mologic significance. In some cases, however, as before 7-.
and in rapid speech generally, this inorganic consonant is not
always distinctly heard; yet in syllabifying words Mr. Orton
completed such inorganic syllables with a consonant with
mechanical regularity. These syllables with inorganic vowel
and consonant are characteristic not only of Chasta Costa
but also of Hupa and Kato and doubtless other Athabascan
dialects as well. The general phonetic tendency to speak in
definite syllables and the further tendency to limit short vowels
to closed syllables explain these characteristic Athabascan
* Kato examples are taken from P. E. Goddard, "Kato Texts," Univ. Cal. Publ.
Amer. Arch, and Ethn., 5, 65-238; and "Elements of the Kato Language," ibid., II,
1-176.
' -a- may be secondarily lengthened from -a-.
32 VI American Indian Languages 2
278
developments. The quality of the inorganic vowel varies
for different Athabascan dialects; it is a{u) in Kato as well
as in Chasta Costa, i {u before voiced or voiceless w, il or e
before post-palatal ^-sounds) in Hupa, apparently e in Galice
Creek, (s (probably identical with our a) in Carrier. Chasta
Costa xAt/VAl/lal "they sleep" is etymologically equivalent
to x/V/lal; X-, third person plural prefix, cannot stand alone
and is therefore followed by a and / borrowed from -/'-, while
-/'- (verb prefix V- reduced from Ve-, therefore not capable of
combining with x- into xaV-) in turn needs a syllabifying a
followed by / borrowed from -lal. Other examples of inorganic
A, with and without following inorganic consonant, are:
VAc/yAc/Ve "I shall go" {t"A- = V- reduced from Ve-)
dd/ya/xAt/t!a "they won't fly" {xAt- = x-)
nd/xAt/dAl/nic "they work" {xAt/dA- = x/d-, d- reduced
from de-)
dd/xAn/uAt/VAc "they go to bed" {xAn/nAt- = x/n-)
t'e/An/yAl/Ul "he is sinking" (jaI- = y-)
Many syllables with final consonant and a- vowel must
be considered as radical or at least unanalyzable elements.
In not all such cases is ^ a reduced form of a; where a seems
a primary vowel, as shown by comparison with other Atha-
bascan dialects, it seems best to consider it an organic element
in the syllable, though it remains plausible that at last analysis
it is but a reduced form of some fuller vowel. Thus, while
-yAc has been shown to represent an original -yac (Hupa -yaiiw),
-Vac contains a primary a, as shown by comparison with Hupa
-tuw "to lie down" (ultimately -Vac is doubtless -V, reduced
from -Ve, and suffix -c).
Inorganic a sometimes becomes palatalized to i, though
there is not enough material available to make it certain just
when this change takes place. Examples of this secondary
i have been found before c (but not before its developments
s and 6) and 5 derived from tc (but not before original 5 or its
Six: Athabaskan and Na-Dene tankages 33
279
development) when itself preceded by m, n, or 6 (preceding
7, however, tends to preserve a). Examples are:
mis/ki'' "gull" (cf. Kato butc/k'ai')
nic/ya "I come" {nic- = cessative 7z- and first person
singular c)
nic/dac "I dance"
t'e/nic/lat "I drown"
t'e/dic/ya "I go" {6ic- = durative 6- and prone minal c;
cf. t'ed/ya "he goes" without vowel after 6)
tclAy/ye/6ic/ya "I eat"
ye/ die /I "I saw him" (cf. c/ycd/l "he saw me")
Bid/ si "I let him"
With -die- contrast -OaO- (both from original *-sac-) in da/ 6 Ad /da
"I am sitting;" with -0zc/- contrast -sasI- (from original -*sAd-
and -*sAd- respectively) in tcId/sAsi/se "I am crying." -yic-
was heard in yd/yic/tia "I fly," but as this is an isolated example
(contrast -yAc- in nd/da/yAct/tlo "I bathe" and -yAd- in yAd/Az
"I sneezed"), it seems possible that this form was misheard
for yd/yAc/tla. Besides -nic- also -nAc- is met with: dd/nAc/t'Ac
"I go to bed" and nd/nAc/ An "I stop him;" it is probable that
in these forms -nA- is a reduced form of ne- (cf. Hupa tcin/ne/tuw
"she goes to bed") and thus not directly comparable with -ni-
of -nic-. Unaccented A, itself reduced from a, has in one case
{-yAc "to go") been found further palatalized to i: do /Vac/
yic "I'll not go," Id/n/yic "don't go!" (cf. t'Ac/yAc/t'e "I
shall go"); this -yic contracts with directly preceding t'A-
into -Vac: do/ Vac "he won't go.^^
Original Athabascan ai has in Chasta Costa become
monophthongized to i. Examples are:
l/gl "white" (cf. Kato i/gai)
'" Should probably be miskH'.
''With ihis-t'Ac Kato ta/cac in dd/la/co* ta/cac "not anywhcrt' I went" (P. E. God-
dard, "Kato Texts," Univ. Cal. Puhl. Amer. Arch, and Ethn., 5, No. 3, p. 182, 1. 17) is in
striking agreement.
34 yf American Indian Languages 2
280
mis/kfi"- "gull" (cf. Kato biUc/k'ai')
/zp'-^ demonstrative "that" (cf. Hupa hai)
ail as organic diphthong seems to occur but rarely in Atha-
bascan. If do "no!" (cf. Hupa dau) may be regarded as distinct
from adverbial do "not" (cf. Hupa do), we would have an
example of the parallel development of au to o in Chasta Costa.
Certain contractions that take place between i of first person
plural -it- and second person plural -o- with preceding vowels
will be spoken of in discussing the pronominal prefixes.
One of the most striking phonological characteristics of
Chasta Costa is the disappearance of an original rj^^ or of its
representative, nasalization of preceding vowel. Its former
presence can always be proved by comparison with other
Athabascan dialects that, like Hupa, still preserve it. In the
case of all vowels but inorganic A nasalization has left no trace
whatever, original q, (from arj), ^ (from et]), and i (from Iry)
being reduced to a, e, and t; originally short vowels, on losing
their nasalization and thus coming to stand in an open syllable,
become lengthened, while originally long vowels in a closed
syllable not only lose their nasalization but are shortened.
Thus, a syllable si may represent an original sj (or sir]) or
si (or sir]), while sil may go back as well to sll as to sjl. Examples
of the absolute disappearance of an original 17 are:
nd/xe "you paddle" {nd- = *nq-, cf. Hupa nun/ya "you
are about")
do/yCl/tla "you won't fly" {yd- = ^yq,-, cf. Hupa yum/ mas
assimilated from *yun/mas "you are rolling over")
tc!dl/se/t'e "you will cry" {tcfal- = *tc!q-l-; cf. tc!dd/se/t'e
"I shall cry" with -c- "I" morphologically parallel
to -.- "you")
'* t is here shortened to i because of following glottal stop.
•' i* denotes long I with weakly rearticulated parasitic i. Such "pseudo-diphthongs"
sporadically occur in Chasta Costa in lieu of ordinary long vowels.
'* i. e., ng of English sing.
Six: Athabaskan and Na-Dene Languages 35
281
id/na/yat/xwl "don't vomit!" {yat- = *yqt-, cf. yd- from
*yq- in na/yd/dit/xwl "you are vomiting")
nel/l "you are looking at him" {nel- = *n^l- ; -I = -'?*,
cf. Kato -in' "to see")
It "dog" (original Athabascan */?, "^li-q; cf. Hupa Lzw,
Montagnais I'in, Hare //'/w, Loucheux /'m, Carrier
//, old form loe^'n}^)
Nasalized inorganic 4 seems to have acquired a palatal
coloring i\ this i then regularly developed to I in open, i in
closed syllables. It thus often seems as though Chasta Costa
I, i is the morphologic equivalent, for instance in second person
singular forms, of Athabascan 77, an equivalence, as has just
been shown, due to secondary phonetic developments. Examples
oil < i < 4 are :
Ve/Bi/ya "you go" {61- = *S4- ; cf. Hupa na/sin/ya "you
are going about")
ni/dac "you dance" {ni- = *W4-; cf. Hupa nin/yauw
"go!")
3'w/u'w f/x/wi "you whistle" {di- = *d4-\ ci. Hupa da /din/ La
"run!" assimilated from * da /din/ La
na/tcll/ilo "you swim" {tcli-^'^'kylA-'}^ cf. Hupa na/kin/-
yun "come eat!")
yd/yi/tla "you fly" (7t- = *74-; cf. Hupa ye/win/ya "you
are going in")
yd/yi/tla "it flies" (7^- = *74-; cf. Hupa na/win/tau "it
will settle down" assimilated from ^na/win/tau)
Vi/lal "you are sleeping" (/'I- = */'4-; cf. Hupa tin/xaiiw/ne
"you take along")
verb stem -si "to make" (cf. Hupa -tcwiii)
'^ Morice, op. cit., p. 210. Carrier has evidently undergone a development parallel
to that of Chasta Costa. All northern Athabascan forms except Carrier (and Chipewyan)
are taken from R. P. E. Petitot, " Dictionnaire de la langue Dcnc-Dindji^."
'* k^l is "fortis" palatal k, Hupa ^1, Morice's q.
36 VI American Indian Languages 2
282
Examples, in closed syllables, of i<i<4 are :
tc!d/Bil/se "you cry" {9il- = *s4-l-; cf. Hupa na/dil/we/-
sil/en/^''
na/dit/t!d/Ve "you will bathe" {dit- = *d4-t-', cf. third
person nd/dAt/t!d/t'e)
t'd/yit/nd "you drink" {yU = *y4-t-; cf. third person
t*d/yAt/nd)
yd/yil/gAd "you climb" {yil- = *y4-l-; cf. third person
yd/yAl/gAd)
t'il/xwAd "you cough" (tHl- = *t'4-l-; cf. third person
t'Al/xwAd)
ne/cli/l "look at me!" {cil- = *c4-l-)
Hupa -;7 (that is, our rj) seems at times to correspond to
Chasta Costa -?z, but comparison with northern Athabascan
dialects indicates that in such cases we are dealing with original
-n. Thus, ?iAn "you," despite Hupa nin, is shown to have
original -n by Montagnais nen and Loucheux 7ia7i; dAn/tdi
"four," Hupa dink {=dir)k^!), does not go back to original
*d4/k^!i but to *dAn/k^!i or *dA7]/k^!i (rj assimilated from n),
as evidenced by Loucheux tan; la/cAn "black" corresponds to
Loucheux del-zen; similarly, dAn "in, at" must have original -n
despite Hupa din and Kato dun (original *d4 would have given
Chasta Costa *dl).
Consonants.
The consonantal system of Chasta Costa, like that of most
Athabascan dialects, is characterized by a lack of labial stops,
though m is common; b has been found in bo/di "cat," a loan-
word from English pussy, but seems not. to occur in native
words (yet cf. tcA/pd/yu "flower"). The consonants of Chasta
Costa are: the labi'al nasal m; the dental stops /', d, t!, and
dental nasal 7t; the back stops g, q' (or qx), q!, voiceless spirant
" In Hupa n (or nasalization) disappears in closed syllables. In such forms Chasta
Costa is etymologically more transparent than Hupa insofar as -i- is a reflex of original
-A-, whereas Hupa -t- is the normal inorganic vowel.
Six: Athabaskan and Na-Dene Languages 37
283
X (as in German Bach) , and voiced spirant y (as in North German
Wagen) ; the labiahzed back stops k'w, gw, q!w, and spirant xw
(sometimes weakened to hw) ; the sibilants s, c (as in EngUsh
ship), 6 (as in Enghsh thin), and z (voiceless lenis, intermediate
between ^ and English z, heard in -az "to sneeze"); the affric-
ative palatal consonants tc\ dj, and tcl; the affricative alveolar
consonants ts, tsl, and affricative dental consonant tdl; the
laterals /, I (voiceless spirantal /, with l, dorsal t followed
by /, as variant), and l/; the glottal stop ('); the aspirate h
(' at the close of a syllable) ; and the semivowels y and w.
Of these /', q\ k'w, and /c' (English ch) are aspirated surds
(^' is not found, k'w has been found but once and may be con-
sidered of doubtful occurrence); (b), d, g, gw, and dj are voice-
less but lenis, intermediate acoustically between surds and
sonants^^ {dj is intermediate between English ch and j) ; //,
ql, td, ts!, td!, and l! are so-called "fortis" consonants, in other
words, they are pronounced with simultaneous closure of glottis
but are released before the release of the glottal chords. q\
q!, gw, and q!w {g has not been found, but very likely exists)
are velar consonants; k! has not been found, ^^ its place being
taken by g/.^° Of secondary origin are syllabically final / and
k, which may be considered as voiceless stops differing from
/' and k' in their lack of aspiration; they are etymologically
equivalent to d and g. It is highly probable that also w, which
does not frequently occur, is but a secondary development
or acoustic variant of y after o-vowels;-^ after o-vowels y be-
comes labialized to y'^, in which both y and w elements are so
weak that one is constantly in doubt as to whether he hears
'" It is possible that these "intermediate" stops are sonant at their moment of releass.
" Unless, as seems possible, k of mis/ki "gull" was misheard for k!.
^° ql corresponds to Hupa kz, g is Hupa k-z. q! is by no means as forcible a sound as
is, e. g., Chinookan q!. There is something decidedly illusive about it; the velar stop
element seems to be reduced to a minimum, the glottal catch element is very strongly
marked, and a weak x seems at times to precede the velar stop (e. g., '^qld/xAO "arrow").
Despite my familiarity with Chinookan q!, I did not often succeed in pronouncing Chasta
Costa ql so as to satisfy Mr. Orton's car. It may well be that ql is really "fortis" or glot-
talized x (x/); cf. Tlingit 5/.
"' In Hupa 7 has become w in every case.
38 VI Atnerican Indian Languages 2
284
7 or w (thus do/ ye- becomes do/y'^e-, dd/'^we-\ similarly,
what was heard as do/wa- may really be do/y'^a-). However,
w occurs also in sa' /wAs/ts!e "sandhill crane;" wAs/xe "good."
This consonant system is only in part a faithful repre-
sentative of the original Athabascan system. Some conso-
nants have become merged with others, while other consonants
have kept distinct but have been changed in regard to place
of articulation. Chasta Costa m, t\ d, tl, n, g, q! (kl), q!w,
y, I, i, lI, ', h, and y seem in practically every case to correspond
to these same Athabascan sounds.
Athabascan k\ as also in Hupa, has become x in Chasta
Costa :
xd'/tc'ii "goose"-^ (cf. Hupa xa-^; Applegate Creek k'q'/-
tc'u; Kato k'a')
nd/xi "two" (cf. Hupa nax; Montagnais nak'e^^)
ts!d/xe "woman" (cf. Carrier tsekhe-'")
tsxd/xe "child" (cf. Carrier cezkhehkhe)
Analogously to this change of k' to x, original Athabascan
k'w has become xw (sometimes heard as hw) in Chasta Costa.
This sound is preserved as such in Kato {k'w) and Chasta
Costa (yXw), but seems generally to have fallen together in other
dialects with original k\ Examples are:
hwd "foot" (cf. Kato kwe*"; Carrier ne-khe; Loucheux
cekpe)
na/yd/dAdt/xwi "I vomit" (cf. Carrier khu "vomiting")
It seems, however, to persist as k'w in:
k'wAs/Vd/7ie "six" (cf. Hupa xos/tan)
Etymologically but not phonetically distinct, both in
Hupa and Chasta Costa, from these secondary x and xw are
^ -Ic u is augmentative.
" See Goddard, "Kato Texts," note 32.
" Petitot's ' represents aspiration.
** Father Morice represents "fortis" stops by means of points belovv characters.
Six: Athabaskan and Na-Dene Languages 39
285
original Athabascan x and xw. A good example of the latter
is:
-xwAd "to cough" (cf. Carrier xwces "cough," as noun)
Athabascan sibilants and sibilant affricatives {ts and
tc sounds) have undergone various modifications in Chasta
Costa. Original s has regularly become 6:
Ba/^aI "grizzly bear" (cf. Carrier sces-e^oel "brown bear")
t'e/dic/ya "I go" (cf. Hupa te/se/ya/te "I am going away")
VeB/ya "he goes" (cf. Hupa tes/ya/te "it is about to come")
-gAB "to climb" (cf. Hupa -k2s)
-xwaB "to cough" (cf. Carrier xwces)
Before / (or its variant l), however, 5 is regularly retained:
ts!a/sASL/se^^ "I cry;" tddsL/'se "he cries;" tc!a/ydsL/se
"they cry" (with these forms contrast tc!d/Bil/se
"you cry")
na/yesL/sl "he tells" (contrast nd/BU/sl "you tell")
cAsl/sl "he lets me" (contrast Bid /si "I let him")
cAsl/Val "he kicks me" (contrast Bid /Veil "I kicked him")
qlwAi/dasL/na "it was lying on it"
Athabascan is would, by analogy, have been expected to
develop into IB (as in Chipewyan), but B seems to be regularly
found instead:
Bi "head" (cf. Carrier n-tsi "your head;" Montagnais
-ihif Hare -kjwi\ Loucheux -tchir^. Kato -sV "head"
seems to indicate that in Kato also, at least initially,
5 and ts fell together.
Ba/^^cl "hair of head" (cf. Montagnais ethi-pa^^)
^* -SASL- is assimilated from *-sacl-, -s- being here prevented from becoming -6- be-
cause of following -5- (before l) of same syllable.
" i. e., -tei. Petitot's th is td. In Hare /5 (or its reflex 16) developed into what Petitot
writes kfw, perhaps to be understood as k(t>, i. e., k plus bilabial/.
'' Petitot's tch is our Ic.
** Petitot's p is 7.
40 VI American Indian Languages 2
286
l/do "yellow, green" (of. Montagnais del-thop "yellow;"
Hare de-kfwoy "yellow," Hupa u't-tso "green;" Kato
L-tso "blue")'
In some cases ts seems to have become s:
se "stone" (cf. Kato se\ Hupa tse; Montagnais the; Hare
kfwe; Loucheux tchi; Carrier tse)
As might be expected, Athabascan ts! has regularly become
td! in Chasta Costa:
dd/de/dil/tB!i "we are sitting" (cf. Hupa na/ya/del/tse,
i. e., -tsle, "they lived as before")
teiAd/dd "story"
Athabascan c is normally preserved as such (e. g., ct "I").
However, it is assimilated to s before 5 and tsl:
s/tsli/de "my sickness" {c- "my")
nes/tsUi/l "I am seen {-c- "I")
As/se/t'e "I shall cry" (from *ac-)
s I tsl An I na I 'Ac "he will bring it to me" (c- "me")
Assimilation of *sac to sas has taken place in:
tc!a/sAsL/se "I cry" (cf. tddci/se/Ve "I shall cry")
Original *sAc>*sic, however, regularly developed to die:
tclAy/ye/eic/ya "I eat"
Original *sac, after being assimilated to *sas, regularly shifted
to OaB, unless, as we have seen, it was protected by immediately
following I:
da/ 6 AS /da "I am sitting" (from *dd/sAc/da)
Ve/BAB/lal "I have been sleeping" (from *re/sAc/lal)
tcfed/t.'d "I swim across" (probably misheard for tcIe/dAd/tfo)
Original s, when immediately following c, also causes it to
assimilate; ss, which thus results, is then regularly shifted to
66:
yd/yA6/6el "I threw" (from '^yd/yAc/sel)
Six: Athabaskan and Na-Dene Languages 41
287
Athabascan tc (sometimes /cw?) is not retained in Chasta
Costa, but appears regularly as s:
l/sAk "red" (of. Kato L/tcik\ Loucheux ditssigY^
mis/k{!)i(') "gull" (of. Kato butc/k'ai*)
sd'/wAs/tsIe "sandhill crane" (cf. Applegate Creek tcd'/-
wdc/tc{!)e)
-si "to make" (cf. Hupa -tcwin; Kato -tcl; Chipewyan
-tsl'')
-se "to cry" (cf. Chetco -swe; Hupa -tcwen; Kato -ke';
Carrier -ssd)
Chasta Costa sx is found in:
sxo/ld "five" (cf. Hupa tcwo/la; Chipewyan sa/so/la/yai^)
Athabascan tc! remains, tc! often being shifted, however,
to ts! (or 5'22) :
tele- verb prefix "across the water" (cf. Hupa tee-, i. e.,
td.e-, "down to the beach, out of the house;" Kato
tce-\ Chipewyan ts'e- "to a body of water")
-ts!An "toward, to" (cf. Hupa -tcin, i. e., -tdit); Kato
-tc'uiV; Chipewyan -ts'un)
ts!i/de "sickness" (cf. Loucheux tssik, i. e., tsUk)
-s'at' "to be hurt" (cf. Hupa -teat, i. e., -te!at, "to be sick,
to become ill")
There is still another set of sibilants in Chasta Costa,
which go back to original palatalized (anterior palatal) ^-sounds
(gy, k^, k^!). In Kato, Navaho, Apache, Chipewyan, and
other Athabascan dialects, as in Chasta Costa, these have
become affricative sibilants, without, however, falling together,
as a rule, with the original Athabascan te- consonants. In
Chasta Costa, k^ has become te\ k^! has become tc! (this te!
*> Petitot's tss is our tsl.
" Chipewyan forms are taken from P. E. Goddard, "Analysis of Cold Lake Dialect,
Chipewyan," Anthr. Papers Amer. Mus. Nat. Hist., vol. X, pt. II. Chipewyan forms
taken from Petitot are referred to as Montagnais.
^- It is quite likely that tcl and ts! are here merely auditory variants of ts! (i is mid-
way between 5 and c). In Kato tc' , ts' and 5' also interchange.
42 VI American Indian Languages 2
288
does not vary, apparently, with ts!) ; for g^ I have no examples.
Chasta Costa and Chipewyan are largely parallel in their
development of Athabascan ts, tc, and k^ sounds:
Athabascan
Hupa
Chasta Costa
Chipewyan
dz
d^, d
ts
ts
e
te, d
ts!
tsl
td!
te!, 6'
dj
dj
dz
tc
tc(w)
s
ts
tc!
tc!
ts!,
ts!
ts!
g'
g'
dj
ky
k^
tc
tc
kyf
ky!
tc!
tc!
There are thus three distinct series of sibilant affricatives (and
of sibilants) in Chasta Costa and Chipewyan, none of which
is in direct accord with the original Athabascan sounds; Hupa,
it is highly important to note, reflects the original sounds almost
exactly .^^ Carrier, it would seem, has also preserved the ky-
series.
Examples of Chasta Costa tc' from original ky are:
dL,tcd/yi "big thing" (cf. Hupa -kya/o "large;" Kato
-tcay, -tea' "to be large")
''In his "Analysis of Cold Lake Dialect, Chipewyan," Goddard treats Chipewyan
ts and tc as though they were one sound corresponding to Jicarilla and Navaho tc (p. 86).
Examination of the various illustrative forms scattered throughout the paper, however,
soon convinces one that Chipewyan ts, dz, and ts! correspond respectively to Hupa,
Jicarilla, and Navaho fc(w), dj, and tcl; whereas Chipewyan tc, dj, and tc! correspond
respectively to Southern Athabascan ts, dz, and ts! and to Hupa k^, g^, and k^!. Thus,
the Southern Athabascan ts- sounds represent both original ts- sounds and k^- sounds;
perhaps there is a phonetic difference that does not come out clearly in the orthography.
As for Kato, Goddard finds no difference between tc- sounds that go back to original
tc- sounds and those that correspond to Hupa k^- sounds (" Elements of the Kato Language,"
pp. 16, 51). However, deictic tc'-, corresponding to Hupa tc!-, varies with ts' and s', thus
suggesting ts! as the true sound ; on the other hand, tc'- (to indicate indefinite third personal
object) corresponding to Hupa k^!- occurs consistently as tc' (contrast examples of tc'-,
ts^-, s'- on p. 50 with those of tc'- on p. 51). It seems plausible, then, that in Chipewyan,
Chasta Costa, and Kato original k^- sounds became true tc- sounds, while original tc sounds
were shifted to ts- sounds (which are apt to be heard as either ts- or tc- sounds).
Six: Athabaskan and Na-Dene Languages 43
289
-tc'u augmentative suffix (e. g., W/tc'ii "horse," literally
"big dog") (cf. Hupa -kyo; Kato -ted)
Examples of tc! going back to Athabascan k^I are:
dAn/tdi "four" (cf. Hupa dink, i. e., di-qkyf)
sic! At I de "seven" (cf. Hupa xo/kit, i. e., -k^Ht)
tddsL/se "he cries" (cf. Hupa kya/teL/tcwii "it cried,
i. e., k^Ia-)
tc!- verb prefix indicating indefinite object (cf. Hupa
k-, ky-, i. e., k^!-; Kato tc'-)
Athabascan possessed sonant sibilants (2, j) and sibilant
affricatives {dz, dj). Of these sounds z has been found in Chasta
Costa -Az "to sneeze;" dj is illustrated in several forms, but,
as we shall see in a moment, does not in these go back to Atha-
bascan dj. dz has not been found, though it may exist. 7, as
in Kato and Hupa, has become c:
ia/cAn "black" (cf. Hupa Lu/hwin<*-cin; Kato i/cmi^;
Jicarilla Ll/zl; Nav. Ll/jin; Chipewyan del/zun;
Loucheux del-zen)
Chasta Costa dj results from / (unaspirated) plus y:
qlwAt/tc/At/dja "table" {<*qIwAt/tc!At/ya "whereon one
eats;" -ya "to eat")
ya/da/yit/dja "we are ashamed" {<*ya/da/yit/ya; cf.
yAc in ya/dAcl/yAc "I am ashamed")
Of the lateral consonants, only three (/, I, and lI) have
been found in Chasta Costa. Original dl may have been pre-
served also, but Athabascan did was heard rather as / (unas-
pirated) plus Id:
yAct/lo "I laugh" (cf. Chipewyan -did, -dldk' "to laugh")
-/- is very probably third modal -/- here; while -did really
appears as -Id. After c and s, I becomes /:
nd/dAcl/nic "I work" (cf. nd/dAl/nic "he works")
nd/xwAcl/ye "I play" (cf. nd/xwAl/ye "he plays")
qlwAt/dasi/nd "it was lying on it"
44 VI American Indian Languages 2
MORPHOLOGY.
Pronouns.
Independent personal pronouns:
ci "I" ne "we" (probably contracted
from *ne/he; cf. Hupa ne/he)
nAn "you" nd/ne "you" (plur.)
yu "he, that one" yu/ne, yun/ne "they, those"
(really demonstrative) (really demonstrative)
Examples of possessive pronouns are:
cic/la "my hand" {cic is independent cl combined with
possessive prefix c-\ literally, "I my-hand")
nAn /la "your hand" (that is, nAn n-, "you your-hand")
hi la "his hand" {hi is demonstrative)
c/na/yd "my eyes"
s/tsH/de "my sickness, I am sick"
n/ts!l/de "you are sick"
nd/ts!l/de "our sickness, we are sick"
no/tsll/de/ha "your (pi.) sickness? are you (pi.) sick?"
{-ha is interrogative)
xd/ts!l/de "their sickness, they are sick"
Many nouns, when limited by preceding possessive pronouns,
suffix -e, as regularly in Athabascan. Thus, from mAn "house:"
cic/mAne "my house"
nAn/mAne "your house"
A noun followed by another with suffixed -e is to be under-
stood as genitively related to it. Examples are:
dAne' III tele "person's dog" {ll/tde from ll "dog," with
(290)
Six: Athabaskan and Na-Dene Languages 45
291
voicing of /- to /-; cf. Hupa Lin "dog," xo/lin/ke,
i. e., xo/lirj/kyfe "his dog")
tAkAc^'^ hlci^/le "bowstring" (literally, "bow's string;" cf.
Chipewyan Vul "rope," possessed form Vu/le)
ga/yu ts!i/de "baby's sickness, baby is sick"
As reflexive possessive is used xd/dAt- (with -d/dAt- cf.
Hupa a/d- ; Carrier cedced-) :
xd/dAt/lt/tc!e "his own dog" (used reflexively)
Of demonstrative pronouns there have been found:
hi* "that, he" (cf. Hupa hai, indefinite demonstrative and
article); hi'/tli "that thing"
yii "that one" (cf. Hupa yd "that")
yu/ne, ytin/ne "those, they"
m- "it" (cf. Hupa m-; Kato b-): niAl "with it"
de seems to be used as relative in:
de ucL/Ve "what I want"
This element is perhaps demonstrative in force and related to
Hupa de in ded "this," hai/de "this."
Totality is expressed by Vwl "all, everything" (cf. Hupa
a /tin "all"). Compounded with this element are:
Vwl/de "everything" {-de is very likely related to Hupa
di- in dl/hwd "something," dl/hwe/e "nothing")
do I Vwl/de "not everything"
Vwi/dAn "everywhere" (literally, ^'all-at;" cf. Hupa a//iw/-
din "every place")
Nouns.
Primitive non-descriptive nouns, as in all Athabascan dia-
lects, are relatively frequent in Chasta Costa. Monosyllabic
nouns are:
Body Parts.
la "hand" (cf. Hupa -/a; Kato -/a*)
** Probably to be understood as lAk/gAc.
46 yi American Indian Languages 2
292
hwd "foot" (cf. Kato -kwe*; Chipewyan -ke)
61 "head" (cf. Kato -si'; Chipewyan -61, -t6i)
-ya "hair" (in 6 Ay a "head-hair;" cf. Kato -ga' "hair;"
Chipewyan -ca, i. e., -ya)
Animals.
tdac "bird"
ll "dog" (cf. Hupa Lin; Chipewyan Ll)
Natural Objects.
se "stone" (cf. Hupa tse; Kato se)
cd "sun" (cf. Hupa hwa; Kato ca)
lAt "smoke" (cf. Hupa hit; Kato uit)
Culture Objects.
mAn "house" (cf. Hupa diminutive min-tc "hut")
Llel "matches" (originally doubtless "fire-drill;" cf. Chipe-
wyan L'eL "fire-drill")
gd6 "camass" (cf. Hupa kos "bulbs")
LloH-e "(its) string" (cf. Hupa Lol "strap;" Chipewyan
L'ilL "rope")
Primitive, at any rate not easily analyzed, nouns of more
than one syllable are:
Persons.
dAtt/ne, dAne' "person, man" (cf. Chipewyan de/ne, dun/ne;
Carrier tcene)
ts!d/xe "woman" (cf. Carrier tsekhe; Kato tc'ek)
dis/ne' "male" (with -ne cf. probably -ne of dAne')
sd/sAs "white man"
tsxd/xe "child" (cf. Carrier oe'zkhehkhe; Kato skl-k "boys,
children")
kel/'e "boy" (perhaps misheard for k!el/'e; cf. Kato
k'il/lek "boy")
gd/yu "baby"
Body Part.
na/y& "eye" (cf. Hupa -na; Kato -7ia*\ Chipewyan -na/ca,
-na/ce)
Six: Athabaskan and Na-Dene Languages 47
293
Animals.
BAyAl "grizzly bear" (cf. Carrier sces-e^cel "brown or cross
bear")
mis/ki!)i{') "gull" (cf. Kato biitc/k'ai')
dis/dac "fawn"
dA/meV /ke "pelican"^^
nat/qH "duck" (cf. Kato nd'/qH'')
ml/tc'd/tsUl/fil "deer"
Ve/q!d/lec/re "mink"
sd'/wAs/ts!e "sandhill crane" (cf. Applegate Creek tcd'/-
wdc/tc{!)e)
tclal/tchis/dje "ruffled grouse, 'pheasant' "
6d/gi "kingfisher"
dAs/nAl "red-shafted flicker"
teiAB/nd/yal/tetde ' 'hummingbird' '
0c I id e "bluejay"
nd/ts!d/le "horned lark"
so's/ga/ga "robin"
ts!d/ts!uk "wren"
kAsis "barn swallow"
ga/lal/'e "crow"
Many of these animal names, as well as some of those that
follow, are probably descriptive verb forms that have become
stereotyped.
Plants.
tcA/pd/yii "flower"'"
mt/tlal/tdAd "arrow- wood"
do'/de "tar- weed" (probably compounded with Atha-
bascan do' "grass;" cf. Hupa Lo/da-itc "an herb")
tc!Al/yat/ts!s "sunflower (?)"
"This word is humorously used to refer to Democrats, Democrat and dA/mel' /ke
exhibiting some similarity in sound.
'* This form was obtained independently.
" This word is remarkable as containing p, a sound that is normally absent in Atha-
bascan.
48 ^f American Indian Languages 2
294
dAl/si "pine" (cf. Kato dul/tcik "yellow pine," from -tclk
"red")
nd/ife "pine-nut"
dA/nAc "manzanita" (cf. Hupa din/nuw; Kato tun/nuc
"manzanita berries;" Galice Creek de/rec)
mAt/tcfi "cat-tail"
cAc/dd' "oak"
Culture Objects.
xAnAd "canoe"
at/ tea "pipe"
tclA/BA/gAl "sandstone arrow-shaft scraper"
tA/kAc "bow" (probably VAk/gAc\ cf. Kato gqc "yew")
*q!d/xA6 "arrow"
det/t!e "arrow-point"
Abstract.
tdlAd/dd "story"
yA/wls "whistling" (cf. Carrier ytiyuz "whistling," as
noun)
tsli/de "sickness" (used with possessive pronouns to indi-
cate "to be sick")
Several animals are designated by words ending in -tc'u,
an augmentative suffix, "big" (cf. Hupa and Kato animal and
plant names in -kyo and -ted respectively). Such are:
ii*/te'u "horse" (literally, "big dog;" cf. Chipewyan
Lln/teo)
xd'/te'ii "goose" (cf. Applegate Creek k'q'/te'u. These
words are formed from Athabascan xa: Chipewyan
xa "goose;" Kato ka')
dAe/te'u, des/te'ti "grouse" (cf. Kato due /ted, dus/ted
"grouse")
Vet/mo/tc'u "pigeon"
eu/dc' /tc'u "bald eagle"
BAe/dA/li/tc'u "owl"
H/tcIe/tc'u "red-headed woodpecker"
ga/sd'/te'ii "raven"
Six: Athabaskan and Na-Dene Languages 49
295
Nouns ending in -til or -tllni denote "one who has so and
so." -ni is, Hkely enough, related to -ne of dAn/ne "person;"
-ne or -n is found in many Athabascan dialects as suffix denot-
ing "person." Examples of -t!i{ni) are:
U/t!l/ni "dog-owner"
do I ail till ni, do J at I ill "bachelor" (literally, "not-wife-hav-
ing-person." do- "not;" at-, i. e., ail "wife," cf.
Hupa ui "wife," Kato aV "sister")
Examples of noun compounds consisting of two noun
stems are:
6a /yd "head-hair" (shortened from 61 "head" and -ya
"hair." Cf. Chipewyan 6l/Ga)
ga/lal gwd/yu "red-winged blackbird" (literally, "crow('s)
brother-in-law." With this cf. Chipewyan da/tsa/-
tcel/le "a small crow," literally, "crow younger-
brother"3«).
An example of a compound noun consisting of verb and
noun is:
aI/Az dAn/ne "sneezer" (literally, "he-sneezes person")
An example of a compound noun consisting of noun and
adjective is:
iclac l/66/e "bluebird" (literally, "bird blue")
A characteristic type of noun in Athabascan is formed by
verbs which, while remaining strictly verbal in form, are used
to refer to objects, in other words, are logically nouns. As
has been already noted, several nouns of more than one syllable
listed above as unanalyzable are doubtless, strictly speaking,
verb forms. Quite clearly verbal in form are:
nd6/Ll6 "paper" (cf. nal/ilo "he writes")
ql'wAt/da6t/gAc "table-cloth" (literally, "it lies or is thrown
^ Goddard, op. cit., p. 1 10.
50 VI American Indian Languages 2
296
down on top;" cf. Hupa -k2as, i. e., -gas, "to throw,"
and wes/kas "it lay"^^)
qlwAt/tclAt/dja "table" (literally, "thereon it is eaten")
mAl/Ve/tc!At/ts!Al/lec "smoking materials" (hterally, "there-
with it is smoked")
Numerals.
1. la, Wjca (cf. Hupa La\ Kato La/ ha')
2. nd/xi (cf. Hupa nax] Kato nqk/ka'); nd/xi la "two
hands"
3. t'd/yi (cf. Hupa tak, i.e. Vak!; Kato tak'; Chipewyan
ta, ta/ce
4. dAn/tdi (cf. Hupa dink, i. e. dirik^!; Chipewyan
dl/Gl)
5. sxo/ld (cf. Hupa tcwo/la; Chipewyan sa/so/la/cai*)
6. k'wAs/Vd/ne (cf. Hupa xos/tan)
7. slclAt/de (cf. Hupa xo/kit, i. e. -K^z7)
8. nd/xAn/do ( = "it lacks two, two less")
9. Ian/ do ( = "it lacks one, one less")
10. hwe'/de
Of numeral adverbs there were recorded:
Idt/dAu "once" (cf. Hupa na/din "twice," min/ Lun/din
"ten times")
la/me/q!e/ca "in one time"
Adjectives.
Of adjectives, or verb stems with adjectival significance,
there have been found:
wAs/xe, wAs/xd "good;" wAs/xe li "dog is good"
txAs/xe/la "rich" {-la is verbal sufhx)
du/An/de "bad" (evidently verbal in form, du-, do- is
negative; -de probably misheard for -t!e "to be, exist;"
cf. Hupa un/te, i. e. An/t!e "there is")
"Goddard, op. ciL, p. 281.
Six: Athabaskan and Na-Dene Languages 5 1
297
AL/tcd/yt "big thing" (cf. Hupa -kya/o "large;" Kato
-tcac)
l/gi "white" (cf. Hupa -L/kai; Kato -h/gai)
la I c An "black" (cf. Kato -L/cun^\ Chipewyan del/zun)
l/sAk "red" (cf. Kato -h/tclk)
l/do "yellow, green" (cf. Hupa lit/tsd "green;" Kato
-L/tso "blue")
"White," "black," "red," and "yellow, green" are characterized
by prefixed I {a)-, which is common as adjectival prefix also in
other Athabascan dialects.
Adverbs.
Adverbs of place are:
xun "there" (cf. Hupa third personal pronoun xon?):
xiin t'e/Bi/ya "there you go"
hV xiin Ved/ya "there he goes"
txun/la "where?":
txun/la VejBi/ya "where are you going?"
do/dAt "nowhere" (cf. Hupa -dit- in hai/dai/dit/din
"where;" do- is negative)
(i^y^/ge "up" (cf. Hare /eg^):
dAk/ge Bicl/Vdl "I kicked him up"
md'^/dAn "on edge" (-dAn is postposition "at;" md^-K
*mq-<*marj-; cf. Hupa ntL/man "each side")
Adverbs of time are:
xat "then" (cf. Hupa xat "yet, right")
xd "quickly" (cf. Hupa xa "yet")
xun/de "tomorrow" (cf. Hupa yis/xiln/de "tomorrow"):
xun/de do/wa/yAc/l "I'll see him tomorrow"
xun/de td/Ad/dd nAl ndcl/si "tomorrow story to-
ycu I-shall-tell"
xun/de t'Ac/yAc "tomorrow I'll go")
52 VI American Indian Languages 2
298
t'wt/dAn "always" (literally, "all-at"):
Vwi/dAti t'Al/dAc "he always runs"
t'wt/dAn As/se "I always cry"
xAi/tsH/dAn "this evening" (doubtless misheard for xal!-',
-dAn is postposition "at." Cf. Hupa xu/Le "in the
night"):
xAL/tsH/dAn do/wan/yAc/l "I'll see you this
evening"
Modal adverbs are:
do negative (cf . Hupa do) :
do/t'Ac "he won't go"
do/rAc/yic "I'll not go"
do/yd/t/a "he won't fly"
dd/As/se "I'm not crying"
do/nd/dACL/nic "I'm not working"
dd/yAc/l "I didn't see him"
do/ned/l "I'm not looking at him"
dd/ucL/t'e "I do not want"
do/na/yAct/xwl "I do not vomit"
. la prohibitive:
Id "don't!"
Id/n/yic "don't go!"
la/yi/i "don't see him!"
Id/nd/xwil/ye "don't play!"
ld/na/dit/t!d "don't bathe!"
Id/na/yat/xwi "don't vomit!"
doldd/qle "unable"
dol LAn "not much" (cf. Hupa Lan "much," do/ian "little")
do I wi I la "of course" (cf. Hupa ddii "it is," he I don "at
least")
do lid emphatic negative (really verbal inform, "to cease;"
cf. Hupa -lan, -lun with negative prefix do- "to quit,
leave, desist") :
do /Id c/yl/i "you didn't see me"
Six: Athabaskan and Na-Dene Languages 53
299
cd'*/djl "all right" (cf. Hupa nil/hwdn/^x "properly"?)
cAl/q!we "to be accustomed to":
cAl/qlwe na/dAct/t!d "I'm used to bathing"
t!V/xun "to keep on:"
tlV/xun ne/cAl/l "he keeps looking at me"
BAk/gwe "in fragments"
ho future prefix (more properly intentive) :
hd/ya/yic/t!a "I'll fly"
hd/tc!AsL/se "he wants to cry"
ho I ill I yit/lo "stop laughing!"
do/wa future prefix (probably with dubitative coloring) :
do/wa/c/yl/i^/Ve "you'll see me"
s/ts!l/de do/wa/ aI/W "I'll get sick" (Hterally,
"my-sickness will-become")
do/wa/ncL/yan/nAl "he will upset them"
dd/wa/it'dt/nTii "they will go to pieces"
Postpositions.
Athabascan is characterized, among other features, by the
use of a considerable number of postpositional elements of
chiefly local force. They are appended to nouns or pronom-
inal, numeral, or adverbial stems; less often to verb forms, in
which case they have subordinating force. Chasta Costa
examples are:
-dAn "at" (cf. Hupa -din):
xAL{!)/tsH/dAn "this evening"
Vwi/dAn "everywhere" (hterally, "all-at")
Idt/dAn "once" (cf. la- "one")
md^/dAti "on edge"
al/dAc/ni/dAn "when I tell him" (hterally, "I-
tell-him at")*«
"Similarly in Hupa -miL "when," as verb suffix, is doubtless simply pronominal
■mi- plus postposition -/. "witli."
54 VI American Indian Languages 2
300
4 "with, to" (cf. Hupa -l; Kato -l):
xAUAd/l 7idc/xe "I paddle canoe" (literally, "canoe-
with I-paddle")
tdlAd/dd riAl ndcl/sl "I tell you story" (literally,
"story you-with I-make")
td/Ad/dd caI na/yesL/si "he tells me story"
(literally, "story me-with he-makes")
mAl/Ve/tc!At/ts!Al/lec "wherewith it-is-smoked, ma-
terials for smoking" {mA-l- "therewith;" cf.
Kato buL "with it;" Hupa mil "with, in")
This same -I is probably also found attached to verbal prefix
a- (used in verbs of saying) :
al/dAc/ni/dAfi "when I tell him" (cf. Hupa ah/-
tcit/den/ne "he talked to")
-tslAn "toward" (cf. Hupa -tcin "toward;" Kato -tc'un*
"to, toward"):
s/tslAn/na/'Ac "to-me he-will-bring-it"
-me "in" (cf. Hupa -me "in;" Kato -hV "in"):
mAn/me "in house"
-me/q!e "in, around in" (compounded of -me and -q!e\
cf. Chipewyan -k'e "on"):
mAn/me/q!e "around in house"
la/me/q!e/ca "all in one time" (cf. la, Wjca "one")
Verbs.
As in other Athabascan dialects, the typical Chasta
Costa verb consists of one or more adverbial prefixes, which
may be followed in order by a deictic or third personal ele-
ment, a first modal prefix, a second modal element, a first or
second person subjective element, and a third modal element or
"class" sign; these, not all of which need of course be present,
are then followed by the Terb stem itself. The stem often
ends the verb form, but may be followed by one or more enclitic
elements of modal or syntactic force. The verb form is fre-
Six: Athabaskan and Na-Dene Languages 55
301
quently preceded by an adverb or postposition which, while
best considered as a non-integral part of the verb, forms a
rather close syntactic unit with it. A pronominal object,
if present, comes after an adverbial prefix but before a first
modal element. Thus, the verb form Vd/yd/Bot/nd/hd "do
you (plur.) drink?" consists of seven elements: Vd-, an ad-
verbial prefix referring to water; yd-, a second adverbial ele-
ment; B-, a second modal element of durative significance;
-6-, second person plural subjective pronominal element; -/-,
a third modal element, probably intransitive in force; -nd,
verb stem "to drink;" and -hd, an enclitic interrogative element.
The various elements that go to make up verb forms will be
taken up in the order indicated.
Adverbial Prefixes, a-, a-, 'a- used with verbs of say-
ing, doing, and being (cf . Hupa and Kato a-) :
d/dJAn "he says"
al/dAc/ni/dAn "when I tell him" (for -1-, see
under Postpositions)
dd/dAt 'An/ tie "there is not anywhere"
This a- is probably equivalent to an indefinite object, "some-
thing," indicating what is said or uttered without definitely
referring to it. This comes out rather clearly on comparison
with a form like yu/wls dAcl/ni "I whistle" (literally, "whistling
I-utter"), where no indefinite object a- is required, what is
uttered being specifically referred to by yu/wls "whistling."
That a- is somewhat in a class by itself as compared with other
adverbial prefixes is indicated by its being followed in forms
with indirect object by postpositive -/-.
yd-, ya- "up (in the air)" (cf. Hupa ya-\ Kato ya^-):
yd/yAcl/gAd "I climb"
yd/ y Ad /del "I threw"
yd/yic/lla "I fly"
It is not clear what significance is to be attached to ya- in:
ya/dAcl/yAc "I am ashamed"
ya/da/yit/dja "we are ashamed"
56 VI American Indian Languages 2
302
ye- "into enclosed space (including mouth)" (cf. Hupa
ye-] Kato ye'-, yV-)
ye/ydt/ne/la "he bit it"
da-, da- "sitting or lying on something above ground"
(cf . Hupa and Kato da-) :
da/ 6 Ad /da "I am sitting down"
dd/de/dil/tB!i "we are sitting down"
dd/nAc/Vac "I go to bed"
q!wAt/dadt/gAc "it lies thrown down on top,"
i. e. "table-cloth")
qlwAt/dash/nd "it was lying on it"
Ve- "in the water" (cf. Hupa ie-; Kato /e'-):
Ve/An/yAc/lAl "I am sinking in the water"
Ve/nic/lat "I drown"
Vd- referring to water (cf . Hupa and Kato ta-) :
Vd/yAct/nd "I drink"
tde- "across a stream" (cf. Hupa tee- "out of;" Kato
tc'e- "out of;" Chipewyan ts'e- "used of approach
to a body of water") :
tcle/Bii/tlo "I swim across"
An- implies disappearance or undoing (cf. Chipewyan
*a-, an- "away," implies "desertion or abandonment"):
Ve/ An/yAc/lAl "I am sinking in the water"
do/wd/ An/nd/yan/nAl "he will upset them"
an- "back, hither" (cf. Chipewyan '5- \in-, ai- "back,
toward home"):
an/yi/al "come here!"
tdd-, tela- of unknown significance (cf . Hupa kya-"^^) :
tcld/ sAsh/se "I cry" (cf. Hupa kya/teh/tcwe "she
heard it cry")
tddsi/se "he cries"
se'- used with verb of smiling:
se' /yAt/lo "he smiles" (cf. yAt/ld "he laughs")
^' Goddard lists forms in kya-, i. e., k^ta-, under ky-; see op. cit., p. 90. It seems better
however, to keep them apart.
Six: Athabaskan and Na-Dene Languages 57
303
tclo- of unknown significance (cf . Hupa kyo-^^) :
tcfo/yit/siL/la "he pointed with his finger"
ne/tc!uc/lec "I'll bet you"«
u- of unknown significance (cf. Hupa verbs in o-**)
dd/ucL/t'e "I do not want"
de/ucL/t'e "what I want"
nd-, na- indefinite movement on surface of ground or
water; horizontality (cf. Hupa and Kato na-):
fiAn/ndd/yd/la "he went around it"
s/tslAn/na/'Ac "he'll bring (it) to me"
yAn/na/Ac/t'e "he will bring here"
nd/ni/An "stop him!"
nd/ya "is going about, living"
i/t'i yAn/na/'d "he brags" (literally, "high, im-
portant he-has")
nd/xwAcL/ye "I play"
nd/dAch/nic "I work"
do/wa/nd/yan/nAl "he will upset them"
ndcl Lib "I write"
xatiaBU ndc/xe "I paddle canoe"
nac/t!d "I swim, bathe"
na/tc!il/de "you wash"
nd/dAcl/de "I washed myself"
nd/dAt/t!d/Ve "he'll bathe"
nd/xAt/dAl/el "they'll bathe"
tdlAd/dd caI nd/Bil/sl "story to-me "you-told,
caused"
na- "back again" (cf. Hupa and Kato na-), followed by
third modal -/-:
na/yd/dAdt/xwi "I vomit"
yAn- of uncertain significance (cf. Hupa wun- "to pursue
*^ Goddard lists forms in kyo, i. e., kyjo-, under ky-; sec op. cit., p. 90. Perhaps k^lo
is compounded of k^l- and o-.
♦' This tclii- is probably better explained as deictic tcl- followed by future imperative
M-; see note 86.
"Goddard, op. cit., p. 115.
58 VI American Indian Languages 2
304
or seek something; to attempt something by per-
sistent effort"):
yAn/na/Ac/t'e "he will bring here"
i/tH yAn/na/'d "he brags" (perhaps "important
he-seeks-to-have")
ne- of unknown significance:
necl/l "I'm looking at him" (cf. ye/ die /I "I saw
him")
nes/tslAl/l "I am seen"
xw{a)- of unknown significance:
nd/xwAci/ye "I play"
xwAcl/I "I believe (it)"
xwAn/ne/diL/ya "you win" (see under first modal
7ie-)
yd- ya-, (y) refers apparently to "mouth":
ye/ydt/ne/la "he bit it" {yat- may, however,
have been misheard for yAt-, with second
modal 7-; see note 92)
na/yd/6Adt/xwt "I vomit"
id/na/yat/xwl "don't vomit!"
t'd/yd/6it/nd "we drink"
Vdlyit/nd "you drink" (or is y- here second
modal prefix?
Verbal prefixes of local force which are doubtless primarily
postpositions and which are prefixed to adverbial prefixes proper
are:
nAn- "around" (cf. Hupa -nat\ Kato -nd)\
nAn/ndS/yd/la "he went around it"
qlwAt- "on, on top" (cf. Hupa -kut "on;" Kato -k'wut'
"on"):
qlwAt/tclAt/dja "whereon one eats, table"
q!wAt/daBt/gAc "it lies thrown down on top,
table-cloth"
qlwAt/dasi/nd "it was lying on it"
Six: Athabaskan and Na-Dene Languages 59
305
Deictic Prefixes. Under this head are grouped a small
number of quasi-pronominal elements of third personal refer-
ence which regularly come after adverbial prefixes, if any of
these are present. They cannot be grouped with first or second
personal subjective elements, as their position is quite distinct
from these; first and second modal prefixes may come between.
Of deictic elements there have been found:
td- denotes lack or indefiniteness of object of transitive verb
(cf. Hupa ^-, ky-, i. e. k^!-\ Kato tc'- ):
tdAy/ye/dic/ya "I eat" (i. e. without specific
object being designated; cf. Hupa yik/kyu/-
win/yan "it ate")
q!wAt/tc!At/dja "whereon one eats, table"
na/tc!il/L!d "you write" (cf. Hupa na/kis/Lon,
i. e. na I k^ lis I hi on "she made baskets")
nd/tc!l/t!d "you swim, bathe"
na/tcHl/de "you wash" (cf. Kato te'/na/tc'ih/deG
"he washed it")
tdAt/Vit/dAl "we wash ourselves" (cf. Hupa wa/-
kin/nin/seL "it was heated through")
mAl I V e I tcl At I tsl aI I lee "wherewith it is smoked"
(somewhat doubtful, as id- here follows first
modal prefix Ve-\ but see note 77)
td All ltd "he sucks" (cf. Kato tc'lh/Vot "[make] it
suck")
It is possible that in this last example td- is third personal
subjective (cf. Hupa tc-, i. e. td-; Kato tc'-, ts'-, 5'-), as sug-
gested by Ad/tlo "I suck" with its lack of td- prefix. No other
plausible case, however, of third personal subjective td- is
available, so that its existence in Chasta Costa must be con-
sidered doubtful as yet.
Generally third person singular subjective forms are dis-
tinguished by the lack of any pronominal prefix, but in certain
60 ^^ American Indian Languages 2
306
cases deictic elements are found which are clearly third personal
(subjective) in value. These are:
dj- (cf. Hupa tc-, i. e. tcf-; Kato tc'-, ts'-, s'-):
d/dJAn "he says" (verb-stem -n; cf. Hupa ai/-
tcit/den/ne "he talked to them")
djAn/la "he says" (cf. Hupa tcin "they say;"
Kato tc'in)
It is quite likely, however, that dJAn is to be explained as from
*dyAii {dy, as we have seen, becomes dj), in which d- is first
modal prefix (cf. dt/nl "you make a sound") and *yAn is reduced
from *yen {ye- as below; -n to say).
ye-, ya- (cf. Hupa y-, yl- referring to other than adult Hupa;
Kato yl-) :
uaI na/yei/si "to-you he-tells" (contrast ndcl/sl
"I tell")
caI na/yesi/sl "to-me he-tells" (with second
modal prefix s-, 6-; contrast nd/dU/sl "you
tell")
ya/qled/ya'^^ "he eats" (contrast tclAy/ye/6ic/ya
"I eat")
tcIe/xAd/tfo "he swims across" (contrast tcfe/-
Bit/t!d "you swim across")
This X- seems to have no parallel in Hupa or Kato (is it con-
nected with third person objective xo- of Hupa, kw- of Kato?).
Were it not that -t!d "to swim, bathe" is Used only of singular
subjects, one might surmise that nc-'is really plural xA- (see
below) .
Among deictic elements are further to be reckored certain
prefixes that serve to indicate either plurality as such or more
specifically third personal plurality. These are:
ya- (cf . Hupa ya- ; Kato ya^-) :
yu/wis ya/dil/ni "we whistle" (contrast yil/-
ims dAcl/ni "I whistle")
" qlc- was very likely misheard for te-.
Six: Athabaskan and Na-Dene Languages 61
307
yu/wls ya/dAl/ni "they whistle" (contrast yn/wls
dAl/nl "he whistles")
ya-, ya-
tc!a/yd/dil/se "we cry" (contrast tcfd/sAsL/se
"I cry")
tcIa/ydsL/se "they are crying" (contrast tcfdsL/se
"he cries")
tc!d/ydl/se/t'e "you (pi.) will cry" (contrast Icldl/-
se/t'e "you (sing.) will cry")
XA- third person plural (apparently not found in either
Hupa or Kato; but cf., without doubt, Chipewyan
he- "used for dual or plural of verbs in third person"):
yd/xAy/yi/tIa "they fly" {yd/yi/tia "it flies")
dd/yd/xAt/t!a "they won't fly" {do/yd/tfa "he
won't fly")
XAs/se/re/ha "will they cry?" {As/se/t'e/ha "will
he cry?")
do/xAs/se "they're not crying" {do/As/se "he's
not crying")
nd/xAt/dAl/nic "they work" {nd/dAl/nic "he
works")
c/xA/yee/l "they saw me" {c/yeS/l "he saw me")
na/xAt/da/yAl/el "they are bathing"
t"e/An/xAy/yAl/lAl "they sink in the water"
{Ve/An/yAl/lAl "he sinks")
xAt/t'Al/lal "they are sleeping" {VAl/lal "he is
sleeping")
dd/xAn/nAt/VAc "they went to bed" {dd/nAt/t'Ac
"he went to bed")
x At /VAI/xwaB'' they cough.'' {VAI/xwaB'' he coughs')
xa/AI/az "they sneeze" (aI/Az ''he sneezes")
First Modal Prefixes. Under this term are comprised
a small number of rather frequently occurring elements which
regularly come after both adverbial prefixes and deictic ele-
ments, but precede another set of modal elements (second
62 VI American Indian Languages 2
308
modal prefixes) which are to be taken up shortly. Their mean-
ing is rather colorless. Besides their position they have this
peculiarity in common, that they lose their vowel in indefinite
tense forms (such as have no second modal prefixes: d-, y-,
or W-) and are thus reduced to single consonants. They are:
t'e- (definite tenses), t'- (indefinite tenses) seems to indicate
durative activity (cf. Hupa te-; Kato te-, t-):
re/dic/ya "I go;" indefinite: do/rAc/yic "I'll
not go;" t'l/yAc/t'e "you must go"
VAci/dAc "I run" (indefinite)
tclAt/Vo/dAl "you (pi.) wash yourselves" (indefi-
nite)
t'/yi/l "he looks around" (indefinite; but see
note 69)
Ve/BAd/lal/la "I've been sleeping;" indefinite:
VAc/lal "I'm sleeping"
VAcl/xwAd "I cough" (indefinite)
mAl/t'e/tc!At/ts!Al/lec "wherewith it is smoked"
(as following td- is deictic, it is more likely
that Ve- here is adverbial prefix, not first
modal; see note 77).
de- (definite tenses; da- before 7-), d- (indefinite tenses)
meaning unknown (cf. Hupa d-, du-\ Kato de-, d-)\
al/dAc/ni/dAn "when I tell him" (indefinite)
yu/wis dAcl/ni "I whistle" (indefinite)
c/na/yd di'/s'aV "my-eyes hurt" (definite; cf.
Hupa du I win Ileal "it got sick")
nd/da/yAct/l!d "I bathe;" indefinite: nd/dAct/-
Ud'/Ve "I'll bathe"
na/da/yil/el "we are bathing;" indefinite: nd/-
dil/el "we'll bathe"
nd/dAci/nic "I work" (indefinite)
na/dAcl/de "I washed myself" (indefinite)
ya/dAcl/yAc "I am ashamed" (indefinite)
ya/da/yil/dja "we are ashamed" (definite)
dd/de/dil/lB!i "we are sitting down" (definite)
Six: Athabaskan and Na-Dene Languages 63
309
ye- (definite tenses) , 7- (indefinite tenses) meaning unknown :
tclAy/ye/dic/ya '1 eat"
ya/q!eB/ya "he eats" {q!e- is probably misheard
for ye-)
ye I Bill "you saw him;" indefinite: do/wa/yl/i/-
Ve "you'll see him"
yedt/lo "he breaks into laughter;" indefinite:
yAt/lo "he laughs"
an/yi/al "come on!"
This ye-, y- should not be confused with second modal 7-,
which will be taken up presently. Two first modal prefixes
(/'- and 7-) occur in V/yi/l "he looks around;" that 7- is not
second modal here is indicated by parallel definite forms with
ye- (see ye/di/l above), further by weak form V- of first prefix
(definite tenses require t'eY"^
ne- (definite tenses), n- (indefinite tenses) meaning unknown
(cf. Hupa ne-, n-; Kato ne-, n-; Chipewyan we-, nil-):
xwAn/ne/dii/ya "you win" (cf. Kato kun/ne/sli/-
yan "you win")
dd/nAc/VAc "I go to bed" (indefinite; cf. Hupa
definite: tcin/nes/ten ''hQlSiy')
n/dd" "it is, stays" (indefinite; cf. Kato definite:
tc'n/nes/dai "he sat down")
nd/nAc/An "I stop him;" nd/ni/An "stop him!"
(indefinite)
n/do "it is not" (indefinite; cf. Kato n/dd*/ye
"there is none")
This ne-, n- is not to be confused with second modal n-, which
occurs only in definite tenses.
Second Modal Prefixes. These comprise three conso-
nantal elements {B- or s-\ 7-; and 71-) which are used only in
definite tenses and which have reference, as far as any definite
*' Moreover, te- in definite tenses seems regularly followed by second modal 6-, not
7-. Yet -I- of yl- causes difficulty; see note 69.
64 yf American Indian Languages 2
310
significance is ascertainable at all, to what may be termed range
or span of activity, but not to tense as such. 6- {s- in certain
forms) is durative or continuative in force (cf. first modal
t'e-, which is regularly followed by 6-) ; w- is cessative, marking
the end of an activity or marking an activity which is con-
ceived as the end point of a previous activity (e. g., "to come"
as contrasted with durative "to go"); y- is the most uncertain,
being apparently inceptive or momentaneous in some cases,
but clearly not so in others/^ They are, it seems, mutually
exclusive elements. In practice their use seems largely deter-
mined by the prefixes that precede, n- and y- always begin
their syllable, being completed either by -i- ( < *-ir}-) or by
subjective pronominal or by third modal elements, which are
joined to them by means of -i- or inorganic -A-; 6- (s-) is similarly
joined to following subjective pronominal elements, if one is
present, otherwise it forms part of the preceding syllable.
Exrmples illustrating 6- (s- before I, l) are:
t'e/dic/ya "I go;" reO/ya "he goes" (cf. Hupa
te/se/yai "I went away")
nAfi/ndd/yd/la "he went around it"
tdAy/ye/dic/ya "I eat;" ya/q!ed/ya "he eats"
(contrast Hupa yik/kyil/win/yan "it ate"
with w-)
tcfe/dit/tfo "you swim across" (contrast na/da/-
yit/tfo "you bathe")
dad /da "he is sitting down" (cf. Hupa sit/dai
"he lived")
dd/de/6il/td!i "we are sitting down" (cf. Hupa
de/soh/tse/te "you will stay")
tc!d/6U/se "you (sing.) cry" (contrast Hupa winj-
tcwu "you have cried")
'^ Goddard somewhat doubtfully assigns inceptive force to its Hupa cognate w-; in
Kate its cognate g- seems clearly inceptive only in certain verbs; while in Chipcwyan
Goddard ascribes continuative value to g-. It would be worth while making a somewhat
extcnilcd comparative study of the second modal prefixes of Athabascan, which form one
of the most difficult but at the same time important chapters of its grammar.
Six: Athabaskan and Na-Dene Languages 65
311
ye /die /I "I saw him" (cf. Hupa te/suw/ifi "I am
going to look")
na I QUI SI "you told story" (cf. Hupa na/sei/tcwen
"I made")
Ve/Bi/lal/la "you've been sleeping" (cf. Hupa
nit/te/sil/lal/le "you would go to sleep")
Bid/ Veil "I kicked him" (contrast Hupa ye/tcu/-
wii/taL "they landed" with w-)
na/yd/6it/xwi "you vomit"
ye/dAdt/lo "I break into laughter"
Lad/d/la one was { = Ld "one" plus d/'d/la; cf.
Chipewyan ^e/'d "was there")
q!wAt/dadt/gAc "it lies thrown down on top,
table-cloth" (cf. Hupa wes/kas "it lay")
q!wAt/dasL/nd "it was lying on it"
xwAn/ne/diL/ya "you win" (cf. Kato kun/ne/-
slL/yan "I win")
Examples illustrating n- are :
nic/ya "I come" (cf. Hupa nei/yai "I came")
nic/dac "I dance" (cf. Kato nuc/dac "I will
dance")
Ve/ni/lat "you drown" (cf. Kato tc'n/ mil flat
"it floated there")
Examples illustrating y- are:
yt/dac "he dances" (cf. Kato tc'/gtm/dac/ kwaii
"he had danced")
yd/yAd/gAd "I climb" (cf. Hupa ya/wii/kas
"he threw up")
yd/yi/tfa "it flies" (cf. Hupa na/win/tau "it will
settle down")
nd/da/yAct/t!d "I bathe"
na/da/yil/el "we are bathing" (cf. Chipewyan
ni/i/oinL/ilL "take through the water")
66 ^/ American Indian Languages 2
312
fe/An/yAc/Ul "I'm sinking in the water" (of.
Hupa da/na/wil/laL "it was floating there")
yAci/Az "I've been sneezing"
tc!d/yit/siL/la "he pointed with his finger"
ya/da/yit/dja "we are ashamed"
Subjective Pronominal Prefixes. There are three per-
sons and two numbers (singular and plural), making six persons
in all. The third persons, as we have seen, are indicated either
by the absence of a pronominal element or by deictic prefixes
which come between the adverbial prefixes and the first modal
elements. There thus remain four persons (first person singular
and plural, second person singular and plural) for treatment
here. In the definite tenses the pronominal elements are
appended to the second modal elements, with which they form
a syllable, an inorganic a or i, if necessary, serving to connect
them. In the indefinite tenses the pronominal elements are
appended to whatever element (adverbial prefix, deictic ele-
ment, or first modal prefix in reduced form) happens to precede
them. They never begin their syllable except in the com-
paratively small number of cases in which the verb form, in-
definite in tense, has nothing preceding the pronominal element
or, in the case of the third person, nothing preceding the third
modal prefix or verb-stem. When this happens, the second
person singular and plural and the first person plural stand
at the very beginning of the verb; the first and third persons
singular, however, begin with an inorganic vowel A-.
First Person Singular -c- (cf. Hupa -w-; Kato -c-] Chipe-
wyan -s-) :
Ve/dic/ya "I go"
dicL/Val "I kicked him"
nic/ya "I come"
dd/nAc/VAc "I go to bed"
Ve/An/yAc/lAl "I am sinking in the water"
y Act /Id "I laugh"
Six: Athabaskan and Na-Dene Languages 67
313
ya/yAcl/gAd "I climb"
VAcL/dAc "I run"
nac/tlo "I swim, bathe"
Acl/t!6 "I suck"
acl/Az "I sneeze"
In definite tenses with 6- or n- as prefix the inorganic vowel
connecting these elements with -c- is regularly i\ this is evi-
dently due to the palatal quality of the -c-. In definite tenses
with 7- as prefix, however, the normal inorganic vowel, a,
is found, due, no doubt, to the velar position of the prefix.
In the indefinite tenses the connecting vowel, if required, is
always a. Where we have nAc- we are dealing with first modal
ne-, reduced to n-, plus -c-, not with second modal n- plus -c-\
contrast definite nic/ya with indefinite da/nAc/VAc.
Before s- sibilants -c- is assimilated to -s-:
do I AS I se "I'm not crying" {<'^Aclse)
Sic- goes back to original *sic- or *sac-. When -c- came to
stand before a dental consonant {d, t, /), it was assimilated to
-S-, and the inorganic vowel preceding it assumed the form a ;
this *sAs- then regularly became 6 Ad-:
da /e Ad /da "I am sitting"
ye/BAQt/lo "I break into laughter"
na/ya/BAdt/xwi "I vomit"
Ve/dAd/lal/la "I've been sleeping"
Before third modal -1-, die- seems to be regularly retained (cf.
Bicl/t'dl above; Sid/ si "I make"). Secondary sas-, not shifted
to OaO-, is found, however, before 1{l) when this element is
secondarily changed from third modal -/-:
tc!d/sAsL/se "I cry"
That sasl- here is equivalent to *sasI-<*sacI- is indicated by
tc!d/6il/se "you cry;" contrast Bil/t'cd "you kicked him,"
eicl/Val "I kicked him."
68 VI American Indian Languages 2
314
It is to be carefully noted that -c- (or its reflexes -s-, -$-)
is in Chasta Costa found in both definite and indefinite tenses.
There is no trace of an element corresponding to the Hupa
.^. (-e-), Kato -t-, Chipewyan -1-, which are found in forms of
definite tenses. It is quite probable that the -c- of the indefinite
forms was extended by analogy.
Second Person Singular i- (cf. Hupa -77-, i. e. -r)-\ Kato
-W-; Chipewyan n-, ne-, or nasalization of vowel:
Vil/dAc "you run"
Vd/yit/nd "you drink"
na/tcHl/Ud "you write"
na/da/yit/t!d "you bathe"
nd/dit/t!d/t'e "you'll bathe"
yd/yil/gAB "you climb"
ya/dil/yac "you are ashamed"
nd/dil/sl "you made, told"
nd/xwil/ye "you play"
xwil/l "you believe it"
xwAn/ne/BiL/ya "you win"
VU/xwaB "you cough"
In all these cases the -i- connects a following third modal ele-
ment {-1-, -/-, or -/-) with a preceding prefix. Examples of
-/- beginning its own syllable are:
tl/Az "you sneeze"
hd/il/l "stop!"
If there is no third modal element, the -i-, lengthened to close
-1-. closes its syllable:
dd/ni/t'Ac "go to bed!"
t'i/lal "you are sleeping"
dd/Bi/dd "you are sitting"
yd/yi/t!a "you fly"
do/l/se "you do not cry"
Six: Athabaskan and Na-Dene Languages 69
315
This i-, 1-, is only secondarily the second person singular
subjective element. The original element was doubtless -17-
(cf. Hupa). which was reduced to nasalization of preceding
vowels; the inorganic vowel, when nasalized, took on i- timbre
Finally, when nasalization disappeared, the /- timbre alone
remained as the reflex of original -rj-. Where, in many indefi-
nite tense forms, the nasalized vowel was other than an inorganic
one, there was nothing left of the -77-:
do/yd/tfa "you won't fly"
nd/xe "you paddle"
nei/l "look at him!" (cf. ned/i "I'm looking at
him")
In such cases the second person singular fell together with the
third, as in dd/yd/t!a "he won't fly."
First Person Plural (i)t-*^ (cf. Hupa it/d-, -d-\ Kato d-;
Chipewyan -/-, -d-^'^) :
Vit/lal "we are sleeping"
tclAt/Vit/dAl "we wash ourselves"
Ve/ nit flat "we drown"
da /nit /Vac "we went to bed"
ye/dit/l "we saw him"
ya/yit/t!a "we fly"
Ve/An/yit/lAl "we are sinking in the water"
dd/it/se "we are not crying"
In Hupa and Kato regularly, and in Chipewyan often, the
first person plural subjective pronominal prefix begins its
syllable; in Chasta Costa it regularly ends its syllable, unless
it has to stand at the beginning of the verb form, when it consti-
tutes a syllable by itself (cf. dd/it/se above; do "not" is inde-
pendent adverb rather than prefix).
" / is here unaspirated, and is thus etymologically identical with d.
'" In Father Legoff's Montagnais paradigms -id- or -//- often, in fact regularly, appeals;
-i- seems, as in Chasta Costa, to be organic.
70 ^f American Indian Languages 2
316
If the prefix preceding the pronominal element ends in
a vowel, the -i- disappears:
do/ydt/tla "we won't fly"
This does not mean, however, that this -/- is to be considered
an inorganic vowel, as is the case in Hupa it/d-. If -it- is followed
by third modal -/-, both -/- elements combine into a single -/-,
and all that is left of the pronominal prefix is the -i- :
t'd/yd/dit/nd "we drink" (contrast t'd/yddt/nd
"they drink")
If the third modal element is -I- or -/-, -/- disappears and -l-
is changed to -/-; thus the first person plural of I- verbs and
/- verbs is always formed alike. In Hupa and Kato third modal
-f- regularly becomes -/-, but d- is preserved; hence Hupa
dil-, Kato dul-. In Chipewyan, however, as in Chasta Costa,
-/- not only becomes -/-, but -/- disappears. For Chasta Costa
this means that the second person singular and first person
plural of /- verbs is identical, provided, of course, that there
is no deictic prefix of plurality in the latter and that the verb
stem does not change for the plural. Examples of /-verbs are:
dd/de/Bil/td!i "we are sitting" (cf. Chipewyan de/dll/d'l
"we are sitting")
nd/dil/nic "we work" (cf. nd/dil/nic "you work")
t'il/xwAd "we cough" (cf. VU/xwaB "you cough")
nd/xwil/ye "let us play" (cf. nd/xwil/ye "you play")
il/Az "we sneeze" (cf. tl/Az "you sneeze")
na/da/yil/el "we are bathing;" nd/drl/el "we'll bathe"
tc!a/ya/dil/se "we cry" (cf. tda/Bil/se "you cry")
Examples of /- verbs are:
yu/wls ya/dil/nl "we whistle" (cf. third person plural:
ya/dAl/nl)
na/tclil/Ud "we write" (cf. na/tdil/dd "you write")
Six: Athabaskan and Na-Dene Languages 71
317
If, in an indefinite tense form, the pronominal element is pre-
ceded by a prefix ending in a vowel and is, besides, followed by
third modal -I- or -/-, both -i- and -/- have to disappear and
there is nothing left of the pronominal element except, in the
case of /- verbs, the change of -/- to -/-;
nel/l "let us look at him!" (cf. nel/l "look at him!")
tcld/yel/se, very likely misheard for tc!d/ydl/se "we'll cry"
(cf. definite: tcfa/yd/dil/se "we cry")
Second Person Plural 0- {ci-Ylupa. o'-; Katoo'-; Chipewyan
rS/lal "ye sleep"
t'e/do/lai "ye have been sleeping"
t'e/nd/lat "ye drown"
yd/yd/ t!a "ye fly"
t'e/An/yd/Ul "ye sink in the water"
t'd/yd/dot/nd "ye drink"
dd/o/se "ye are not crying"
No aspiration was heard after o in Chasta Costa. This does
not seem due to faulty perception, as /- verbs keep their -/-
after o-, whereas, under similar circumstances, Hupa, Kato,
and Chipewyan change -/- to -I- {o'-l- becomes ol-). Indeed,
in Chasta Costa I- verbs change their -I- to -/- after second
person plural 6-. Examples of o- before /- verbs are:
nd/dol/nic "ye work"
na/da/yol/el "ye bathe;" nd/dol/el "ye will bathe"
t'dl/xwAd "ye cough"
nd/xol/ye/le "ye play" (for -xwol-)
ol/Az "ye sneeze"
Examples of -/- becoming -/- after 3- are:
ya/ddl/7il "ye utter, make a sound" (cf. third person plural
ya/dAl/nl)
ne/xd/ol/l "ye look at him" (cf. nci/i "you're looking at
him")
72 VI American Indian Languages 2
318
When, in an indefinite tense form, o- is preceded by a prefix
ending in a, a and o contract to long a (which, it would seem,
remains long even in closed syllables) :
do/yd/ t!a "ye won't fly" {<*yad'-; cf. third person singular
do/yd/tla with original yd- ; and second person singular
dd/yd/t!a < *yq- < *yarj)
tc!d/ydl/se/t'e "ye will cry" (cf. definite: tcIa/yd/Odl/se
"ye cry;" and contrast tcld/ydl/se "we'll cry" with
short -a-)
Third Person. As already noted, the third person, apart
from possible deictic prefixes, is marked by the absence of any
pronominal element. If the element preceding the third modal
prefix or the stem consists of a consonant which must begin
its syllable, an inorganic -a- is found between the two; if a
third modal prefix is absent, the syllable preceding the stem
is closed by a consonant borrowed from the first consonant of
the stem. Examples of third persons with -a- before a third
modal prefix are:
ya/dAl/yAc "he is ashamed"
VAl/dAc "he runs"
nd/xwAl/ye "he plays"
yd/yAl/gAd "he climbs"
yu/wls dAl/ni "he whistles"
Vd/yAt/nd "he drinks"
Examples of third persons with -A- followed by an inorganic
consonant are:
dd/nAt/VAc "he went to bed" (-/- is not third modal; cf.
second person singular dd/ni/VAc)
Ve/Ati/yAl/lAl "he is sinking in the water" (-/- is not third
modal; cf. second person singular Ve/An/yl/lAl)
First modal n-, reduced from ne-, however, has in several cases
been found without following inorganic vowel and consonant.
Six: Athabaskan and Na-Dene Languages 73
319
In such cases it closes the preceding syllable, which may even
belong to another word. Examples are:
dd/dAt 'An/tie "not-anywhere there-is" (cf. Kato qn/t'e
"it is;" Hupa un/te "there is")
Idn/do "nine" (really la n/do "one is-lacking") ; nd/xAn/do
"eight" (reduced from na/xi n/do "two are-lacking")
(cf. Kato 71 /do' /bun "it will not be," but also niit/do'
"all gone")
cic/mAnen/dd' { = ci c/mAne n/da') "I my-house is" (cf.
Chipewyan ne/da "she sat")
If the verb form consists, properly speaking, of the stem alone,
without prefix of any kind, an inorganic a- completed by a
consonant that depends for its form on the first consonant of
the stem is prefixed for the third person:
As/se/Ve "he must cry" (<*j^//V; cf. dd/As/se "I'm not
crying" <*Ac/se)
do / wa / aI / W "he will become" (<*/^; do/wa is adverb
not influencing form of verb proper. That -/- is here
no third modal element is shown by forms like Hupa
o/le "let him become")
This /I- at the beginning of a third personal form appears also
when the verb begins with a third modal element:
aI/Az "he sneezes"
In this respect Chasta Costa differs from Kato, which need
have nothing preceding the stem; with As/se compare Kato
tee' "he cried."
In the third person of definite tenses with second modal
7- or n- prefix this element is followed by -I, in case there is no
third modal prefix present. This goes back, without doubt,
to nasalized -i- or -A-, in turn reduced from original -it)- (or
-At]-). This nasal element, characteristic of definite third
personal forms (except such as have 6-, Athabascan s-, as second
74 VI American Indian Languages 2
320
modal prefix) is found also in Hupa (-in-), Kato {-un-), and
Chipewyan (-«-, -in-). Examples are:
Ve/ni/lat "he drowns" (cf. Kato tc'n/nul/lat "it floated
there," nul- assimilated from nun-; Ve/ni/lat also
"you drown")
yi/dac "he dances" (cf. Kato tc' /gun/dac/kwan "he had
danced")
yd/yi/t!a "it flies" (cf. Hupa na/win/tau "it will settle
down;" ya/yi/ila also "it flies")
di*/s'aV "it pains" {dl*-<*dir]-, contracted^^ from *de/-
yir}-; cf. Hupa du/win/tcat "it got sick")
Rather hard to understand is:
Ve/An/yAl/lAl "he is sinking in the water"
One would have expected -7!-, not -yAl- (as seen above, -/-
is not third modal, but inorganic). Is yA- reduced from first
modal ye-, this form being indefinite in tense?
In Hupa this -in- does not seem to be found before third
modal prefixes; in Chipewyan -n- {-in-) may, however, occur
before -I- and, as inferred from Father Legoff's Montagnais
paradigms, also -/-. As for Chasta Costa, what examples are
available on this point show that -i- does not occur before -/-
(e. g. yd/yAl/gAB "he climbs"). For I- verbs I have no safe
example. Before -/- it seems that -i- is present in some cases,
not in others:
tcld/yit/sih/la "he pointed with his finger"
but, without -i-'.
na/da/yAt/t!d "he is bathing"
On the whole, it seems possible that Athabascan -Ar]- (or
-An-) was originally a more freely movable element than it has
" Parallel in form to Hupa verbs belonging to Class I, Conjugation 1 D, in which
prefixed first modal d- or deictic k^l- contracts with -it}, w- (Athabascan 7-) being lost.
See Goddard, op. cit., p. 113.
Six: Athabaskan and Na-Dene Languages 75
321
become in e. g. Hupa, being required 'by certain verbs in their
definite tenses, but not by others. This is suggested also by
Father Legoff's Montagnais paradigms.
Third Modal Prefixes. There are three of these: -/-,
-/-, and -t-\ they always complete a syllable immediately before
the stem. -I- is characteristic of many verbs which are either
transitive or, at any rate, imply activity directed outward;
in some cases, however, this significance is not obvious. As
we have seen, this -I- becomes -/- in the first and second persons
plural. Examples of -1-{-l-) are:
Bid/Val "I kicked him"
tB!AB/da caI nd/6il/sl "story to-me you-told, made"
necl/l "I'm looking at him"
ndl/de "he washes (something)"
nail Lib "he writes"
tc!Al/t!d "he sucks"
yu/wis dAl/nl "he whistles;" yu/wls dAcl/nl "I whistle"'^
xwAn/ne/diL/ya "you win"
If nothing precedes this element, it seems (unlike -/-) to begin
its word without preceding inorganic A-:
dd/wa/l/t'dt/fiAl "they will be broken" {do/wa is merely
proclitic)
l/tH "he is important"
Verbs in -/- are regularly intransitive; they denote states
of mind or bodily activities that may be thought of as self-
contained, not directed outwards. A reflexive meaning is
sometimes apparent. After first person subjective -c- (s-)
it always appears as -/-. Hence the first person singular, the
first person plural, and the second person plural of -/- verbs
and /- verbs are always alike (but contrast did- < *sAd- with
5Asi<*sAd-). As -/-, when standing after s, becomes -I- also
in the third person, the second person singular alone remains
'- This verb is irregular, inasmuch as -/- does not occur in the second person singular:
yA/wis di/nt "you whistle."
76 VI American Indian Languages 2
322
as an infallible criterion of whether a verb belongs to the l-
class or /- class. Examples of -/- are:
ya/dAl/yAc "he is ashamed"
xwAcl/1 "1 believe;" xwil/i/ha "do you expect?"
VAl/dAc "he runs"
yd/yAl/gAd "he climbs"
nd/dAl/de "he washed himself"
aI/Az "he sneezes"
VAI/xwaB "he coughs"
na/xAt/da/yAl/el "they are bathing"
nd/xwAl/ye "he plays"
nd/dAl/nic "he works"
tddsh/se "he cries;" tela I HI I se "you cry"
tc!dcL/se/Ve "I'll cry;" tc!dl/se/Ve "you'll cry"
hd/yAci/i "I stop, cease;" hd/il/i "stop!"
Verbs in -/- are also intransitive. Examples are:
Vd/yit/nd "you drink"
ye hat I ne I la "he bit (it)"
tdd/yit/sii/la "he pointed with his finger"
qlwAt/dadt/gAc "it lies thrown down on top,
table-cloth"
yAct/ld "I laugh" {-t/lo may, however, represent
original Athabascan -did "to laugh")
ya/da/yit/dja "we are ashamed"
There may be a passive significance in :
q/wAt/tclAt/dja "whereon it is eaten, table"
With iterative na-:
na/yd/6Adt/xwl "I vomit"
Verb Stems. The stems that have been determined for
Chasta Costa are:
-'d,-'a "to have position, to be" (cf. Hupa -ai, -a; Kato
-*ai*, -'a'): d/'d/la "(one table) was"
Six: Athabaskan and Na-Dene Languages 77
323
-'AC "to bring" (cf. Hupa -an, -uii, -auw "to transport
round objects"; Kato -*qn, -^qc): yAn/na/'Ac "he
will bring it"
-'An "to bring to a halt, stop" (perhaps another form of
preceding stem): 7td/ni/An "stop him!"
-Az "to sneeze": acl/Az "I sneeze"
-ai "to come" (cf. Chipewyan -^as, -*az, -*ais "to travel,
used of two persons only"?) an/yi/'al "come on!"
-ci, -el "to bathe (plur. subject)" (cf. Chipewyan -el, -eL,
-uL "to move on the surface of water"): na/da/yil/el
"we are bathing"
-'I "to see, look at" (cf. Hupa -en, -in; Kato -*ln' "to
look"): ye/dic/l "I saw him"
-i "to stop, cease" (cf. Hupa -en, -in "to do, to act, to deport
one's self"?): hd/yAch/i "I stop (laughing)"
-I "to believe, expect": xwAcl/i "I believe"
-ya, -yAc, -yic "to go, come" (cf. Hupa -yai, -ya, -yauw;
Kato -yai, -ya, -yac): Ve/6ic/ya "I go;" t'Ac/yAc/t'e
"I must go"
-ya "to eat" (cf. Hupa -yan, -yun, -yauw; Kato -yan^ -ytl*):
tclAy/ye/Oic/ya "I eat;" qIwAt/tcfAt/dja(<-t/ya)
"whereon one eats, table"
-ya "to win" (cf. Kato -yan, "Kato Texts," p. 146, 1. 13;
not listed in "Elements of the Kato Language"):
xwAn/ne/dii/ya "you win"
-yan "to upset" (cf. Kato -yan "to clear off"?): do/wd/An/-
nd/yan/nAl "he will upset them"
-ya, -yAc "to be ashamed" (cf. Kato -yafi "to be ashamed"):
ya/dAl/yAc "he is ashamed;" ya/da/yit/dja {<-t/ya)
"we are ashamed"
-ye "to play" (cf. Hupa -ye "to dance"): nd/xwAl/ye "he
plays"
-lal "to sleep" (cf. Hupa -lal, -lai; Kato -lal, -Iqi): t'cd/-
lal/la "he's been sleeping"
-IaI "to sink in water;" -lat "to drown" (cf. Hupa -lat,
78 VI American Indian Languages 2
324
-la "to float"): re/Ari/yAc/Ul "I am sinking;" fe/-
nic/lat "I drown"
-le "to become" (of. Hupa -len, -lin, -hi, -le; Kato -lin%
-le): do j wa I aI / le' "it will become"
-lee "to wager, bet": ne/tclilc/lec "I'll bet you"
-lee "to smoke": mAl/Ve/tc!At/ts!Al/lee "wherewith it is
smoked"
-Id "to laugh, smile" (cf. Chipewyan -did, -dlok'): yit/lo
"laugh!"
-lIo "to write" (cf. Hupa -Lon, -Ld, -Low, -Loi "to make
baskets, to twine in basket-making;" Kato -Loi,
-Ld, -Lon): nail Lib "he writes"
-na "to drink" (cf. Hupa -nan, -nun-, Kato -7iqn): Vd/-
yAct/nd "I drink"
-na "to lie" (cf. Kato ndLjiin/na'' "were left"?): qlwAt/
dasL/nd "it was lying on it"
-ne "to bite, seize with one's teeth": ye/ydt/ne/la "he
bit it"
-nl, -71 "to make a sound, to say" (cf. Hupa -ne, -n "to
speak, to make a sound;" Kato -nl, -ne, -n, -nee, -nlL):
yu/wls dAl/ni "he whistles;" a/dJAn "he says"
-nic "to work": nd/dAl/nic "he works" (cf. Montagnais
-7ii "exprime Taction des mains"^^
-dAl "to wash oneself (plur. subject)" (cf. Hupa -sel, -scl
"to be or to become warm;" Kato -sll "to steam,"
-sUl, -suL "to be warm"): tc/At/t'it/dAl "we wash
ourselves"
-del "to throw": yd/yi/del "you threw"
-se "to cry" (cf. Hupa -tcwu, -tcwe "to cry, to weep;" Kato
-tcec, -tee'): tcIdsL/se "he cries"
-61 "to cause" (cf. Hupa -tcwen, -tcwin, -tewe "to make, to
arrange, to cause;" Kato -tcin, -let, -tclL): ndcl/sl
"I cause"
■sil "to point with one's finger": tc!d/yit/siL/la "he pointed
with his finger"
" Father L. Legoff, "Grammaire de la Langue Montagnaise," p. 139.
Six: Athabaskan and Na-Dene Languages 79
32.5
-da', -da "to sit, stay" (cf. Hupa -dai, -da; Kato -da, -dai):
da I Bi I da "you are sitting"
-dAc "to run" (cf. Hupa -dai, -daL, -dauw "to pass along,
to go, to come;" Kato -dac "to travel"): VAl/dAc
"he runs"
-dac "to dance" (cf. Kato -dac "to dance"): nt/dac "you
dance"
-de "to wash (sing, subject)" (cf. Kato -dec, -de'): nd/-
dAl/de "he washed himself"
-t'di "to kick" (cf. Hupa -taL, -tul, -tuL, -tal "to step, to
kick;" Kato -tal\ -tqL): dicl/t'dl "I kicked him"
-t'Ac "to lie down, go to bed" (cf. Hupa -ten, -tin, -tuw "to
lie down;" Kato -tin, -tuc): dd/nAc/VAc "I go to bed"
-Vat "to break, go to pieces" (cf. Chipewyan -taL, -till "to
break"): dd/wa/l/t'dt/uAl "they will be broken"
-Ve "to want" (cf. Hupa -te "to look for, to search after"?):
dd/ucL/t'e "I do not want;" de/uci/t'e "what I want"
-t'J "to be, make valuable" (cf. Carrier til/thi "thou makest
him valuable, treatest him as important"): l/Vi
yAn/na/'d "he brags"
-t.'a "to fly" (cf. Hupa -tau; Kato -t'ac, -t'a'): do/ydc/tla
"I won't fly"
-tie "to be of (that) sort" (cf. Hupa -te; Kato -t'e): do/dAt
'An /tie "there is not anywhere (one like him)"
-t!d "to swim, bathe (sing, subject)": nac/t!d "I swim,
bathe"
-t!d "to suck" (cf. Kato -t'ot): tclAi/t/d "he sucks"
-tdfi "to sit (plur. subject)" (cf. Hupa -tse; Chipewyan
-d'i): da/de/dil/tdii "we are sitting"
-tshit "to hurt, pain (intr.)" (cf. Hupa -teat, -tea "to be
sick, to become ill"): di*/s'at' "(my eyes) hurt"
-7« "to go about, live" (cf. Hupa -wai^ -wa "to go, to go
about;" Kato -ga, -gai): nd/ya "he goes about, lives"
-xe "to paddle" (cf. Hupa -xen, -xuw "to float, used only
of plural objects;" Kato -ke' "to bathe (plural only);"
80 VI American Indian Languages 2
326
Chipewyan -kl "to paddle a canoe, to travel by canoe") :
ndc/'xe "I paddle"
-xwAd "to cough" (cf. Kato kos "cough," as noun; Carrier
xwces): VAl/xwAd "he coughs"
-xwl "to vomit" (cf. Carrier khti "vomiting," as noun):
na/yd/dAdt/xwi "I vomit"
-gAd "to climb" (cf. Hupa -kas "to throw"): yd/yAl/gAd
"he climbs"
-gAC "to throw"? (cf. Hupa -kas "to throw"): qfwAt/dadt/-
gAc "it lies thrown down on top, table-cloth"
It will be observed that several verb stems are restricted
in their use as regards number of subject (or object). This
trait is characteristic of Athabascan, as also of other American
linguistic stocks.
Definite and Indefinite Tenses. My material on
Chasta Costa is not full enough to enable me to give a satis-
factory idea of its tense-mode system. It is clear, however,
that absolute time (present, past, future) is quite subordinate
to whether activities are thought of as taking place at some
definite time (generally present or past) or are more indefinite
as to time occurrence. Indefinite forms are apt to be used
for general statements that apply irrespective of any particular
time, for future acts, for negative (particularly negative future)
acts, and regularly for imperative and prohibitive forms. The
contrast between definite and indefinite present forms comes
out in:
definite: nd/da/yAct/t!d "I bathe" (i. e. am. now engaged
in bathing)
I indefinite: cAl/qlwe na/dAct/t!d "I'm used to bathing"
[ (here bathing is not restricted as to time)
f definite: tc!d/sAsL/se "I cry"
[indefinite: Vwi/dAn As/se "I always cry"
definite: yeSt/lo "he breaks into laughter" (i. e. laughs
at one particular point of time)
indefinite: ydt/lo "he laughs"
Six: Athabaskan and Na-Dene Languages 81
327
[definite: xAt/Ve/lal/la "they have been sleeping" (may
\ be said of them at moment of waking up)
[indefinite: xAt/VAl/lal "they sleep"
Futures, as we shall see, are explicitly rendered by suffixing
-Ve to present (generally indefinite) forms; but simple indefinite
forms, particularly with adverbs pointing to future time, may
often be used as futures in contrast to definite present forms.
Examples are:
\
definite: na/da/yil/el "we are bathing"
indefinite: nd/dil/ei "we'll bathe"
definite td!AB/da caI na/dU/si "story to-me you-told"
indefinite: xiin/de id! Ad /da uaI ndd/sl "tomorrow story
to-you I-tell"
definite: n/ye/Bic/l "I saw you"
I indefinite: xAL/tsH/dAn do/wan/yAc/l "this-evening I'll-
■[ see-you"
I definite: Ve/dic/ya ''I go''
[indefinite: xun/de Vac/yAc "tomorrow I'll-go"
Negative presents or futures are regularly expressed by pre-
fixing do "not" to indefinite forms; when more explicitly future,
-Ce is suffixed to them. Examples of indefinite forms preceded
by do are:
definite: tdac yd/yi/t!a "bird is-flying"
indefinite: dd/yd/t!a "he won't fly"
definite: na/yd/SAdt/xwi "I vomit"
indefinite: do/na/yAct/xwi "I do not vomit"
definite: tc!d/sAsL/se "I cry;" tc!d/dil/se "you cry"
indefinite: do/As/se "I'm not crying;" do/l/se "you're not
crying"
definite: ye/Bic/i "I saw him;" c/ye/Bi/l "you saw me"
indefinite: do/yAc/t "I didn't see him;" do/ld/c/yl/l
"you didn't see me"
[definite: Ve/Bic/ya "I go;" VcBjya "he goes"
indefinite: dd/VAc/yic "I'll not go;" do/VAc "he won't
I go" {<*VAlyAc)
82 VI American Indian Languages 2
328
Imperatives are simply second person subjective indefinite
forms. Examples are:
yi/l "see him!" c/yi/l "see me!"
nel/l "look at him!" (identical with indefinite present:
nei/l "you're looking at him"); ne/cil/l "look at me!"
nd/ni/An "stop him!"
Prohibitives are simply imperative forms preceded by ia:
la/yt/l "don't see him!"
First person plural indefinite forms may have hortatory signifi-
cance:
nd/xwil/ye "let us play!"
As regards form, definite tenses are primarily distinguished
from indefinite tenses by the presence of second modal prefixes
in the former, often also by the appearance of the first modal
prefixes in a fuller form than in the latter; the presence of
-I- or -i- in certain third person definite forms may also be
recalled. It seems, further, that certain adverbial prefixes
which have a short vowel (even though in an open syllable)
in definite forms lengthen it in corresponding indefinite forms:
definite: nd/da/yAct/t!d "I bathe;" na/da/yit/t!d "you
bathe;" na/da/yAt/t!d "he's bathing"
indefinite: nd/dAct/t!d' /Ve "I'll bathe;" nd/dit/t!6/Ve "you'll
bathe;" nd/dAt/t!6/Ve "he'll bathe"
definite: na/da/yil/el "we are bathing;" na/da/ydl/ cl
"ye are bathing;" na/'xAt/da/yAl/el "they are bathing"
indefinite: nd/dil/el "we'll bathe;" nd/dol/el "ye will
bathe;" nd/xAt/dAl/el "they'll bathe"
[definite: tc!a/yd/Bil/se "we cry;" tc!a/yd/ddl/se "ye cry"
undefinite: tc!d/ydl/se "we'll cry;" tdd/ydl/se/t'e "ye will
' cry"
These changes of quantity, however, are doubtless only second-
arily connected with change of tense, as indicated, e. g., by
Six: Athabaskan and Na-Dene Languages 83
329
tela- in definite singular forms: tcId/sAsL/se "I cry;" tcld/-
dil/se/ha "do you cry?" It is very likely that we are dealing
here primarily with considerations of syllabic and quantitative
rhythm or balance.^*
In Hupa Goddard has exhaustively shown that verb stems
often assume different forms for different tenses and modes.
This is very likely also true to a considerable extent of Chasta
Costa, but I have but little material bearing on this point.
A quantitative change is found in:
j definite -el: na/da/yil/el "we are bathing"
(indefinite -el: nd/dil/el "we'll bathe"
-c characterizes indefinite forms in :
definite -ya: Ve/Bic/ya "I go;" Ve/Si/ya "you go"
indefinite -yAc: t'Ac/yAc/Ve "I must go;" Vl/yAc/Ve "you
must go"
negative indefinite -yic: d5/VAc/yic "I'll not go;" Id/t'l/-
1 yic "don't go!"
f definite -ya: ya/da/yit/dja{<-t/ya) "we are ashamed"
^ indefinite -yAc: ya/dAcl/yAc "I am ashamed"
Pronominal Objects. Pronominal objects are regularly
prefixed to the verb. They come before deictic and first modal
elements, but after adverbial prefixes. Thus, while not as
thoroughly immersed in the verb form as the subjective pro-
nominal elements, they cannot well be considered apart from
it. The third person singular object is not designated. In
form the objective elements are, on the whole, identical with
the possessive pronominal prefixes of the noun. They are:
Singular 1. c- Plural 1. 7i6-
2. n-, ne- 2. no-
3. — 3. xo-
" Hardly stress accent as such. I cannot help feeling that such rhythmic phenomena
will turn out to be of fundamental importance for Athabascan generally.
84 VI American Indian Languages 2
330
"He — them" or "they — them" is expressed by means of xl-.
c- and n-, when standing at the beginning of a verb form, take
no inorganic a- before them (contrast subjective Ac-).
The definite forms of yed/l "he saw him" with combined
pronominal subject and object are:
With first person singular object:
Sing. 2. c/ye/di/l "you saw Plural 2. c/ye/do/l
me
3. c/yed/t 3. c/xA/yed/i
With second person singular object:
Sing. 1. n/ ye/Sic/ 1 "I saw Plural 1. n/ye/dit/l
you"
3. n/ yed/l 3. u/xa/ yed/l
With first person plural object :
Sing. 2. no/ye/dd/l "you saw Plural 2. no/ye/do/l
us"
3. no /yed/l 3. no /xa/ yed/l
For no/ye/dd/l "you saw us" one would have expected *«<?/-
ye/dl/l. It seems that "ye saw us" has been extended in its
usage to embrace also "you (sing.) saw us." It may indeed
be that my data on this point rest on a misunderstanding, but
there seems to be something analogous in Hupa. "You (sing.)
are picking us up" would be expected in Hupa to be *yun/nd/-
hil/luw {Ml- assimilated from hin-). Instead of this form,
however, Goddard lists yun/no/hol/luw, which is not identical
with but seems, as regards its second 6- vowel, to have been
influenced by ymi/nd/ho/luw "ye are picking us up."^^
With second person plural object:
Sing. 1. no /ye /die /I "I saw Plural 1. no/ye/dit/l
you (pi.)"
3. no /yed/l/ 1 1 3. no /xa/ yed/l /la
"Goddard, op. cit., p. 186.
Six: Athabaskan and Na-Dene Languages 85
331
With third person singular object:
Sing. 1. 7^/^^cA "I saw him" Plural 1. ye /Bit/ 1
2. ye /Si/ I 2. ye/ 66 /I
3. ye6/l 3. xi/yeB/i
With third person plural object:
Sing. 1. x6/ ye/ die/ 1 "I saw Plural 1. xo/ye/dit/l (heard
them" also as xo/we-)
2. ye /Bo /I 2. xo/ye/Bo/l
3. xl/dA/yeB/l 3. xi/yeB/i/la
Here again, one would have expected *x6/ye/Bi/l for "you
(sing.) saw them." As it is, "you (pi.) saw him" seems to be
used also for "you (sing.) saw them," both forms being logically
parallel in that both involve a second person — third person
relation, only one of the two persons, however, being plural.
Objective forms of indefinite tenses of this verb are:
With first person singular object:
do/wa/c/yl/i^/Ve "you'll see me"
dd/ld,/c/yi/i "you didn't see me"
c/yi/l "see me!"
c/ya/i*/Ve "he'll see me"
With second person singular object:
do/wa/n/yAc/l "I'll see you"
With third person singular object:
do/wa/yAc/l "I'll see him"
do/yAc/i "I didn't see him"
do/yAc/l/Ve "I won't see him"
do/wa/yl/i/Ve "you'll see him"
yi/l "see him!"
la/yi/i "don't see him!"
Objective forms of indefinite tenses of ne-l-'l "to look at" are:
With first person singular object:
ne/cU/i "look at me!"
ne/cAl/l "he looks at me"
86 VI American Indian Languages 2
332
With third person singular object:
neci/l "I'm looking at him"
do/ned/l "I'm not looking at him"
nel/i "you're looking at him;" "look at him!"
nel/l "let's look at him!"
ne 1x6 1 oil I "you (plur.) look at him"
This last form may, likely enough, have been mistranslated
for "you (plur.) look at them" (of. xd/ye/Bo/l above).
Other forms with first person singular object are:
cAsl/sl "he lets me, causes m.e to"
cAsl/t'dl "he kicked me"
With second person singular object:
ne/tc!ilc/lec "I'll bet you"
Passives. As in Hupa, pronominal subjects of passive
verbs are objective in form. From ne-l-'l are formed:
nes/ts!Ai/i "I am seen" ne/no/tslAl/t "we are seen"
nen/tslAl/l "you are seen" ne/no/tslAl/l "ye are seen"
ne' I is! All I "he is seen" nelxblislAlll "they are seen"
IsIaI-, which appears in these forms, probably contains third
modal -I- preceded by deictic tsl- implying indefiniteness of
logical subject: "man sieht mich." Apparently connected
with this IsIaI- is IsIaI- in:
mAllt'eltclAtltslAlllec "wherewith it is smoked, smoking
utensils"
Verbal Suffixes. A number of enclitic elements of tem-
poral or modal significance are found rather loosely suffixed
to verb forms. These, so far as illustrated in our material, are:
-t'e future particle (cf. Hupa -te, -tei; Kato -tei, -telle):
aclI AzIVe "I shall sneeze"
VAclldllVe "I shall sleep"
naldAcililo' iVe "I shall bathe"
nedlllVe "I'll look at him"
dolyAclilVe "I won't see him"
Six: Athabaskan and Na-Dene Languages 87
333
nd/dAci/nic/t'e "1 shall work"
do/nd/dACL/nic/t'e "I shall not work"
tcIdcL/se/re "1 shall cry"
do/As/se/re "I'll not cry"
tc!dl/se/t'e "you will cry"
-t'e seems to imply obligation to some extent, as well as simple
futurity, as is shown by its translation as "must" in some
cases :
As/se/t'e "he must cry"
rAc/yAc/re "I must go"
t'l/yAc/t'e "you must go"
All forms with suffixed -t'e, it will be noticed, are indefinite;
none has been found that is definite.
-nAl seems to be used for future acts:
dd/wd/An/nd/ya?i/nAl "he will upset them"
do I wa HIV at I nAl "they will be broken, go to
pieces"
-ha, -hd interrogative:
nd/xwil/ye/ha "are you playing?"
Ve/Bo/lal/ha "have ye been sleeping?"
no/ye/Bd/l/ha "did you see us?"
nel/i/ha "did you look at him?"
tcld/dil/se/ha "did you cry?"
do/o/se/ha "are ye not crying?"
As/se/Ve/ha "will he cry?"
tcId/ydl/se/Ve/ha "will ye cry?"
fd/yit/nd/hd "do you drink?"
ha seems to both precede and follow in :
ha/xwil/i/ha "do you expect?"
-/a probably inferential (cf. Hupa -xo/laii, -xo/liiii):
t'e/OAd/lal/la "I've been sleeping" (said on wak-
ing up)
Ldd/d/la "there was one (table)"
txAs/xe/la "(evidently) rich"
88 VI American Indian Languages 2
334
Probably also in :
do I wi I la "of course"
-la seems also to be used of simple narrative in past time, with
very weak, if any, inferential force:
tc!d/yit/siL/la "he pointed with his finger"
ye/ydt/ne/la "he bit it"
nAn/ndd/yd/la "he went around it"
no/yeB/l/la "he saw you (plur.)" (cf. no /yed/i "he saw us")
no/xA/yeB/i/la "they saw you (plur.)" (cf. no/xA/yeB/l
"they saw us")
xl/yeB/l/la "they saw them" (cf. xi/yeB/l "they saw him")
I do not know whether the contrasts in person and number
found in the last three pairs of forms are real or only apparent.
-le of unknown significance
nd/xol/ye/le "you (plur.) play"
Syntactic Combination of Verbs. Two verb forms some-
times combine syntactically, one depending on the other. The
second verb is subordinate to the first in:
dd/ucL/Ve nd/xwACL/ye "I-do-not-want I-play," i. e. "I
don't want to play"
hd/yACL/l yAct/lo "I-stop I-laugh," i. e. "I stop laughing"
hd/il/l yit/lo "stop laugh!" i. e. "stop laughing!"
ha/xwil/i/ha yAn/na/Ac "do-you-expect he-will bring?"
It seems that sometimes the first verb, which is then a third
personal form, acts as a sort of complementary infinitive to
the second:
yd/yi/t!a Bicl/sl "he-flies I-make-him," i. e. "I let him
fly"
ya/yi/t!a cAsl/sl "he-flies he makes-me," i. e. "he lets me
fly"
l/t'i yAn/na/'d "he-is-important he-has-for(?)," i. e. "he
brags about him"
Six: Athabaskan and Na-Dene Languages
89
TEXT: THE GOOD DOG.^^
l/t'l" yAn/na/'a^** xa/dAt^"
Make important he has his own
'An/t!e«2 na/ya" a/dJAn.«^
is like him moves about," he says.
li/tcle.^"
dog.
de/ucL/t'e"
"What I want
d6/dAt«i
'Nowhere
hi/t !i«6
that thing
** Wolverton Orton claimed not to know any regular Chasta Costa myth texts. The
following, which is merely an English joke anecdote taken from a popular periodical that
happened to be lying about and translated into Chasta Costa by Mr. Orton, will at least
serve to give some idea of Chasta Costa word order and sentence construction.
"^, third modal element, -/'t, verb stem. Cf. Carrier tU/thi "thou makest him
valuable, treatest him as important."
**7/in- and na-, adverbial prefixes, -'o, verb stem. For na/'a- "to have," cf. Hupa
naii/a/te "you will have." "He has his dog made valuable, treated as important," i. e.,
"he brags about his dog." Indefinite tense, because statement is general and does not
refer to any one point of time.
" X-, third personal pronominal element. -d/dAt, reflexive possessive element.
*" Possessed form of it "dog." Observe change of I- to /-, and suffixing of -tele. Cf.
Hupa -lin/k{yi)e; Chipewyan Ltn/k'e.
" do, negative adverb. -dAl, postpositive element.
" 'a-, reduced from 'a-, prefix used with verbs of saying, doing, and being. It is
probably equivalent to indefinite demonstrative: "(there is of) that (kind)." -n-, first
modal element, -//e, verb stem. Cf. Hupa Mn//c "there is;" Kato qn/^'e "it is;" Chipe-
wyan an/t'e/hi/k'e "it was." Indefinite tense, because statement is general.
" nd-, adverbial prefix, -ya, verb stem. "Moves about," i. e., "is living, is to be
found": "there is no (dog) like him anywhere." Cf. Hupa na/wa "they were there;"
Kato na/ga/kwqn "he had walked;" Navaho na/oa, i. e., na/ya, "he is going about"
(quoted from Goddard, Analysis of Cold Lake Dialect, Chipewyan). Indefinite tense;
general statement.
" d-, prefix used with verb of saying; see note 62. dj-, third personal deictic prefix;
or perhaps dJA- = *dyA-, reduced from *dye-, first modal prefix d- and third person deictic
prefix ye-, -n, verb stem. Probably definite in tense, though it shows no second modal
prefix; cf. Hupa present definite third singular a /den.
" de, apparently relative in force, u-, adverbial prefix, -c-, first person singular
subjective element. -L-, third modal prefix, -t'e, verb stem. Indefinite tense; general
statement.
•• ht, demonstrative stem, -tli, suffix applying, it would seem, to things. Perhaps
ht/tii is assimilated from "hx/tla; for -tla, cf. Chipewyan t!a "that; often used to point
out one of several persons or things characterized by a descriptive phrase or clause."
(335)
90 VI American Indian Languages 2
336
s/ts.'An/na/'Ac" a}/dAc/nl/dAn.«» do t'wi/de«» La a/dJAn.
to me he brings when I tell him." "Not everything," one says.
H/tli/ni^" mAn/me/q !e" t'/yl/i." xat q!wAt/tc!At/dja"
Dog-owner around in house he looked around. Then table
Ld^/a/la^^ mAn/me" wAs/xe q !wAt/da^/gAc" na^/L!6"
one there was in house. Good table-cloth, paper
hi q!wAt/dasL/na^» i/yi/tc'u^" mAl/t'e/tc!At/ts!Al/lec.»«
that was lying thereon, that wherewith it is smoked.
" S-, assimilated from c-, objective (or possessive) first person singular pronominal
element. -Is! An, postposition, no-, adverbial prefix, -'^c, verb stem. Cf. Hupa do/-
xo/lin/na/ta/auw "he won't carry." Indefinite tense; general statement.
** a-, as in notes 62 and 64. -1-, postposition; refers to implied third person indirect
object of verb, -d-, first modal prefix, -c-, as in note 65. -nl, verb stem. -dAn, post-
position; here used to subordinate verb. Cf. Hupa Hn/niL/duw/ne "I am telling you."
Indefinite tense; general statement.
^' t'wi, pronominal stem denoting totality, -de, indefinite demonstrative stem.
'" /t "dog." -tli/nt "one who has;" evidently contains common Athabascan suflfix
-n, -nl "person."
" m/4n "house." -me/q!e, compound postposition.
" /'-, first modal prefix reduced from t'e-. 71-, first modal prefix 7- reduced from ye-,
-I- remaining unexplained. -I, verb stem. According to this analysis, t' /yt/i is indefinite
in tense; this seems hard to understand, as it refers to one act in past time. Another
analysis seems more likely: /'-, instead of or misheard for t'e-, form regularly used in
definite tenses; 7-, second modal prefix; -?-, definite third person ending for 7- verbs.
Cf. Hupa tcii I te I we I in I il "he looked about as he went along."
" q/wAt-, postposition "upon" used as adverbial prefix, k!-, deictic prefix here indi-
cating indefiniteness of object. -/-, third modal prefix presumably with passive force.
-dja, from -ya after -/-, verb stem "to eat." "It is eaten thereon," i. e., "table."
^* Ld, numeral "one," to which verb proper, 9/d/la, is attached. 6-, second modal
prefix, a-, verb stem, -la, verb suffix. Definite past tense, because referring to definite
point of time in narrative. Cf. Chipewyan de/*a/ht/k'e/lai "(lake) was there."
" -me, postposition.
^^ qlwAt-, as in note 73. da-, adverbial prefix. 6-, second modal prefix. -^, third
modal prefix. -gAc, verb stem. Verb form ("it lies thrown on top") used as noun.
" nd-, adverbial prefix, d-, second modal element. -lIo, verb stem. Verb form
("whereon there is writing") used as noun.
'* qlwAt-, as in note 73. da-, adverbial prefix, s-, second modal prefix. -L-, third
modal prefix; doubtless original -/- changed to -1-, -L-, because of preceding s-, which in
turn is prevented by it from changing to d-. -na, verb stem. Definite past tense.
" Analysis uncertain, presumably demonstrative in force.
^ ntAl- "therewith" consists of pronominal stem m- followed by postposition -1-.
t'e-, adverbial prefix, tc!-, deictic prefix indicating indefiniteness of object. -/-, consonant
borrowed from following -ts!, to complete syllable begun by k!-. tsUl-, apparently passive
in force, -lee, verb stem. Verb form used as noun: "smoking materials."
Six: Athabaskan and Na-Dene Languages 91
337
txAs/xe/la^i d6/at/t!L82 q!wAt/tc!At/dja xa s/tslAn/na/Ac
He was rich bachelor. "Table quickly he'll bring to me,"
dJAn/la«3 H/t!i/ni. d6/LAn»^ xwacl/P» djAn/la da/at/t !i/n!.
said dog-owner. "Not much I believe it," said bachelor.
ne/tc!uc/lec.»« c6"/dji." an/yl/aLS* djAn/la H'/t!i/ni tclo/yit-
"I'll bet you." "All right!" "Come here!" said dog-owner, he pointed
/siL/la«« H q !wAt/tc !at/dja lat/dAn'" nAn/na^/ya/la'^
with his finger. Dog table once he went around.
xat ye/yat/ne/la^^ ma^/dAn.^^ la djAn/la d6/at/t!i/ni
Then he bit it at edge. "Don't!" said bachelor,
t'wi/de do/wa/na/yan/nAl.^" do/wl/la»*djAn/laq!wAt/tc!At/dja
"everything he will upset." "Of course," he said, "table
^^txAs/xe, adjective stem "rich;" perhaps related to wAs/xe "good." -la, verb
suffix of probably inferential value.
^"^ do, negative, at = at! "wife." -t!i, noun suffix denoting "one who has." "One
who has no wife," i. e., "bachelor."
*^ dJAn, as in note 64. -la, verb suffix.
^* do, negative, lah, adverb "much."
** XW-, adverbial prefix, -c-, first person singular subjective pronominal element.
-/.-, third modal prefix; from -/-, because of preceding -c- (cf. note 98). -I, verb stem.
Indefinite present, negative adverb preceding.
^ne-, second person singular objective pronominal element, tclu-, adverbial prefix;
very likely really compound of deictic element tcl- (indicating lack of specified object,
namely wager) and modal 6-, il- denoting future imperative, -c-, as in note 85. -lee,
verb stem. Indefinite present, because of future or slight hortatory meaning: "let me
bet with you!" Cf. Chipewyan tus/be "let me swim."
*' With co"-, cf. Hupa -hwon "good;" Kato -con "to be good."
*' an-, adverbial prefix. 7-, first modal prefix, -i-, second person subjective pro-
nominal element. -aL, verb stem. Indefinite tense, used as imperative.
*^ tclo-, adverbial prefix; perhaps compound of deictic element tcl- (object pointed
out is not specified) and first modal 6- of unknown significance, y-, second modal prefix.
-t-, connecting element between second and third modal elements, characteristic of third
person of definite tenses with 7-. -/-, third modal prefix, -sit, verb stem, -la, verb
suffix. Definite past; marks point in narrative.
^ Numeral adverb of la "one." -dAn, postposition.
" HAn- and na-, adverbial prefixes. -0-, second modal prefix, -yd, verb stem, -la,
verb suffix. Definite past; refers to definite point of time in narrative.
^ ye-, adverbial prefix, -ya-, second adverbial prefix. -/-, third modal prefix, -ne,
verb stem, -la, verb suffix. According to this analysis, this verb is indefinite in tense,
which is difficult to understand. More plausibly, yat- may be considered as misheard
for yAt-; 7- second modal prefix. In that case, it is definite past.
^^ ma*^-, noun stem "edge." -dAn, postposition.
" dd/wa, proclitic adverb indicating futurity, probably not with absolute certainty.
na-, adverbial prefix, -yan, verb stem. -nAl, verb suffix. Indefinite in tense, because
future in meaning.
" Adverb containing inferential -la.
92 VI American Indian Languages 2
338
do/wa/An/na/yan/nAl9« t'wi/de do/wa/1/t'at/nAl" 0Ak/gwe
he will upset, everything will go to pieces, in fragments
s/tslAn/na/Ac. ha/xwil/i/ha^^ AL/tca/yi»« yAn/na/Ac'""
he will bring to me. Do you expect big thing he will bring here
}a/me/q!e/ca.'"' wAs/xe If. na/ni/An'"' na/nl/An djAn/la
all in one time? Good dog." "Stop him, stop him!" said
d6/at/t!i/ni. d6/da/q!e>03 na/nAc/An'o" djAn/la H/tlT/ni
bachelor. "Unable I stop him," said dog-owner,
t'wi yAn/na/Ac/t'e'"^ xwAn/ne/^iL/ya'"« djAn/la d6/at/t!T/ni.
"all he will bring here." "You win," said bachelor
** As in note 94, except that another adverbial prefix, ah-, is present.
" do/wa and -tiaI, as in note 94. i-, third modal prefix. -I'at, verb stem.
'* ha, interrogative adverb, xw-, adverbial prefix, -i-, second person singular sub-
jective pronominal element. -/-, third modal prefix. -I, verb stem, -ha, interrogative
suffix. Indefinite present in tense.
" A-, of unknown significance. L-, prefix common to several adjectives, -tcd/yi,
adjective stem "big."
^'"^yAtt- and na-, adverbial prefixes, -ac, verb stem. Indefinite tense, because point-
ing to future time.
"" ia, numeral stem "one." -me/q!e, compound postposition, -ca, found also with
la alone: Id'^/ca "one."
'"' nd-, adverbial prefix, n-, first modal prefix, -l-, second person singular subjective
pronominal element. -An, verb stem. Imperative mode.
103 Adverb containing negative do-. Perhaps -qle is postposition (cf. -me/qle).
^^ no.-, n-, and -ah, as in note 102. -c-, first person singular subjective pronominal
element. Indefinite in tense, because of preceding negative adverb.
"* As in note 100. -Ce, future suffix; here used because idea of futurity is more explicit.
">* xtf /in-, adverbial prefix, we-, first modal prefix. 6-, second modal prefix, -i-,
second person singular subjective pronominal element. -L-, third modal prefix, -ya,
verb stem. Definite present in tense.
Six: Athabaskan and Na-Dene Languages 93
APPENDIX.
A few Galice Creek words were obtained from Mrs. Punzie,
a few Applegate Creek words from Rogue River Jack. These
two Athabascan dialects are probably practically identical.
5 indicates something acoustically midway between 5 and c;
r (tongue-tip trilled) and / occur as reflexes of Athabascan n;
nasalization (indicated by ') seems to occur, k' and k'w are
found as contrasted with Chasta Costa x and xw.
Galice Creek.
ya'/k'ds "seeds (sp.?);" said to be called bdnax or b4yu
in Chinook Jargon
tcla/ba/d/k'wa's "brush used for medicinal purposes (sp.?)"
L!d'/ddi "tar- weed" (cf. Chasta Costa ild'/de; Hupa
Lo/daitc)
yel/yat/ts!ai/yt "sunflower" (cf. Chasta Costa, tc I Al/y at /ts/e)
giis "camass" (cf. Chasta Costa god; Hupa kos "bulbs")
ddl/si "pine" (cf. Chasta Costa dAl/si; Kato dul/tclk)
Id I lH "pine-nut" (cf. Chasta Costa nd/de)
del res "manzanita" (cf. Chasta Costa dAJnAc; Hupa
din/nuw; Kato tun/nuc)
md'/ts!i "cat-tail" (cf. Chasta Costa mAt/tdi)
sds/da' "oak" (cf. Chasta Costa cAc/dd')
i/dd/ge "acorn" (perhaps misunderstood; cf. Kato i/taG
"black oaks")
Applegate Creek.
k'q'/tc'u "goose" (cf. Chasta Costa xd'/tc'u; Kato ka')
dAc/tc'u "grouse" (cf. Chasta Costa dAc/tc'u; Kato dAcf-
tco)
(339)
94 y^ American Indian Languages 2
340
dAclVe' Itc'u "bob-white, quail"
k!ai'/dic/tca/we "ruffed grouse, pheasant"
k'dn/ta/tc'u "pigeon" (cf. Kato kwl/ylnt)
do/s'An/tsfa/ya "screech-owl"
si lick lies "kingfisher"
tddlkeldi "red-headed woodpecker"
tea lwdcltc{!)e "sandhill crane" (cf. Chasta Costa sd'lwAsl-
tsle)
Editorial Note
Originally published in University of Pennsylvania, Anthropological Pub-
lications 2(2), 271-340 (1914). Reprinted by permission of the University
Museum, University of Pennsylvania.
The Athabaskan languages of southwestern Oregon — which are now (1989)
virtually extinct — constitute a distinctive subgroup whose relationship to the
northwestern California subgroup is surprisingly distant, although the two are
usually classified together as "Pacific Coast Athabaskan" (Hoijer 1960). The
Oregon group seems to have comprised four distinct, though closely related,
languages: Upper Umpqua, Galice-Applegate, Chetco-Tolowa, and Lower
Rogue River. The last was spoken in a variety of local dialects, the best known
of which are Chasta Costa (on Rogue River about 30 miles upstream from the
mouth), Tututni (near the mouth of the river). Euchre Creek (along the coast to
the north of the river mouth), and Upper Coquille (still farther north, in the
inland area behind Coquille Bay). No full grammatical study exists of these or
of any Oregon Athabaskan language. Besides Sapir's description of Chasta
Costa, the fullest published documentations of Oregon Athabaskan are of
Galice (Hoijer 1966), Tututni (Golla 1976), and Tolowa (Bright 1964; Bom-
melyn et al. 1989). Considerable manuscript documentation also exists, partic-
ularly of Galice (by Melville Jacobs; see Seaburg 1982) and of Lower Rogue
River (by J. P. Harrington; see Mills 1981: 69-76). Some of Harrington s Lower
Rogue River material (including a number of aluminum disk recordings) was
obtained from Wolverton Orton, Sapir's Chasta Costa consultant.
Corrigenda to Father Morice's "Chasta Costa and the
Dene Languages of the North."'
In view of the fact that Father Morice has reviewed my Notes on
Chasta Costa Phonology and Morphology in so evidently a friendly spirit
it may seem a bit churlish to point out what seem to me to be either
slips or misunderstandings in his recently published paper on Chasta
Costa and more northern Athabaskan dialects. If, nevertheless, I
venture to do so, it is not because of any desire to minimize the value of
Father Morice's paper or to attach an overweening importance to my
own very scanty contribution to Athabaskan linguistics, but to help
advance our understanding of the problems of Athabaskan phonology
and morphology. The chief value of Father Morice's paper seems to
me to lie in the further light it throws on the Carrier language, of which
previous papers have already shown he has an admirable mastery. I
earnestly hope that Father Morice will not be content with the rather
sketchy papers he has hitherto given us on the Carrier language, but will
eventually publish a complete presentation of the intricacies of its
phonetics and grammatical structure.
I. " Dr. Sapir's a is my as, almost the sound of u in ' but,' more
exactly that of e in the French ^e, te, le" (p. 560, footnote). My a of
Chasta Costa words is to be pronounced like a of- German Mann and thus
in sound corresponds to Carrier a, not (b. My A is another vowel alto-
gether, though often phonetically reduced from original a. It is prac-
> See American Anthropologist, N. S., 17, 1915, pp. 559-572.
96
VI American Indian Languages 2
766
tically identical in sound with u of but and doubtless corresponds phone-
tically, largely also genetically, to Carrier o^.
2. "This [C. C. tclac 'bird'] is evidently none other than the
Carrier tscrz [Morice's s and 2 are described by him in a letter as sibilants
midway between 5 and c, and z and j, respectively; they correspond histori-
cally to Chipewyan^and t? < Ath, s and z], which in that language means
not ' bird ' but 'feather-down' " (p. 560, 1. 15). This is not so evident. In
fact it is phonologically impossible. Father Morice's Carrier form is
clearly cognate with Chipewyan -d'Hd (Goddard; my -dUd) "soft feathers,"
Montagnais, tthoepdh (Petitot; tdlA-yd^ in my orthography) "plume
servant k broder, cote de plume " (possibly mistranslated for " duvet "),
Hare kkwew (my k^\ew), Loucheux tthaw (my td\aw), Navaho ts'os
(Franciscan Fathers; my /5!o5) "down feather." These forms imply
original Athabaskan *ts\ez (reduced *ts\ez, *ts\es) and *ts\os "down."
This would be expected to appear in Chasta Costa as td\Ad (very likely
actually found in td\Ad-nd-yal-td\od " humming-bird," perhaps literally
"soft-feathers fiy-about-making-a-humming-sound"; for verb stem
•tdlod cf. Hupa -tsots, -tsos " to make a kissing-Iike noise, to smack one's
lips," Nav. beets' OS " whistle " as noun). C. C. tc\ac can have nothing
to do with this tdlAd, but must go back to Ath. *k^\ac or *tc\ac}
3. " This possessive [C. C. -lUc\e '(his) dog '] entails in Chasta Costa
the accretion of a sort of suffix which he writes tele, the equivalent of
my tse. Now litse means in Carrier, not somebody's dog, but she-dog! "
(p. 561, 1. 3). Father Morice's quandary is of his own seeking. -ts\e
of his Carrier form is not at all connected with -tele of my Chasta Costa
one. Carrier li-ts\e is simply compounded of li "dog" and ts\e
" woman." This ts\e (Ath. *tc\e) is found in Carrier tsekhk, Montagnais
ttsh-kwi, ttse-k'^ (i. e., tsle-k'e), Hare ttse-Une, ttse-k'u, Loucheux ttse-ndjo,
Kato tc'ek, C. C. ts\dxe. Carrier, Montagnais, Kato, and Chasta Costa
point to Ath. *tc\e-k'e " woman." C. C. -tele of -li-tcle, however, goes
back, not to Ath. *-tcle, but to Ath. *-k^le with glottalized anterior palatal
k^l; cf. Hupa -liii-ke " (his) pet " (read- k^le). How this peculiar possess-
ive *-k''le (cf. normal -e in, e. g.. Hare se-llin-e " my dog ") arose I can
not say. Its isolated character stamps it as probably archaic. Perhaps
-jfe"!- of Ath. *-liiiki'\e is connected with -g-, -k of Ath. *lik, *hg-, which is
parallel to *ltrj; cf. Anvik hlik (Chapman; my lik), Loucheux V egce-tsellce
> Since this was written Dr. Goddard has kindly called my attention to Hupa
klyawM (read k^lauW) " birds " which corresponds exactly to Ath. *k''!ac. This elimi-
nates Ath. *lc!ac as possible prototype and still more effectively disposes of Father
Morice's analysis.
Six: Athabaskan and Na-Dene Languages 97
767
" petit chien " (i. e., legA-). (Incidentally, Petitot gives for " chienne "
Montagnais I'intse, Loucheux Ventsi. Are these forms errors for -ttsi,
-ttsi})
4. " The- . . . , at least in the north, does not merely mean ' in the
water,' as Sapir would have it . . . but it hints at the ' bottom of the
water ' " (p. 561, 1. 25). C. C. t^e- was defined as " in the water."
That, in Pacific Athabaskan, it has no necessary reference to " the
bottom of the water " is showji by several of Goddard's Hupa and Kato
examples. Thus, some Hupa verbs with t^e- are translated " in the
water it seemed about to tumble," " in the water she floated back,"
" when he put his hand in it (i. e., the water)," cf. Kato <'e'- verb " she
washed them." Evidently Goddard's definition of Hupa /'e- as re-
ferring " only to motion into water and under its surface " seems quite
adequate. In Carrier thentltlat " thou art precipitately brought to the
bottom, thou sinkest," second modal -n- (to use Goddard's terminol-
ogy) is terminative in function, so that t^e-n- . . . -tlat necessarily
denotes " to sink to the bottom of the water," even if Ce- by itself be
assumed to mean " in the water." A wider range of significance for
Ath. *t^e- than Father Morice would assign it is implied also by such
Navaho forms as tqe-li "water horse"; Navaho tqe- (i. e., my txe-)
regularly corresponds to Ath. *Ce-. Navaho tqe- forms are likely to be
archaic, as Ath. *Ce- has been almost entirely superseded in Navaho by
Ath. */'a- (Navaho tqa-); see below. Even in northern Athabaskan I
do not find Father Morice's remark to apply without qualification, even
if correct for Carrier. In Anvik Ath. *t^e- appears as te-, tH- (Chapman's
orthography); note teyldoutel "floating in the water." In Loucheux
Ath. *Ve- appears as tchi- (Petitot's orthography), i. e., tcH- (Loucheux
Ve-, i. e., /'e-, does not correspond, according to Loucheux phonetic laws
that I have worked out, to Ath. */'e-, but to Ath. *t'a-)\ note tchi-dhitlle
"gtre k flot."
5. " Tha- . . . , which he gives as ' referring to the water ' has
really that signification [i. e., ' in the water '] " (p. 561, 1. 27). Ath.
*Ca- is undoubtedly found employed locally (" in the water ") in prac-
tically all Athabaskan dialects; indeed it seems in some dialects (e. g.,
Navaho, to some extent apparently also Chipewyan) to have extended
its sway at the expense of parallel Ath. *t^e-. There is, however, plenty
of evidence to show that Ath. *Ca- is frequently substantival, not local,
in force, that it is, in fact, an old noun stem (" water, wave, sea," parallel
to more wide-spread Ath. */'o " water ") that has become a verb prefix.
A local meaning is impossible, for instance, in the numerous verbs of
98 VI American Indian Languages 2
768
drinking found in Pacific Athabaskan (e. g., Hupa ta-win-nan " he
drank it," Kato ta-ya'-o-nqn " let them drink," C. C. t'd-yAct-nd " I
drink "). A direct substantival meaning, moreover, is obvious in such
compound nouns as Navaho tqd-bd " shore " (lit. " water-edge "),
while in several Mackenzie Valley dialects Ath. */'a even occurs as un-
compounded noun stem (Montagnais t'a, i. e., my t^a, " flot, onde ";
Hare t'a; Loucheux t'i:).
6. " As to the verbal stem -al, which he believes to mean ' to come,'
I more than suspect that it is but a corresponding form of -ya, which he
represents as expressing the idea of ' going, coming,' and should be
-yal " (p. 562, 1. 11). That C. C. an-yi-al really means "come on!"
and not, like Carrier 'oen inyal, "go on!" (as suggested further on by
Father Morice, 1. 26) is conclusively proved by two facts. In my brief
Chasta Costa text (p. 337, 1. 3) an-yi-aL occurs as translation of English
" come here! " (this text, it should be remembered, was translated from
English) under circumstances in which " go on!" would be quite out
of place. Further, during my residence among Chasta Costa Indians
in Siletz Reserve, Oregon, in 1906, I distinctly remember that an-yi-al
was often used by elders in calling children to them. As to whether
recorded C. C. an-yi-al, an-yi-aL is to be understood as an-yi-yal, as
claimed by Father Morice, or as an-yi-'al, as I had assumed, I now incline
to think that -yal is correct. .This is because of such forms as Hupa
ivin-yaL " come on " and Kato gUn-yaL " walk," which seem to cor-
respond exactly to C. C. -yi-{y)al. Should -'al prove to be correct, it
would probably have to be compared with Loucheux -'a " to go," Ath.
*-ac " two go " (with dualic -c).
7. " Dr. Sapir furthermore quotes the verb stem -t\o as denotive of
the act of swimming, while, according to him, that of paddling is rendered
by the radical -xe. Now, in most northern Dene dialects, the former
refers to paddling, while the latter indicates the act of navigating, or
moving about in a canoe" (p. 562, 1. 28). True, but there is plenty of
evidence to show that Ath. *-k^e, *-Ven, *-k'^er}, *-kH frequently refers
to or implies paddling. Thus, Goddard translates Chipewyan -ki by
" to paddle a canoe, to travel by canoe " (e. g., ta-kt-hwu " when he
paddled "). This is confirmed by Father LegofT, who defines 'ke'l
(i. e., -k'el): " est le progressif de 'ki, et signifie proprement naviguer, en
ramant " (e. g., pes-'ke'l " je rame, j'avance en ramant "). Similarly,
for " ramer " in Hare Petitot gives not only e-ttoh, but also e-k'e. And
in Anvik I find -kan, -kahl often translated as " to paddle " (e. g., xH-kahl
'" he is paddling," ti-qH-kan " he paddled on "). For C. C. -t\o " to swim,
Six: Athabaskan and Na-Dene Languages 99
769
bathe " I find no exact parallels. It may, as Father Morice suggests,
have primarily meant " to paddle " (denominative verb from Ath.
*t\os " paddle ") and changed its force dialectically.
8. " The verbal stem -lal . . . , to which our author attributes the
sense of ' to sleep,' has in Carrier the value of ' to dream of ' (with a
complement). Might not Dr. Sapir's informant have misunderstood
his questioner and thus unwittingly misled him? " (p. 563, 1. i). I
think we can manage without this hypothesis of misunderstanding. In
Hupa -lal, -laL means not only " to dream," but also " to sleep " (e. g,
nit-te-sil-lal-le " you would go to sleep "). Both meanings are given also
for Kato -lal, -IqL (e. g., n-to-lqL " let him sleep "). Turning to northern
Athabaskan, we find that in Chipewyan (Montagnais) -lal, -lal, -lal,
-lal regularly means " to sleep, to fall asleep, to put to sleep " (e. g.,
Goddard's hi-teL-lal " he is asleep"; Petitot's in-t'es-l'al s. v. " dormir
debout," es-l'al s.v. "endormir"; Legoff's in-'tes-la'l " je m'endors "),
while for " rever " Petitot gives quite another stem in Montagnais
and Hare. C. C. t'e-d-lal-{la) " he's been sleeping," with its prefixed
elements t'e- and (in definite forms) -6-, corresponds remarkably to
Hupa -te-sil-lal- quoted above, Kato {n-)te-s-laL " he went to sleep,"
and Chipewyan {ln-)te-d-Lal " he is asleep." In Hupa and Kato -lal,
-lal, when meaning " to dream " seems regularly preceded by 7ia- (e. g.
Hupa kin-na-is-lal " he dreamed," Kato u-na-s-laL " he dreamed
about ").
9. . . . '^ -lal, or rather -tlal. ... As may be seen by Sapir's
rendering: t'Ulal. The double consonants tl and ts are of frequent oc-
currence in Dene and form as many indivisible groups. The syllables
of all Sapir's verbs in the first person plural are wrongly cut up: the t
which he attributes to the penultimate syllable should commence the
last one: -tlal, -tsoel, -tlat, -thcec, etc. Hence several of his verb stems are
incomplete. For instance, -se, 'to cry,' should be -tse (Carrier -tso)\
-si, ' to cause,' should be -tsi (Carrier -tsi); -lo, ' to laugh,' cannot be
understood without its /. . . . Were he familiar with the Dengs' syl-
labic way of writing their own language, he would have been spared this
little inaccuracy " (p. 563, 1. i and footnote i). There are several com-
ments I wish to make on this passage.
(a) That, in Chasta Costa, not -tlal but -lal is to be considered as
true verb stem is quite unmistakably evidenced by such forms as t'l-lal
" thou art sleeping," I'o-lal " ye sleep." Were -/- part of the stem,
there would be absolutely no reason for its disappearance in these forms
(cf., for phonetic comparison, t'd-yit-nd " thou drinkest," I'd-yd-dot-nd
100 VI American Indian Languages 2
770
" ye drink," in which third modal prefix -t- is preserved between personal
element -i-, -6-, and stem consonant n). These remarks apply also to
C. C. -lat, not -tlat, as Father Morice would have it. Comparison with
Hupa and Kato, also Chipewyan, abundantly confirms my own analysis.
(6) There is no point whatever in quoting C. C. t'it-lal (not t'ltlal,
as Father Morice has it) " we are sleeping " as evidence of a stem -tlal.
In Chasta Costa -{i)t- regularly appears as first person plural subjective
prefix, except, as in Chipewyan, before third modal -/- and -/- (e. g.,
ye-dit-'i " we saw him," where Father Morice would hardly claim that
-t'i, not -'i, is the true stem). That this -t- is indeed an organic element
in the first person plural prefix is shown by comparison with Hupa -d-
(e. g., na-dtL-Le " we are painting "), Kato -d- (e. g., na-ddl-yic "let
us rest "), Chipewyan -it-, -d- (e. g., e-gU-'t " we saw it "), Loucheux -di-
(e. g., i-di-kwoll " nous amenons "). What has apparently misled Father
Morice in his analysis of Chasta Costa is that Ath. first person plural
(and dual) *-{i)d{e-)- has in several dialects, perhaps by analogy of third
modal -I- and -/- verbs where -d- regularly disappeared (I am inclined
to think that in Hupa and Kato -d- has in these cases been restored by
analogy), been replaced by -i-, -i-, e. g.. Carrier nd-i-ta " we are both
sick " (Carrier first personal plural prefix tsoe-, i. e., ts\A-,'^ does not seem
to be a widespread Athabaskan element, though Father Jette has
recorded for Ten'a ts- forms of like meaning, perhaps to be understood
as ts\-); Hare i-ssi "nous deux faisons " {ya-issi "nous faisons");
Navaho ch'i-ne-i-ka " we two carry milk out " {ch'cB-de-i-kd " we carry
milk out "). Even in these dialects, however, survivals of the old -d-
are found in such cases as allowed of its carrying over into the stem
syllable, e. g.. Carrier nt-tas " we two walk," i. e., nl-tlas, morphologically
equivalent to n-U-'as (cf. noe-hce-'as "they walk two together"); Hare
witta " nous deux allons," i. e., un-tla (Ath. *zi-tlas), morphologically
equivalent to w-it-'a (Ath. *z-id-'as); Navaho da^-dt-t'a^ "we put a
round object on " (equivalent to da^-d-it-'a'), bd^-ni-'nil " we two put
several things on " (equivalent to ba^-n-U-nil), l-gye " we two marry"
(equivalent to id-ye, cf. i-ye " he marries "). There can be no talk in
such cases of stems -t\as, -t\a\ -'nil, and -gye but only of phonetic resul-
1 1 strongly suspect, as is indeed indicated by Father Morice himself (see "The
Dene Languages," 1891, p. 193) that this /foj- is really impersonal in meaning, ta
begin with; with Carrierna;-/5a5-/»7 "we walk"; cf. nas-tsi-ya "one walks." For reasons
of modesty, perhaps, it may have become customary to say "people (in general) do
so and so" instead of " .ve do so and so." This indefinite Carrier tsce- is quite possibly
cognate with Hupa U-, i. e., tc!-, of third personal forms applying to adult Hupa (e. g.,
tce-xauW "he is catching," originally perhaps "one catches").
Six: Athabaskan and Na-Dene Languages 101
771
tants of first person plural -d- plus stems -as, -'a\ -nil, and -ye (Ath.
*-7e).
(c) As regards syllabic division, I must emphatically disagree with
Father Morice. It is quite clear that in many Athabaskan dialects the
-/- {-d-) of the first person plural belongs to the following syllable, as we
have just seen. This proves nothing, however, for Chasta Costa, nor
do I see how familiarity with the Carriers' syllabic way of writing their
own language would have materially helped my ear in its perception
of Chasta Costa sound combinations. As a matter of fact, my informant
Wolverton Orton was particularly careful in syllabifying and I was prac-
tically never in doubt, in recording his forms, as to where syllabic di-
vision was to be placed. Hence t'ii-lal is not to be " corrected " to
tH-tlal, t'e-nit-lat to t'e-ni-tlat, yit-lo to yi-tlo, tclAt-t'it-dAl to tc\At-Ci-tBAl
(Father Morice's -tscel). Father Morice's -thcec is simply another ortho-
graphy for my -t'^Ac. I may point out that were we dealing in these
Chasta Costa forms with such consonantal groups (affricatives) as //
and tS, preceding i would appear lengthened to I, because completing its
own syllable. That I have consistently misheard both vocalic quantity
and syllabic division in these words (e. g., -iHt-dAl for -Cl-tdAl) I cannot
admit. Naturally I do not deny that td and td\ also occur in Chasta
Costa as true affricatives, though Id seems to have regularly developed to 6.
As to C. C.-t-lo (I have no example of C. C. affricative il, dl, nor
does it seem to occur in Hupa or Kato), I am uncertain, as I pointed out
in my " Notes," whether this is to be understood as directly representative
of Ath. *-dlo (as preserved, e. g., in Carrier, Chipewyan, and Navaho)
or as analyzable into third modal -t- and stem -Id. For Ath. dl: I, cf.
Ath. *-dlo, *-dli " to be cold ": Montagnais e-llu " cold," Kato Id " frost."
C. C. -Id " to laugh " might well correspond to Kato -Id " to deceive "
(primarily ".to laugh at"?).
(d) As regards C. C. -se " to cry " and -si " to cause," Father
Morice is going altogether too far when he says these stems are incom-
plete for -tse and -ist. As I have already stated in my " Notes " (p.
287), Ath. tc\ of which Carrier ts is reflex (Ath. /c' is preserved as such
in Navaho, Jicarilla Apache, Kato, and, as labialized tc'w, in Hupa),
has always been simplified in Chasta Costa to 5 (Ath. tc^>ts>s; cf.
Ath. ts>td>d). Hence to " correct " C. C. -se to -tse is as justifiable
as it would be to " correct " French chef to *kep because, as no one
denies, derived from Latin caput.
ID. " If Dr. Sapir will allow me, I will also observe that the desinence
-tc'ac [probably misprint for -t^Ac], which he gives as a distinct verbal
102 VI American Indian Languages 2
772
element, is nothing else than the plural stem of the same [Carrier] verb
nanisthi, whose derivative nthcesthih effectively means ' to lie down, go
to bed ' " (p. 563, 1. 7). Father Morice then goes on to compare C. C.
-t^Ac directly with Carrier -this, -thez, -thcES (plural stem). These state-
ments are misleading. It is evident from my Chasta Costa material
(e. g. dd-nAc-t^Ac " I go to bed ") that -t^Ac applies to singular subjects.
This is confirmed by comparison with Hupa -tiiW (which corresponds
regularly to C. C. -t^Ac) and Kato -tUc (e. g., Hupa tcin-ne-tiiW " she
goes to bed," Kato na-nHn-tilc " lie down again "). This Ath. *-t'ec is
characteristic in Pacific Athabaskan of " indefinite " ^ forms as contrasted
with " definite " Ath. *-t^in, *-t^er}, *-t'i (Hupa -ten, -tin; Kato -tin;
Carrier -thi; Montagnais -Vi; Navaho -tqi); in most Athabaskan dialects
*-i'e or *-t^el is used in " indefinite " forms (Hupa imperative -te; Mon-
tagnais eventual -Ve; Navaho present -tqe, future -tqel). " Indefinite "
-c forms are in general apparently characteristic of Pacific Athabaskan.
C. C. -t^Ac is phonetically practically identical with Carrier -thces, but
not morphologically. Carrier " proximate future " -thces is reduced
from present -thes; in other words -5 (Ath. -c) is here found in all forms,
" definite " and " indefinite." This plural (and dual) stem Ath. *-t^ec,
*-t'ej, *-t^ec (cf. Hupa -tetc; Montagnais dual -/'ez; Navaho -iec; Jicarilla
Apache -kec) contains dualic -c (cf. Ath. -'ac " two go "; Navaho -'esh,
-'ezh, -tsh " to act upon two animals "). We now see clearly that Father
Morice was misled by a phonetic convergence of morphologically distinct,
though genetically related, forms.
II. " Unless I am very much mistaken, what he adduces as the equi-
valent of ' I am seen, you are seen, he is seen,' etc., really means simply:
' people see me ' (French: on me volt), ' people see thee,' etc." (p. 563>
1. 22). Father Morice is, in my opinion, quite right. In fact he merely
repeats what I had already pointed out (p. 332): " tslAl-, which appears
in these forms, probably contains third modal -I- preceded by deictic
^5!- implying indefiniteness of logical subject: ' mann sieht mich.'
Surely Father Morice knows that German man sieht mich is identical in
force with French on me voit.
I take this opportunity of modifying my analysis of C. C. ts\A-. I do
not now think that it is comparable to C. C. deictic tc\-, which denotes
lack or indefiniteness of object, not subject, of transitive verbs; this tc\-
goes back to Ath. H!(e)-, which is preserved as such in Hupa. C. C.
ts\- (Ath. *tc\-) of such forms as nes-ts\Al-'i " one sees me, I am seen,"
' I use "definite" and "indefinite" in Goddard's sense. "Definite" tenses are
present definite and past definite, all others are "indefinite" (including present in-
definite, imperative, eventual, and other forms).
Six: Athabaskan and Na-Dene Languages 103
773
however, is clearly subjective and impersonal in force and very probably
corresponds to Carrier indefinite ts{oe-), see 9 b above. Goddard's
Chipewyan ts'- is probably identical with this Chasta Costa and Carrier
impersonal prefix (e. g., ts'e-Lu " he was caught," i. e., " one caught
him "); this means that Goddard's comparison of Chipewyan ts'- with
Hupa ^(^!)- is incorrect (see p. 133 of his " Analysis of Cold Lake Dialect,
Chipewyan "), and indeed we should in that case, as in Kato and Chasta
Costa, have expected tc\-. I do not know how to reconcile with these
subjective impersonal forms (Ath. */c!-> Kato tc\-, ts\-, s'-; C. C. ts!-;
Carrier ^5!-; Chipewyan ts\-) Petitot's Montagnais, Hare, and Loucheux
indefinite tse- forms. Could he have throughout misinterpreted is\e-
(in his orthography Use-) as tse-? E. Sapir
Editorial Note
Originally published in American Anthropologist 17, 765-773 (1915).
Reprinted by permission of the American Anthropological Association.
This reply to Morice's criticism of Sapir's Chasta Costa work (Morice 1915b)
prompted a rejoinder (Morice 1917). Both of Morice's papers, together with his
original review of Sapir's monograph, are reprinted in the appendix to this
volume.
THE NA-DENE LANGUAGES, A PRELIMINARY REPORT ^
1"^HE problem attacked in this paper is that of the genetic
relationship of Athabaskan, Haida, and Tlingit. Important
morphological, to a less extent also lexical, resemblances
between Haida and Tlingit have long been pointed out by Boas and
Swanton, resemblances which have led them to assume, though
rather hesitatingly, genetic relationship between these languages.
Boas has also somewhat vaguely hinted at fundamental resem-
blances in structure between Athabaskan and Haida-Tlingit, but
no concrete evidence has been given on this point. A full presen-
tation of the comparative lexical, phonological, and morphological
evidence that serves to show, beyond all reasonable doubt, that
Athabaskan, Haida, and Tlingit are indeed but divergent represent-
atives of a common prototype is given in an extensive paper on
"The Na-dene Languages" now in course of preparation as a
memoir of the Anthropological Series of the Geological Survey of
Canada. The present sketch, prepared at the request of Dr P.
E. Goddard, is merely a rapid abstract of some of the leading
points involved. I wish expressly to emphasize the fact that it
does not present all the evidence at my disposal. While, however,
it does not constitute the complete demonstration of my thesis, I
believe that enough is here given to remove this thesis beyond the
realm of the merely probable. The term "Na-dene," which has
been chosen to designate the hypothetical prototype of Athabaskan,
Haida, and Tlingit, will be explained in the latter part of the
paper.
I. Morphological Features
Stem and Word Form. — The most typical and doubtless his-
torically primary type of stem form found in the Na-dene languages
is the monosyllabic stem consisting of consonant plus vowel; in
1 Read in substance before the Anthropological Association at Philadelphia.
534
106 VI American Indian Languages 2
535
Haida (H.) the consonant may be replaced by a cluster of two
consonants which, in cognate words, appears contracted to a single
consonant in Athabaskan (Ath.) and Tlingit (Tl.). Examples
are: Ath. *-tsi^ "daughter," *t'o "water," *tl!o "grass," *-k'e
"foot," *-ne "to speak," *-ya "to stand (plur. subj.)," *-'a "to
find," *-ya "for," *-na "around," *na- "again;" H. tc/W "cedar,"
q/a "harpoon," st/a "foot," Iga "rock," t'a "to eat," q!a "to
sleep," xa "to follow," sa "above," gu "at," q/o- "by means of
the teeth;" Tl. t^a "stone," nu "fort," xa "enemy," ha "to dig,"
g'a "to say," ci "to hunt for," t!a "behind," k^a "on," dJL- "quick-
ly." Many, perhaps all, elements consisting of a single con-
sonant (or cluster of two consonants) are phonetically reduced
owing to the loss of a vowel; e.g., Ath. *-n, *-77 "person" <.*-ne;
H. stU- "with the fingers" <stl!a "hand;" Tl. / "to" <de.
In all Na-dene languages, however, a large number of stems is
found consisting of consonant plus vowel plus consonant; e.g., Ath.
*-ts!en "bone," *-tIas "to cut;" H. k'un "point," sgol "to hide;"
Tl. dis "moon," tsi'n "to be strong." In a very large number of
cases there is clear internal evidence to show that the final consonant
is an old suffixed element whose original meaning has doubtless
generally been lost. Examples of such "petrified" suffixes are:
Kato hts "clay," Navaho h'c "dirt, ground," cf. Navaho h'-
in compounds; Anvik /'aZ "bed" (< Ath. */V/), Hupa-f'e/c "several
lie," cf. Hupa -/'c "to lie (sing.)," past definite -t^en; H. (Masset)
s'ail "to weep," cf. s'ai-ga; H. xal- "by means of fire acting from
without," cf. xai "sunshine;" Tl. t'i'n "to see," t'i'sl "to look
for." While a considerable number of such stem finals corres-
pond in Athabaskan and Haida or Tlingit (e.g. Ath. *-del "several
go," H. dal "many persons go by land;" Ath. *-k!an "to burn,"
Tl. qia'n "fire"), numerous cases are found of stems that cor-
1 Forms given as Ath. are reconstructed on the basis of the actual forms found in
various Athabaskan dialects. The general methodology of linguistic reconstruction
and the sounds reconstructed for Athabaskan specifically are dealt with in the longer
oaper above referred to. The phonetic system employed in this paper is the one
worked out by the Phonetic Committee of the American Anthropological Association;
this report will be published in the near future.
* Indicates reconstructed forms.
^
Six: Athabaskan and Na-Dene Languages 107
536
respond according to regular phonetic law except for the final
consonant; sometimes two of the three Na-dene languages agree
as against the other; often the simple vocalic stem is found in
one or two, but extended by a final consonant in the other. Exam-
ples of these cases are : H./'a7 "year," Tl. Z'o"^ "year;" Chipewyan
dial "moss," Louch. tdek, Tl. slate (Chip. d!a- and Louch. tde-
point to Ath. *ts!a-, which points, with Tl. sla-, to Na-dene *ts!a-) ;
Ath. *-lad "end," H. tlan "end;" Ath. *-yel "night passes," H.
ga-l "night," Tl. get "to get dark;" Ath. *-ca, *-cal "to catch with
a hook," H. djd "bait," Tl. cat "to seize;" Ath. *xes "mountain,"
Tl. ca "mountain;" Ath. *-t'an "to eat," H. t'a "to eat;" Ath.
*-t!o "to shoot," Tl. t!uk "to shoot." Examples of this sort
make it fairly obvious that many of the stems with final consonants
that are yielded by a purely descriptive analysis are ultimately
reducible to vocalic stems followed by what was originally a suffixed
element. That all Na-dene stems with final consonants are of
such origin cannot be demonstrated, but it does not seem at all
improbable. The characteristic Na-dene stem may thus be
symbolized by cv, of which c and cv-c are further developments.
Reduplication is a grammatical process that is conspicuous
in Na-dene by its absence. It is found neither as a word-forming
nor purely grammatical device. The only possible widespread
Athabaskan exception that I have been able to find is the demon-
strative stem didi "this," alongside of unreduplicated di. With
this it is interesting to compare the probably reduplicated Haida
interrogative stem gu'gu-s "what?" A negative feature of this
sort is not in itself very indicative, but gains in weight when the
Na-dene languages are contrasted with the Tsimshian, Kwakiutl-
Nootka, and Salish languages to the south, in all of which redupli-
cation plays an extremely important part.
The typical Na-dene word is built up of a number of monosylla-
bic elements (in most cases of form cv), one of which is the main
stem, about which cluster a number of subsidiary etymological and
grammatical elements that may be termed prefixes and suffixes.
The various elements of a word, aside from certain ones that are
perhaps best considered as proclitic and enclitic particles, make up a
108 yi American Indian Languages 2
537
coherent enough morphological unit, but are far from welding to-
gether in a manner suggestive of such form units as we are accus-
tomed to in Indo-germanic or are found also in manyAmerican Indian
languages (e. g., Kwakiutl, Eskimo, Yana, Southern Paiute). Most
of the elements preserve a considerable share of individuality, while
many can, indeed, be shown to be identical in origin with or special-
ized forms of independent stems. Thus, an Athabaskan word like
Kato t^aya'o'nat] "let them drink" readily falls apart into four
perfectly distinct elements: the main stem -narj "to drink"
and three subsidiary elements that may be described as prefixes,
but which are far from fusing either among themselves or with this
stem into a close morphological unit; the prefixes are /'a-, an
element having reference to water (cf. independent forms Kato
t^o' "water," Montagnais ^ t^a "billow"), a demonstrative element
ya'- indicating the plurality of the implied (but not definitely ex-
pressed) third personal subject of the verb, and o'-, a hortatory or
"future imperative" modal element: " water- they-shall-drink."
The "word" Caya^o'tia-q feels decidedly like an old "sentence" of
monosyallabic constituents, t^a ya' o' na-n, the first three elements
of which have lost their absolute independence and all four of
which have settled down to a rigidly prescribed order relatively to
each other. This same type of sentence-word (we may either
think of it as very closely knit sentence or, as we are more accustom-
ed to do, as relatively loosely knit word composed of easily analyzed
elements) can be abundantly illustrated also in Haida and Tlingit.
A Haida example is tWalgu'ldayarj "(she) had put (it) (on her son)
as a blanket," consisting of a primary stem -Igii'l- "to go around,
to be wrapped about," an instrumental prefix tl- denoting activity
with the hands (identical with the verb stem tl "to touch"), a
classifying nominal prefix t!a- which defines the object of the verb
as belonging to the class of flexible things thought of as crossing or
coiled (cf. //a-da " to wear ; " an original Haida stem *^^a "blanket"
'Another name for Chipewyan. I use "Montagnais" to indicate that I am
quoting from Petitot, "Chipewyan" from Goddard. All other Athabaskan forms
are quoted from Goddard and The Franciscan Fathers; all Haida and Tlingit forms,
from Swanton.
Six: Athabaskan and Na-Dene Languages 109
538
becomes very probable on reference to Ath. *tfe "blanket"),
a causative suffix or auxiliary verb stem -da " to cause, to have as, "
and two temporal-modal elements, a perfective -y- (or -i-) and a
suffix -aj] indicating that the statement is not made on one's own
authority: hand-blanket-be wrapped about-cause-d-as experi-
enced by others. Finally, a typical Tlingit example is afforded by
q!a-o-dt-sa "(he) blew (upon the raft)." As in the Athabaskan
example given above, the main stem, sa "to blow," comes at the
end of the word ; it is preceded by three elements, an instrumental
prefix qia- "with the mouth" (identical with the noun stem q!a
"mouth"), a modal element 0- whose exact significance is unknown
(it seems to be found only in active past temporal forms with third
personal subject), and another modal or "aspect" suffix rfi- of
apparently inceptive or momentaneous force: mouth — in past time
(?) — momentaneously — blow.
One of the incidental consequences of this type of structure is
that, while the analysis of the word into its parts is in most cases
easily undertaken, a just idea of the actual value or content of the
word as a whole cannot be obtained by merely summing the values
of the analyzed elements. There is, equally in Athabaskan,
Haida, and Tlingit, a great deal of idiomatic usage involved; in
many cases all we can say is that it is customary for a certain
perfectly definite idea to be expressed by a stem of fairly wide
range of significance preceded by such and such not always evidently
applicable prefixed elements. Thus, the purely etymological
analysis of the Hupa do' ma{k)-k!ai na-si-ri-a "not-it-after over
surface of ground-continuously-thou-have in possession" conveys
as good as no notion of the actually well determined idea conveyed :
"thou didst not want to (go home)." Similarly, the Haida verb
k^wa-lgi'-sta-sga- "in a stream-large cylindrical objects-remove
from (a place)-toward an open place" really means " (olachens)
run in a stream toward the sea."
Noun and Verb. — The relation betwen noun and verb is quite
parallel in all three languages. While verbal and substantival
forms are throughout clearly kept apart (verb forms may be substan-
tivized in various ways), the radical element of a word may often
110 VI American Indian Languages 2
539
be indifferently used as predicating or denominating stem. Thus,
the Haida stem na indicates both "house" and "to dwell," go't
is used either as a noun meaning "buttocks" or an adjectival verb
"to be last." In Tlingit this elasticity of usage is apparently less
marked, though examples occur (e.g., sa "voice, name; to name,
call;" CL "song; to sing"). Denominative verbs of this sort are
particularly common in Athabaskan, e.g., Kato k!ari' "withes,"
-k/ar]' "to twist;" Chipewyan xal "club," -xal "to use a club;"
Hupa tlfo "grass," -tUo "to make baskets;" Chipewyan t'an
"ice," Kato -t^arf "to be cold;" Navaho si'l "steam," Kato -si'l
"to steam." Under these circumstances it is perfectly natural
that stems which are found used only as nouns in one of the Na-dene
languages have become specialized as verbs in another. Examples
are: H. xao "liquid," Ath. *-k^a "liquid has position;" Tl. qia'n
"fire," Ath. *-k!an "to build a fire, to burn;" Ath. *tc'el "steam,"
H. sgal "to steam;" H. das "live coals," Ath. *-das "to burn, to
singe;" Ath. *ts!ai, *ts!a- "dish," Tl. sliql "dish," H. sq.'ao "to
put in a dish."
A peculiarity of many Na-dene verb stems is that they are
limited in their range to a particular class or number of objects.
The simplest type of these is formed by verbs applying specifically
to a singular, dual, or plural subject or object; e.g., Tl. gu "to go
(one person)," at "to go (plur.);" H. q'a "to go (one person),"
dal "to go (plur.);" tHa "to kill one person," tUda "to kill several;"
Ath. *-ya "one person goes," *-del "several go;" Hupa -yen "to
stand (sing.)," -ya (plur.); -'a "one object is in position," -'e/^'
"several objects are in position." Still more characteristic are
distinctions based on the shape of the object affected; e.g., Tl. fan
"to carry a long thing," t\ "to carry a round thing;" Hupa -t'an
"to handle or move a long object," -k^os "to handle or move any-
thing that is flat and flexible;" Navaho -\c "to lead (by a rope)
a pair of animals," -los "to lead (by a rope) a single animal;"
-tW "to act upon an animate object," -djol "to act upon such
objects as hay, wool, or hair. " In Haida such verbs do not seem to
be found, but it is interesting psychologically to observe that
corresponding classifications are here expressed by another means,
Six: Athabaskan and Na-Dene Languages 111
540
namely by the use of a long series of classifying nominal prefixes;
e.g., /c/1,5- "cubic objects, such as boxes, " sq!a- "long objects, like
sticks and paddles," ga- "flat objects."
Verb Structure. — In all Na-dene languages the verb consists of
a series of elements, which may be grouped into certain classes that
have fixed position in the complex relatively to each other; the
verb stem gravitates towards the end of the complex, particularly in
Athabaskan and Tlingit. The typical Athabaskan verb may be
analyzed as consisting of: adverbial prefix (including original noun
stems, ^ local postpositions,^ petrified demonstrative stems of
chiefly objective reference,^ and certain other elements which do
not occur in other connections'*) + objective pronominal prefix
+ demonstrative element referring to subject of verb + "first
modal" element + "second modal" element^ + pronominal
subject + "third modal" element^ + verb stem + temporal-
modal sufifix + syntactic suffix (these are best considered as enclitic
particles). Any of these elements but the stem may, in a particu-
lar form, be missing; two or more of the same general type may be
exemplified in a single form. The order of elements as given above
varies slightly for different dialects.
Quite similar in its general features to the structure of Atha-
baskan verb forms is that of corresponding forms in Tlingit. The
analysis may be given as: pronominal object (best considered as
proclitic to verb form) + nominal prefix of instrumental signifi-
1 E. g., Hupa t'a- "water," sa'- "mouth."
* E. g., Hupa ye- "into," wa- "to," xa- "after."
' E. g., Hupa a- used as indefinite object with verbs of saying and doing, xa-
"same as before."
* E. g.. Hupa da- "resting on," no'- "coming to rest," na'- denoting indefinite
movement over surface, na- "again." At least some of these may be independent
verb or other stems in origin. With da- of. Ath. verb *-da "to sit (sing, subject)";
na- may be identical with *na- "two" (found, alongside of absolute *nak'e, in, e.g.,
Hupa na-diT) "twice," na-nin "two men").
'These two sets of "modal" elements are not easy to define. They are best
considered as indicating certain "aspects," i. e., as defining range of activity with
reference to such notions as inception, continuation, distribution, cessation, and
indefiniteness of object.
« These "modal" elements also are difficult to define and, like "first" and "second
modal" elements, are largely bound up in usage with idiomatic factors. Their primary
significance is to define voice, i. e., such notions as transitive, intransitive, and passive.
112 VI American Indian Languages 2
541
cance* + "first modal" prefix ^ + pronominal subject + "second
modal" prefix + "third modal" prefix' + verb stem + quasi-
temporal suffix + syntactic suffix.
Differing more widely from the Athabaskan pattern of verb
structure is that of Haida. In Haida the pronominal subject and
object are not as closely welded into the verbal framework as in
Athabaskan and Tlingit and are best considered as independent
elements of speech. . However, as they occupy definitely deter-
mined positions immediately before the verb form proper,
their structural difference from the corresponding elements of
Athabaskan and Tlingit is more apparent than real. There is
involved here merely a difference of degree of coalescence of originally
distinct elements. The Haida verbal scheme may be represented
as follows: pronominal object -+- pronominal subject + instru-
mental prefix (most of which are in origin noun and verb stems
capable of being used independently) -j- classifying nominal prefix
(several of which, perhaps all, are old noun stems) + prefixed
adverbial element '' -f- main verb stem -f- auxiliary verb stem (doubt-
less independent verb stems in origin which have become specialized
as quasi-suffixes) + adverbial element (in origin independent
noun, verb, adjective, or adverb stems)* + locative suffix + tem-
poral-modal suffix.
This analysis of the Haida verb is not complete. It should
• E. g., qfa- "mouth," lu- "nose, point."
* These elements do not form a well-defined class. They embrace such notions
as causation, aspect, voice, tense, and indefiniteness of subject. They correspond, in
Athabaskan, partly to certain adverbial prefixes, partly perhaps to "first modal"
elements.
'These two sets of "modal" prefixes seem primarily to define various aspects
(perfective, progressive, completive, inceptive, repetitive, momentaneous, transi-
tional). They correspond to Athabaskan "first modal" and more particularly to
"second modjil" elements. As far as known, Athabaskan "third modal" elements
find no counterpart in Tlingit.
♦ These elements, of which Swan ton lists four, are termed by him "stems in initial
position." It does not seem to me that there is any real necessity for the setting up
of this class. Two of the elements are best regarded as nominal classifiers, one as an
instrumental prefix, the other as a verb stem regularly compounded with other stems
(see below).
' Swanton classifies these into three groups of "stems in terminal position," but
this sub-classification, even if justified, is of no particular consequence here.
Six: Athabaskan and Na-Dene Languages 113
542
be observed, first of all, that the order of elements fluctuates in
accordance with their logical relation to each other; thus, locative
suffixes directly follow the main stem and are followed by auxiliary
verb stems, if the locative element does not logically apply to the
latter (e. g., q'a-dl-da " to take aboard," literally " to go-into
canoe-cause"). Secondly, two or more independent verb stems
may combine into a compound verb which is held together by the
preposed pronoun (or pronouns) and the suffixed temporal-modal
suffix (or suffixes) at the end of the complex. Each member of the
compound may be itself attended by derivative prefixes or suffixes
(including even certain temporal-modal elements, like continuative
-gar]-). If we assume, as internal Haida evidence makes more than
probable, that all auxiliary verb stems and suffixed adverbial
elements are nothing but compounded originally independent
stems, we may reduce the above analysis of verb forms to: pronom-
inal object + pronominal subject + I. -f- II. + • • • -f- temporal-
modal element, in which I., II., • • • stand for complexes of type:
instrumental prefix -j- classifying nominal prefix + verb stem +
locative sufifix + continuative suffix).
Naturally, in any given verb form only a comparatively small
number of theoretically possible positions are filled. I. generally
contains the predominant stem of the whole verb form. Haida
verb composition in its present form is doubtless largely a special-
ized development, though probably based on Na-dene processes.
For this reason the typical Haida verb form in its older form must
be defined, eliminating II., . . . , as: pronominal object -\- pronominal
subject + instrumental prefix + classifying nominal prefix +
verb stem -+- locative suffix -f temporal-modal element. This
scheme, despite its peculiar features, more nearly resembles the
Athabaskan and Tlingit schemes than the one first given. ^
Comparing the three verbal analyses given, we find that the
Na-dene languages have several important traits of verb morpho-
logy in common. These are:
• I do not, of course, mean to imply that all instrumental and classifying nominal
prefixes are older, as verbal elements, than all "stems of terminal position." Analogy
always operates to feed a type already in existence.
114 VI American Indian Languages 2
543
1. Noun stems are included as prefixes in the verb complex.
They are partly of instrumental (or local) significance, partly,
more particularly in Haida, general classifiers of subject or object.
2. Both pronominal subject and object elements regularly
precede the verb stem. Of these, the object comes first in the comp-
plex. In Haida the degree of coalescence of pronominal elements
with the verb complex is much less than in Athabaskan and Tlin-
git. In these languages the subjective pronominal element is an
integral part of the verb-form, being often separated from the
objective element by an adverbial prefix.
3. Local affixes are found in both Haida and Athabaskan,
though they are suffixed in the former, prefixed in the latter. At
least some of the Athabaskan local prefixes are postpositions in
origin; these, as regards their position after pronominal objects,
offer striking analogies with corresponding elements in Haida and
Tlingit, as we shall see later.
4. Athabaskan and Tlingit possess a large number of prefixed
"modal" elements, which define adverbial notions, to a less extent
temporal ideas, but primarily aspects. They are divisible into
several position-classes, according to whether they precede or
follow pronominal object and subject. These elements are in some
respects the most characteristic of Na-dene morphology, though
their presence is hardly traceable in Haida.
5. The verb stem is a generally monosyllabic element clearly
marked off from the rest of the verb complex. It is nearly always
preceded by a number of originally independent modifying elements.
In Athabaskan it undergoes internal phonetic and morphologic
changes as it passes from one tense (present and past, definite and
indefinite) to another. Such changes have not been indicated by
Swanton for either Haida or Tlingit. Dr Boas, however, on the
basis of material recently secured from a Chilcat Indian, informs me
that internal stem changes for tense, analogous to those found in
Athabaskan, are characteristic also of Tlingit.
6. A series of temporal-modal elements is found suffixed to the
verb stem. Some of these are firmly united with the verb stem
(e. g., continuative -/ in Hupa tc!u-wi-l-Ce-l "he was bringing;"
AM. ANTH., N. S., I7 — 35
Six: Athabaskan and Na-Dene Languages 115
544
usitative -tc in Tlingit u-q^ox-tc "he kept coming in;" perfective
-y- in Waida. su' -da-y-agam "had said"), others are ni.ore in the
nature of enclitic particles (e. g., emphatic -ht in Hupa do' a-du-
win-ne-he "don't say that!" imperative -de in Tlingit na-at-de
"(for firewood) go!" Haida hortatory -dja-q in I gLti-gat-dja-q "let
me adorn [you]!").
7. Still more loosely suffixed, in most cases, to the verb form is
a series of syntactic particles, largely used to subordinate it in
various ways. In part these elements, as we shall see, are post-
positions in origin.
8. While compounding of verb stems is most luxuriantly de-
veloped in Haida, indications are not lacking of the presence of
the process also in Tlingit and Athabaskan. Thus, the Tlingit verb
stem CL "to desire" may be prefixed to another verb stem to form
its desiderative (e. g., CL-t^an "to desire to pick berries"); cf. such
Haida compounds as gi'da-yu'an-SLtj-ga " to wish to give much food,"
in which the auxiliary verb stem -sir} "to wish," however, is suffixed
to the main stem. Nearer the Haida type is the class of Athabaskan
verbs in *-tsfe (cf. independent verb stem *-ts!e "to hear"), indicat-
ing that the action of the main verb is heard or otherwise perceived
(e. g., Hupa -Cuw-tsle "to hear one splitting logs," Chipewyan
-ni-d'e "to hear one say"). Differing morphologically from, yet
psychologically similar to, the Haida type of verb composition is
the use in Tlingit and Athabaskan of two independent verb forms
to form a logical unit; e. g., Tl. gax gax-yi'-sa-tH "cry you-will-be,"
i. e., "you will cry" (cf. such Haida compounds as t^a'-ga "to
eat-be," i. e., "to eat"), and Hupa. tc/i[n]-nL- rj-y a wt-n-tfe "he-came
he-was- (thus)," i. e., "he always came."
9. A highly important feature found in all Na-dene languages is
the use of subjective or objective pronominal elements, according
to the nature of the verb, to indicate its logical subject. This
feature will be referred to again in connexion with the pronouns.
More important than any of the specific features we have named
is the similar manner in which the various elements going to make
up a verb-complex are linked together. The resulting structure
may be termed a sentence in miniature, not only psychologically,
but, as is much less often the case in America, also morphologically.
1 16 VI American Indian Languages 2
545
Noun Structure. — There are a large number of monosyllabic
noun stems, which may be used absolutely, in all three Na-dene
languages. These are both of type cv and cvc.
They often enter into composition, the qualifying noun regularly
preceding (e. g., Chipewyan tUo-bec "grass-knife, mowing machine;"
Tl. xa't-sla'x"* "root-hat"). In Haida, however, simple composition
of noun stems does not seem to be often found (in such compounds
as Masset i'tHadas 'ai "chief's blood," possessive -i or -a seems to
have contracted with the final vowel of the second noun). Another
type of composition which is particularly characteristic of Na-dene
is the suffixing of a possessive element (Ath. *-e, *-ye, *-ye; Tl. -i,
-71, -w, -wu; H. -ga, -l, -a) to the second member of the compound;
this element indicates that the second noun governs the first, in
other words that the first is genitively related to it (e. g., Chipewyan
k^a-t'uwe "goose-lake;" Tl. slate a'n-i "moss town;" H. xo'ya
tlu'-ga "raven's canoe, beans"). It is important to observe that
finite verb forms may be nominalized or turned into relative
clauses in Athabaskan and Tlingit by the suffixing of this possessive
(better relative) element (its different forms in Athabaskan and
Tlingit are due to phonetic factors) ; cf. Montagnais td!ai gay-e
"plate which-is-white" with Tl. at-ci'-yt "those who can sing."
Possessive pronouns are prefixed to nouns; they are identical
in form with the objective forms used with verbs. Most nouns but
terms of relationship and, generally speaking, those indicating parts
of the body are in Athabaskan followed by the relative particle
discussed above (e. g., Chipewyan be-di' "his head" but be-'ay-e
"his snowshoes"). Similarly in Tlingit the relative suffix is regu-
larly used with a possessive prefix except in the case of terms of
relationship and, though not consistently, body part nouns (e. g.,
Tl. du-tla "his mother" but du-tc^u'n-i "his dream"). Haida
possessive constructions are on the whole analogous to those of
the other two Na-dene languages, the relative suffixes -ga and -i,
-a being used in some cases, omitted in others (e. g.,/ dja'-ga "his
wife," Masset / tla'l "her husband," Masset / k!u'g-i "its heart").
Derivative affixes (aside from nominalized verb forms) are quite
sparingly used in Na-dene. A diminutive suffix is found in both
Six: Athabaskan and Na-Dene Languages 117
546
Athabaskan and Tlingit (e. g., Hupa (/je/o'-Zc "small storage-basket,"
Tl. a'-^/" "little lake"). Noun plurals (aside from Tl. collective
-q!) are not often formed, though special plural forms for terms
denoting human beings are found here and there (e. g., Kato
skH'-k'^ "boys," Tl. du-k^a'ni-yen "his brothers-in-law," H. kfwai-
ga-lar) "elder brothers").
Pronouns. — There are two classes of pronouns in Na-dene.
Subjective pronouns are used as the subjects of active verbs (in
Athabaskan of most verbs generally) ; objective pronouns as the
objects of transitive verbs, subjects of neuter verbs (which may
best be interpreted as objects of impersonal verbs), and possessive
prefixes with nouns. While the two series are distinct as such, not
all the respective forms are etymologically unrelated; in Tlingit
there are one or two minor differences between the objective and
possessive series. In both Athabaskan and Tlingit, as we have seen,
the subjective and objective pronominal elements are integral parts
of the verb complex, the possessive elements of the noun complex;
in Haida the pronominal elements may be considered as inde-
pendent words or, at least in part, as proclitic elements. A third
series of pronouns is found in Athabaskan and Tlingit; these are
independent denominative terms, which, however, have no influence
on the form of the verb or noun.
The employment of objective pronouns with verbs denoting
states has been rather obscured in Athabaskan by the spread of
subjective forms, but there are enough cases to make it clear that
the impersonal verbs with objective pronominal elements char-
acteristic of Haida and Tlingit were at one time better represented
also in Athabaskan.
The contrast between verb forms with subjective and objective
pronominal subject is exemplified, e. g., by Tl. gu-x-t^u-si't "we
will cook it" (with subjective pronominal element t^u- "we") and
ha-k"-gu-wa-t!a "we will be warm" (with objective pronominal
element ha- "us": "it will be warm to us"); by H. I q^a-t!al-gan
''I got off" (with subjective pronominal element I "I") and di'
skfistl-djdi'-ga " I am truly full" (with objective pronominal element
di' "me": "it is truly full to me"); and by Hupa o-r]-xai "thou
118 VI American Indian Languages 2
547
art buying" (with subjective pronominal element -77- "thou")
and ni[k]-^!-o-u<ar} "go to sleep!" (with objective pronominal
element « I- "thee": "let it sleep to thee!"). That the verb forms
with objective pronominal subject are indeed impersonals with pro-
nominal object is made clear by comparing them with such transi-
tive forms as Tl. ha-u-si-ne'x "they have cured us;" H. di' dala-q
tl-gaxa-gd-ga "you tire me with your handling;" and Hupa yafn]-
ni-l-t\r] "he picks thee up." The possessive use of objective pro-
nominal elements is illustrated by Tl. ha'-q'aha'gu "our eggs;"
H. di' gi'da "my daughter"; Hupa nL[t]-t''ai "thy paternal uncle."
Postpositions. — Very characteristic of Haida, Tlingit, and Atha-
baskan is a set of local and relational elements which regularly
follow the noun or pronoun that limits them (e. g., H. st'al-ai st'a
"the cliff from;" Tl. xa'na-de " evening-towards ; " Hupa nLfi-tcht]
"ground-toward"). These postpositions offer remarkable morpho-
logical and etymological analogies in the three languages. No less
than about thirty-five Athabaskan postpositions and local verb
prefixes (which, as w^e shall see, are in all probability postpositions
in origin) can be more or less confidently stated to be cognate with
corresponding Haida, Tlingit, or Haida-Tlingit elements. Out of
twenty-five Hupa postpositions listed by Goddard, at least fifteen
seem to be related to similar elements in Tlingit, Haida, or both.
These facts show that the postpositional elements of Na-dene reach
back, aside from certain later dialectic developments, into the
earliest period of Na-dene linguistic history that it is impossible to
arrive at by comparative evidence.
In some cases it is possible to show that postpositions are
nouns in origin, the complex of noun -+- postposition forming
originally a compound noun. Thus, Hupa and Kato -lai, Chipewy-
an -laye "on top of" is simply the noun stem for "end, top" com-
pounded with the preceding element; Chipewyan -bQ. "around,"
the similarly employed stem for "edge" (Ath. *mar], *man). With
Tlingit t!a "behind," Haida t!al "behind, back of," and Tl. k'a
"on" compare respectively Chipewyan -t!a-ze "back" (body-part),
Kato -/.''a "tail," and Navaho -^'a "surface." Hence it is intelligible
that the same noun stem may in some cases have developed inde-
Six: Athabaskan and Na-Dene Languages 119
548
pendently into distinct postpositions in different Na-dene languages;
e. g., Ath. *man, Tl. wan "edge" means "around" as Athabaskan
postposition, "close to" as corresponding Tlingit element. The
nominal origin of postpositions is further made very probable by
the fact that they are frequently preceded by possessive pronouns:
Hupa mi-ye "under it" (originally perhaps "its bottom") like
mL[n]-nLr] "its face;" Tl. hasdu-q^a'nax "after them" (originally
perhaps "their following") like hasdu-cayi'tia-yi "their anchor;"
H. di' ga "to me" (originally perhaps "my vicinity") like di'
go'ri-ga "my father." Whether we shall ever be able actually to
demonstrate the nominal origin of all Na-dene postpositions is
doubtful, but there can be little doubt of the correctness of this view.
Postpositions often occur compounded among themselves. In
some cases the analysis is evident (e. g., Chasta Costa -me'-qU
"inside of" < "therein-at;" Tl. -k'a-q! "on" < "on-at;" H.
gei-st^a "out of" < "in-from"); in others the two (or more) ele-
ments have grown into a unit that can be analyzed only by com-
parative evidence (e. g., Tl. t^a'yi "under" contains Tl. yi' "down
in," but /'o* does not occur alone; comparison with Tl. t^a'-k "in
the middle of," t\-n "with," and particularly Ath. *-/'a "among,"
shows fa'yi to have originally meant "down among").
Postpositions combine with verb forms in two ways, as local or
relational prefixes and as syntactic suffixes. We have already
indicated that several of the local prefixes of Athabaskan are merely
postpositions in origin that have become somewhat firmly attached
to the verb complex. Thus, Hupa xa- in xa-n-t^e "look for it!"
is evidently etymologically identical with -xa in no'-xa "after
us." In some cases the postposition comes after elements which
can hardly be disconnected from the verb form, e. g., a-ya-l-tc! L[t\-
-dtn-nt "he told them" (here -I "with" appears immersed in the
verb, which demands the indefinite objective a- "it" as constant
prefix; morphologically parallel is Tl. da- "to" in verbs of saying,
e. g., ye da-ya-dii-q^a "thus to-him-spoke" like Ath. a-l-, which is
doubtless identical in origin with postpositive -de "to," Masset da
"to," Ath. *-d, *-de, * de-n "to, at"). These facts are not sur-
prising when we bear in mind that the indirect object, nominal or
120 VI American Indian Languages 2
549
pronominal, precedes the verb and is followed by its postposition
(e. g., Hupa xo'-xa t^(-r]-\n-t'e "him-for thou-wilt-look;" Tl. a-da
a-o-H-t'aq! "it-around they-drifted;" H. la-gei la sk' U-nana-q-xida-
i-as "it-into he began-to-chop-up"). It needs only the removal
of the object (which then remains understood) from the postposition
to bring the latter into closer touch with the verb. In the last
Hupa example the removal of the expressed object {xo'-) leads to a
form like the xa-n-tU first quoted. In Tlingit this use of the
postposition as verb prefix with unexpressed object does not seem
to be common, but examples abound in Haida, e. g., gei la q^a'-tdi-
gan "into he went-in." In the last example gei is morphologically,
as well as etymologically, parallel to Hupa verb prefix yt- "into"
(cf. Chipewyan postposition -ye "in"). Here again we observe that
Haida has allowed distinct elements to coalesce to a less extent
than Athabaskan. In view of the tendency in Athabaskan for
postpositions to become specialized as verb prefixes, it is not sur-
prising if we find cases of such prefixes, no longer used as post-
positions, corresponding etymologically to Haida postpositive
elements. Such an example is Ath. *tse-, *tsi-r}- "away from, out
of;" H. st'a "from, after" (for H. st': Ath. ts, cf. also H. st'a-
classifier for ring-shaped objects: Navaho tsa-bg.'s "ceremonial
hoop," yo-s-tsa "ring").
A verb form as such is sometimes conceived of as nominalized
and is followed by a postposition which serves to subordinate it.
Thus, in a Chipewyan form like hi-l-tc!t-t!a "because he was angry,"
postpositive -t!a "with, on account of" (cf. he-tla "with it")
nominalizes and subordinates hi-l-tde "he was angry" ("he-was-
angry because-of," "because of his being angry"). Such syntactical
developments have taken place independently in the Na-dene lan-
guages, to some extent even in the various Athabaskan dialects.
This is indicated, among other things, by the fact that even where
two Na-dene languages have employed the same postposition for
syntactic purposes, the use to which it is put is different (e. g., Ath.
*-de "if" and Tl. -/ "in order to" both go back to Na-dene post-
positive *-da "to, at"). Tl. -ya and -«, which make subordinate
clauses out of verb forms (e. g., has a-ga-ca'-n "when they marry;"
Six: Athabaskan and Na-Dene Languages 121
550
a-t^e'-x-ya "when she slept"), are doubtless identical with post-
positive -ya "in the neighbourhood of" and -n "with, at."^ In
Haida temporal clauses are formed by nominalizing verb forms by
means of suffixed demonstrative {g)ai "the," these being then
followed by postpositive dliC. Subordinate clauses formed by
means of postpositions without preceding {g)ai also occur (e. g.,
Masset / k!ota'l-an sd-e-t "after he died," literally "he died place-
the-to"). The degree of coalescence of postposition and verb
is again much less in Haida than in Tlingit and Athabaskan.
Summary. — It has become evident that the morphologies of
Haida, Tlingit, and Athabaskan present numerous and significant
points of comparison. Despite not unimportant differences of
detail, the same fundamental characteristics are illustrated in all
three. In not a few cases elements (or even processes) which are
thoroughly alive in one of the languages linger on merely as sur-
vivals in another (e. g., -xa, freely used in Haida as distributive
suffix with numerals, postpositions, and nouns, lingers on in Tlingit
as compounded -na-x after numerals and as sporadic noun plural
*-k\ *-k'e, *-k'ai in Athabaskan).
Considerable specialization must, of course, be allowed for.
Peculiar to Haida are the development of a large class of nominal
classifiers, a great exuberance of composition of verb stems, the
development of a set of local suffixes in the verb, and greater
looseness in the treatment of pronominal elements and postposi-
tions. The synthetic tendency has gone farthest in Athabaskan, in
which, e. g., pronominal subject and "modal" element often unite
inextricably (there are, however, analogies to this in Tlingit).
Tlingit, on the whole, seems to have the smallest number of purely
distinctive morphologic features. It shares with Athabaskan a
lesser degree of independence of pronominal elements, a great
development of verb prefixes denoting aspects and, it would seem,
the employment of internal stem changes for tense differences.
As in Haida, the distinction between active verbs with subjective
pronouns and static verbs with objective pronouns is better pre-
served than in Athabaskan.
> My interpretation of Tl. -/, -ya, and -n as syntactically specialized postpositions
diffeis from Swanton's, at least as far as expressed in his grammatical sketch.
122
VI American Indian Languages 2
551
II. Comparative Vocabulary
The lexical evidence bearing on the genetic relationship of
Athabaskan with Haida and Tlingit comprises, at the moment of
writing, over three hundred distinct Athabaskan stems and gram-
matical elements which can be, with greater or less probability,
assigned to the reconstructed Na-dene language. Only a selection,
comprising less than one third, of this lexical material is here
presented. The arrangement is alphabetical, from the point of
view of Athabaskan.
Athabaskan
I. a- demonstrative stem
2.
Haida
a- dit.
-ade "elder sister"
3. -ca-rj obligatory future
4. -ca, -cal "to catch with a
hook"
5. -d, -de "at, to"
6. da "what?"
7. -do, -da/ "to go, to travel"
8. -da-f) "to drink"
9. -das "to burn"
10. -de/ "several go"
11. del "crane"
12. di "this"
13. -dja hortatory
14. djarj "mud"
15- -ga?7 "to be mouldy"
16. -go " toward"
17. -gid, -yid "to dive"
18. -7a "for," -7an "to"
19. -ya "to go"
20. -ye, xe "grease"
21. -yed "to run"
22. -yel "to be dark, night
passes"
-sa-rj infallible future
djd "bait"
-da "to"
-dal "to move along"
da'dj, {das) "live coals"
dal "several go by
land"
di'la dit.
dei "just that way"
-dja-T) dit.
tc^a'n dit.
gu'na "decayed"
gtia, gui dit.
gi' dit.
ga "to," gan "for"
-ga "to go in order to"
ga't dit.
ga'l " night"
Tlingit
a- dit.
a't "father's sister,
father's sister's
daughter"
cat "to seize"
-/, -dt "to"
da' "what?"
dana "to drink"
at "several go"
du'l dit.
de "now"
ga "for"
-ya "to go to"
e'x "grease"
get "to get dark'
Six: Athabaskan and Na-Dene Languages
123
552
Athabaskan
23. -ye, -yel "to kill, to fight"
24. -ywo "tooth"
25. hai "that"
26. he-, xe- "they"
27. X0-, he- "he, him"
28. -^'a "liquid has position"
29. -k'an, -k'a "to fish with a
net"
30. -k^a "on"
31. -^'e personal noun plural
32. k'ene "friend"
33. kla "arrow"
34. kleri "withes"
35. -klan "to burn"
36. -k!e "on"
37. -k'os, k'es "to tie"
38. 4 "with"
39. 1-, la negative
40. la "one"
41. -la "to jump"
42. -lad "end"
43. lo', lok' "fish"
44. -///a "butt; behind"
45. me- "he, it"
46. man "edge"
47. mes "cheek"
48. -n, -77 local postposition
49. -na "to die"
50. -nan "to drink"
51. -ne, -n "person, people"
52. -neg, -leg "to relate"
53. -ne "to play"
54. -ni "to touch, to do with
one's hands"
55. no "place of retreat, is-
land"
56. -onay{e) "older brother"
57. 5- durative verb prefix
Haida
gail "to fight"
hao " that "
xao "liquid"
xao "to fish"
Tlingit
-xa distributive suffix
q!a' "harpoon'
qian "grass"
k^u "to tie"
al "with"
ila'- "the first'
tla' "to dive"
tlan "end"
-dlga "after"
wa- "that"
-n, -rj general post-
position
ntal, nil dit.
na "to live; house"
nar] dit.
ux "tooth"
he " this "
has "they"
hu "he"
-k'a "on"
-na-x distributive num-
eral suffix
xo'n "friend"
q!a "point"
q!a'n "fire"
-q! "at"
I negative
tie'- "one"
tUu'k! "cohoes"
we "that"
wan "edge"
wac "cheek"
-n "with," also local
postposition
na "to die"
na "people"
nik "to tell"
ni "to put"
mi "fort"
hu7ix "man's older
brother"
s- modal prefix
124
VI American Indian Languages 2
Athabaskan
58. -sen "to hide"
59. sil "steam," -sil "to
steam"
60. -/'a "among"
61. /'a "wave" /'a- verb prefix
referring to water
62. -t'an "to eat"
63. -t'e "to look for," -t^an "to
look"
64. -/'e "to be cold," /'en "ice"
65. -t'e "to lie"
66. i^ez "night"
67. -tla "because of"
68. tla "feather," -t!a "to fly"
69. -tla "tail," tIa-Tj "back-
wards"
70. -ties "to step"
71. -tlo "to shoot"
72. -tlod "to rub"
73. tsa "ring-like object"
74. tslai "dish"
75. -tslen "bone"
76. tsli "again"
77. -/c't "grandfather"
78. -/c'ot; " mother-in-law "
79. tela "hat"
80. tcio "fir, spruce"
81. -ifa "for"
82. xa- "up, out of"
83. xfl' "goose"
84. xa-T) "quickly"
85. -xan, -yan "to grow up,"
xan "old age"
86. xin "song"
87. -ya "to stand ' (plur.)
88. -yan "to eat"
Haida
sil "to steam"
t^a-oan "alongside of"
t^a'ria "sea-water"
fa dit.
/'a/ "cold"
/'at, /'t "to lie"
-tla dit.
tiagun "feather," tlao-
" feather-like object '
-tial "back of "
553
Tlingit
Sin "to hide"
si't "to cook"
-Ca'-k^ "in the middle
of"
/'a'g/ "to chew"
tiln "to see"
t^a'dj "cold"
/'at "to lie," /'a "to
sleep"
/'c*/ "night"
tia-wu "feather"
•tla "behind"
tla "to step"
st^a- "ring-shaped ob-
ject"
sqlao "to put in a
dish"
tc\n dit.
djo'n dit.
tcju "cedar"
k'wagi "above"
xaha "mallard"
xao- "to do a thing
quickly"
ga "to stand"
tluk "to shoot'
tlus "to rub"
shql "dish"
sla'q "bone"
tslu "again"
tc^a'n dit.
sja'x" "hat"
-xa-n "to"
k^e "upward"
can "old person"
ci' "song"
ya'n " to eat"
Six: Athabaskan and Na-Dene Languages
125
554
Athabaskan
Haida
Tlingit
89.
ye- "that," y- "he"
gai
'this, that, the"
ya "this"
90.
-ye "at the foot of, under"
-yi' "down in"
91-
92.
-ye personal noun plural
ye "supernatural being,"
-yen "to practice sha-
manism"
-ye-n plural of terms
of relationship
ye'k "supernatural
helper"
93.
-ye "in"
-gei
"into"
•ge' "inside of," -yi-k
"inside"
94-
-ye suffix making relative
clause
-7i suffix making rela-
tive clause
95-
yo "that yonder"
gu'
there"
yu "that yonder"
96.
-'a "to go"
q'a
'to walk"
97-
-a "to tell, to sing"
g'a "to say"
98.
-'in "to see, to look"
q\v
dit.
gen dit.
III. Phonology
The phonetic systems of Athabaskan, Haida, and Tlingit,
despite a good many differences of detail, present important points
of similarity. Three types of stops are found in each : intermediate
(or sonant), aspirated surd, and glottalized (fortis). Sibilants and
sibilant affricatives, k-spirants, and laterals are well developed.
A remarkable phonetic feature held in common by the three Na-
dene languages is the paucity of labials; b, p\ and p! were clearly
not found in Na-dene (b and />' are rare Haida sounds), m existed
only doubtfully (Ath. m, whence b in certain dialects, is not equiva-
lent to Haida m, but to Haida-Tlingit w), while w was certainly
found. Athabaskan has lost the old velar series of stops as such,
while Haida and Tlingit have preserved them; on the other hand,
the Na-dene anterior palatals, best preserved in Haida, have been
lost as such in Tlingit.
In the more elaborate paper on the Na-dene languages in course
of preparation, the historical relationship of the Athabaskan
sounds to their Na-dene prototypes and Haida and Tlingit corre-
spondents is systematically worked out on the basis of all the
evidence available. Here it will suffice to point out some of the
more important correspondences, referring to the numbered entries
of the comparative vocabulary for illustrative examples.
126 VI American Indian Languages 2
555
Stopped Consonants
1. Ath. d: H. d, -t: Tl. d, -t (nos. 2, 5, 6, 7, 8, 9, 10, 11, 12, 21)
2. Ath. /': H. /": Tl. /' (nos. 60, 61, 62, 63, 64, 65, 66)
3. Ath. //: H. //: Tl. // (nos. 67, 68, 69, 70, 71, 72)
4. Ath. g: H. g: Tl. g, -k (nos. 15, 16, 52)
5. Ath. /fe": Tl. k' (no. 30)
6. Ath. /fe": H. x: Tl. y (nos. 28, 29, 31, 32)
7. Ath. x: H. jfe'(M^): Tl. k' (no. 82)
8. Ath. k': H. fe' (no. 37)
9. Ath. 3/,(^): H.^ (nos. 17, 87)
10. Ath. ife/pH. g/PTl. g/ (nos. 33, 34. 35, 36)
11. Ath. ': H. g': Tl. g' (nos. 96, 97, 98)
12. Ath. 7: H. g: Tl. g (nos. 18, 21, 22, 23)
13. Ath. y: H. g: Tl. g, 7 (no. 93)
Continuants
14. Ath. m: H. w: Tl. w (nos. 45, 46, 47)
15. Ath. n, -7)-. H. n, -77: Tl. n (nos. 3, 8, 14, 15, 32, 34, 35, 46, 48, 49, 50, 51
52, 53- 54. 55- 56, 58, 63, 78, 85, 88, 98)
16. Ath. /, I: H. /, I: Tl. I (nos. 4, 7, 10, 11, 22, 23, 38, 39, 59)
17. Ath. s: H. s, dj: Tl. 5 (nos. 9, 57, 58, 59)
18. Ath. c (j): H. 5, dj: Tl. c (nos. 3, 4, 47)
19. Ath. x{> c in most dialects): Tl. c (nos. 85, 86)
20. Ath. y: H. g: Tl. y (nos. 88, 89, 95)
21. Ath. y (before front vowel): H. g (g): Tl. 7 (nos. 90, 91, 92, 93, 94)
22. Ath. x: H. X, x: Tl. x (nos. 20, 81, 83, 84)
23. Ath. h, x: H. h: Tl. h (nos. 25, 26, 27)
24. Ath. 7: H. g: Tl. 7, -x (nos. 19, 20, 24, 56)
Affticatives
25. Ath. I: H. /^: Tl. // (nos. 40, 41, 42, 43)
26. Ath. dj: H. dj, tc' (nos. 13, 14)
27. Ath. /c': H. tc\ dj: Tl. tc' (nos. 77, 78)
28. Ath. tcl: H. /c/: Tl. s! (nos. 79, 80)
29. Ath. ts!: Tl. /i/, 5/ (nos. 75, 76)
30. Ath. tsl: H. til: Tl. tl! (e. g., Ath. *-/5.'e "penis": Tl. tllel dit.; Ath. *-tsIi
"to sit [plur. sub].]": H. tl!a-{o-) dit.)
Consonant Clusters. — The study of Na-dene sibilants and sibilant
affricatives is rather involved and presents several difficulties. The
summary given above (17, 18, 19, 26, 27, 28, 29, 30) exhibits some of
the main developments. An important group of cases is afforded by
Haida stems or elements beginning with consonant clusters whose
first element is a lateral (I, tl!, dl) or a sibilant (5). Swanton, in
Six: Athabaskan and Na-Dene Languages 127
556
commenting on these clusters, surmised that they were perhaps
due to the prefixing of an old morphological element (e. g., 1-, s-).
There is, however, no evidence whatever to support this. On the
other hand, I have at my disposal upwards of twenty such examples
which point clearly to the inference that these Haida clusters were
found in Na-dene and correspond to lateral and sibilant affricatives
in Athabaskan and Tlingit. The following relations can be estab-
lished :
A. 31. H.lg-: Ath. ///- (?)
32. H.lq'-: Ath. tU-
33. H. dig-: Ath. tU- (no. 44)
34. H. It'-: Ath. /c'-
35. H. H.'x-: Ath. tcl- (?)
36. H. tUd-: Ath. fsl- (cf. 30.)
B. 37. H. sg-: Ath. Is- (?)
38. H. sq!-: Ath. ts!-: Tl. si- (no. 74)
39. H. sg-, (sk'w-): Ath. tc'-: Tl. tc'-
40. H. sqlw-: Ath. tcl-
41. H.St'-: Ath. ts- its!-): Tl. t'-
42. H. St'-: Ath. tc'-
43. H. St!-: Ath. id-
Vowels. — The great majority of vowel correspondences is per-
fectly intelligible; a certain number of unsolved problems still
remain. In comparing Tlingit with Haida and Athabaskan. forms,
it is necessary to bear in mind that, under as yet undetermined
circumstances, Tlingit a has developed to e (e. g., Tl. xa "to eat":
a-xe'-x "he ate"; Tl. de'x "two": daxa-ducu "two plus five,
seven"; Tl. /'a "stone": /V-5.' "stones"). A more important prob-
lem is presented by Ath. e (doubtless open in quality), which is not
to be directly compared with Tl. e. It is clear, both from internal
Athabaskan evidence (e. g., Ath. *-t^e and *-i^a "to look for";
Ath. *t'e'- "in the water," *t'a- "water") and, still more, from
comparison with Haida and Tlingit, that Ath. e (which must be
assumed for the earliest Athabaskan period) has developed from
Na-dene a; less frequently Ath. e goes back to Na-dene i. Under
what phonetic circumstances, however, Na-dene a has remained as
such in Athabaskan or become e is not clear for the present. This
I believe to be one of the most important problems of Na-dene
phonology.
128 VI American Indian Languages 2
557
Some of the more important vocalic correspondences are:
44. Ath. a: H. a, a: Tl. a, a', a, {e) (nos. i, 2, 3, 4, 6, 7, 8, 9, 13, 14, 18, 19,
25. 28, 29, 30, 33, 35, 40, 41, 42, 44, 46, 49, 60, 61, 62, 67, 68, 69, 73,
74, 79, 81, 82, 83, 84, 85, 87, 88. 96, 97)
45. Ath. e (sometimes reduced to e^)'- H. a, a, «: Tl. a, a, a, (g) (nos. 5,
10, 20, 21, 22, 26, 31, 34, 45, 47, 51, 53. 64, 65, 66, 70, 75, 89, 92)
46. Ath. i: H. i, t, i, ei: Tl. t", , e e (nos. 12, 17, 54, 59, 77, 86, 98)
47. Ath. 0: H. M, 0': Tl. m, «• (nos. 16, 24, 27, 37, 43, 55, 56, 71, 72, 78, 80. 95)
Whether or not Na-dene possessed pitch accent must remain un-
decided for the present. Its presence in Tlingit and a few remarks
by Morice and Legoff as to its possible existence in Athabaskan
make this not improbable. Should this prove to be the case, some
of the phonological difficulties in Athabaskan and Tlingit vocalism
may be solved (e. g., Ath. e < Na-dene d, Ath. a < Na-dene a).
All this, however, is quite vague as yet.
IV. Conclusion
The main conclusion to be derived from the selected morpho-
logical, lexical, and phonological evidence that we have passed in
review is, I believe, obvious. Athabaskan, Haida, and Tlingit
must be considered genetically related. The correspondences are
of so intimate a character that mutual borrowing of words and
morphological features seems out of the question. It is, however,
no less obvious that each of these languages is very distinctive and
represents a highly differentiated form of the Na-dene prototype.
In no sense can Haida, Tlingit, and Athabaskan be said to form a
continuum comparable to that of the Athabaskan dialects when
these are compared among themselves. Each Na-dene language has
evidently passed through a very long period of development in
linguistic isolation from its sister languages. It would be rash, in
the present state of our knowledge, to dogmatize on the relative
conservatism of the Na-dene languages. I would venture to sug-
gest, however, that Haida has remained the most faithful to the
original sound system of Na-dene, but that, on the whole, the
original morphological features are best preserved in Tlingit.
' This weak vowel is differently colored in different dialects; e. g., Hupa i (less
frequently a). Carrier a, Navaho ».
Six: Athabaskan and Na-Dene Languages 129
558
Several facts suggest that Tlingit and Athabaskan may have had
a common Hnguistic history after Haida had become differentiated,
Imt too much should not be made of this.
The name that I have chosen for the stock, Na-dene, may be
justified by reference to no. 51 of the comparative vocabulary.
"Dene," in various dialectic forms, is a wide-spread Athabaskan
term for "person, people"; the element *-ne (*-w, *-ri) which
forms part of it is an old stem for "person, people" which, as suffix
or prefix, is frequently used in Athabaskan in that sense. It is
cognate with H. na "to dwell; house" and Tl. na "people." The
compound term "Na-dene" thus designates by means of native
stems the speakers of the three languages concerned, besides con-
tinuing the use of the old term Dene for the Athabaskan branch
of the stock.
An important ethnological consequence of our linguistic results
is that a demonstration is at last given of the northern provenience
of the Athabaskan-speaking peoples. So long as Athabaskan was
counted a separate linguistic stock, there was no conclusive k
priori reason for considering its Pacific and Southern branches as
having spread out from the northern group. Under the present
circumstances a southern drift of Athabaskan-speaking tribes can-
not seriously be doubted. The center of gravity of the Na-dene
languages is clearly in the northwest, in southern Alaska and
adjacent parts of northern British Columbia and southern Yukon
Territory. Owing to the great linguistic gulf separating Haida and
Tlingit, I should be inclined to consider the coast of southern
Alaska, the present home of the Tlingit Indians, as the most likely
region in which the Na-dene languages developed. The Athabaskan
branch of the stock undoubtedly formed a relatively undifferentiated
unit long after Haida and Tlingit had become differentiated from
each other. The Athabaskan dialects have so many distinctive
traits in common that it is perfectly evident they have had a
long history in common. They may be considered a specialized
interior offshoot, just as Haida is a specialized island offshoot.
130 Vf American Indian Languages 2
Editorial Note
Originally published in American Anthropologist 17, 534-558 (1915).
Reprinted by permission of the American Anthropological Association.
Nineteenth-century Russian linguists speculated about a possible rela-
tionship among Athabaskan, Tlingit, and Eyak, an isolate of the Copper River
area (Krauss 1964; Pinnow 1976). Sapir was almost certainly unaware of this
work, and saw data on Eyak only quite late in his career, after Birket-Smith and
de Laguna's ethnographic field work in 1933. The germ of Sapir's Na-Dene
hypothesis, as he himself notes (p. 534), lay in the speculations of Boas and
Swanton (especially Swanton 1908: 472-485) about a historical connection
between Haida and Tlingit. It was the position of Haida, rather than that of
Athabaskan or Tlingit, that was central to Sapir's view of the relationship.
Haida, with its relatively uncomplex morphology, seems an improbable con-
gener for languages so thoroughly "polysynthetic" in spirit as Athabaskan and
Tlingit, but in Sapir's view the convoluted morphosyntax of Athabaskan-
Tlingit was a late development, and the more open structure of Haida words
represented either an archaic situation or a quite different development from
the "isolating" Na-Dene proto-language. Most of Sapir's lexical comparisons
with Haida depended on the assumption of morphosyntactic changes of this
magnitude.
Sapir's Na-Dene proposal was not well received. Both Boas (1920) and God-
dard (1920) disparaged Sapir's use of a genetic hypothesis to explain
resemblances — particularly morphological similarities — that they felt could
better be explained as borrowings. The issue of Athabaskan-Tlingit-Haida
"morphological borrowing" remained alive until at least the 1950s, when Dell
Hymes in an important series of papers proposed a technique of "positional
analysis" to assess the historical value of structural resemblances of this sort
(Hymes 1955, 1956).
Research in recent decades has clearly established the genetic relationship of
Athabaskan and Eyak (Krauss 1964, 1965). A genetic relationship between
Athabaskan-Eyak and Tlingit is accepted by most scholars, but is seen as dis-
tant and problematic (Krauss 1968, 1969; Krauss and Leer 1981; Pinnow 1966).
In the recent literature, it is this grouping that is usually referred to as "Na-
Dene." Krauss has called Sapir's belief that Haida belongs in Na-Dene "an
illusion" based on "mistranscription, misanalysis, mistranslation, and/or mis-
interpretation" (1979: 841). Levine, in an influential paper (1979), has given a
detailed critique of the "spurious" evidence on which Sapir based his hypoth-
esis, tracing many of the supposed errors to Swanton's deeply flawed descrip-
tion of Haida (1911). In his recent general treatment of linguistic relationships in
the Americas, Greenberg (1987: 321-330) rebuts Levine and wholeheartedly
accepts Sapir's more inclusive version of Na-Dcne , which he sees as an intrusive
group unrelated to any other American Indian linguistic stock. Pinnow also
regards the relationship of Haida to Athabaskan-Eyak-Tlingit as likely to be
Six: Athabaskan and Na-Dene Languages 131
genetic, and has published considerable lexical evidence for the relationship
(Pinnow 1985a, b,c, 1986a, b, 1988, 1990), as well as a comprehensive history (in
German) of Na-Dene research from 1798 through 1976 (Pinnow 1976).
Sapir's ms. corrigenda on his own offprint are as follows:
Original For: Read:
p. 545, 1. 32 -ga -ga
p. 546, 1. 6 ga-laer) §a-laer)
p. 553, entry 63 t'i!n t'in
p. 553, entry 79 sja-x" s!a-x"
p. 553, entry 80 tcju tc!u
ga ga
p. 553, entry 87
On p. 535, lines 1-3, the text from "the consonant ..." to "Tlingit (Tl.)." is
bracketed with a question mark by Sapir. On p. 541, footnote 3, Sapir indicated
the last sentence of the note with the word "revise!" On p. 547, last paragraph,
Sapir questions the statement "Chipewyan-/7fl. . . *man)/' On p. 549, end of first
paragraph, Sapir questions the sentence "Such an example . . . 'ring')." In the
Comparative Vocabulary, Sapir questions the Tlingit forms in entries 22 and
33 and the Haida forms in 60, 74, 89, and 95; adds Tlingit na 'to drink' to entry
50; and recalculates the total entries from 98 to 95. On pp. 555-556, Sapir
questions nos. 30, 33, and 38 in the list and states that "This paragraph [headed
"Consonant Clusters"] should be eliminated. Swanton is right." On p. 557, first
text paragraph, Sapir questions the sentence "Should this prove . . . Na-dene
^/)." On p. 558, Sapir notes, beside the first three lines, "More certain now."
The Sino-Dene Hypothesis
[excerpt from a letter to A. L. Kroeber]
I have long wanted to write you about Nadene and Indo-Chinese, but my
evidence accumulates so fast that it is hard to sit down and give an idea. Let me
say this for the present. If the morphological and lexical accord which I find on
every hand between Nadene and Indo-Chinese is "accidental," then every anal-
ogy on God's earth is an accident. It is all so powerfully cumulative and inte-
grated that when you tumble to one point a lot of others fall into line. 1 am now
so thoroughly accustomed to the idea that it no longer startles me. For a while I
resisted the notion. Now I can no longer do so.
The chief stumbling-blocks in the way of a general acceptance of the syn-
thesis would be: 1. Failure to realize the very exceptional type of language to
which Nadene belongs. It is really quite alone in America, so far as I can see.
The contrast between it and Eskimo, Wakashan, and Algonkin is tremendous.
2. Failure to realize that the Nadene languages are not one-third as synthetic as
they look. Go at analysis half-way decently and get into a critical perusal of
connected text and you soon realize that the complex verb breaks down into a
cluster of very live elements, each of which has a syntactic or positional value,
not merely as "affix" but as radical element. Haida in particular, I find, is
extremely analytic. It knows no "prefixes" and "suffixes" at all except for certain
important survivals that Swanton does not even mention (e.g. causative and
denominative s- in s-kit to 'handle a club, to club' < kit 'club', cf. Tlingit sl-,
transitivizing prefix; also Haida /- = Tlingit it- = Athabaskan "3d modal" /-,
also primarily transitivizing). What Swanton calls affixes are all independent
stems entering into composition, or even little verbs. There is no "tense suffix"
in Haida, merely a series of enclitic phrases consisting of demonstrative +
particle verb of being. His "continuative" -gAn, e.g., is simply 'that-is
(duratively)', his "imminent future" -asan is really -'a-sa-i} 'this-will-be
(durative)', and so on. It all crumbles to pieces at the least touch. I think the
same will prove true of Athabaskan-Tlingit, though here the integration is
more thorough. But I no longer seriously believe we have the right to consider
anything preceding the "second modal" elements (Ath. 7-, n-, and v-) as part of
the verb and am beginning to doubt if even these elements and the subjective
pronominal "prefixes" are part of the true verb. I think it more than likely that
such an Ath. form as *yasectk'os (purely theoretic form ad hoc\ Not to be mis-
taken for genuine Ath.) 'I picked up a flexible object' is to be analyzed as *ya se
c l-k'os 'up it-is (that) I handlc-a-flexible-object'. *l-k'os is the verb; the rest is a
series of somewhat reduced independent elements that follow in a definite
order. 3. The third prejudice to overcome is the nature of Indo-Chinese itself.
134 VI American Indian Languages 2
Modern Chinese is a very secondary development. The most typical represent-
ative of the earlier stage is Tibetan — which is startlingly Nadene-like. It has
those fundamentally important "3d modal" elements of Ath., Tlingit, and
Haida (e.g. du- 'to be together': s-du- 'to cause to be together, to assemble'; in
fact, 4 of its more important verb prefixes, which are "voice" elements, seem to
me to correspond in form and meaning to Nadene elements — s- to Tlingit-
Haida s-; r- to Tlingit-Haida-Ath. /-; d- (medio-passive) to Tlingit-Ath. d-,
survivals also in Haida; nasal prefix to Ath. ''3d modal" -n-, -r)-, of mysterious
value but probable active intransitive). Moreover, Tibetan has vocalic ablaut in
its verbs (e.g. X]-gex]-s 'to fill', perfect b-kax], fut. d-gax}, imperative /c'oij).
Again, the transitive verb is really passive, as in Tlingit. In both, for instance,
you would say 'Man-by horse kill' = 'The man killed the horse', Tlingit agen-
tive -tc corresponding exactly to Tibetan instrumental -s. In both Indo-Chinese
and Nadene, postpositions are of extreme importance and serve to subordinate
preceding verbs and clauses. Indeed, reading Tibetan text gives you precisely
the same feeling as reading Haida text. I wish I had time to illustrate. In both
groups the fundamental element is really a noun, the verb a kind of
denominative structure. In brief, I should say that the similarity in feeling
between Tibetan and Nadene is at least as close as between Latin and English,
probably closer. Thus the theoretical road to a synthesis is clear. And the lexical
evidence is startling. You would be amazed at some of my material. Things like:
1. Tlingit k'a 'surface'; Navaho k'd 'surface': Tibetan k'a 'surface'
2. Chinese fan 'charcoal': Haida s-fan 'charcoal'
3. Old Chinese ti 'this': Ath. di 'this' (Ath. di really means ti)
4. Old Chinese ti 'pheasant': Ath. di 'partridge'
5. Nadene k'u 'hole' (TI. k'u-q'" 'hole', t'a- tu-k'^ 'cave' = 'rock-interior-
hole', yoL-k'o 'to fall into a hole'; Nav. k'o, e.g. ts'e-k'o 'rock-hole' = 'canyon'):
Indo-Chinese k'u 'hole' (dozens of forms, e.g. Tib. /:'w-r) 'hole', Karen k'u,
with falling tone, 'to dig a hole').
These are only a drop in the bucket. Naturally it is a big problem and there
are going to be hundreds of knotty points to unravel. But I do not despair. My
present plan is to proceed as follows. First, to prepare part 1 of a Nadene com-
parative study, to consist of my present lexical material (about 300 comparable
radical elements, to which I add constantly). In this I would give reconstructed
Athabaskan but also actual Ath. dialectic forms. Before publishing parts 2 and
3, on morphology and phonology, which need much preliminary work, I intend
to publish special papers on selected portions of Nadene grammar, e.g., certain
archaic post-positions; or demonstrative stems; or general points of syntax. In
this way I shall be keeping the problem live and accumulating experience for
the definitive Nadene study. Of course I shall have to do Ath., Haida, and
Tlingit in the field. I want particularly to gather a large amount of purely lexical
material. People do not realize how scanty is our material, and for my purpose,
which is comparative, I need stacks of it. What Goddard gives us is a miserable
pittance — and wretchedly analyzed or not analyzed. Did you read his attack on
my Nadene? You can have no idea of the laughable errors he commits. It is the
Six: Athabaskan and Na-Dene Languages 135
work of an utter groundling that does not know his own material. What do you
think of a man who expects you to unravel the complex phonology of Nadene by
drawing up an alphabetical list of Tlingit ''stems" and "matching" them with
random "stems" from Kato or Ten'a ad libitum'? Very much as though you
"matched" Sanskrit words now with French, now with Portuguese. Great
method, what? And what do you think of a man who rules out comparisons
because he does not "know" the Ath. form in question? Particularly when the
form occurs in a book (Chapman's Ten'a) which he has "edited"! I may reply to
Goddard, but it is really no use. He is a man of no more than average linguistic
ability, completely at the mercy of his local sentimental memories, and abso-
lutely without vision as to the older drift of Ath. He probably imagines his lists
of stems are the last word on the whole subject. The degree to which he has
failed to analyze his material is shocking in the extreme.
I shall not broach the Indo-Chinese part of the problem till I have moored
myself more completely in Nadene. The final plan is: 1. a Nadene comparative
grammar to be published in 3 parts (possibly an Ath. etymological dictionary as
a side-show); 2. a Nadene-Indo-Chinese demonstration; 3. a more general
treatment of the evolution of the whole group, showing how old types have been
replaced by new ones. As a starter, I am at work now on a paper on Haida
phonetics, which may interest you when you see it.
P.S. I cannot resist the temptation to give a somewhat livelier idea of the
remarkable way in which lexical elements are interwoven in Nadene and Indo-
Chinese. I have some cards along, so don't need to trust to memory. I shall give
an idea of the richness of some of my entries by dealing with a group of related
words.
In Ath. we have a stem *fu, post-vocalic *-Iu, which may be rendered as 'coiP
or loop'; e.g. Nav. to loop', as vb.: Nav. -lo 'to catch with a rope', Jic. Apache
-lo' 'to lasso', Chipewyan -lu, -tu 'to be caught in a net or noose'. So far, so
good. Here our friend Dr. P. E. Goddard would end. But it is difficult to believe
that Ath. */76>/'rope, strap' (found in all dialects) is unconnected. How? With-
out going into details (it would take too long), I may say that I feel justified in
analyzing *t'lo-l into *t-'lo--i. How *-'lo-, -'lu is related to *{u, -lu I cannot yet
tell, but I strongly suspect Nadene had both / and '/, and in related stems. As
you will see from my Haida paper, Haida has both / and '/; in Tlingit '/ probably
became /'. I should guess that 7 is causatively related to /,- Ath . *///, -lu is intr. :
'loop; to lie coiled', *- '/w would be 'to cause to be coiled, to make a loop, to tie
around' (possibly -lo' is a secondary form of *-'lu). Now -/ we know to be con-
tinuative or usitative; and /- is medio-passive. Hence *t-'lo-l is what is always
caused to loop around, what loops about something', in other words, rope,
strap'. We learn important things from such an analysis: that "3d modal" ele-
ments were welded with verb stems and appear in nominal derivatives; that
there was an old alternation /.• '/ whose significance remains to be discovered.
That we are on the right track is confirmed by another common Ath. stem
136 ^^ American Indian Languages 2
whose formation is precisely parallel to that of *t'to-i. This is *t'iei 'fire-drill'.
Fortunately we are here not dependent on Goddard's material alone. From
Petitot we learn that in Hare and Loucheux there is a verb *-/e (-d-le, -t-le, -l-le)
meaning 'to revolve' (words involving it are: 'virer au cabestan', 'cylindrique',
'tourbillon', 'tourner', 'se tourner'). Hence *r-7e-/ is 'what keeps turning itself,
what revolves drill-like'. This parallelism of *r7o-/ 'rope' to */7e/ 'fire-drill' is, of
course, highly suggestive. It shows that many of Dr. Goddard's "stems" may not
be pure father-Adam radicals. And we see that Ath. fl fails to correspond to
Haida and Tlingit fl for a reason. It is a secondary development in probably all
3 groups. Such a sound as /' appears in cognate words throughout; not so fl —
which fact alone casts a reflex light on our analysis.
Let us proceed. To Ath. *hi, -lu is clearly related Ath. *-lui 'to wrap around':
Hupa -loi 'to tie, to wrap around', Kato -//' 'to tie up' (old Ath. form possibly
causative *-'lui). And further, having once allowed Ath. t'{ to analyze itself into
medio-passive t- (d-) + '/, we do not feel we are doing anything ungodly to
analyze Ath. *fVw' 'grass' into *t-'lu' 'what is wound (in basketry)'. This analysis
of 'grass' is helped by denominative Ath. verbs, e.g. Hupa t'h-n, -t'lo, -t'lo-W,
-t'lo-i 'to make baskets, to twine in basket-making'; Nav. -t'lo, -t'lg-, -t'lo-l 'to
tie (e.g. the hair)'.
We may summarize all this as follows:
Ath. *lu, -lu 'coil, loop'
— *lu 'to be caught in a noose'
— *'lu 'to catch in a noose'
— *'lui 'to wrap around'
r-7w' 'what is twined' > 'grass' (denominative vb.: 'to twine in basketry; to
braid hair')
*t-'lo--l 'what is always looped' > 'rope, strap'
Now comes the fun. Indo-Chinese lu as follows:
Tibetan lu-i] 'a strap, slung over the shoulder or round the waist, for carrying
things'
Miao /o-r) 'bridle' (close o)
Tib. metaphorically: caus. s-lu 'to cause to be snared', i.e. 'to ensnare,
beguile, seduce'
Angami Naga te-rhu 'sly' {rhu < h-ru; Tibeto-Burman hi-, hr- > Ih-, rh- is
exceedingly common; h- is common as causative prefix, e.g. Tib. s-l- often par-
allel to Ih-)
T'ai group: Siamese roi^^^ 'enfiler' (numbers indicate tones in H. Maspero's
orthography); White Tai roi, loi 'enfiler'
And now Chinese: ///- group: lui]^-^ (numbers for tones according to usual
Modern Pekinese system) 'a cage, to snare' < Old Chinese (i.e. 7th Cent,
forms, which I have worked out carefully from Karlgren's tables) lu[o]x]- (- =
level; / = rising; \ = falling tone). luv]<-^ (in hi^^^> t'ou(-> 'halter', i.e. 'snare-
head') < Old Chinese /w/o/r)-;/w^-^^ 'girdle gem' <luo\ (words with falling tone
seem frequently to be old passive derivatives: 'what is looped around one's
waist'?); /m^^'' 'thatched hovel' < luo-; lu<^> 'hempen thread' < luo-; lo^^^ 'conch,
Six: Athabaskan and Na-Dene Languages 137
spiral, screw' < lud- {a is a dark-timbred a- vowel); lo^-^ 'lines in the palm' < lud
?lei'^^ 'to creep, cling to', same character also read lei^^^ 'series, connected'
(words evidently refer primarily to creeping vines) < ludi-, ludil. These Chi-
nese lu, lud forms are paralleled by another set in lui- (cf. Ath. *-lui above):
lei^^^ 'to bind' < lywie- (< *lwie-);lei''''> 'acreeper (asof melonor pea)' (not sure
of reconstruction, but phonetic element in character suggests initial lu or lywi);
lii^^^^ 'silken thread, a hank' < lyhi/. And metaphorically we have luj}^'^^ 'foolish,
to impose on' < lu[o]r\\; lii^'^^ 'deceitful' < />'h'/o\ (i.e. 'winding about, ensnar-
ing with blandishments'; cf. Tibetan s-lu above).
Other probably connected Tibetan words are: Ihub (i.e. causative h-lub) 'to
bind, tie, fasten (e.g. ornaments to the ear)'; k-lub 'to cover (e.g. the body with
ornaments)'; Iwa-ba 'a woolen blanket'.
Observe how well the Ath. forms integrate with the numerous Tib. and Chi-
nese forms. But we are not done. Very likely connected with Ath. *-lu is Ath.
*-lu-s 'to drag an animal by a rope': Nav. -Ids, -lo-z, -Ids classifier verb "denot-
ing a single animal as an object: the inference is that the animal is led by a rope"
(Franciscan Fathers); Hupa -los 'to drag, to pull along'. With these forms I feel
inclined to compare (though here I feel far more hesitant) Chinese lo^^> tsz<^^
'mule' {tsz<-^^ is merely 'son', often used to make nouns) < lud-; also lu 'donkey'
< lywio-. The parallelism between Ath. and Chinese would be a convergence
from related radicals rather than a specific etymological parallel.
But we are far from finished. Perhaps related to Ath. *-lu is an important
classifier verb *-/e, *-la\ Nav. -/e, -la, -let 'to handle a long, flexible object, as a
rope, quirt, leather, hide, etc' (Franciscan Fathers). Parallel to this is a set of
Indo-Chinese forms in *la, *le (Ath. e is often parallel to a; just how related I do
not yet understand, possibly reflex of old alternation a: [a):
Tib. causative s-le, Ihe (< h-le) 'to twist, plait, braid the hair, to make a
basket, to knit'; s-le 'a coarse basket'; s-le-po, s-le-ba, s-le-bo 'a flat basket'
{-po, -ba, -bo are "articles"); Iha-s, Ihe-s < h-la-s, h-le-s 'braid, wicker-work,
texture; twisted cake or bun'; Ihe-s-ma < h-le-s- 'the act of twisting, plaiting'
(-ma is "article"); /a«-/?w 'braid, plait, tress of hair' (-/?« is diminutive); lan-ts'ar
'ornaments worn in the hair'; le-brgan 'diapered design of woven fabrics'; le-na
'the soft downy wool of goats below the long hair; fine woolen-cloth'; Ida-ldi 'a
kind of ornament of silk or cotton, a fringe or tassel' < d-la- {dl- regularly > Id-
in Tib.; very easy to illustrate); Idan-mgo 'the yarn-beam of a loom' < d-lah-
{mgo 'head'); Idem-Idem < d-lem 'flexible, supple, elastic, pliant'; Ideb- < d-
leb- 'to bend round or back, to turn round, to double down'. 1 am not so certain
of this last, which brings us into a large set of forms in la- and lo- referring to
'turning, turning back', which may well be related to our present set but which I
prefer, for brevity's sake, not to go into just now.
Now Chinese. We have two series: *la and *//. Based on *la are: lan'-> 'basket
with handle' < Idm-; lan^^^^ 'rope, hawser' < Idml; lao'-^> (tezi) 'netted case' <
Idk, same character also read h^-^^ 'joined, to tic up'; lei^'K also read h<'---'^> 'to
rein in, to strangle' < hk (perhapsbcttcr to /fv- scries above); /o'^'^ 'net, sieve' <
ld-\ lo<^) 'shallow open basket' < Id-. Based on *// are: li<~> 'ornamented girdle'
138 ^I American Indian Languages 2
< bie- (?); W^^ 'basket' < />'/-; //^- "^^ 'rope to tie a boat', character also readi/'^-^''
'a well woven gauze' < xyil (?), which probably means older *h-li (loss of post-
consonantal / is now well established for Chinese).
Is it not impressive that Ath. and Indo-Chinese *lu seem to have reference
chiefly to "looping," Ath. */e, la and Indo-Chinese *la, li to "handling a long
flexible object, twining, basketry"? Now let us return to Nadene. In Tlingit we
have //r' 'fine basket'. Remember that Tlingit has no voiced /, only voiceless /.
Further, experience shows abundantly that many Tlingit forms in final conso-
nant (particularly if glottalized) are compounds. Hence we may suspect that lit'
really means old *li-f(a). Reference to Ath. gives us *-t'a 'receptacle': Hupa -fa
'sack'; Kato -fa 'pocket, blanket fold'; Nav. -dzis-fa 'pocket' {-dzis 'hollow,
semi-tubular' ?). This somewhat unsatisfactory parallel is buttressed by Haida
fao-fa 'box' (almost certainly misheard for tao-fa 'food-receptacle'). Tlingit ///'
is therefore probably li-fa 'receptacle (for small objects) of twined basketry'.
We have further /'/ 'woolen blanket' < *// or *//' (final glottal stops seem regu-
larly to affect the first consonant; I have good evidence for this). Based on *la or
*lia we have also in Tlingit fleq' 'tentacles of squid' *t-'le-q' (with medio-pas-
sive r-; Tlingit e is always related to a- forms) < 'what is long and twists itself?
(cf. fundamental meaning of Ath. *-/e, *-la above).
And in Haida we have dli-n 'arm of devil-fish', which I analyze as d-li--n {d-
medio-passive; -n perhaps causative, to which there are good Ath. and Tibetan
analogies; -//•- < Im- ?) < 'long things that twist themselves about'?
I am quite likely to have misinterpreted here and there, but the Nadene and
Indo-Chinese parallels seem highly suggestive to me. Of course, this sort of
group-parallelism is not isolated.
I shall refer to another interesting group, without going into details. I spoke
of Ath. *-/e 'to revolve' and its derivative *flel 'fire-drill'. To these forms belong
Haida hi 'to surround, move around'. In Indo-Chinese we have a well-cham-
bered Tibetan set: re and ril (e.g. res 'change, turn, time, times'; i}-g-re 'to roll
one's self ; causative s-g-re 'to roll' ; ril 'round' ; causative s-g-ril 'to wind or wrap
round; to roll, wrap, or wind up; to wag (the tail); to roll (a stone)'. Here may
also belong Haka kut-hrer 'ring' {kut- 'hand'; h- common Tibeto-Burman
causative prefix), though Conradi sets this to Tib. forms in la-, le- referring to
"twining."
Editorial Note
Originally published as part of letter 332 (October 1, 1921) in Victor Golla
(ed.). The Sapir-Kroeber Correspondence. (Survey of California and Other
Indian Languages, Report 6.) Berkeley: Department of Linguistics, University
of Cahfornia (1984).
Six: Athabaskan and Na-Dene Languages 139
Sapir's manuscript Sino-Dene "dictionary," including more than a hundred
lexical comparisons, is in the Library of the American Philosophical Society
(manuscript 497.3 B63c Na20a.3, vol. 2; cf. Kendall 1982: 28). Nearly all the
entries date from the early 1920s, and it is clear that Sapir went little further
with the idea than the speculations contained in this letter. Of these specula-
tions (see also 1925o, later in this volume) Krauss has written: "Sapir was . . .
carried far beyond any objectively justifiable conclusions by his enthusiasm for
the idea" (Krauss 1973: 963). Few would challenge the accuracy of this assess-
ment. The connection is, however, a plausible one, both on linguistic and
anthropological grounds, and it continues to attract attention. Robert Shafer, a
Sino-Tibetanist, published two papers in support of the Sino-Dene hypothesis
(Shafer 1952, 1957), but, being based neither on Sapir s Sino-Dene files nor on
extensive comparative Athabaskan or Na-Dene data, his work cannot be con-
sidered an authoritative statement. Greenberg, Turner and Zegura (1986) have
proposed that Na-Dene (including Haida) represents the language of an Asi-
atic population that entered the New World about 7,000 years ago, and that it is
distinct, both physically and linguistically, from all other American Indian pop-
ulations. While this is consistent with Sapir's Sino-Dene hypothesis, Green-
berg, in his recent world-wide linguistic classification, does not link Na-Dene
with Sino-Tibetan or with any other linguistic group, considering it to be one of
15 fundamental linguistic families in the world (1987: 332-337). Pinnow (1990),
relying mainly on Greenberg 's own evidence, believes that a good case can be
made for viewing Na-Dene as intermediate between Old World languages
(especially Sino-Tibetan) and New World languages (especially Greenbergs
"Almosan-Keresiouan" phylum).
Athabaskan Tone
A part of the summer of 1922 was spent by the writer at Sarcee Reserve,
Alberta, in studying the language of the Sarcee Indians. A series of texts was
obtained as well as supplementary grammatical material. The most important
single point that appeared was the fact that Sarcee has a well-developed system
of pitch accent. Fundamentally this system has a striking resemblance to the
Tlingit tonal system described by Dr. Boas, though secondary developments
have complicated the Sarcee system considerably. The tonal resemblances
between Tlingit and Athabaskan constitute an important further argument in
the Nadene theory recently put forward. Hints on Athabaskan tone are to be
found also in Father L. Legoffs study of Chipewyan (Grammaire de la Langue
Montagnaise). Father Morice makes a few isolated references to tone in Car-
rier, where it is almost certainly a feature of importance judging from brief MS
linguistic notes taken by C. M. Barbeau among Carrier Indians at Hagwelgate.
Some years ago P. E. Goddard noted tonal differences between otherwise iden-
tical second person singular and third person subjective forms in the Hupa
verb; these observations, based on the study of [391] Rousselotgraphs, agree
with the Sarcee results obtained. So fundamental is tone to Sarcee morphology
that it is well nigh inconceivable that it should be entirely absent in any other
Athabaskan dialect.
Editorial Note
Originally published in American Anthropologist 24, 390-391 (1922).
Reprinted by permission of the American Anthropological Association.
Sapir's references to earlier work are to Boas (1917) for Tlingit, to Legoff
(1889) for Chipewyan, and to Goddard (1907) for Hupa "Rousselotgraphs."
Morice later commented on Carrier tone more extensively in his full treatment
of the language (1932).
At the time of Sapir s Sarcee work, tone systems had been described for many
African and Asian languages but only rarely noted among American Indian
languages (Sapir's own description of a pitch accent in Takelma being one of
these instances; see 1912h, Volume VIII). We now know tone to be a wide-
spread feature in the Americas, particularly in eastern North America and in
lowland South America. Ironically, comparative evidence in Athabaskan (and
Athabaskan-Eyak) now indicates that tone was not present in the proto-lan-
guage but developed in several Athabaskan subgroups (but by no means in all)
as syllable-final consonants were simplified or lost (Krauss and Golla 1981:
142 Vf American Indian Languages 2
69-70). Sapir, in later fieldwork, encountered at least two Athabaskan lan-
guages without tonal systems, Anvik and Hupa. In the latter case, where Sapir
had gone to the field with every expectation of confirming Goddard s earlier
reports, he was clearly nonplused (see Sapir 1928i, reprinted in this volume,
and Sapir s letter to Kroeber, 28 June 1927, quoted in Krauss 1986: 163).
A TYPE OF ATHABASKAN RELATIVE
As is well known by students of Athabaskan
linguistics, the Athabaskan adjective is in form
a verb. Even the simplest, non-pronominal or
third-personal, form regularly contains either
a " first modal " prefix - (generally de- ^, d;
ne-, n-; t'e-, t'- ; or ie-, i-), a " second modal "
prefix {se-, s-), a " third modal " prefix (gene-
rally -/- or -i-), or a first (or second) modal
prefix followed by a third modal prefix (forms
in de-l-, ne-l-, nl-i- are typical). It is unneces-
sary to give examples here, as they may be
readily supplied from the special dialectic
monographs (see, e.g., Goddard, Legoff, Mo-
rice, Petitot),
Father Legoff, however, in his Grammaire de
la Langue Montagnaise \ calls attention to an
interesting class of adjectival derivatives
(" nouns formed from'adjectives ") in which
the prefixed elements are lost and the bare stem
appears in the relative ^ form. Some of his
examples are :
1. I follow Dr. Goddard's convenient terminology,
without thereby committing myself in the least as to the
term ""modal ".
2. I use e as a formula for the reduced or " pepet "
vowel which has different forms according to dialect or
according to varying phonetic circumstances in one
dialect (e.g. i, less frequently a, e, a. or u, in Hupa ; a
in Kato and Chasta Costa ; e, a in Chipewyan). It may
often be shown to be a reduced form of an older Atha-
baskan e or /.
5. Montreal, 1889. See p. jo. Petitot and Legoff use
*' Montagnais " in the sense of Chipewyan.
4. By " relative '" I mean the form assumed by nouns
when they are qualified by preceding elements (nominal
or pronominal) and by verbs when they are used as rela-
tive clauses. The fundamental Athabaskan relative suffix
is probably -e, -e (-ye, -y^), but the actual dialectic
forms are often involved by the operation of various
di-l-ba (to be) gray : bay-z the gray one
dz-l-gai (to be) white : gay-z the white one
m(d)-du-i (to be) short : du-z the short one
dz-yzl (to be) squat : yzl -s the squat one
dz-bal (to be) round : bao the round one
dz-l-zzn (to be) black : :^zn the black one
dz-yo' (to be) shaggy : yo' the shaggy one
dz-l-Bo' (to be) yellow : Oo' the yellow one.
Such forms as bg^, ;(en, yo' , and 60' look for
all the world like unmodified stems, but there
is every reason to believe that they are relative
forms, like gay-z and yel-z, that have either
fallen together with or that differ in some res-
pect from the parallel stem forms found in the
adjective-verbs. Possibly the phonetic record
is defective ^.
Legoff says of these forms : " This kind of
phonetic laws. Frequently the e, e is dropped but is
then apt to leave a trace in the voicing of the prece-
ding consonant, now final (e.g. the relative form of
Hupa -fa AMONG < Ath. *-t'ax is -t' au < -t' av.\ the
regular Hupa development of Ath. *-t'a-{, reduced from
*-t'a-f-e).
5. I write ^a^-e rather thad ba-y^, though Legoff speaks
of a suffixed -ye, because comparative evidence demons-
trates the existence of the stem-form -bay- a.s well as -ba' .
-ba'-i and Petitot's Montagnais -ba-a seem to presuppose
Ath. *-mah-e. Navaho -bat, e.g. gim i-bai gray hawk,
is Ath. *-may-i; d. further Hupa -nmi, Kato -bai. I am
modifying Legoff's and Petitot's orthography and that
of the Franciscan Fathers so as to bring them into har-
mony with current Americanist usage.
6. Goddard writes di-l-{7.ti to be black, but tia^iiz-
^an' BLACK K0.\ (Texts and Analysis of Cold Lake Dialed,
Chipewyan, apamnh, X, 1 10). Does this mean that -iin
is the relative form of -inn ? The relative form of the
corresponding Navaho -/t« black is -/««, with lengthened
vowel. The formation of the relative in Athabaskan has
never been properly studied. It is one of the fundamental
problems of the language.
144
VI American Indian Languages 2
137
substantive is hardly used except as sobriquets
which men give one another or as names which
people give to animals, in order to distinguish
them. And then they are always followed by
the word ya:(^t, which means little". As a
matter of fact, the type is illustrated in other
connections, e g. :
d--k\oTO BE BALD I t^i-k'zit THE BALD-HEADED
ONE, BALD-HEADED (Pet.)
(h)o-rt-iur '< to be slippery (Pet.) : t' an'-xurz
SLIPPERY ice (Leg.)
These two examples differ in an important
respect. In t'xn'-:(urs and numerous other com-
pounds of its type the first member (ice) desig-
nates the properly denominating concept of
the group, which is then qualified by a relative
form (jur-i) of Legofi's " sobriquet " type. In
t^i-k\lt the first member of the compound
(head) is not the properly denominating con-
cept of the whole, though it is itself qualified
by a relative form (^k\l-t), again of the sobri-
quet type. The group tHi-k\l-., taken as a unit,
is to be understood as qualifying a third, un-
derstood, noun. We can express tliis by saying
that while slippery is relative to ice and ice
not relative to another concept, bald is relative
to head and bald-head to person. As far as
such a form as tOi-k'eoi is concerned, it makes
no difference whether the qualified noun is
expressed or not. Obviously the difference bet-
ween t'oin'-iure. and thi-k'zle is analogous to
the English difference between red breast and
(robin) redbreast. Whether there is a pro-
sodic difference (one of stressor pitch) between
the two Athabaskan types does not appear
from the evidence, but it is possible that they
are not formally identical.
As there is no genuine line of demarcation
in Athabaskan between " adjective " and
" verb ", one may expect that forms both of
7. Petitot writes 0- for /;o- (see Goddard and Legon).
-r- is a postvocalic form ol d- in Chipewyaii.
the type bays, the gray one and t'ixn'-:(tm
slippery ice may be based on " verb stems ".
This is exactly what we find. A few examples
are :
Mont. -Y? TO MELT ; ds.-l--(j to be melted
(Pet.) : dlts '(in-z grease easy to melt,
meltable grease (Leg.)
Mont, -ti'a^ to cry : ts'ay-e. the whimpering
one, grumbler (Pet.)
Hare -k'^i to lie habitually : k'^t-t liar (Pet.)
It scarcely needs to be pointed out that
LegofT's " sobriquets " are simply qualifying
terms in the relative form, the noun referred'
to being unexpressed. It is not a far cry from
compounds like t^i-k'tlt the bald-headed one
and dUs -^in-e. meltable grease to such clipped
forms, say, as k'toe. the bald one and -(ins.
WHAT melts (easily). Such forms are in type
identical with LegofFs hayt the gray one. Ot
ts'a-^t grumbler, Petitot remarks that it is deri-
ved from ts'ay tears; we would then have to
interpret ts'ayz as the one with tears rather
than as the whimpering one. Petitot's analysis
is perfectly credible, for forms of the type of
bayz may be directly formed from noun stems,
as I shall show from Navaho evidence. But I
hope also to show that the difference between
-ts'a-( TO CRY and ts'a-; tears is purely a matter
of translation, not of intrinsic Athabaskan
form.
In Navaho there are a great many prefixless
or radical qualifying terms in relative form.
They are freely used with or without preceding
nouns and are based on " adjectival ", " nomi-
nal ", or " verbal " stems. They all denote
permanent or characteristic attributes and easily
take on the character of sobriquets or regular
clan or personal names. Examples of " adjec-
tival " origin are ^ :
8. My Navaho examples are taken from tiie Vocabuhry
of the Navaho Language of the Franciscan Fathers (2 vols.,
St. Michaels, Arizona, 1912).
Six: Athabaskan and Na-Dene Languages
145
138
n-jun-i NICE (relative form) ; «i-/p" to be gocd :
jun-i NICE
n-nt'i TO BE LONG : nts, nf^-i long
n-t't'lro BE WIDE : t'd WIDE
h-ts'o to BE YELLOW : h ts'o-i YELLOW EARTH
(place name)
ii-jin TO BE BLACK : hwo' ji'n-i tooth-black,
DECAYED TOOTH J ya'' ji''n black LOUSE
a-i-i'sg's-i to be slim (also -t'sgs, -t'sos-i,
-t'so's) : ts'i' t'so's-i head-slim, the slim-
haired ONE (man's name)
hwo' c-gi'j-i TOOTH wHicH-is-MissiNG (c- assimi-
lated from S-, " second modal " prefix) : hwo'
gi'j-i MISSING TOOTH ; bo-hwo'' gi'j his-tooth
MISSING
di-tc'oc-i stubby (relative form) : ts'i' tc'o'c-i
head-stubby, the stubby-haired ONE (man's
name)
a-gud-i SHORT ; Montagnard i-gor-z (Pet.) : ga
gu'd-i ARM-SHORT, ARMLESS ; k\ gud-i SHORT-
FOOTED
ia-gai to be white : ga' ts'o gai' rabbit-large-
WHITE, WHITE JACK-RABBIT
Examples of *' verbal" origin are :
-t'lo, -t'ig', -t'ioi to weave : Hog-i 9 grass-
weavers, SIA INDIANS
-y^'d, hwud, -hwul to run : t'o' hwul rapid
WATER (place name)
This type is doubtless actually well repre-
sented in Navaho, but the material is scanty or
not easily accessible. To it belongs probably
Tucson (Arizona place-name), said to mean
BAD-SMELLING WATER ; the Navaho (or Apache)
form is probably something like to tc'a (cf.
Nav. -tc'iUy -tc'a, tc'ji to smell).
Examples in which the related word is a
noun are quite numerous. The reference is not
to the concrete content of the noun as such
but to a person or object, expressed or implied,
that IS conceived as the possessor of or as in
9. Or directly to t'io' grass?
some way related to the thing defined by the
noun. Examples are :
k'ai WILLOW : k'a''-i those who have (or are
CONNECTED WITH) WILLOWS, WILLOW CLAN
yo' BEAD : yo-0 those who have beads, bead
CLAN
fca' HAT : t'cah-i he with the hat (man's
name)
gic CANE : gic-i HE WITH THE CANE (man's name)
ca'c KNOT : ca'j-i knotty
xvi burden, bi-yf'l his burden : ye'l-i the one
WITH A burden, he WHO IS SLIGHTLY HUNCH-
BACKED (man's nick-name)
fcoc WORM : hwo' fco'c-i tooth -wormy, a hol-
low TOOTH
yii SUPERNATURAL BEING : t'o yf WATER THAT
HAS SUPERNATURAL BEINGS, DANGEROUS WATER
(place name)
t'o WATER : na' t'o'-ho enemy that is connected
WITH WATER (?), ENEMIES AT THE WATER,
ISLETA INDIANS
cac BEAR : na' cdc-i ennemy that is bear, bear
ENEMIES, HANO INDIANS
Relative forms like t'cah.-i bear an obvious
formal similarity to such English derivatives
in -ed and -y as bearded and knotty. But the
resemblance is more apparent than real. The
relative form of the noun is not a true " adjec-
tival " derivative of the noun, as is shown by
the fact that morphologically parallel forms are
built on stems conventionally set down as
" adjectival " or " verbal ". The genetic
relation between these Athabaskan relative
forms and the possessed form of the noun (e.
g. Montagnais t'sah-i his cap [Leg.], si-t'sa-a
MY CAP [Pet.]; Hupa hWm song : xo-hW\.n-[n^'.
HER song ; Chasta Costa c-man-e my house) on
the one hand and the subordinated form of the
verb (e.g. Mont, i^-yt one grows up : {s--ye-
h= GROWTH, m-^^-i-l-dyd we are afraid : ns—,"-
t-l-dyed-i we who are afraid [Leg.] ; Hupa
-lal TO FLOAT continuously : na-na-t-lnl-B it
floating; Nav. di-c-Jnuuc i yell : d\.-la-hwuc-i
146
VI American Indian Languages 2
139
ONE WHO YELLS MUCH, howler) OH the Other
is obvious. In Navaho the relative form with
final vowel (generally -/ ; old -e assimilated to
-cJ after radical o)is probably no longer felt as
identical with the possessed nominal form with
final consonant (cf. bi-ye-'l his burden with
yz'l-i above), but the general consensus of
Athabaskan evidence makes it higly probable
that such alternations as -yi'l and yd-i go back
to Athabaskan alternations of type *^z'l-e:
*'(el-e. In other words, even in Navaho the
possessed forms of the nouns are simply redu-
ced relatives (perfectly analogous to xe'i : -yfl
is d:(d MOUNTAIN : yo-l-gai' d^i'l shkll-white
mountain). We have already seen that there
are analogous doublets in Navaho for the rela-
tive forms of " adjectival " stems (cf. gi'j-i
and gi'j .missing, jm-i and ji'n black above).
It is not the purpose of this paper to discuss
the functions and the fundamental significance
of the Athabaskan relative. I hope to show in a
future paper that it is a feature that goes back,
both in general form and as an actual phonetic
element, to the Nadene period and that it con-
sists in essence of an old particle, probably a
demonstrative stem, that could be freely added
to any word or group of words to relate it to
an expressed or understood person or thing.
The primary function of the Athabaskan (and
Nadene) relative is thus an exceedingly wide
one, of which the particular usages listed in our
grammars are but specific applications or rather
English (or French) translations. The lengthen-
ing of the stem vowel and the voicing of a
final voiceless spirant '° are merely secondary
phonetic phenomena due to the presence of
10. In part no doubt retention of originally final voiced
spirant. 1 believe it to be probable that in such Athabas-
kan alternations as '/Vo/ strap : *tlol-e, *t'iol-e STRAP
OF (one) it is the -/- which represents the old conso-
nant, preserved because of the following relative element,
and that the -i of the absolute form is du^ to a secondary
unvoicing of'the old -/.
the relative element. As this element became
reduced to zero, these secondary phenomena
tended to take over the properly relative func-
tion.
Just as we have the alternation of final -i :
-/-, of -X (-') : -Y-, of -X (-') : -y-, of -c : -/-,
and of -s : -;(-, so also these alternations occur
initially " ; e. g. Mont, iue fish : se-llut my
FISH (Pet.), Hare xi burden : se-yzl-t my bur-
den, Hupa hi smoke : m'.[l]-h[t]d-t his smoke,
Nav. Sin song : bi-yi'n his song (in Athabaskan
terms *xen.: *-yen-i ; cf. Mont, cen : -yzn-Zy
Hare ci : -yin-e). Here too the alternation
could only have been due to phonetic circum-
stances to begin with. If a word was closely
connected, in thought and in position, with a
preceding word or element, the voiced spirant
(say /) was retained or the voiceless spirant was
voiced. Athabaskan *xei burden : *cB-yel-e my
burden thus originally meant no more than
that two radical elements {*ce me and *xel bur-
den) were united into a phrase with the help
of a following denominating element *-e, *-e :
me-burden the, i.e. my burden. The *-e
preserved the -/ of *xel, ordinarily *xei,
while the voiceless x passed to -y- in intervo-
calic position '^ It is very doubtful if there
was any specific function connected with the
X- : -y- interchange. In time, however, there
11. In final position note also -r; (or nazalization) :
12. Possibly an old y- was here preserved, but passed
to unvoiced x- when not protected by an immediately
preceding vowel. The alternations listed in the text na-
turally apply to primary Athabaskan. In certain dialects
some of these alternations ceased to operate freely be-
cause phonetic laws divorced the consonants that had
originally belonged together. Thus, in Hupa the old
Athabaskan interchange of x- : -y- (preserved in Chipe-
wyan, Hare, and Navaho) had ceased to be a live pro-
cess because of the falling together of Athabaskan .v and
k' into Hupa x and the change of Athabaskan y to Hupa
w ; the corresponding final alternation of -x : -y lingered
on as -' : -w, -u. Note also that in Pacific Athabaskan
voiced sibilant spirants have been leveled with voiceless
sibilant spirants.
Six: Athabaskan and Na-Dene Languages
147
140
can be no doubt that the voicing of initial spi-
rants came to be felt as intrinsically, not me-
rely mechanically, connected with the relative
function. Hence such detached forms as "^^eUe
THE ONE WITH A BURDEN (Nav. yel-i '^), "jUH-e
THE GOOD ONE (Nav. jun-i), *jen-e the black
ONE (Mont. :(enn). It is probably because of
the intrinsically " relative " significance of
" adjective " stems that these regularly begin
with a voiced spirant if the initial consonant is
a spirant.
Such forms as Hare ^'^'e-e liar and Navaho
t'^'I WIDE bring home to us the highly impor-
tant fact that the actual " radical elements "
of Athabaskan verb and adjective forms are
more freely isolated than one might at first be-
lieve to be possible. A careful study of all the
available material would tend to show that
these radical elements have a considerable mo-
bility, that they are not far removed from the
status of independent monosyllabic " words ",
and that the complex " word " of our Atha-
baskan texts and paradigms feels a great deal
more like a closely knit phrase or sentence
than has yet been suspected or, at any rate,
explicitly demonstrated. I hope to show in
due time what is the true nature of the various
" prefixes " and " suffixes " that render the
morphology of the Athabaskan verb so complex
in appearance. It will appear that each and eve-
ry one of these elements is a relatively self-
contained unit in the sentence, either a deter-
minative or an actually predicating element.
Much of the " vagueness " of meaning or
function that we feel to attach to many of these
elements is simply an index of our inability to
carry over the Athabaskan manner of expres-
13. Athabaskan y is preserved in Navaho only before
a. It appears as y before i and e, labialized to w before
0. The alternation x : y, however, leads one to suspect
that this " y " is not phonetically identical with the y
that corresponds to common Athabaskan y.
sion into precisely equivalent English (or
French) form'"*.
For the present I shall content myself with a
few random examples suggesting the mobility
and essential concreteness "> of the verb stem.
The Hupa verb stem for '' to flow " is -Un,
-hr„ which has numerous cognates in other
dialects. Forms like t'cewishnt't it will flow
OUT and nzdhn it always flows do not seem
to suggest the possibility of combining the bare
stem freely with other concrete elements. I
have found no examples in Hupa of -/sn or -hr,
so used, though they may 'of course exist. But
this stem (Athabaskan *-lin, *-leri) is clearly
related to Athabaskan *-// (e.g. 'Mont, -^a-i-d-li,
Hare yi-d-li couler a terre [Pet.]), probably
also to Athabaskan *-lej (tears) flow (e. g.
Mont, dc-l-hz, Hare dt-l-W [Pet.]). Athabas-
kan *-li appears in Hupa as -k (reduced to -/) ;
it is found, without formal prefixes of any
kind, in certain compounds : no-U dam, wa-
terfall (lit. down-flow or halt-flow), iz-l-
d'.r, together-flow-place (village name). Si-
milarly, -xa{u)W, an " indefinite "form of-.vd
liquid has position, is directly compounded
14. Very much as one who tries to see in a French lo-
cution like Qu'est-ce qu'il a ? an exact equivalent of our
English What has he ? would find it difficult to get the
proper form-feeling of the elements est, ce and qu\
15. In a large number of Athabaskan verb forms it is
impossible to assign to the radical element (that is, the
" verb stem " proper) the kind of concreteness that would
in our eyes yield the actual concrete significance of the
form. It does not follow that the stem has not a defini-
tely concrete significance of its own, clearly apprehen-
ded by the native form-feeling.. All that we have a right
to say is that the summing of significances inherent in
the theoretically independent elements of the complex
does not seem naturally to lead to the precise idea that
we express in such and such a way. The true difficulty
from our naive standpoint is in such cases not that the
stem is a « vague » element but that it seems inapplic-
able, just as to one unacquainted with English idiom
the notion of " fall " in such a sentence as The friends
had a falling out is inapplicable. A a matter of fact, I be-
lieve that there are few languages in America that feel
their " stems ",.and elements generally, as definitely and
as concretely as Athabaskan, Tlingit, and Haida.
148
VI American Indian Languages 2
141
with sa MOUTH : sa-xa(u) H^ liqjjid which has
BEEN PUT IN THE MOUTH, ACORN SOUP. In Na-
vaho, again, the verb stem -ua to live (e. g.
X'.-n-c-na 1 live) may be used as an unmodi-
fied element in a compound ; k'a'-na''-ni
ARROW-LIVE-PEOPLE, LIVING-ARROW PEOPLE (clan
name). Such examples could be multiplied con-
siderably.
If the monosyllabic *' verb stem " may thus
be isolated in practice as a more or less freely
movable element, capable of conveying a definite
notion in its own right, we cannot but con-
clude that the purely formal difference between
verb (and adjective) stem and noun stem be-
comes a tenuous one. What is to prevent us
from interpreting the -h of Hupa no-lz as a
noun meaning flowing or current, no-h and
it-l-d\.rj meaning properly down-flowing (not a
secondarily aominalized form of an inherently
verbal to down-flow) and reciprocal-cur-
rent-locality ? Might not the Navaho k'a'-
na'-ni be just as well interpreted arrow-life-
people ? As a matter of hex., I cannot see that
anything seriously stands in the way of such
an explanation, and its adoption would at once
make clearer a number of morphological pecu-
liarities. Among such peculiarities are : 1. the
ease with which a great many evident nouns
are transformed into " verb stems " (e. g. Hu-
pa ^'fl DRESS : -k'a TO wear a dress ; Kato djit]
DAY : -djvr^ TO BE day) ; 2. the frequency with
which " verb stems " with aclearly defined verbal
lorce, if we may trust all appearances, take on,
when isolated, an abstract or concrete nominal
significance (e. g. Athabaskan *-yan to pass
THROUGH life : *xan old age, maturity ; *-dlo'
TO LAUGH : *dlo' laughter ; *-lo to snare, to
BE CAUGHT IN A NOOSE : N'av. lo'^ SNARE, LOOP);
3. the fact that a number of verb stems refer
not to specific activity but to a class of objects
1 6. Voiceless spirants initially, voiced spirants in post-
vocalic, or originally postvocalic, position, according to
the typical Athabaskan rule.
(e. g. *-'a« TO HANDLE A ROUND OBJECT ; *-^'oj
TO HANDLE A CLOTH-LIKE object). I hope later
to take up this fundamental question and to
show that in a verb form it is not the " verb
stem " that is the distinctively verbal element
but, where found, the" third modal " element;
that all " verb stems " are in fact nouns not
only in theoretical origin but in actual usage ;
and that verbs translated according to the forms
TO DIE, TO BE SEEN, and TO KILL fall iuto patterns
more accurately rendered by death is, sight
TAKES PLACE, and TO MAKE SLAUGHTER. If this
interpretation is correct, an element like -an is
not properly a " verb stem " indicating some
kind of activity or state with reference to a
single round object but is actually a noun which
means, or originally meant, a round object. The
three classes of verbal usage listed above would
fall into a single category applicable to all other
verb forms as well. Indeed, it will appear that
this theory of the, essentially nominal character
of all " adjective " and " verb " stems simpli-
fies enormously the whole aspect of Athabas-
kan (and Tlingit) morphology.
Meanwhile, whether or not we are willing to
go so far in the present stage of our knowledge
as to accept the nominal theory of verb radicals,
it is clear enough that the Athabaskan relative
forms discussed in this paper belong together. If
a Navaho form like Ccah-i is to be interpreted
as THE one HAVING (ot CONNECTED WITH) A HAT,
we may venture to interpret a verbal deriva-
tive like Hare k''^t-z as the one having false-
hood, an adjectival derivative like Navaho
jun-i as the one having goodness. There is
certainly no serious point of morphology that
would make such an interpretation impossible.
As it is, it is sometimes an arbitrary matter
whether we assign a given relative form to an
adjectival or to a nominal source. In Navaho
ga k\s-i one-armed, armless, k\s-i short,
crippled, DEPRIVED OF may be looked upon as
an adjectival (or verbal) formation (cf. gnd-i
Six: Athabaskan and Na-Dene Languages
149
142
above). Identical, however, with Navaho k\s,
which I have not found in its bare form, is
Anvik Ten'a k'xf) piece (oF)(Athabaslian *k'es).
Clearly it makes little or no difference, from
the Athabaskan standpoint, whether we ana-
lyze ga k'ls-i verbally as the one whose arm is
CUT OFF(cf. Navaho verb -k'z, -ki'\ -k\ to cut
WITH A knife), adjectivally as the one who is
SHORT OF AN ARM, or nominally as the one
WHO HAS AN ARM-PIECE, AN ARM-SEVERANCE. In
the same way, it seems an indifferent matter
whether we interpret Montagnais ts'ay-t ver-
bally as THE ONE WHO IS (aLWAYS) CRYING Or
nominally as the one who (always) has tears,
WEEPING ; or Navaho yil-i (a sobriquet for a
hunchback) verbally as the one who is wont
TO carry (a burden) (c(. Navaho -xe or -xd,
-yr, -xi-i, TO carry on one's back, to handle
A burden), or nominally as the one who (al-
ways) HAS A BURDEN.
Editorial Note
Originally published in International Journal of American Linguistics 2,
136-142(1923).
THE PHONETICS OF HAIDA
TABLE OF CONTENTS
Introductory.
I. Consonants.
The Consonant System.
The Intermediates.
The Unaspirated Hard Stops.
The Aspirated Surds.
The Glottalized Stops and AfFricatives.
The Voiceless Spirants.
The Nasals and Voiced Spirants.
The Glottalized Nasals and Voiced Spirants.
The Laryngeal Consonants.
Secondary Consonantal Processes.
Initial Consonant Clusters.
II. Tht' Syllable.
Syllables with 1-Vowel.
Syllabic Nasals.
III. Vowels.
Qualitative Changes.
Vocalic Quantity.
Diphtongs.
W. Stress and Pitch.
INTRODUCTORY
The following notes on the sounds of tlie
Skidegate dialect of Haida are based on mate-
ri.il which I was fortunate enough to secure
iVom Peter R. Kelly, a well educated Haida
Indian who is at present engaged in missionary
work among the Indians at Nanaimo, Vancou-
ver Island. Mr. Kelly visited Ottawa in March,
1920, as member of an Indian deputation to
the Canadian Government and was too much
occupied to give me more than a few hours.
In spite of the brevity of my notes I believe
the insight gained into Haida phonetics is suf-
ficient to warrant this paper. I cannot, of course,
give an adequate account of the Haida sound-
system, but purpose merely to present data
supplementing Dr. Swanton's brief statement '.
The phonetic system employed in this paper is
explained in « Phonectic Transcription of In-
dian Languages » (Smithsonian Miscellaneous
Collections, vol. 66, n° 6).
A remark or two on the general impression
produced by Haida maybe of interest. I took
several opportunities to have Mr. Kelly speak
Haida connectedly and was thus enabled to
hear it long enough to form a definite image
ot its acoustic quality. It is one of the most
remarkable languages that I have ever heard.
Indeed, I cannot recall having at any time
heard connected speech that appeared more
definitely possessed of individuality. The great
frequency of nasal consonants {n, r,), the con-
stant occurrence of sonorous ^-sounds, the pro-
fusion of /-syllables (see below, p. 152), and
the musical cadences are probably the chief
determinants of this individuality. Haida is
very far from being a harsh language. On the
contrary, it was voted a beautiful language by
all who heard Mr. Kelly's recital of a Raven
myth. Several of us in Ottawa heard connect-
ed Mohawk, Tsimshian, Nass River, Thomp-
son River, Shuswap, and Danish at the same
time. If we were asked to rate these seven
languages on the score of acoustic appeal, I
believe the consensus of opinion would be a
division into loin' groups : Haida as an easy
I. See pp. 210-215 of I. R. Swanton, HuiJu, tin Illiis-
lialive Sketch (Bureau o( American luhnology, Bulk-tin
^O, pt. I, pp. 20)-2S2 [lc>10]).
152
VI American Indian Languages 2
I
144
first; Mohawk as a fairly pleasant, but none
too close, second ; Tsimshian, Nass River, and
Danish as a moderately uneuphonious third ;
and Thompson River and Shuswap as an exe-
crable last. These remarks are of no great scien-
tific value, but they may be of some interest
none the less as serving to bring home the fact
that the « harshness » of certain West Coast
languages results from the printed page rather
than from their actual articulation.
I. — Consonants.
Thk Consonant system. — Swanton recog-
nized 28 organically distinct consonants in
Haida. I believe his table errs in two respects :
in not including a number of sounds which he
recognizes as existing but does not consider
as elements of the fundamental sound pattern
of the language ; and in neglecting to take ac-
count of certain sounds that he did not hear.
To the former class belong the anterior palatals
and the labialized gutturals and velars. To the
latter class belong the glottal stop ' and a set of
glottalized nasals and semivowels.
As to the anterior palatals, Swanton remarks,
" An anterior palatal series might be added to
these, but the sounds to be so characterized
seem only palatals followed by a close [read
" front "J vowel. " I do not know if this is
originally true or not, but I think there can
be no reasonable doubt that the anterior pala-
tals are felt as a primary series. They occur
before /-, »-, and rt-vowels, which last they co-
lor to ci. It is true that in such a form as xdi
ARM one sometimes hears a slight z-glide (x'di),
but I do not think we have the right to
conclude that xdi is felt as a secondarily modi-
1. lam not referring to the glottal stop, if it is a
glottal stop, of Masset Haida, which corresponds to
Skidegate velar ^, but to a primary glottal stop in Skide-
"ate itself.
•
fied form of xiai, though it is of course pos-
sible that it may go back to such a form. At
any rate, I prefer to consider the anterior pala-
tal series as a phonetically well defined prima-
ry group of consonants and the /-glide, when
it occurs, as the secondary fact. I do not hear
gia''da.i the blanket, for instance, but gd''da.i.
As a matter of fact, Mr. Kelly's ear proved ex-
tremely sensitive on the diflfrence between the
k- and ^-series. The word for eyebrows was at
first recorded sk'd-'dji, but this pronunciation,
in spite of the a-vowel, did not satisfy him.
The correct form is sk'd''dji. There is even
reason to believe, if my record is to be trusted,
that there is a difference between the ^- and the
^-series before /-vowels. Thus, I heard go'xxg'H-
gx TO BE starting TO BURN, not -^i/- (misheard
for -^J- ?) ; t'a'gr'gx TO be through eating,
not -W"- as in p'sdi'r, two blanket-like
OBJECTS. However, I have not enough evidence
on this point to be justified in speaking with
confidence.
As regards the labialized gutturals and velars,
Swanton sometimes writes according to the
form Java (i.e. k'zva) to strike, at other times,
according to the form sgod'nsir, (i. e. sgwa''-
nsvr^ one. Here again I think there is no rea-
sonable doubt that we must look upon the la-
bial element (whether written lu, u, or in
Swanton's material) as a constituent element
of a primary labialized ^-sound. Counting the
anterior palatals, the two sets of labialized k-
sounds, and the new glottal and glottalized
consonants, we have 47 primary consonants in
Haida. They may be arranged as follows (see
p. 145).
Of these consonants, I did not myself obtain
an example of aspirated p but set it in the
table because of Swanton's form djA'pAt to
SINK SUDDENLY, which I interpret as dju'p'at.
It may, however, be really djxp'at (i.e. djxp -\-
\t; cf. X3.T,\' eye), in which case p' would
have to be removed from the table. Aside
Six: Athabaskan and Na-Dene Languages
153
145
Aspira- Glottal-
Intermediate ted ized
Stop Surd Stop Surd Stop
Voice- Glottalized
Glottalized less Voiced Voiced
Nasal Nasal Spirant Spirant Spirant
Labial
b
p'
m
'jn
Dental
d
t'
/'
n
'n
Palatal Sibilant
(AfFricative)
dj
tc'
t'c
Anterior Palatal
i
Js'
¥
Guttural
g
)i
k'
n
'■n
Labialized Gut-
tural
gw
k'w
k'w
Velar
S
f
>
Labialized Velar
gw
q'w
qw
Lateral '
dl
ti
fi
Laryngeal
'
X
X
xw
y
yw
i
h
w
w
from w and 'w, which are common, and
m, which is not rare as syllabic final, labials
are very uncommon in Haida. Most, if not all,
examples of initial m and 'w may be suspected
of occurring in Tsimshian loan-words. Besides
the 47 primary consonants that 1 have tabulat-
ed, three secondary ones must be noted : syl-
labically final p and /, which are true surds but
unaspirated, and spirantal velar y ; p and / are
secondary forms of b and d, while 7 is some-
times heard as a variant of g between vowels.
It is barely possible that we should distinguish
between guttural q and velar r,, but I did not
hear the latter sound.
The INTERMEDIATES. — By " intermediates"
are meant unaspirated " voiceless lenes, " con-
sonants pronounced with the voicelessness of
typical French surds (/?, /, k) but with the les-
ser energy of our sonants {b. d, g). To the ears
of English-speaking persons they are more apt
to sound like sonants than surds, while French-
men would be almost certain to set them down
as typical unaspirated surds. The Haida inter-
mediates are identical with the corresponding
series heard in Iroquois, Athabaskan, Takelma,
I. First three laterals are affricative.
Yana, Achomawi, and Miwok. A recent oppor-
tunity to study Mandarin Chinese phonetics
has made it clear that these typical American
" intermediates" are absolutely identical with
the unaspirated surds of Chinese, which sound
distinctly " softer "than the surds of French or
Italian. It is possible that the Haida interme-
diates are sonant at the moment of release, but
their general effect, if carefully heard, is cer-
tainly not that of sonants. They are here written
as sonants because it is convenient to adhere
to establisjied usage and because it is advisable
to. keep p, t, ^ for the " harder" secondary
forms that may result from them under appro-
priate circumstances. The iv of giu and giv
and the / o( dl are fully voiced. I supect that
the d o{ dl'is at least partly voiced also. A word
as to g. This is generally a very firmly pro-
nounced stop, hut it seemed to me that a faint
uvular trill could sometimes be detected as a
glide between the g and a following vowel.
Examples of intermediates are :
a''bx TO CHEW FOOD FOR A CHILD
dxr, THOU
sdx'lISvr, FOUR
k'u'dx LIP
djr'djW TO BE SPLIT UP
154
VI American Indian Languages 2
1^6
dja-'dy. WOMAN
t'/r EAR
digud/x'rga MY DAUGHTER
t'a-'gx TO EAT
giva'i ISLAND
ga''.'na bucket
ga'xa'' CHILD
gtva- sea-bird (sp.)
dla'i peace, quietness
After an accented short vowel Swanton
heard a / before dj and d. Hence he writes
klA'tdju and xA'tdjii small ' for what I heard
as k'xdjir ; similarly, klA'tdAla small ones ^
The firm, voiceless attack of the dj after a
markedly short vowel created the illusion of
a syllable-closing /. As his classifiers k !At- and
xAt- are illustrated only before dj, it is almost
certain they should be read k'a- and xol- (Swan-
ton's x and .V, it should be remembred, are my
X and X respectively). He himself writes xA-
in xA'dAla, the plural of xA'tdju '. That our
analysis is correct is demonstrated by k'oLSgwan-
S\:r^ ONE little OBJECT, k\sdir, TWO SMALL
objects. The point is of some interest for
N.idene, as it leaves Haida ^'a-, classifier for
small objects, identical with its Tlingit cognate
-k'x to be s.mall.
The unaspirated hard stops. — When b
and d appear at the end of a syllable, they sound
much more like our/) and / than do the ordi-
nary intermediate ^ and d. They impreess the
ear as the normal English p and/; in which the
breath release has been suppressed. Examples
are :
t\'pdjlf TO BE TALL AND STRAIGHT
dx'pdju TO BE VERY SHORT "»
1. Op. cit , p. 231.
2. Op. cit., p. 241.
3. Op. cit., p. 276.
4. Swanton gives t.'Ap- as a classifier for short and
protruding objecis (op. cit., p. 234), which does not
correspood to my data. He does not give dAp- as a clas-
di''gltga MY CHILD
q'e't SPRUCE (cf. q'e-'d-x to be spruce)
Sget RED
^i''x\t to. PICK UP (a canoe)
That the final stops are not aspirated is a
noteworthy fact in view of the common Ame-
rican Indian rule that final stops, whether in-
termediate or aspirated surd in origin, are re-
leased by a markedly audible breath (e. g Oji-
bwa, Tlingit, Nootka, Comox, Takelma, Yana,
Paiute). In this respect Haida difl"ers from its
remote relative Tlingit and is in accord, it
would seem, with certain Athabaskan dialects.
Goddard writes, e.g., Hupa Lit smoke, while
the Franciscan Fathers write lid in Navaho. It
is likely that 'an unaspirated surd is tneant in
each case. If I may trust my memory of Chasta
Costa, there too final stops are unaspirated, e.g.
l-t'at TO GO to pieces.
The aspirated surds. — These are "hard"
and markedly aspirated surds, much like our
English p, t, k when initial before vowels ex-
cept that they are even more strongly aspirated.
They are identical in every respect with the
aspirated surds of Athabaskan, Tlingit, Takel-
ma, Yana, and Chinese.
Examples are :
t'a- TO eat
st\ from
tc'i'nga grandfather
tc'i''djlC TO BE a BIG, BLOWN-UP THING
k'd'xwx OUTSIDE
Vmi POINT, NOSE
sk'w'djW TO BE SMALL AND ROUNDISH
k'wa'i HIP
q'a''dji' head
(J ' I Tig XT, TO BE SEEING
sifier. Apparently there was some confusion between
these rhyming classifiers.
5. Swanton defines the classifier tci- (op. cit., p. 227)
in terms of « such objects as full sacks and bags, pillow, s,
etc. ». Mr. Kelly stated Ic'i- referred primarily to the
blown-up stomach of a seal used asm float in fishing.
Six: Athabaskan and Na-Dene Languages
155
M7
q wa'i ROPE
tia'S BRANCH
tc'a'tidju'gx TO BE FAT
Swanton remarks, " It is doubtful whether
d and / [i. e. /'] and dj and tc [i. e. Ic'] really
exist as recognizedlys eparate sounds " ' . I do not
see how there can be any reasonable doubt on
this point, da' thou and t'a' to eat ; dji--,
classifier for cleft objects (like hands), and
tc'i'-, classifier for blown-up objects ', are as per-
fectly distinct as g and k' or g and q\ (Swanton
does not list dji'- as classifier. He either did not
isolate it as a classifying element or he confused
it with tc'i'-. Examples of its use, besides dji''-
djw already quoted, are sila'i dji'-sgwa''nsir,
ONE hand [literally, hand cleft-one], stki'i
dj'.-sdi'r, TWO hands, k'u'dx dji'-sd\r, TWO lips.
There is no doubt that there are plenty of
examples of dji'- in his material. Note, for
instance, dji'wAl [Ma.sset] roots of fallen
trees).
The glottalized stops and affricatives.
— These are the well known stops and affri-
catives pronounced with simultaneous glottal
and oral closure and with glottal release follow-
ing upon the oral release. Swanton says,
" Some speakers bring these out very forcibly,
while others p.iss over them with considerable
smoothness ^ ". Mr. Kelly pronounced them
quite as smoothly as any other consonants.
Their essential nature is certainly not to be
explained as due to " urging more breath against
the articulating organs than can at once pass
through 5 ". If there is a true " fortis " series
in Haida and Athabaskan, it is the aspirated
surds, which are indeed pronounced with an
excess of breath.
Examples of glottalized surds are :
1. Op. cit., p. 210.
2. Op. cit., p. 210.
3. Op. cit., p. 210.
t'a' gun feather
t'a''r,OLl TONGUE
st'a'i FOOT
/Vw red cedar
qoLuCcdoL cheek
q'a't'car, to be going in
^'d'll leg *
k'i''dji seal-stomach
k'a't DEER
k'w'dx LIP
k'wd'i TO WAIT
q'a'ri HEMLOCK
q'e'sdir, two spherical-like objects
qWCl'Tt TO BURST
fia''doLn gorge
C'tll WE '
I may note that/V tends to move front in its
position — either to that of an nnterior palatal
c- sound (/V) or even to that of an .f- sound (e.
g. i''knCsui3. dla'' msdir, two gigantic people ;
but also q'y.'nt'c:dx cheeks. Does t^adx corres-
pond to Swanton 's -djit (op. cit., p. 260) ?
The VOICELESS spirants. — These require
no special comment, x is pronounced like ch
of German ich. As Swanton remarks, s often
interchanges with dj^, which does not normally
occur as a syllabic final. Swanton states that " s
becomes dj before most vowels ". Inasmuch as
both dj and s occur between vowels (e. g. i'dji'
to BE : isdxr, TO BE CAUSING TO BE ; but also
" participial " -asi, -as'.\ would it not be better
to say that -dj- becomes -s just as -b- and -d-
become -p and -/ .'' We would then have one
final (-j) representing two distinct intervocalic
consonants (s- and -dj-).
Examples" of voiceless spirants are :
sdw'nsOLT^Xy. EIGHT
a'k'oSU THIS THING
xdi ARM
4. See p. 152 regarding syllabic /- sounds.
5. Op. cit., pp. 214, 215.
156
VI American Indian Languages 2
h8
t'a-'xidlgJ. TO BE ABOUT TO EAT
xa-'ya sunlight
xwi' to be cold
Xa' DOG
X7Lr,\' eye
xd neck
xiua-'igcr a thing that is loose
la\ ia, il i
igiinnl THREE
The nasals and voiced spirants. — Of the
three nasals, n and c may occur as either sylla-
bic initial or syllabic final. In such a word as
'l(i''(ar,a THEIRS the guttural nasal q must be
considered as belonging to the final syllable, in
such a word as xoc'r^'c eye to the first, m, as we
have already seen, is rare as an initial but not
uncommon as a final.
Examples of final m are :
t^'mdjtr TO BE something THIN AND ROUNDED
fy-m lice (Mr. Kelly considered this word
as connected with the classifier t\m-. This mat
be only a folk etymology, however.)
dld''nidjn' to be a gigantic person
ga''mdju- TO be a wide thing
^d'mdjir TO be a large (canoe, blanket)
Several Haida syllables ending in m seem to
belong to a set of classifiers :
t'xm- THIN and rounded
dlam- gigantic, corpulent (?)
ga'm- wide and rounded
^iim- large (in reference to canoes and
blankets)
igum- (Swanton) large and roundish (e.
g. rattles)
Fa;//- (Swanton) small and roundish
js,'rt'm- (Swanton) large around (?)
dla'm-, ga'm-, and gam- are not listed by
Swanton. My ga'm-, however, may be the
same as his ga-w- (op. cit., p. 235) ; I am cer-
tain of the velar g, for the element was also
recorded zs -(am-, e. g. ^w -(am-sgiva'nsu, one
big ear.
Although the evidence is far from complete,
I would risk suggesting that this set, of which
there are doubtless other, members, contains
a common element -;//- indicating something
like rounded, all around. This view is strongly
supported by the fact that several of the m- clas-
sifiers are clearly related to other classifiers
without -m or with final -p. Thus, to t'am-
may correspond t'a- (Swanton) coiled and
flexible; to ga'm- evidently corresponds ga'-
FLAT (e.g. pr -(a'sgiva-'nsir, one flat ear);
^dm- is clearly related to p'- canoe, blanket-
like (e.g. tin' p'sgiua'fiS'.T, one canoe and
^d''at grsdi'r, two blankets : tiw ^d'mdjus
A large canoe, tiw gd-'mdxh sdvr^ two large,
SPREADY CANOES, m'' ul gamdoiJa gi'sdir, two
large blankets) ; igxm- probably belongs to
iga- (Swanton) branching objects, k'xm-, as
Swanton himself points out, is derived from
k'x- SMALL ; and slla'm- belongs with sthp-
(Swanton) slim.
Parallel to the -m- series is a -p- series
(fo^p-, doip-, stixp- [Swanton], t'ioip [Swanton],
audita/)- [Swanton] = probably sk'oip-^. Of
these, tap- may go with t\m- ; stiap-, as we
have seen, with slia'iii-; t^ixp-, judging from
Swanton's one example, which refers to the
surface of the moon, goes well with his L!-
thin and FLAT (as I shall show later, this must
be interpreted as t'il-, a reduced form of /'fa-,
t'ia-); and skap-, which in his one example
refers to the curled tail of a dog, may go with
sk'a'- roundish (used of eyes, water-drops,
berries). The meaning of the -p- series is more
obscure than that of the -m- forms ; possibly
stubly protruding covers the case. Ond" sur-
mises that the complete set of Haida classifiers
is a more complex subject than Swanton's data
represent. Further, that the i- and s- which
begin so many of the classifiers (and other noun
and vtrb stems) are vestiges of an older classi-
ficatory system that was related to the /i- and
s: classifiers discovered by Boas in Tlingit. The
Six: Athabaskan and Na-Dene Languages
157
NOS. 3-4
whole subject of Haida classifiers needs a rene-
wed and intensive study.
Examples of n and r, are so numerous in the
forms scattered in this paper that I do not need
to give further examples here.
Unglottalized w does not seem to be as com-
mon a sound as glottalized w. Examples of w,
y, and /.are :
ha- Wit HURRY !
t'sCWx'ga TO BE A CEDAR-BOARD
yd-'goLlari ancestors
ga''yir sea
q'a'fcaya''g<xn went in
t'a'lx'rj WE
xd NECK
sk'a''s'oI%r, A round thing
The glottalized nasals and voiced spi-
rants. — The sounds coming under this head-
ing are identical with the corresponding sounds
in Kwakiutl, Nootka, Nass River, and Tsim-
shian. I have not heard the Kwakiutl sounds
of this type, but the Nootka series ('w, '«, 'w,
'j), the Nass River and Tsimshian series ('w,
'w, 7, 'w, y), and the Haida series ('w, *w, 'r,,
'w, 'y), sound perfectly analogous to me.
I hear no difference, for instance, between the
'i^ of Nootka W"}7' HIGH HILL, of Nass River
'wr GREAT, and of Haida 'wxsi' that thing.
I speak of this because Boas has differentiated
the 'm, for instance, of Kwakiutl from the 'w
of Tsimshian, which he writes m ! and consid-
ers a '' fortis " m analogous to " fortis " p ! .
(our p'). In classifying these sounds with the
glottalized stops and affricatives (*' fortes ") of
Tsimshian I believe he is perfectly correct, for
they all belong together psychologically, but
this grouping applies fully as well to Nootka.
When I first taught a Nootka Indian to write
phonetically and explained the meaning of such
symbols as p ! and / ! (for which I now write
/)' and /'), I was interested to find that he wrote
m ! of his own accord where I was in the habit
of writing 'm. He seemed puzzled to find that
I was not using an analogous orthography for
the glottalized stops and affricatives on the one
hand and the glottalized nasals and semivowels
on the other. This instance demonstrates pretty
clearly, it seems to me, that the native phonetic
feeling of Nootka finds the essential peculiarity
of the " fortes " in their glottalization and not
in their supposedly '* increased stress of arti-
culation , " for in such Nootka sounds as 'tn
and '« there seems to be no increase of stress.
In the glottalized stops and affricatives the
closing of the glottis lasts during the whole
oral articulation of the consonant and beyond.
In the glottalized nasals, semivowels, and voi-
ced lateral, however, the glottis is closed simul-
taneously with the oral contact but released
instantly thereafter, the voiced continuant the-
reupon becoming fully audible. The acoustic
effect, therefore, of such a sound as m is very
nearly of a glottal stop followed by ni-, yet
not quite, for a conscious compounding of '-|-
m nearly always fails to satisfy the Indian's ear.
That the glottal element is felt to inhere in the
consonant is clear from the syllabification. In
such words as Haida ga''tia bucket and fa'^na'
child the glottal stop belongs to the second
syllable, not to the first. In neither Haida nor
Nootka, as a matter of fact, can a syllable end in
a glottal stop.
So far as I know, '•/; has not yet been record-
ed for any other Indian language. It is not a
common sound, nor are '/// and ';/ frequent in
Haida;'///, as I have already indicated, probably
occurs mainly in Tsimshian loan-words. On the
other hand, \i\ '\, and '/ are exceedingly com-
mon sounds, appearing in some of the most
important stems in the language (e.g. \L'a-
that, 'j)7r- big, la he). How important is the
distinction between 'v and v, for instance, may
be seen from the fact that when I pronounced
Swanton's yu'An big as yir'xn, Mr. Kelly had
not the remotest idea what it meant; it should
158
VI American Indian Languages 2
no
have been ';y7r''an. The finding of these sounds
in Haida was unexpected. They are not found
in Tlingit and seem also to be absent in Atha-
baskan — with one interesting exception. In
Navaho the Franciscan Fathers have recorded
■;/. It is the regular correspondent there of an
etymological t -\- n.
Examples of glottalized nasals and voiced
spirants are :
'ma' (exclamation of pain)
'madjx OCHRE (Probably borrowed from
Tsimshian. Cf. Tsimshian niES- reddish, wE5-
ans OCHRE ; perhaps Boas' m'Es- is to be read
'W3tj-.)
ga''na bucket
W,a'g3,'r,ga'r, to vie with one another
'wa'ngx TO LOOK for food at ebb tide
'tt'aJi* THAT THING
'yu'"oin BIG
'la' GOOD
la', 'h, 7/ HE
The laryngeal consonants. Every stem that
apparently bet;ins with a vowel really begins,
in all probability, with either a glottal stop or
with /;. Examples ol glottal stops beginning
syllables are :
'a'dja SMEARED WITH PREPARED SOAPBERRIES
'fl'7/ PADDLE
'a'uga .MOTHER
'a''tlgxu HERE (Swanton writes d'LgAn,
but I heard //, not dl. This may be a mishea-
ring on my part.)
'(tllod-'a OURS
'l-7'.Y] MAN
7 'i''i:r,x ifc'dju'gi HE IS A STOUT MAN
na'g"'i -77 'a in the house
yif'txn BIG
{gu'nuf three
iH-fu'nui si.K
tia'"ai ten
sk'a's'olxT, A round thing
In a few cases the ^lottal ston was not record-
ed, e.g. t*7tr,/"iu/a 5iii-/; two MEN, ga'xa. i'hri<x
CHILD-MALE, BOY, but these are either mishear-
ings or secondary slurrings on the part of the
speaker.
There are, however, at least two important
elements that begin, or seem to begin, with a
vowel unpreceded by a glottal stop. These are
the demonstrative u, o (w , o') and the verb
i''dii\ i''dj'., a. Examples of the demonstrative
are :
di' sk'ddj u' v'dji it is my eyebrows
§d'gxn na'i {y)u my house (Note the glide
)', indicating clearly that no glottal stop has
been slurred, gd'gxn is probably assimilated
from gd'gxr,, see Swanton, op. cit., p. 259,
who wites gia'gAfi)
'la'o' he <C 'la o'
ia'o- I <iiao'
'WXSI' gO'Xxgl'lgTr, THAT IS STARTING TO
burn
na''gai yf ia q'a'tcar, i go into a house
a'k'os u this thing"
'ivx'sga.i sd:r, fs'.'zvx'gx those two are
CEDAR BOARDS.
The reason for this absence of the glottal
stop has been indicated by Swanton. The gene-
ral demonstrative of reference is haii (S>vanton's
hao), often contracted to ir , u, 0. The vocalic
hiatus without glottal stop is thus the etymo-
logical equivalent of an old h.
Examples of the verb i''dj\., '.5 are :
la 'o' na'.t -;a c'dj-.-ri i am in a house
tllgU".'' U tx isdxr, I CAUSE IT TO BE AWAY, 1
PUT IT AWAY
See also the first example under demonstra-
tive u, 0. The consistent absence of the glottal
stop in this verb may possibly be explained as
a slurring, but I think it more likely that it is
to be interpreted as due to an old /; that has
disappeared. I would suggest that i'dj\, is is a
I. a- slurred from 'a-, the demonstrative stem this
corresponding to ';tw- that.
Six: Athabaskan and Na-Dene Languages
159
151
later form of hi'dj'., his, and that this verb is to
be understood as composed of a demonstra-
tives stem hi'-, hi- (cf. hi''-dll-goLl come here !
and Swanton's hi-t!A-gA'n then, hi nan
only). Justasthis /;/- parallels /;fl-(cf. Swanton's
hx-n LIKE, AS FOLLOWS ; general demonstrative
hao^if , u,o; " article "*hai, parallel 10 gai, in
e. g. *na hat house-the > na'i), so *hi-dji,
*hi-s TO BE parallels an old *ha-s, preserved as
Swanton's " participial " -as, -is, -es. Such a
phrase as Swanton's nxn gaxd'gas one who was
A child is to be interpreted, it seems to me, as
«ar( gaxa-'-g a-s one child-be it-is, contracted
from an older gaxa'-ga has. Such endings as
-^-5? (S wanton) are probably to be interpreted
as -ils\ <Z -l-y-S'; i-asi, which occurs as phonet-
ic parallel. See below on /- syllables.
Though /; has demonstrably dropped out in
certain cases in intervocalic position, it is a
clearly articulated consonant when preserved,
e.g. :
hv'dllgy.1 COME here !
haw't HURRY !
Secondary consonantal processes. Final
vowels are, as a rule, released without breath.
This is in keeping with the phonetic forms -p
and -t instead of/)' and t . A few cases of breath
release have been noted, however, in absolu-
tely final position :
na'' house (but na t'c'-sdir, two houses)
go'gx' (it) is on fire
go'dxgy.' sets fire to
Rather frequent is the spirantal voicing of g
to velar y between vowels :
dx'r,^d^a yours
'la''(a'r^a ^d-ra theirs
'zua'n-^oi. TO GO down to look for food at
ebbtide
da la '(O'da you start the fire!
na''gai yc into a house
I. Op. cit., p. 254.
ga-t'a-' 'hfai"';a something-eat keeper,
eater
d j c' go' '.;7i seven
gw -(%'sgu<a''nsvr^ one (flat) ear
q'o''-(a rock
Less often g is spirantized and unvoiced after
initial / :
ixa' stone (for iga')
Before s, t\, n, and i there is sometimes assim-
ilation of r, to n :
sdx'nsir, four <C *sd7:r,-sd<.r, TWO-TWO
r'hnfsuix dla''msdvr, two gigantic people
(cf. i-'hT,a man)
dxn II I THEE. . . (Irom dxr, iV)
gd'gxn na'i my house
But this rule is not invariable : -y; is preserved,
e.g., in :
'r'/i-/; fsda. dx'psdir, TWO STUBBY MEN
On the other hand, n does not assimilate to
r, before g :
tc'i'nga grandfather
'llgx'nga cousin
Initial consonant clusters.— Every Haida
syllable begins with one of the forty-seven
consonants listed in the table, or with a cluster
of two consonants, or with a vowel (originally
preceded, it would seem, by /;). The first
element of the cluster is always i- or s-. The
second element is an intermediate, aspirated, or
glottalized stop or (after 5) a //, never, I believe,
a sibilant affricative or a spirant ; ix- is merely
a secondary form of /^^-. Examples are :
igi'- (classifier for fairly big, roundish objects)
iga' STONE
igu'nid three
Ik'a'i chin
sdir, TWO
sget red
sgiva''ns:r^ ONE
st'v'wai THE sea-egg
sk'a'- (classifier for round objects)
sq'aii GROUSE
160
VI American Indian Languages 2
152
St a i FOOT
sq'a- (classifier for long objects)
I obtained no example of i before a glottalized
stop; Swanton gives Ik'.A'mAl needle of co-
N'lFEROUS TREE. I suspcct that his in- is really
llr-, as it would be an isolated example off
(or s) -\- nasal. Swanton fails to distinguish
initial sd- from j/'-. He writes both st-. An-
other error is his group sL !-. It may exist, but
his instrumental sL!- with the fikgers and
sL!di HAND are really stll- and stia'i, probably
related, with ^-prefix, to ///- to touch (Swan-
ton's L-, p. 226).
Swanton's clusters with initial /, dl, li, and
/'/ do not really exist. These /-sounds are in
every case to be interpreted as syllabic //, 'II,
dll, til and /'// (see below).
II. — The syllable.
Before taking up the vowels, it will be con-
venient to define the Haida syllable. There is
no doubt that the language, like Athabaskan,
has a strong and well-defined feeling for the
syllable as an integral phonetic and psycholo-
gical unit of speech. It therefore becomes
important to understand its structure. Aside
from the secondary loss of h and the slurring
of ', every syllable begins with a consonant
or an 5- or /-cluster of two consonants. It
may end in a vowel (long or short), a diphthong
(long or short, but I suspect that all long
diphthongs are felt as the equivalent of two
syllables of form *-ahai, *-ahau), an /-vowel,
a consonant, or a cluster of two consonants
ending in / or s (these clusters are likely,
however, to be secondary forms of older disyl-
labic forms, e. g. i'djins [Swunion] <^i'djiT,7Ls).
The final consonant can be only rn, n, yj, 5,
unaspirated p or /, /, /, or //. All of Swanton's
examples of final dl and /'/ are to be inter-
preted as dll and /'//. Swanton speaks of non-
vocalic stems like " L to touch " or '' sL to
PLACE IN A certain DIRECTION. " These ele-
ments are syllabic : ///, stiL Many of his non-
vocalic groups are even disyllabic, e. g. /'//'//
(Swanton's Z,//).
Syllables with l-vowel. — In normal
English pronunciation the second syllable of a
word like w^/a/ consists of a consonant follow-
ed by an /-vowel. In other words, there is
no pure vowel in the syllable at all, not even
a " mid-mixed " d. The tongue does not
release its stop position but merely adjusts
itself on the spot to a lateral articulation. Such
words as metal, medal, flannel are phonetically
mi'tl, mz'dl, fld'nl.
The peculiarity of the Haida /-syllables, of
which there are a vast number in the language,
is that they always begin with a lateral con-
sonant, voiced or unvoiced. The following
table shows the relation between the simple
laterals and the syllables with /-vowel :
Syllable luith y.-vcnvel Syllable with l-vowel
I- n a-,
*/« '// ('/•, '/)
dU dll {dl, df)
h il
tix til
fix t'il
The syllables in the second, column are
reduced forms of those in the first, with which
they vary in Swanton's orthography (e.g. Lga
or LA'ga land, both to be understood as
lil'ga). They also interchange actually accf>rd"
ing to accentual or other phonetic circumstances
with full syllables in a', a, a (e. g. la', 'la in
'In' .va'r/'." HIS EYES, na''i •;'. 'la ga't'car, HE is
GOING INTO THE HOUSE : '/ 'i''ii.r,7. tc'r'djw HE
IS A BLOWN-UP (obese) MAN ; /fl I : // ; -dxloL
adjective plural : -dj.il). That Swanton too
heard, though he did not explicitly record, /-
Six: Athabaskan and Na-Dene Languages
161
1)3
syllables is shown by such accentuations as Z,'-,
/'-, I'-andL!'-.
Examples of /-syllables are :
k^all LEG
da''ll RAIN (recorded as dal by Swanton,
which failed to satisfy Mr. Kelly's ear. This
word is not only clearly disyllabic, but the
accented vowel is higher in pitch than the /-
syllable. Cf. Swanton's ddla-ge'ii-si rain fell,
BBAE 29 : 12, 1. 8.)
'a'll PADDLE
qui'' FOREHEAD (should have been record-
ed q'u'Il or qui')
'IJgx'nga COUSIN
'h;a'i'^';a keeper
dx'n U YOU I. . . (not to be confused with
dxnl swelling)
Sge-'il TO CRY
titil TO RUB one's hand against
tlc'll five
q'w'dll to go aboard a canoe
hi'dllgal come here!
tilga' earth
tilgWV' AWAY
til^u'nui SIX
t'il'^d' to soak (a dry salmom)
'i''t'ilgd-fa ours
When an /-syllable beginning with /, 7,
or dl is pronounced with very weak stress,
the length of the / is of course reduced, though
it never loses its syllabic character. We may
then write simply /, '/, dl. The syl'able -ilr,
(reduced from -/ay;) was heard as -ir,, with syl-
labic q, in lia'^ait, (i. e. tia'"al avj) sgu'a''ris:T,
gou ten it-is one missing, nine.
In interpreting the phonetics of Swanton's
texts, there is no genuine syllabic ambiguity
in the case of initial L-, L-, L!-, and /- before
consonants and of final -L and -L/, which
necessarily represent /-syllables. It is different,
unfortunately, with initial / before consonants
and with final -/, -/, and -L, which may repre-
sent non-syllabic or syllabic laterals. It is im-
possible to tell offhand whethersuch orthograph-
ies as igeda-i and ga'lxua represent igc'daJ
or ilge'da.i, ga''lxu'a or go'lhwa. It is hardly
conceivable that the morphology of Haida can
be adequately understood without an exact
knowledge of its syllabification, for the Haida
" word " is essentially a group of significant
syllables. The recognition of /-syllables is likely
to put many points of grammar in a new light.
Thus, we may surmise that the two forms
of the first person plural (subjective fa-la-r,
and objective ''/'-t'il) are closely related, ''/'-til
being a reduced form of *\''-t\xya. The true
basic forms would then be fa and 'c-fa, -la
and -Ix-r, being pluralizing elements.
Syllabic nasals. — It is quite possible
that we should recognize also syllables of type
n and syllabic q, alternating on phonetic
grounds with syllables of type w. and r,a. The
alternation of-r; (-n) and -/;« in cases like i"7.-/;
(or i''i:.n) fsdx sdvr, two men
and
i''hr,x sgiua''nsiri one man
should perhaps be interpreted as an alternation
of i"'/iy;y; {'.''hnn by assimilation) and i'7tr,a.
This consideration may explain the constant
interchange in Swanton's material of forms like
-gAn and -gAni, -agAri and -agAni, -gin nnd
-gtni, in which -/ (i. e., -i) can hardly be a
" perfective " element, as Swanton assumes.
For the present, I cannot say whether Haida
distinguishes final -n and -r, from -n and sylla-
bic -Y). All I can say is that I feel strongly that
while Haida reduces the quantities of syllables
freely (e. g., in such a gamut as 7^/', 'la, 7a,
'//, '/). it resists the extinction of syllables, and
that if syllables ar.^ actually lost, they are lost
phoneticallybut not psychologically.
HI. — Vowels.
I am able to give only a preliminary idea
162
VI American Indian Languages 2
1)4
of the Haida system of vowels. The vocalic
nuances seem to be due primarily to secondary
phonetic causes rather than to basic etymolo-
gical differences. It is quite probable that there
are only three organically distinct vowels :
a, z, //. Each of these runs through a gamut
of quantities and qualities that give the lan-
guage a far greater vocalic variety than the
simplicity of the fundamental vowel scheme
would argue.
Qualitative changes. — The a-vowel
seems to be the most liable to phonetic
change. Its fundamental quality is that of
German a in Mann, e. g. da''ll rain, but
when short, whether accented or not, it very
frequently assumes the duller timbre of a.
Thus, -ga TO BE and fa i vary with -gx and
// (i. e. /x) in :
ia ga l'a''s[a i am eating something
a'k'os u il l'a-''(xsgx i shall go and eat
THIS thing
The a-timbre is particularly common before
nasals, e. g. -gx-r, (continuative of -ga^ and
-^a-n('.) (past definite oi -ga), /'aw lice.
(It seems fairly evident to me that Swanton's
temporal -gxn [op. cit., p. 247], future -sga
[p. 249], and continuative -gah are merely com-
binations of " declarative " -ga to be or to be
so and certain particles that he has not properly
isolated : -r,, -r,a to be continuously ; -s, -sa-
ABOUT to bk ; and -n -««. to be at a given
MOMKNT OF TIME, TO HAPPEN. It is clear that
they occur also without -ga. Boas has already
pointed out the analysis oi -s-ga- [see p. 249].
I hope to show at a future opportunity that the
whole tense-modal system of Haida is nothing
but a loose compounding of demonstrative
elements and particle verbs and that the synthe-
tic nature of this scheme is more apparent than
real. Thus, Swanton's " infallible future
-asah [p. 249] is merely a verb phase : 'a-sa-r,-
[a] this-will-be [durativkly]).
After anterior palatals and y, a (a) appears
as a (as in English mat), perhaps even as z (as
in English met), e. g. :
k'd'xwx outside
^'d'll LEG (also heard as k'e'l)
Xdi ARM
t'il'^d- to soak (a dry salmon)
sk'd-'dji eyebrow
^d'gxn MINE
^d''da.i THE blanket
^dm- wide and sprhady (classifier)
yd''gxlxq ancestors
After sibilant afFricatives and also before r,
there is a tendency for x, particularly if unac-
cented, to pass into i (English i of bit).
Before yj this seems to take place particularly
after laterals (/, /) and sibilants. Examples are :
sdvr, TWO (d. sdxnsi-f] four, dissimilated
from sdx'n-sdvr, <C *sdxr, -sdxr,)
daUr,^d''(a yours (cf. dxixr, ye)
There is no doubt that many Haida syllables
in I are old a-syllables, as variants prove ;
e. g. 'r'hrix MAN is evidently to be understood
as an original 'i''ixr,x (perhaps actually pro-
nounced 'i''iir,x, as Swanton's variant ff/^a
seems to indicate), as shown by certain of
Swanton's forms, e. g. iia'n-, iia'h-.
After anterior palatals there is good reason
to believe that an original a is sometimes com-
pletely palatalized to r, i. This is not a mere
matter of nuancing, like the change of a to a
or of X to I, but a definite phonetic process that
disconnects the new vowel from its old catego-
ry. Under what circumstances this change
takes place I do not know. I suspect that an
old a' is merely colored to d' {a) after anterior
palatals, but that the corresponding short
vowel was completely palatalized to i (secon-
darily also /■). A good example is ^i'- (clas-
sifier for blanket- like objects), which is almost
certainly related to gd'-, gd- of gd''-d- blan-
ket and ^d-m- (see page 148). The original
quantitative relations were probably as follows :
Six: Athabaskan and Na-Dene Languages
163
1)5
^a- > gi--
ga-d- >■ gd'-d-
^a'-m- >> gdui- (closed syllable, hence with
shortened vowel)
Again, g'rw-ai the ear : ^w ear can be
best explained as palatalized from an old ''gaw-,
final -an o{*gaii contracting to gw . This inter-
change of -/'tf- : -w seems to take place also
after s, e. g. Skidegate sfi (Swanton) to say :
siw-As (contrast Masset sdw-^. A basic saw-
best explains the alternations. Another alter-
nation of similar form is that of tin' canoe :
tilw-a'i the canoe (Swanton always writes
Lu-a'i). This again is doubtless the reflex of an
old tlaii : tlaiu-.
For the /-vowel, /, i, and e were heard as
variant timbres. The t-timbre is quite frequent,
perhaps normal, for the long /- vowel, e. g. :
'c'til^dya OURS
A-a'r/f EYE
tilglUf' AWAY
djr'gO'-(OL SEVEN
'.■ ilTfT. MAN
Examples of the less common ^-timbre are :
c'dji-q IS
xei REGION OF THE NECK (cf. xd neck).
The characteristic timbre of the u-vowel is
u (as in English put), here written u. The cor-
responding long vowel was heard partly as o'
(e. g. go'dx TO START A eire), partly as w
(as in English fool). The latter seems to occur
chiefly after sibilants and anterior palatals
(e. g. -dju- TO BE so AND SO, gU' EAR, 'y/r"Qt«
big).
A glide a was noted in 'lyai^-^a keeper.
In ^d''at BLANKET, the long a vowel, modified
to d after g, reasserts its proper quality before
the final / and thus appears as a broken vowel.
Vocalic quantity. — Quantity is a diffi-
cult matter in Haida. It is likely that there are
etymological quantitative distinctions, but it
is impossible to be certain, as the actual quan-
titative variations are clearly largely due to
secondary lengthenings and shortenings of
the fundamental vowel. Thus, as already
pointed out, 'la he may be shortened to 7a,
'II, '/or lengthned to 'la' . The determinants of
this quantitative variation are probably pho-
netic rather than morphological, but I doubt
if stress accent is the only or even the decisive
factor, as both the long and the short vowel
may occur in an accented or unaccented syl-
lable. I suspect that the distribution of quanti-
ties is the resultant partly of inherent quanti-
tative distinctions (e. g. long a' in yd''gxhr,
ANCESTORS : short a, a in k\'dju' to be short),
partly of a tendency to establish a rhythmic
equilibrium. This equilibrium seems to depend
on several factors, chief among which are the
nature of the syllable (a closed syllable tends
to be short, an open one long), the place of
he accent, and the grouping of the syllables
in phrases. It is impossible to give rules at
present ; the subject is evidently complex. At
the same time I do not feel that the quantities
are distributed ad libitum, rather that they
remain to be discovered.
The Haida type of quantitative variation is
somewhat similar to that of Athabaskan, if
I am not mistaken. Presumably the Tlingit
feeling for quantity is analogous to that of
Haida. Boas merely remarks that " the quan-
titative value of Tlingit vowels varies conside-
rably " '. In the body of his grammar he unfor-
tunately pays no attention to the subject of
quantity. The quantitative peculiarities of the
Nadene languages are in striking contrast to
those of the neighboring languages. In Es-
kimo, Wakashan, and, I believe, in Tsimshian
as well, the inherent quantitative value of a
vowel is clearly felt and, on the whole, rigidly
preserved. In these languages vocalic quantity
I. F. Boas, GiamiiiatiCid Notes on the Laiu:;iiaQy of llie
Tliu^it Indians, Univ. of Pa. Anthr. Publ., vol. VIII,
no. 1, 191 7, p. ; I.
164
VI American Indian Languages 2
156
is as much a matter of etymology as is the
consonantal framework of the word. Phonetic
variations in the quantity of vowels such as
are found in Haida would be well nigh un-
thinkable in Xootka.
Examples of quantitative alternations are :
a — g^i'-y^'' CHILD : ga'Xx i''i':r,x CHILD
MAi.K, boy; gaxx .ija'dx child female, girl
s^'d''dji' EYEBROWS : di' sk'ddj w v'dji'
IT IS MY EYEBROWS
sda-'nsar,xj. eight : sdy.'ns'.r, four
na-' house; na'ga'i yfl 'a in the house :
na I'e' 'sgwa'nsvr, one house
sgwa''nsi.T, OYiE ; q'a-'dji q'e'sgwa-'nsir, one
head; k'un dji'sgwa-'iisir, one nose; k'nn
k'xsgwa-'uy.r, ONE LITTLE NOSE, POINT ; ou'
''^d-sgwa-'ns\r, ONE (flat) ear : gir -'amsgwa'n-
S'.r, ONE BIG EAR ;^M' k' x'sgwanS:r, ONE LITTLE
E\R;m-'dai ^rsgwa'nsir, one blanket, shawl
i — ; q'a''dji head, hair of head : q'a'dji
7/ — ; la-'o nw'gai vt- q'a't'car, he is going
INTO A house : na''i vi 'la q a't'car, he is
GOING into the HOUSE
dji-'djir TO BE SPLIT up, CUT UP : stia'i
djisgwa-'ns'.r, one band ; stia'i djisdir, two
HANDS
/•V/V" TO BE, e''djirt TO BE (duratively) ;
ia' '.sdxr, I CAUSE TO BE, PUT
// — sk'a''dju' TO BE SMALL AND ROUNDISH ;
^i'dju-'gx (it) is LONG AND MASSIVE, SPREADY;
tilwa'i u gdmdju'gxn the canoe is large :
xd k'lf'dJH neck IS short; dx'pdjti very
short; litr gaiiid/us a large canoe
di'sk'ddjw MY eyebrows; ia''o' la'gd^'a na'i
godxgx I PUT HIS HOUSE ON FIRE ; a'k'os
U {( l'a''gx I AM EATING THIS ; '■lVx!s\' gO'Xx-
g'.'lgxr, THAT IS STARTING TO BURN
Diphthongs. — There are two diphtongs
in Haida, (7/ and an. Each of these exists in
two quantitatively distinct forms — short (a/,
often contracted to e' ; an, ou^ and long (a'/, a.i,
sometimes heard simply as ^Z; a'o\ a'o, ao').
The latter type seems to result from contraction.
The uncontracted short ai is well illustrated
by the enclitic articles -ai (probably <; *hai),
when appended to a stem ending in a con-
sonant, and gai, e.g. :
St'i''iuai THE SEA-EGG
na''gai y<.' o into a house
According to Swanton ', the Skidegate ai
appears in Masset as e', but Mr. Kelly, a Skide-
gate Haida, pronounced a number of fl/-forms
with ^"-vowe!, e.g. :
qe'- classifier for large roundish things (cf.
Swanton s q!ai-, p. 232)
yd''ge' PARENT <^ yd'' gai (cf. T^\\iv.yd''gxlxr\)
t'e'- classifier for bulky lying objects (cf.
Swanton's t'ai-, p. 227)
q'e't SPRUCE (cf. Swanton's^flf/TREE, p. 271)
tic'il FIVE (probably contains iia- ; cf. tlx,
til- TO touch, s-tla hand)
Possibly the Masset-like g'-forms are gaining
currency among the younger people at the
expense of the older az-forms. The contracted
e' is, of course, not to be confused with the e'
which is merely a variant of r.
The long a'i (a.i) probably always results
from the contraction of fl -f- ai, e.g. :
na'i THE house <C na-(h)ai
sCa'i foot <C st'a-{h)ai (cf. Swanton's ins-
trumental sl!a- by kicking)
dia'i auiETNESS < dla-(h)ai
Just as the primary ai may contract to e' , so
it is likely that a'i may contract to ai. Thus,
it seems probable that ^d''da.i the blanket is
to be understood not as gd't -\- ai but rather
as contracted from ^d'da-{hyn ; q\r'-(ai the
ROCK, as contracted from q'o'''(a rock-|- (h)ai.
Similarly, xdi arm >< xd'i < old *xa-(h)ai ; *xa
itself passed to xi (Swanton's xi^) as *^a passed
to^^r'.
The short ^/<-diphthong is written ao by
1. Ot). cii., p. 213.
2. Op. cit., p. 226.
3. See page 155.
Six: Athabaskan and Na-Dene Languages
165
f$7
Swanton. I heard it as an ordinary a -\- u and
did not feel that the labial element was low-
ered to o-position. Examples are :
'a'uga MOTHER (Swanton's aoga^
sq'au GROUSE (Swanton's sqao)
It seemed to me that it was rather the a that
tended to become modified in the direction of
u ; thus, I heard gm to be wanting (in tia-"air,
sgwa'' ns'.Tt gJu nine) and 'o''tiga, almost 'o''ga, as
variant of'a'uga. This tendency o( au to con-
tract to a monophthong is carried through,
e.g., in ga''yu' sea (cf. Swanton's gayao,
p. 272), ^tf ear (*^aw, *gau, and in w, 0', u,
as reduced form of demonstrative hau.
The long a'o, also heard as wo' and ao', is
markedly distinct from- short au. It seems to
result always from the contraction of a -|- (/?)
au, e.g. :
'la'O', 'la'o HE (emphatic form ; «< 'la-[h\aii)
ia'o', lao' I (emphatic form; <C la-\h\au)
IV. — Stress and Pitch.
I was not able to arrive at definite conclu-
sions in regard to Haida stress and pitch and
their interrelations. It is clear that a given
word is regularly accented on a certain syllable.
Generally this is the first, e.g. r7'.v;a man,
go''xxgxr, (it) is flaming, 'yw'djir to be large ;
but the accented syllable does not need to be
the first, e.g. ga'xa'' child, ha'tv'.'t hurry!
Further, it is evident that the stressed syllable
may lose its stress, as it reduces its quantity,
in a given setting in the sentence or when the
addition of one or more syllables changes the
rhythmic pattern of its syllables. Thus, beside
go''xoLgxr, we hive go'xxgx'r,i^oi. (rr) is burning,
go-yxgilgx (n) is starting to bukn ; beside
/'fl"'^a TO EAT and a'k'os ti it t'a-'xidigxn i was
ABOUT TO EAT THIS we have i'l ga t' a'g'r'gyji i
HAVE FINISHED EATING. These alternations of
stress may be purely rhythmic phenomena for
the most part, but I doubt if they are entirely
so. Functional alternations seem to be illus-
trated in g(i''yu' SEA : ga'yw' smoke ; q'a''(ij\
HAIR OF HEAD : q'a'djx HAIRY. The same -a'
PROVIDED WITH is fouud in 'madja ochre and
nia'doL HAVING mountain-sheep, but we find
also 'a''dja smeared with soapberries.
The question of stress is complicated by that
of pitch. The stressed syllable is higher in pitch
than the other syllables of the word. At the
same titne it seemed to me that a low-pitched
syllable might very well bear a secondary stress
so strong as to bring up the question whether,
after all, what I have been making stress in
this paper is not primarily a matter of relatively
high pitch, only secondarily one of stress.
Though I should not like to commit myself
at present, I consider it very likely that pitch
alternations are a primary factor in the dyna-
mics of Haida, though it is perfectly clear that
a given syllable is not uniformly high-pitched
or low-pitched. The actual musical effect ot
Haida is so marked and the sonority of the
more weakly stressed syllables so great that the
operation of a pitch principle is to he looked
for. Yet I cannot say that I succeeded in finding
one beyond doubt.
A few sentences will have to do for the
present to illustrate the alternations of pitch
observed. A high pitch is indicated by an acute
accent over the vowel (a), a low one by a
grave accent over the vowel {a). A fiilling tone
(^) was observed only in 'nid' (exclamation of
pain). If there are significant pitch alternations
in Haida, they are probably of the simple Tlingit
high-low type observed by Boas. In the follow-
ing e\amples, the un^iarked syllables following
a high-pitched syllable did not impress the ear
as being as definitely low-pitched as the one so
marked. The pitch of unmarked syllables after
a low-pitched syllable is low. A stressed syllable
wliosc tone was not markedly high or low is
166
VI American Indian Languages 2
n8
inaicatcd by an acute accent after the vowel (<7').
la\i^ii'';a tia-i ge 'la (]'(rt\:ayn-'gy.ii he went
INTO HIS HOUSE
}:agan iia-i ()•)» go'g^-' my house is on
FIRE
'la-0- gligxii na-igo-dxgx' he puts my house
ON riRE
hro- la-gd'(a na't go-drgx' i put his house
ON FIRE
dd in ■;d-dx you start the fire!
girxxgrrigx IT IS BURNING
gi'rxxgilgy. it is starting to burn
'vjxsi go-xxgUgxr, that is starting to
BURN
iid-gdi '(d \i IN the house
ido- nd'i -.-a c'djir^ i am in a house
nd'gai ^'Z o la qa't'car, i GO into a house
'Id-Q na-'gai yf q'd'fcxr, he is going into
A HOUSE
nd-iyi 'Ja q'd'fcar, he is going into the
HOUSE
id ga t'd'gd i am eating
a'k'os u U t'd-gd i am eating this
tilgiui u ly. isdxTt it put it away
'lUTiSga.i sdir, o t'sVwigx those two are
cedar BOARDS
tlVwdi U djl'-qgxr, THE CANOE IS LONG
X'.'l sq'd'sdir, TWO (long and narrow)
NECKS
Editorial Notes
Originally published in International Journal of American Linguistics 2,
143-158 (1923).
There are two, highly divergent dialect areas of Haida: Northern Haida, a
group of closely related dialects spoken in the northern Oueen Charlotte
Islands, particularly at Masset, and in adjacent parts of Alaska; and Southern
Haida, a dialect chain, the best attested member of which is Skidegate. Sapir's
informant spoke Skidegate, which was the focus of Swanton s study (1911) and
has more recently been the subject of extensive work by Levine (1977). The
Masset-Alaskan dialects (Lawrence and Leer 1977; Enrico 1986) have a mark-
edly different phonology from Southern Haida, including phonemic tone and
two pharyngealized consonants (see Krauss 1979: 839-840).
Sapir's ms. corrigenda to his copy are as follows (note that we use the char-
acter g for Sapir's slashed or underlined g):
Six: Athabaskan and Na-Dene Languages
167
riginal
For:
Read:
144, col. 2, 1. 41
djap'at
djap'at
146, col. 1, 1. 19
remembred
remembered
147, col. 2, 1. 16
we'
we^
147, col. 2, 1. 21
q'dnt'cida
q'dnt'ctda
148, col. 1,1. 11
n and c
n and r)
148, col. 1, 2nd last
7am-s^wa'nsLr)
7am-sgwa'nsir)
148, col. 2, 1. 36
stubly
stubby
148, col. 2, 1. 36
On
One
149, col. 1, 1. 11
g
§
149, col. 1, 1. 16
sk'a-'s'olocr)
sk'a-'s'olar)
149, col. 2, 1. 26
and t'd'na'
and Nootka t'ana'
150, col. 1, 1. 15
^a-'na
ga-"na
150, col. 1, 1. 36
nag^ . . .
na-ga . . .
150, col. 1, 1. 38
tgu'nul
Igu'n'ul
151, col. 1, 1. 23
haw-'t
hawL-'t
153, col. 1, 1. 14
'llga'nga
'llga'nga
153, col. 2, 1. 18
syllabic
syllabic
153, col. 2, 1. 25
L-'kr)a
L-'liga
154, col. 1, 2nd last
'a-sa-g-
'a-sa-r)-
154, col. 2, 1. 25
il,r)a
il'rja
155, col. 1, 1. 19
'f'tt...
'L-'t't...
155, col. 2,1. 5
'11,
'U,
155, col. 2, 1. 20
he
the
155, col. 2, 1. 23-24
they remain
they obey rules which
remain
156, col. 1,1. 13
t'e- 's...
t'e-'s...
156, col. 1, 1. 20
i — ;
i —
156, col. 1, 1. 21
71—,
[delete]
156, col. 1, 1. 25
one band
one hand
156, col. 1, 1. 27
(duratively);
(duratively):
156, col. 1, I. 34
di-s . . .
di- s . . .
156, col. 1, 1. 34
la-'o-'la- ...
la-'o- 'la-...
156, col. 1, 1. 35
fire;
fire:
158, col. 2, 1. 2
qat'carj
q'at'cari
158, col. 2, 1. 9
it put
I put
PITCH ACCENT IN SARCEE, AN ATHABASKAN
LANGUAGE,
Introductory
In the summer of 1922 the author of this paper undertook a field
study of the language of the Sareee Indians, now located at a reserve near
Calgary, Alberta, Canada. The Sareee and the Kiowa Apache were the
only Athabaskan (Dene) tribes that descended into the Great Plains and
assimilated the distinctive buffalo-hunting culture of that area. Both of
these tribes were affiliated with more powerful neighbors — the Kiowa
Apache with the Kiowa, the Sareee, in a less formal way, with the three
Blackfoot tribes. Like other Athabaskan languages, Sareee shows prac-
tically no influence from neighboring languages of alien stock. The num-
ber of Blackfoot, Gree, or Stoney (Assiniboine) loanwords in Sareee is
practically nil, while the morphological influence exerted by these lan-
guagjes is entirely nil. Sareee is thoroughly Athabaskan in its sound-sys-
tem, its morphology, and its vocabulary. It has developed distinctive
dialectic peculiarities, but these do not remove it appreciably from the
linguistic companionship of Navaho, Hupa, Carrier and Chipewyan.
Perhaps the most interesting fact that emerged from the summer's study
w^as the presence in Sareee of a well-developed system of pitch accent or
tone. The interest of this fact far transcends Sareee itself, for so funda-
mental is tone to the phonetic and morphological understanding of Sareee
that it is inconceivable that it should not be shared by the other Atha-
baskan dialects as well. As I propose to show in a moment, we have
some good evidence in the literature on this point, though very little
explicit information has been given on tone in the large body of Atha-
baskan grammatical, lexical, and text material so far published. That
Athabaskan is a tone language is of great comparative interest because
of the existence of tone in Tlingit (S. Alaska), ^ a language to which
Athabaskan is genetically related, though only remotely so '-.
1. See F. Boas, Grammatical Notes on the Language of the Tlingit Indians (University
of Pennsylvania Museum, Anthropological Publications, vol. VIII, n° I, 1917).
2. See E. Sapir, The Na-dene Languages, a Preliminary Report (American Anthro-
1 70 ^f American Indian Languages 2
186
As early as 1876, Father E. Petitot * remarked of certain apparent
Chipewyan monosyllabic homonyms: « Tout depend de I'accent ou esprit
que Ion donne aux consonnes et du ton qui accompa^ne la prononcia-
tion. Par exemple, sha [i. e. 6a1, prononce sur un ton eleve, signifie
longtemps, tandis qu'il veut dire martre lorsqu'on le prononce sur un ton
plus has. » This is clear and suj^gestive enough, but unfortunately Father
Petitot ignores tone in his grammatical sketch and dictionary . A later
student of Chipewyan, Father L. Legoff '^-^ is even more explicit. He
states : ^ « Tout mot, meme toute racine, en montagnais, a sa note ouson
ton bas ou eleve qui en determine la signification. Et ces mots, ces racines
conservent ce ton, meme lorsqu'ils s'allient a d'autres mots, k d'autres
racines, a d'autres elements pour former des mots composes. II faut
done, de toute necessite, en suivant les conversations, s'^fforcer de saisir
cette note ou ce ton, afin de ne pas detonner soi-meme en parlant, et de
ne pas s'exposer ainsi ou k n'etre pas compris, ou a faire entendre tout le
contraire de ce qu on voudrait dire. Caril y a des mots, et beaucoup, qui
se prononcent tantot sur un ton bas, et tantotsur un ton eleve, et qui, de
la sorte, servent a nommer deux choses, Prendre deux idees contradictoires ,
suivant le ton qui accompagne la prononciation. » But Father Legoff too
contents himself with a general statement and neglects to provide his
Chipewyan syllables with their appropriate tones. Hence it is impossible
to gather from his material anything definite as to the tone-system of
Athabaskan. Here and there iri his grammar he throws out a suggestive
hint, but that is all. Thus, he notes that nan-[\. e. ng,-] as a verbal prefix
meaning " across " is pronounced with a high tone, but that as an inde-
pendent low-toned word it means " thy mother " ^. Again, the verb stem
kkeih [i. e. -k\^] is high in the present ("imperfective") forms, low in the
past ("perfective") forms'.
It is a pity that Fathers Petitot and Legoff, who evidently had consi-
derable feeling for Chipewyan, did not treat tone with Ihe detail that it
pologist, N.S., vol, 17, 1915, pp. 535-558). I hope to publish fuller studies on the
Nadene languages (Athabaskan-Tlingit, Haida) in the future. Since the preliminary
report was published much new evidence has come to light.
1. In his monumental Diclionnaire de la Langue Dene-Dindjie, dialectes Montagnais
ou Chippexuyan, Peaux de Lievre et Loucheux. See p. xiii of the " Avant-Propos. "
2. See his Grammaire de la Langue Montagnaise, Montreal, 1889. The French mis-
sionaries use the term " Montagnais " for the more current Chipewyan. This Atha-
baskan " Montagnais " is not to be confused with the Algonkian Montagnais, of the
lower St. Lawrence.
3. Op. cit., pp. 10, 11.
4. Op. cit., p. 120.
5. Op. cit., p. 321.
Six: Athabaskan and Na-Dene Languages 171
187
deserves. As it is, most other recorders of Athabaskan have either over-
looked it or made light of it, so that at this late day it probably comes
as a surprise to most linguists that Athabaskan is indeed a tone lan-
guage. Father A. G. Morice, our authority for Carrier, is not entirely
unaware of the existence of the tone problem. " Change of meaning by
intonation or vocal inflection ", he remarks, "obtains also among some
— not all — of the Dene tribes. Some of these intonations are even
proper to fractions of tribes only. Thus ya which means 'sky' in almost
all the dialects becomes 'louse' to a Southern Carrier when pronounced
in a higher tone. Northern Carriers have another vocal inflection which is
combined with the final hiatus and is also peculiar to them ". ' This pas-
sage suggests dialectic variability in tone. As Father Morice gives no fur-
ther information on Carrier tone, I asked my colleague Mr. C. M. Bar-
beau, who was carrying on ethnological researches in 1921 among the Git-
kshan Indians near Hazelton, B. C, to try to secure some linguistic
material from the neighboring Carrier Indians of Hagwelgate. Mr. Bar-
beau kindly furnished me wnth a small amount of Carrier material in
which the attempt was made to note the high tones. Very suggestive.
among others, are the following forms : -
^Is'd'i
I am eating
^ina'i
you are eating
^b-d-i
eat ye !
gkid-l
I have eaten
gknd'l
he has eaten
The tones of these forms are no doubt accurate in the main, for they
have close Sarcee analogues. Sarcee too has a large number of verb stems
whose present or "imperfective" form is high-toned, perfective low-toned
(Father LegofT's Chipewyan -k'io : -k'ki and Mr. Barbeau's Carrier -'af :
-'d'l correspond to the Sarcee pattern -ts^os : -ts'o':( " to handle a
cloth-like object '" ; -^ii : :(l- I "to be warm" ; - k^as: -ka-^ " to be cold
weather " ; -la:-la "to do"; -tc'o : -Ic'o "to take, get hold of"). Again,
the high tone on the first syllable of "you are eating" is undoubtedly
correct, for in Sarcee, as we shall see, and in other Athabaskan dialects
as well, the second person singular of the verb frequently demands a
high tone on the syllable preceding the stem.
i. A. G. Morice, The Dene Languages (Transactions of the Canadian Institute,
vol. I, 1889-90, pp. 170-212; see p. 182). In Sarcee, " louse " [ya] has the low tone
with a final glottal stop, as in Northern Carrier.
2. ' = high tone; ' = low tone ; •=: long vowel.
172 ^f American Indian Languages 2
im
Dr. P. E. Goddard's extensive and fundamental researches in Athabas-
kan linguistics (comprising materials for Hupa, Kato, Chipewyan, Sar-
cee, Beaver, Apache, and other dialects) nowhere include a treatment of
tone '. Dr. Goddard has recently informed me, however, that upwards
of lifteen years ago, when working on Rousselot tracings of Hupa words,
he found that in apparently homonymous forms for the second and third
person singular .(e. g. fczn:r,yai "you have gone out" or "he has gone
out" ; mtndxt "you are finishing" or "he is finishing' ~) the second
person has a higher tone on the .syllable preceding the verb stem than the
tliird person. This important observation of Dr. Goddard's has apparent-
ly not been published before.
The Franciscan Fathers have given us an excellent dictionary of Na-
vaho, 3 but not a word is said about tone in their introductory gram-
matical notes. However, they are careful to mark the main stress of
each word. It seems very likely that many, perhaps most, of these stress-
ed syllabes are really high-toned (the English and German speech habit
associates greater stress w^ith higher tone and tends to consider the form-
er the primary factor). It is remarkable that in a great many verbs the
stem is stressed in the present, unstressed in the perfective. These stress
alternations are analogous to, and probably identical with, the tonal
alternations of Sarcee (and of Chipewyan and Carrier) verb stems. To
take but one instance, the Sarcee alternation of imperfective -ts'os with
perfective -/5'o';{ is clearly the same phonetic and morphological fact as
the Navaho alternation of stressed -ts^os (present) with unstressed -ts'o's
(perfective), e. g. namsts^o's ''I give you (a buckskin)": na'n'ils^o's ^'' I
have given you (a buckskin) ;" xflj^fj'o'j "I take out(a blanket)" : xa'its''o'S
"I have taken out (a blanket)" *. Again, the verbal paradigms given by
the Franciscan Fathers indicate that, as in Sarcee and Hupa, the second
person singular sometimes dilTers from the homonymous third person
in bearing a stress (i. e. high tone) on the syllable preceding the verb
1. For a passing reference to the possible significance of tone in Chipewyan, see
P. K. Goddard, Analysis of Cold Lake Dialect, Chipewiyan (Antliropological Papers of
the American Museum of Natural History, vol. X, pt. 2, 1912), p. 83.
2. See P. E. Goddanl, The Morphology of the Hiit>a Language (University of Cali-
fornia Publications in American Archaeology and Klhnology , vol. 3, 1905), pp. 117,
ir.o.
3. The Franciscan Fathers, A Vocabulary of the Navaho Language, 2 vols., Saint
Michaels, Arizona, 1912.
4. Op. cit., vol. I, sub" give " and " lake ". I have adapted the orthography of
the Franciscan Fatiiers to my own. In my orthography accents following vowels are
indicative of stress; wlien placed on the vowel, of tone.
Six: Athabaskan and Na-Dene Languages 173
189
stem, e. g. i'yi "you marry": iyt "he marries;" txadi'gis "you wash
yourself" : txa'digis "he washes himself" *.
I shall now proceed to a brief discussion of the Sarcee tone system. It
is impossible in a brief compass to do justice to the subject, which is
diflicult and involved. Sarcee tone cannot, properly speaking, be discuss-
ed as an isolated chapter of the phonology. It interpenetrates each and
every aspect of Sarcee morphology. None of the elements of the language,
whether radical, prefixed, or suflixed, can be accurately understood with-
out a consideration of its tone and of the tone shifts which it undergoes
in combination with other elements. All that I can attempt here is a
preliminary treatment, designed to give some insight into the tonal
peculiarities of an Athabaskan language. 1 hope to give a full and system-
atic study of Sarcee tone in a later work on the phonologv and mor-
phology of this language. My method at the present time will be, in
the main, to list and discuss a series of apparently homonymous sets of
forms. This informal and inductive method will probably convey more
to a public necessarily but little acquainted with Athabaskan morpho-
logy than a more compact and generalized style of treatment. It will be
convenient at times to refer to the general or reconstructed Athabas-
kan form of a word. Aly reasons for the reconstructions - can hardly be
given here but must be reserved for later comparative studies. In any
event, the Athabaskan dialects do not dilTer greatly on most points; the
degree of ditTerentiation is comparable to that of Romance, Slavic, Semi-
tic, or Bantu rather than of Germanic, Indo-Iranian, or Hamitic.
Phonetics.
A few preliminary remarks on Athabaskan phonetics are required. The
Sarcee sound system does not reflect the original Athabaskan system as
well as do those of Chipewyan, Hupa, and Navaho. There have not only
been important shifts of vowels and consonants, but levelings of certain
consonants (e. g. the old ^-sounds and r-sounds ^ have combined into a
single i-series, the Sarcee c-series being, in the main, of anterior
1. Op.cit., vol. I, pp. 13, 14.
2. In my comparative Athabaskan researclics I have used chiefly the following
dialects and authorities : Chipewyan (Petitot, Logoff, Goddard); Hare (Petitot)
Loucheux (Petitot); Beaver (Goddard); Sarcee ((ioddard, Sapir); Kulchin (Sapir)
Ten'a of central Alaska (Jetle^ ; Ten'a of Anvik, hjwer Yukon ((-liapnian, Sapir)
Carrier (Morice) ; Hupa (Goddard) ; Kato (Goddard) ; Chasta Costa (Sapir) ; Navaho
(Franciscan Fathers) ; and Jicarilla Apache and Apache proper (Goddard).
3. My c is Lepsius' s.
174 yf American Indian Languages 2
190
palatal origin ; Athabaskan y and y are distinguished only before origi-
nal a) and vowels (e. g. old i and i, the pepet vowel, have become
leveled to i) ; weakenings or- disappearances of certain syllabically final
consonants (e. g. -n drops if preceded by a short vowel, which was pro-
bably nasalized at one time ; -g and -d are weakened to -', the glottal
stop), with resulting interchange of lost, or weakened, and retained con-
sonant (e g. Ath. *-en : *en-C> Sarcee-i : -in-a ; Alh.* -e^: *-td-e^ Sarcee
-a': -ad-i) ; and contraction of vowels (e. g. Ath. -ai-, -e'e-, -e'a-, -fe-,
-ee- may all now appear in Sarcee as -a'"-, see below). Once the phonetic
changes peculiar to Sarcee have been worked out, however, it is seen
that this dialect is a perfectly regular, if somewhat disturbed, represen
tative of its prototype.
The Sarcee sounds are :
I. Vowels :
a. Short: a (a), fl, i, u {o)
b. Long (or half-long) : a' , a' , i", o' {w)
c. Over-long (with glide-like rearticulations) :
a-", a-', '.••■, o-" («•«)
d. Diphthongs (<original vowel -|- -y, -7):
a{-)i, a{-)i, u[-)i
II. CoiNsoNANTs :
a. Stops and AfFricatives :
Intermediates : b (very rarely), d, g, d^, dj
Aspirated Surds : t' , k\ is\ tc'
Glottalized Surds : /', k\ t's, t'c,
h . Spirants :
Voiceless: s, c., x, x, x, ^w, h (-')
Voiced: i, y, i, 7-, y, T^ "^ (t"')
c. Laterals: /, /, dl, ti, fi
d. Nasals : m, n
Of the vowels, a is a duller form of short a and is pronounced very much
like u of English but ; a is a velarized, dark-timbred a with 7^-glide after
preceding stopped consonants (aside from'), aflricatives (including J/,
ll, H), and sibilant spirants (5, :(.'c, j)\ t is open, as in English it; u,
open as in English put, varying with close (French eau). Of the two a-
sounds, which are difficult to distinguish only in certain positions and
which must be clearly kept apart in theory, a generally corresponds to
Ath. e (open) and to Ath. a in certain cases, g. nearly always to Ath. a
Six: Athabaskan and Na-Dene Languages 175
191
(e. g. Sarcee -^'fl " to be so"<Ath. * -/'e, Sarcee fa' " feather, wing "
<Ath.* fax). Vocalic series c. is particularly common in open syllables
and results either from contraction (e. g. sa"'thk'a " my brother or sister "
<isi-ati:k'a) or from short vowel -(- syllabically final -/- (e. g. xa-'^m " it -
tastes " <i xa-l-ni, cf. Nav. ha-l-ni).
The " intermediate " stops [b, d, g) and the stopped part of the inter-
mediate afTricatives [d^, dj, dl) are essentially voiceless, or voiced only at
the moment of release, unaspirated, and of lesser stress than the normal
voiceless stops of French or Slavic ; they probably correspond to the
" tonlose Medien " of many German dialects. The aspirated surds are
more strongly aspirated than the voiceless stops of English. The glotta-
lized surds, afTricatives, and lateral (/', k\ fsyfc, ^7) are pronounced with
synchronous oral and glottal (or epiglottal) closure and with a sudden oral
release which precedes by a perceptible moment of time the release of
the closure in the larynx. The intermediates and aspirated surds seem
respectively identical to the surds {t,-k, ts, tc,) and aspirated surds (/', k\
/5', tc'), of Chinese; the glottalized consonants (sometimes known among
Americanists as " fortes ") are peculiar to many aboriginal languages of
America,
In this system c and ; are used to indicate the voiceless and voiced
sibilants of French choix {Eng\is\shoe) andjoie; dj, tc\ /V are the correspond-
ing affricatives of the series. The voiceless series of spirants x-, x,
X, 'zv and, correspondingly, the voiced series y [y), y, y, w/ [y^) are
conditioned by the preceding vowel, x (as in German ich) and y (close
to English y, yet not strictly identical with it before i, where it is more
definitely spirantal than English y ; I have written y, however, as there is
no possibility of confusion) occur before i, y also before a; x and y (mid-
guttural spirants) occur before a ; x and v (velar spirants) before a; 'w
(approximately like English wh inwhen as pronounced in America, but with
more nearly spirantal : ^w, almost xtv) and tv (properly, labialized y :
Y"' or yw) before (and after) u. i represents a voiceless spirantal /, much
like Welsh //.
In pronouncing Sarcee, or any Athabaskan language, it is important
to give each syllable its due weight. Ihe syllable as such has a more
clear-cut phonetic distinctness and a more individual morphological signi-
ficance than the syllable of English or French, even if it is not always
possible to assign it a concrete meaning. The syllables of a word do not
generally differ greatly in stress, though the stress is of course not uni-
form. Certain elements, like third person plural gi- " they " and the
radical syllable of the verb, are more strongly stressed than others (e. g.
iuk'agits'a' '■ they cease to cry ", ^i- and the stem -ts'a' are more strong-
176 yi American Indian Languages 2
192
ly stressed than k'a- '' off "> " to cease ", which, however, must not
be slurred over). Variations of stress do not seem to be coordinated with
differences of tone.
Sarcee syllables are of five types :
a. Open syllables ending in a short vowel (e. g. wi, t'a).
h. Closed syllables with short vowel followed by a voiceless consonant
or, far more rarely, by two voiceless consonants (e. g. nd, Va\ the
only finals allowed are -t\ -is', -tc' , -', -s, -c, -x,-\ -i, -st\ -ctc\ -lie' \
more rarely -k\ -t's, and -ft).
c. Heavy syllables w^ith long, or half-long, vowel followed by a voiced
consonant (e. g. ni'l, t''a'^\ the only finals allowed are -:(, -/, -/, -« ;
historically speaking, syllables in -r [<C Ath. *-r7, *-i"y, *-£>', *-^y]> ^"
-0' (<C Ath. *-o'y), and in -a'i, -a'i, -wz [< Ath. e' , a', o , -\ — y or
-y,J also belong here)
d. Heavy open syllables with over-long vow-el (e. g. «■.•', /'d"") result-
ing from contraction of vowels or reduction of type a -(- /.
e. Heavy syllables with over-long vowel followed b}^ a voiceless conson-
ant (e. g. m''s, fa'^s) resulting from contraction of type a -|- type b.
The first three types of syllables are old Athabaskan types, the last
two are largely of dialectic origin.
The point of syllabic division of a non-final, open syllable, particu-
larly if the vowel is short, lies /// the following consonant, which
therebv becomes greminated. Thus, the forrh k'>£'Js^a'' jjiven above is
to be read k' xg' \.f s'' a'' {-g' zzz -kg-, i. e. unaspirated yoiceless ^releasingin
intermediate g). We shall not indicate these purely mechanical gemina-
tions. A final glottal stop, whether an old Athabaskan element or of Sar-
cee origin, is absorbed in the following geminated consonant, e. g.
a"'sdiy'.nC " he has said to him" -j- inferential particle la becomes
a'^sd'-j'-U'd'a. There are other important rules of sandhi which we must
neglect here.
Tone
The tone system followed in this paper is that used by Father Schmidt
in " Anlhropos " and by Westermann and other Africanists. In this sys-
tem a indicates a high tone, a a low tone, and a a tone of intermediate
pitch. The intlections are indicated by combinations of these symbols : a
represents a tone falling from high to low, a falling from high to middle,
a falling from middle to low, a rising from low to high, a rising from
low to middle, a rising from middle to high. In writing Sarcee I have
found it convenient to leave the middle tone unmarked [a = <i) ; my a
Six: Athabaskan and Na-Dene Languages 177
193
therefore does not mean an indeterminate or unknown lone, but one
that is midway between a and a.
The primary Athabaskan tone system may prove to be one of register
alone, possibly recognizing but two grades, high and low. In Sarcee,
however, it was found absolutely necessary to distinguish also a middle
tone and inflected tones. The middle tone may have originated as a pho-
netic compromise betw^een two conflicting tone principles, one of which
demanded that the syllable receive a high tone, the other that it be pro-
nounced on a low tone. However that may be, there are many verbs
w^hich require that in certain aspects the radical element be pronounced
on a middle tone, e. g. dlsi'ya " I have gone ", in which the -ya " to
go " falls from the preceding high syllable but does not drop to the low
tone of a syllable like -tia in dstla '' 1 have done ". With discya con-
trast also dlc'd " 1 shall go " <C *d\-s-yd. Here the -yd has a definitely high
tone, maintained throughout the " imperfective " forms (cf. di'yd " he
will go ", hut di'ya " he has gone "),
The syllables of the language belong to two tonal categories. Certain
syllables have a fixed or inherent tone value of their own ; e. g., '('/
" pine bough with needles ", xdi "' burden ", dxi '' crane ", mii " moth,
sleep ", ts'in " dirt ", tii " dog ", ts'i " red clay ", ts'd " rock ", t'6
" water "', ts'Vl " snowdrift ", ^di " saliva ", Id " sore ", -dd'l " seve-
ral go " (perfective), -t^d'^ " two go " (perf.), -yd " one goes
(imperfective), -i " to act " (imperf.), nd- " again, repeatedly ", k'u-
"■ inside, into ", and d- (demonstrative prefix with verbs of doing, being,
becoming, and saying) have a fixed high tone ; t'cis " powdered char-
coal ", {sl-:{ " a boil ", tlifs " clay ", k'ii " fire ", th " smoke ", mas
" hoop ", did' " laughter ", ya" " louse ", wa' " war-raid ", did' " gum ",
so " star ", ts'a " ordure ", md " snare ", -ddl " several go " (imperf.),
-Cas " two go " (imperf.), -In " to act " (perf.), d^nd. da- (prefix indicat-
ing distributive subject of verb) have a fixed low tone ; and -la (inferent-
ial sufTix), -V' (suffix of relative clause implying absence of subject), -ya
" to go " (perf.), -^t. " to call " (continuative), -k\' " to throw a cloth-
like object " (perf.), and -w. " to use the hand '' (imperf.) have a fixed
middle tone. Other syllables have no inherent tone of their own, but adopt
a tone in a particular form in accordance with certain rules of contrast
or of morphology. These syllables have no assignable tone as such, but
this does not mean that they do not possess a defined tone in a given word.
On the contrary, the intonation of syllables with variable tone is impor-
tant, but the rules governing their tone ar£ not always easy to define.
Thus, such elements as perfective-durative .ri-, perfective yt-, and possess-
ive-objective s\- " my, me ", vary their tones from word to word, but
not optionally. This is illustrated by the three forms :
178 ^l American Indian Languages 2
494
sit'\ " the calf of my \e^
sit^i " you are lying- (in position)
ji/'i " he is lying
The syllables -/'I " calf of the leg " and -t'i " to lie " (perf.) have inhe-
rent tones, low and high respectively. The syllables ji- " my " of the
first form and si- (perfective-durative) of the other two have no tones of
their own. In ji/'i the high tone of 5-!- " my " is due to the contrastive
influence of the low tone of the radical element (examples of other tones
for 51- " my " are sr(dha " my grease ", s\la " my brother [woman speak-
ing] ", contrast ji/^' " my hand "). In sit^i the high tone of perfective-
durative si- is due to the fact that the subject of the verb is a second per-
son singular. The old personal prefix *-«- (Ath. *se-n-^ Sar. si-), which
has been lost in Sarcee, seems regularly to have required a preceding- high
or, in other forms, relatively hig-h, say middle, tone, although the syno-
nymous prefix *ne- (> Sar. m-), which is used in certain forms, is itself
variable in tone (e. g. amid " you do ", nits\' " you are crying"). In
sit^i the si- is just low enough to bring oiit the inherent high tone of the
stem -t'i. Just what tone a variable syllable takes cannot be foretold on
the basis of simple rules ; the various syllables of this type do not all fol-
low identical patterns. The tones so far illustrated belong, then, to three
categ^ories : inherent or fixed tones, tones controlled by definite grammatical
requirements, and tones controlled by contrastive requirements. It is
naturally impossible to draw a hard and fast line between the last two
categories.
All, or nearly all, primary syllables have a level tone. This applies to
syllables with a long vowel as well as to short-voweled syllables; e. g.,
in forms like n'lts^i' " you are crying " or gdisyc:{ " they have run olT "
OT wiictc^o' " leg-big " or yi'an " it has lain " the long vowel of -ts^c,
-)'i';(, ~tc'd\ -a'n must be pronounced on a fixed register — middle, high,
or low — without the slightest upward or dow^nward inflection. The
numerous inflected tones ^ of the language arise from the contraction of
two, rarely three, vowels of different registers. The theory of these
inflected tones is simple enough. The vowels melt into a single long
(over-long) vowel without losing their tones, which tlius combine into
inflections whose tendency and range is determined by their component
elements. Hence, a -\- a becomes a' (half-falling from high level), a -\-
{. Yet far less nnmerous Ihan in Chinese or in certain other American lang-nages,
e. g. Takelma, in which we have primary rising and falling inflections. The quick
change from level lo level, with comparatively few slurs, is highly characteristic of
the general How of Sarcee speech.
Six: Athabaskan and Na-Dene Languages 179
195
a becomes d' (fuU-fallingrfrom high level), a -\- a becomes a' ( half- falling-
from middle level), a -\- a becomes a' (half-rising from middle level),
a -\- d becomes a' (full-rising from lo^v level, a -\- a becomes a' (half-
rising from low level) '. If the contracting vowels have the same tone
there results a long (over-long) vowel of level tone ((i", a' ", or a");
long level-toned vowels of the same phonetic type result also from vowel
-|- /. Only a few examples are needed to illustrate these tonal rules :
/Wi isfscx, blanket I-wear >> t'sidi'sfsix^
k^d adi moccasin without, barefoot>» k^a'di
nd-\-c-atc again-it (indef.)-I-keep-handling-one-object, I handle a
certain object time and again > nd'cqlc
Si-l^una my sickness>> sd':(i'ma
da'" Ictc'o here I-seize-it>> da'ctc^o
gu-d-s-tid thus-I-do> giva'stid
\-nl-i-n'.-i it (indef.) -he-admires-the, the one v^ho admires^ Inl'nl:
k^u dstid fire(wood) I-do, I make a rire>> k'wa'stid (note disappear-
ance of -' in crasis)
n'.sgdywlsl' dnd downward he-becomes, he stoops^ nisgd^(iu\sa"nd
mit'uwa iss'd his-water I shall-heat > mit'-iava-ss'd
t'd:(d-d-yi-s-\ grief-thus-him-treats, he is in mourning>- f'^i^a'^yu't
t's4-i-s-sl noise-he-is-sounding, crow>> fsd'ssi
t^u-^{w\-\-a water-in-it (indef.) -he-handles-one-object, he gets water
The only certain example of an inflected tone in Sarcee that I know of
which cannot be analyzed into two simple tones is the low falling tone,
-a'-, of the first person plural subjective, e. g. gumca'Cd'j " we suckle
them", nxsa'io' " we get moist ". Full historical knowledge would proba-
bly indicate that this element is contracted from two simpler elements
of middle tone and low tone respectively.
It goes without saying that the high, middle, and low levels have no
absolute fixation. Nor are they definitely fixed relatively to each other.
The interval between high and middle or between middle and low or
between high and low varies widely according to the care or emphasis of
articulation. What under some circumstances constitutes a movement from
high to low may under others be equivalent to no more than a half fall.
1. These formulae do nol refer to vocalic quality. The quality of the contracted
vowel is dependent on a number of factors, which do not concern us here.
2. As several of these examples show, the rules of contraction apply not only within
the " word " (a somewhat elusive concept in Athabaskan) but in sandhi between
words (" crasis ").
180 VI American Indian Languages 2
196
Each level and each inflection must be judged or intuitively felt in the
context of the preceding and following tones. Now one of the most cha-
racteristic and also one of the most puzzling things about the Sarcee tone
systtMU in practice is the tendency, within a phonetic phrase or breath
group, to gradually lower the absolute register. This means that a high
tone following on a low or middle tone that is itself preceded by a high
tone tends to be a little lower than the first high tone and to be perceived
by the ear as a middle tone. Many of the ''middle " tones that I have
recorded in running text or in specimen sentences or even within the
single word are really pseudo-middle tones — " dropped high tones ",
as we might call them. I feel that the}' are not normally identical with
true middle tones but keep above the normal middle register. The form
dn\ld " you do " already quoted was often recorded as dnda, with a
secondary middle tone, instead of the theoretically required high tone,
on the last syllable, yet, unless my ear deceives me, the tone cadence of
this word is not identical with that of words like dit'ibdja' " it is flabby ",
diszuuga' '' it is velvety ", nUc^it'ca' " it is tiny ", in which the diminut-
ive -a' falls regularly on the middle tone (cf. , with another tone cadence,
nidot'ia' " it is light in weight ", S'.sowa' " it is a little sour "). In other
words, the " dropped high " really falls between the high and the true
middle level. Where there is an immediately adjacent middle tone to
serve as a standard of comparison, this difference can frequently be direct-
ly perceived. In the phonetic group ayd t^unihi ^' the one who wrings
out water " the cadence of the second word is not adequately suggested
by the orthography. In making the record I noted that the group t^umhi
was really t'linihi on a high register (i. e., not equivalent in cadence to
forms like dnlla with true low -n\). In fiimhi, in other w^ords, the -m is a
true middle and the -hi a " dropped high ". The analysis is not difficult
here, because we have an immediately preceding -|*a with a clearly mark-
ed low tone to serve as gauge ; moreover, we have the analogy of other
related forms in middle-tone -m and of hundreds of analogous relative (or
participial) forms in -L
A good example of the progressive fall in register is the sentence
k'iyldji. ytca '* coat in-I-go, I put on a coat ", in which i indicates a lower
tone than the normal low tone fixed by the preceding wnrd. The theore-
tically correct form of the sentence is k\yidji y\cd, but actually no two
tones are quite alike in it. Neither of the " middle " tones as the sen-
tence was recorded {-dji and -c'(i) is a true middle, but merely a " dropp-
ed high ", the second a trifle lower than the first. Thus, the three high
tones of the sentence are graduated downward, and the two low tones
are also graduated downward in a parallel curve. Functional register
Six: Athabaskan and Na-Dene Languages 181
197
chang-es are compounded, in other words, with a rhetorical register change.
Just as there is a " dropped high " tone, there ia also not rarely a " dropp-
ed middle " (or secondary low) tone, further a " raised middle " or
secondary high tone. These niceties can hardly be properly studied with
the unaided ear. They require the assistance of mechanical devices^. In
any event, how^ever, it Avould be a great mistake to lay too much stress
on variability of tone in Sarcee. It is probably no more disturbing a
factor in the understanding of the basic tone theory of the language than
in Chinese or any other tone language, each of which has its special laws
or tendencies of tone sandhi.
Tone differences in otherwise identical words.
We shall devote the rest of this paper to a brief analysis of a number
of cases of forms which dilfer only in tonal respects. In this way we
shall gain a cumulative idea of the importance of tone in the study of
Sarcee structure. In many cases the forms in question originally dilTered
also in other respects and fell together completely in the consonantal and
vocalic framework because of the operation of various phonetic laws.
Thus tone was left as the last reflex of outward distinction between such
forms. One must beware of makinsr the error, which is so commonlv
made with Chinese and other tone languages that possess numerous near-
homonyms, of supposing that the tone differences arose for the purpose oi
keeping apart words which would otherwise be indistinguishable. As a
matter of fact, perfectly analogous tone distinctions prevail in groups of
non-homonymous forms, w^here tone is not " necessary " as a gran^mati-
cal device. F'urthermore, there are many examples in Sarcee of true
homonyms, of two or more w^ords which belong to distinct form cate-
gories but are identical in all phonetic respects, tone included.
1 . a. -^a'^dd'l he has eaten the berries
b. -^a'dd'l we have eaten the berries
Form a. goes back to *^alddl, in Athabaskan terms *ys.-l-di'l, '' in
1. I hope to secu e phonograpliic samples of ottier Athabaskan dialects, so as to
be able to work out some of the more elusive tone problems -with the help of strictly
objective material. [Since this article was written such material has been secured
from a Kulchin informant].
2. These Athabaskan reconstructions, it is hardly necessary to say, do not imply
that we are certain that precisely such forms were in early use, but merely that,
if they existed, tiiey probably had such a phonetic form. The reconstructions are
practically Athabaskan formulae for the dialectic forms.
182 y^f American Indian Languages 2
198
which *)'£- is contracted from *yc-ye'. Form b. g-oes back to *'(a'ldd'l, in
Ath. terms *-(-€'-l-di'l (perhaps contracted from *'^e-\-l-di'l). -l-dd'l is
perfective of imperfective -l-dal " to eat berries". In form a. *^a- <i*yt-
consists of third person objective *ye- (Sar. y\.-) and perfective prefix *--^e-
(Sar. -yi-) ; as usual, forms without specific subjective pronominal prefix
are used as definite 3 d personal forms. In form b. -^-a'- consists of per-
fective y(^)- and first person plural subjective -rt"-. Transitive verbs with
first or second person or indefinite person as subject have no expressed,
but clearly implied, third personal definite object in many forms, but in
verbs with definite third person as subject third personal definite objects
must be expressed. All that remains in Sarcee of the marked phonetic
difference between Ath. *ye-(eldi'l and *'(eildi'l is the fact that in the latter
form the voice sinks by a slig^ht interval, perhaps no more than a semi-
tone, before leaping to the high- toned verb stem.
2. a. k^u'sdal I have many (e. g. buffalo) coming in
b. k'ii'sdai he has many coming in
c. k'u'^sdai you have many coming in
Form a. is contracted from k'ulsdai, Ath. *k'iuen-'e-c-i-dki, literally : in
(k^wen-) l{c-) cause {i-) a certain one [or certain ones, 'e-, indefinite object)
to go in a group [-dki). -dai is imperfective, -dd'l perfec1,ive ; cf. no. 1.
Form b. reconstructs to Ath. *k^iuen-e-i-dzi. In Old Sarcee, the older form
of the language still spoken by the oldest men and women of the tribe,
forms a. and b. are still distinguished SiS k^u'sddi and k'H'idai (contracted
form k'uddal) respectively. In the former of these forms the element -i-
dropped out after the sibilant, as in so many Ath. dialects (e. g. Navaho
c- for theoretical c-i-\ Ghasta Costa has c-i- but -I- may have been restor-
ed because of the analogy of the other forms). In New^ Sarcee, as spoken
by my chief informant and most of the other Indians, syllabically final
-I- has shifted to -s- ; hence in form a. -s- represents older Sarcee -s-
<i-c- <C-c-i-, but in form b. -s- represents older -/-. Form c. goes back to
Ath. *k'wen-e-n-l-dii.
Syllabically final -n- (here "thou") has disappeared after a short vowel
(there was doubtless nasalization of the preceding vowel, oldest Sarcee
*-i-, as in Chipewyan, before all trace of the old -n- was lost) and, as
in a. and b., m + i or older u + ^has contracted to an over-long M-vowel.
k^u- has fixed high tone. Objective -i- is variable in tone; in c. it takes
the high tone because of the second person singular subject (cf. p. 194).
3. a. m-^i he will grow up
b. ni-^i. you will grow up
c. niYa he has grown up
Six: Athabaskan and Na-Dene Languages 1 83
199
Forms a. and b. correspond to Ath. *ne-yi and *ne-n-yi respectively ;
cf. nic^(7. '^ I shall grow up ". Form c. is a perfective in -i-, cf. nl^iY^^
" you have grown up, " nlcu:vi (-^ic- assimilated from 5 u:- ') " 1 have
grown up ". Moreover, -vi of c. has lost a final -n (contrast ni-^in-i
"the one w^ho has grown up" with n'.-yoL-hl " the one who will grow up").
Hence c. reconstructs to *ne-z-yin. Perfective *-yin, however, is modified
from the normal Ath. *-ydn by the analogy of imperfective *-yi (cf. Chi-
pewyan imperfective -yt : perfective -yan) ; the older *-ydn is preserved in
Sarcee in the causative : -c-ca <^*-i-ydn " to cause to grow up, to raise "'.
Note that imperfective -yi, perfective -yi (n-) belongs to the type of verb
stem with fixed tone in all aspects ; cf. also continuative —/di. The prefix
ni-, however, of variable tone, is only one step lower than the stem in
the imperfective forms, but two steps lower in the perfective forms
(contrast nic^fi " I shall grow up " with nlcic^^i " I have grown up '"l. In
other words, even where the stem is invariable in tone, other elements
in the verb complex are not necessarily treated in a uniform tonal manner
in the varying aspects. It is quite probable that originally the difference
between m- of a. and m- of c. was purely mechanical. The aspect prefix
{*se-*ye-) tends to adopt the middle tone and, if the stem is high-toned,
a prefix of variable tone which precedes the aspect prefix generally
takes the low^ tone (the cadence nlcic-^fi is very typical). It is likely, there-
fore, that form c. is reduced from an old trisyllabic Ath. form of type
*ne-se-ydn, while a. goes back to a disyllabic form *ne-yi. Now, in Sarcee,
where forms a. and c. have come to have the same syllabic structure,
the difference of tonal approach to the stem reflects the old difference in
phonetic build.
4. a. gof^ddlmsts'C I shall punch a hole (.with an awl)
b. goy'^'ddlnists^C I have punched a hole
5. a. goy'^ddini''ts^i you will punch a hole
b. goy'^ddim''ts\'' you have punched a hole
6. a. goy'^'ddl"ts^C he will punch a hole
b. goy'^'ddl'ts'i' he has punched a hole
In these forms the momentaneous aspective prefix m- is used in both
perfective and imperfective forms, so that the only dilference between the
1. In perfective forms in which the pronominal subject " I " is directly follow-
ed by the stem or by the element -/-, Ath. * -c- is regularly replaced by *-i-, reflexes
of which occur in most of the dialects, but in certain dialects, such as Sarcee and
Chasla Costa, the * -c- is used by analogy with the imperfective forms.
184 VI American Indian Languages 2
two aspects, for each of the three persons, is in the tonal treatment of
the syllable or syllables following the first three syllables {-yd- " hole,
through '', labialized to -y'^d- because of preceding o, has a fixed high
tone, to which the variable elements go- " it, indefinite locality " and ^t-
accommodate themselves). In 5 and 6 the i" is due to the loss of an old
/ after the i, in 4 the / dropped out after s without leaving a trace. The
element -w. <C *-ne- became reduced to -«- in 6 and eventually disappeared.
Aside from the first two syllables, these forms reconstruct to Ath. :
4 . a . *de-ne-c-l-ts*ed
b. *de-nt-c-l-is''ed
5 . a . *de-ne-n-l-ts'ed
b . *de-ne-n-l-ts^ed
6. a. *de-n-l-ts'ed
b. *de-n-l-ts'ed
In 4 the prefix -ne keeps one step below the stem, hence middle tone
in a., low tone in b. In 5 the -ne- needs the high tone because of the
second person singular subject. The verb stem, which means " to
handle a pointed object (like an awl) », is *-ts^ed in the imperfective,
*-ts^ed in the perfective (the final -d, reduced to -' in Sarcee, reappears
in certain forms, e. g. goy'^'dd'c'ts'idi. " the one who will punch a hole "
g(yf"ddi''ts''idi " the one who has punched a hole '). It is probably a
denominative in -d from the noun stem *ts'i (Sarcee ts'd) " stone '".
There is evidence to show^ that the imperfective aspect uses the more
primary form of the verb stem and that the stem forms appearing in
other aspects are secondary formations. It is therefore no accident, in all
probability, that imperfective *-ts"ed is identical in tone with *ts'i. It
should be remarked that a large number of the Athabaskan verb stems
are either identical with noun stems or clearly derived from them.
7. a. iditi'jl I shall hide myself
b. idinis'i he hid himself
Form a. uses the imperfective form of the verb stem, -l, " to hide ;
form b. , the perfective form, -'i ; -/- has dropped out after -s- in both a and
b. The reconstructed forms are *ede-ne-c-l-\n and *ede-ne-s-l-'hi. The third
person corresponding to a. is idin'/'\ <C *ede-ne-l-ln ; the first person
corresponding to b. is idimsts'i <C *ede-ne-se-c-l-' iji . Hence -s- of a. is a
pronominal element, while -s- of b. is an aspective prefix. With the
reflexive prefix [di- as well as with the perfective stem - { contrasts the
Six: Athabaskan and Na-Dene Languages 185
201
low-toned -nl- of b. In a. the -m- takes a middle tone as a transition
between the high idi- and the low-toned imperfective -'i.
8. a. :i^is^(d I shall kill him
b. :(is^(a kill him !
These are imperfective forms based on --rd " to kill one person or ani-
mal) ". The middle tone in b. is organically a high tone which has been
depressed to accentuate the high tone of :{{- (cf. also s\.-{i5^(a " kill me ! ")
used in verbs of killing and probably referring to " death. " It is variable
in tone. In a. it is one step lower than the stem, in b. it is raised to the
high tone because of the second person singular subject. In perfective
forms, based on-^i <; Ath. *-'(in, the :(i- sinks to a low tone (e. g. yi;(U^t
" he has killed him "). The Ath. prototypes of a, and b. are :
a . *:(e-c-i-^i
b. ■{e-n-i-'{i (or -^£ ?)
9. a. yi/'ii*;^ you have worn it
-b. yi/'5i';( he is wearing it
Form a. is based on the perfective form of the stem, -/Vi*:( (cf. yisfs\'7
" 1 have worn it ",yVyi/Vf;( " he has worn it ").
Form b. is based on the continuative form of the stem, -t'si':(, with
middle tone. The imperfective is -fsist\ e. g. Ist'sist^ " I shall wear it ",
yifsist' he will wear it ". In a. yi- is an aspective prefix ; the old
pronominal -«- has dropped out (yi- << *-^e-n-), but the yi- is not raised to
the high tone, as it is already high relatively to the stem. In b. y- is the
definite third personal object, Ath. *ye-. Its high tone, as contrasted
with the low yl- of equivalent forms in other aspects (ylyj'' si :( " he has
worn it ", ylt'sist', " he will wear it ") is characteristic of certain conti-
nuative forms, which demand a high tone for the pronominal object ;
analogous to form a., as compared with imperfective yU'sist\ is sIs^ols " he
is throwing (bags) at me " : s\s'(xs " he will throw (a bag) at me ",
yisk'a' " he keeps hitting him (with a cloth) " : ylsJid' " he will hit him '',
yigx " he is poking him " : y\ga. ' " he will poke him '\ yi:(i " he is
calling him " : y'-i: " he will call him" (perfective yi«i:{i " he has called
him "). Observe that a. and b. are identical in cadence but dill'er in
register.
10. a. nayi:(idi the one who stood
b. nayi:(id'. he will stand later on
Form a. is a relative (or participial) in -t formed from the perfective
185 ^^ American Indian Languages 2
202
nayiii {-nd-)- The stem, -iid-, means " to become ", the idea of " stand-
ing " being contributed by the low-toned prefix na- " erect, on the
ground ". The imperfective form of this important verb stem is high-
toned, -71 {-Z'-d-), e. g. na:(C " he stands ". y.- is perfective and takes the
hio-h tone by contrast with the preceding prefix and the following stem.
Form b. is a " delayed future ", characterized by aspective yt- and the
middle-toned suffix-i (analogous forms are dylstldb'. " I shall do it after a
while ", yisdam " I shall drink it later on ", Vu-('^l''yis'ak " I shall get
water later on ^'). Besides the relative suffix -i and the delayed future -i,
there is a low-toned -\ which makes gerundives. It is not properly a
suffix, however, but an enclitic particle attached to the pause-form of the
preceding verb, e. g. nayi^t' -|- i, whence, with contraction, ndyb^j,' " he
having stood.
11. a. *n7.y\sk\' he threw it (a cloth-like object) away
b. n6Lyisk\' I threw it down
c . nxyisk'i' you threw it down
All three forms are perfectives, based on the perfective form, -■ ,k\ ot
the stem ; its imperfective is -^', continuative -k'a'. The Ath. proto-
types are :
a. *nd-ye-s-i-k\'y
b. nd-^e-c-i-k\'y
c. tia-^e-n-i-kU'y
Form a. is really of a very different pattern from b. and c. It has a
high-toned prefix nd- " away " that requires *-s(/)- as a following per-
fective element, whereas the na- of b. and c. is a prefix of variable tone,
generally low, na- " down ", which is followed in perfective forms by
*-v(^)-. The -yl- of a. is the definite third personal object, contrasting in
tone with the high-toned prefix and the middle-toned stem. In the other
persons the -yi- must be replaced by an objective -1-, which contracts
with nd- to nd'-, e. g. nd'slskc " I threwit away", nd'sisk\' " you threw
it away " (Ath.* nd-'e-se-c-i-k'e'y^*' and nd-'e-se-n-i-k\'y). Contrast noLyiyisk" i
'' he threw it down " (Ath. *na-ye-ye-i-k't'y). Note that the only difference
in tone between b. and c. is that the latter form begins a step lower. This
has the effect of accentuating the height of the following syllable,
dependent, as usual, upon the second person singular subject.
12. a. xdy\sdl^' I pulled it out quickly
b . xayLsdlg,' you pulled it out quickly
c. xdyisdla' he will pull it out quickly
Six: Athabaskan and Na-Dene Languages 187
203
The verb stem is-dla {dlad- ) in the imperfective, -dig.' {-dlgd-) in the
perfective, xa- " out '" has a fixed low tone. Form b., having a second
person singular subject, has a higher tone on the perfective prefix -yt-
than form a., in which -yi- takes the middle tone, transitional between
the low-toned prefix and the high-toned stem; cf., fora parallel cadence,
form 4 a (last 3 syllables). Form c. assigns ^ middle tone to the variable
syllable -yi- (definite third personal object, Ath. *-ye-) by contrast to
the fixed low tones which precede and follow.
13. a. nxdiT^i you will disappear
b . nadlif he will disappear
c. nidlX} he disappeared
d. nodi:(C he will appear
(e. nadi:(C you will appear)
I have put form e. in parentheses because I have no actual record of it
under my hand at the moment, but the general analogy of numerous
other forms makes it a safe enough form to accept. The stem, " to be-
come ", is high-toned {-:(i) in the imperfective forms (b. , d., e.), middle-
toned {-:(}') in the perfective (c.) ; cf. no. 10. Forma, is remarkable. It is
an imperfective parallel to b. , yet it seems to have a middle tone on the
verb stem, as in form c, where a middle tone is justified. The reason is
simple. The form nid<.:(iis a secondary development oi n6di:(l.\ with fixed
high tones on na- " away " (cf. no. 11) and -:(C and with a high tone on
the prefix -di- because of the second person singular subject. Certain ele-
ments with a fixed high tone, like our «a-, depress the following syllables
to a slightly lower register, so that an immediately following high
tone takes what seems to be a middle position. (A similar effect is often
produced by the common demonstrative prefix i-, e. g. dsfa " I am thus "
«< theoretical dsfd). I strongly suspect that the -di:(i of a. is not in a true
middle position but is midway between high and middle (cf. the " dropp-
ed high" tones, discussed on p. 196, which arise in another way) ;
while I often noted tones that fell between high and middle or low and
middle, I did not fully realize in the field the theoretical importance of
these secondary " dropped" and " raised " tones. Thus, the syllable -;(i'
of these five forms probably occurs on three registers — high, " dropped
high ", and middle, hut not low. The element noc- of forms d. and e. is
a totally distinct prefix from the ni- of a. , b., and c. In b^ and c. variable
-di- has contrastive tone, in d. it has transitional tone (cadence as in
12a.), in a. and e. it has functional tone. Forme, is one of the curious third
personal perfective forms in -s- in which this element has disappeared
VI American Indian Languages 2
204
(cf. 3 c) ; analogous forms are found in other Athabaskan dialects. The
reconstructed Ath, forms are :
a . *nd-dt-n-7^d
b . "nd-de-^ed
c . *nd-de-ied
d . *na-de-x^d
e. *na-dt-n-/^M
It must not be imagined that the examples which we have selected
are in any way exceptional . The number of such homonymous sets —
homonymous except for tone — is legion in Sarcee and our difficulty has
been to reject rather than to find examples in the recorded material. But
enough has been given to indicate the nature of the tone problem in Sar-
cee. Tone is not a matter entirely of the inherent pitch of a given word
or element nor is it entirely a matter of grammatical symbolism. Both
types of tone function are inextricably interwoven in Sarcee in a system
of considerable complexity, from a morphological standpoint. Phonetic-
ally the complexity is only moderate.
What is the fundamental nature of this tone system ? It is much too
early to speak with assurance, as comparative material bearing on tone
has still to be gathered from other Athabaskan dialects. It is already
reasonably certain that the inflected tones of Sarcee are of secondary
origin and cannot be credtied to the original Athabaskan tongue. In other
words, Athabaskan is in all probability a tone language of varying regis-
ters (high, middle, and low; or, possibly, fundamentally high and low),
not of inflections. The ultimate theory of Athabaskan tone would have
to take account of three important problems : 1 . How are the syllables
with variable tone to be explained ? Did these syllables originally have
fixed tone and did they later, having sunk to the status of relatively weak
proclitic or enclitic elements, lose their tone and take on new tones
according to certain rules of tone sandhi? 2. How did tone come to have
grammatical function (e. g. the high tone, or relatively high tone, so often
found on the syllable preceding the stem in second person singular forms ;
the high tone on the objective pronominal elements in certain conti-
niiative paradigms) ? Are these functional uses the consequence of merely
mechanical tone principles and not directly symbolic in origin ? 3. How
explain the characteristic changes of tone in the verbal aspects? Why do
certain verbs allow of no tone changes in the stem, while others have
one in the imperfective, another in the perfective or continuative ? May
we suppose that originally a verb stem had only one inherent tone,
preserved say in the imperfective stem, and that the other aspects deve-
loped secondary tones on certain mechanical principles, say the amalga-
Six: Athabaskan and Na-Dene Languages
189
S08
mationofthe stem with other elements that have disappeared as such but
have left their trace behind in tone alterations ?
It is not likely that all of these questions can be answered from the
standpoint of Athabaskan alone. The same questions, paiticulary 1 . and 3.,
arise in Tlingit as well, whence it follows that the ultimate explanation
of Athabaskan tone will be given by the comparative study of Atha-
baskan and Tlingit, possibly also of Haida. Should it appear that Tlin-
git-Athabaskan tone originally depended entirely on the inherent
tone value of independently significant syllables, we should still have to
ask what were the determinants of these tone values. But we are very
far from having arrived at the point where such a question is profitable.
Editorial Note
Originally published in Journal, Societe des Americanistes de Paris 17,
185-205 (1925). Reprinted by permission of the Societe des Americanistes.
The Sarcee tone system is more complex than that of most tonal Athabaskan
languages, which normally have only two tones (high and low). The Sarcee mid
tone apparently results from the partial lowering of an original high tone in
certain phonological circumstances, a fact that may be historically connected
with a similar lowering (from high to low) in Navajo and the other Apachean
languages (Krauss and Golla 1981: 85-85). The most recent work on Sarcee is
that of Eung-Do Cook, who has carried out extensive field work. He has pub-
lished a reanalysis of the Sarcee tonal system (Cook 1971, 1984), but does not
challenge the essential correctness of Sapir's description.
Sapir's ms. corrigenda in his copy are as follows:
Original
For:
Read:
p. 190, lib
p. 190, lib
p. 191, 1. 22
[second z]
X.,
[delete]
p. 195, 1. 1
a + a becomes a
a + a becomes a-
p. 195, 1. 15
inini-
ini-ni-
p. 195, 1. 19
mit'u'wa-ssit
mit'u'wassil
p. 197, 1. 2
ia
is
p. 202, 1. 14
p. 202, 1. 17
*ndyLsk\-
ndylsk'f
-k'f.
p. 202, 1. 18
-ka«^
-k'a^
p. 202, 1. 20
*na-...
na- . . .
p. 202, 1. 31
p. 202, 1. 31
p. 204, 1. 22
[1st form]
naytyisk'!
credtied
[delete final superscripts
naytyCsk't-
credited
The Similarity of Chinese and Indian Languages
New light has been thrown on the ancestry of the American Indian by Dr.
Edward Sapir, the Canadian anthropologist now on the faculty of the Univer-
sity of Chicago. Dr. Sapir said that his research work on Indian linguistics has
convinced him of the identity of the language of certain Indian tribes with that
of the primitive Chinese.
The similarity of the two tongues and the linguistic distribution of tribes scat-
tered at random over the Americas have convinced Dr. Sapir that these groups
must have entered this continent as a wedge from Asia. By a close comparison
of the primitive Chinese, Siamese and Tibetian, all in the same language cate-
gory, with the language of the "Nadine group" of North America, Dr. Sapir has
found the same peculiarities of phonetics, vocabulary and grammatical struc-
ture on both sides of the Pacific Ocean.
The American Indian groups speaking the language of the Nadine group are
found in all parts of the North American continent from northern Mexico to the
southern boundary of Alaska, widely distributed among other Indian tribes
whose language and customs are entirely different.
With minor changes the Navajo of New Mexico speaks the language of the
Sarcee in Alberta, and the linguistic stock of the Tlingit, just south of the
Eskimos in Alaska, is much the same as that of the Hupa in California.
It is probable, according to Dr. Sapir, that the migration of Asiatics speaking
primitive Chinese or Tibetian took place at some time in the past, and that
these immigrants settled or moved over the mountains and plains, some
remaining in northwestern Canada to become the Tlingits, and others moving
out to the Queen Charlotte Islands off the west coast to form the Haida group,
and still others penetrating to the deserts of the Southwest.
From the modern Chinese, which in academic circles is considered relatively
simple, students of linguistics can reconstruct primitive Chinese which is far
more complex than any of the dialects known to the Mongolian layman of
today. Dr. Sapir has discovered not only that the Indians of the Nadine groups
speak with a tonal accent, raising or lowering of the voice to give certain mean-
ing to words, in a manner similar to the tonal peculiarities of the early Chinese,
but also that the meanings of certain words arc identical. Further, he has dis-
closed the fact that the Indians have retained certain prefixes and suffixes that
long ago have disappeared from the Chinese speech, but which are clearly dis-
cernible in the early forms.
192 VI American Indian Languages 2
Editorial Note
Originally published in Science 62 (1607), supplement of 16 October, xii
(1925). Reprinted by permission; copyright by the American Association for
the Advancement of Science.
This anonymous, and somewhat shoddy, piece of scientific journalism seems
to have been based on a brief interview with Sapir shortly after his move to
Chicago in the fall of 1925. Certain phrases sound like Sapir s, but most of the
text must have been reconstituted from hasty notes (hence "Nadine," "Tibe-
tian," and other gaffes). The article appeared in the separately paginated
Science Supplement section of the journal, indicating that it was not intended as
a formal announcement of scientific results, but rather as news of work in
progress.
Review of
Berard Haile: A Manual of Navaho Grammar
A Manual of Navaho Grammar. Arranged by Father Berard Haile, O.F.M.,
of the Cincinnati Province of St. John the Baptist. Santa Fe, New Mexico: Santa
Fe PubHshing Co., 1926. Pp. xi + 324. $6.00.
This is an invaluable work, useful both to the practical student of Navaho and
to the scientist interested in comparative Athabaskan linguistics, emanating
from the enthusiastic band of Franciscan missionaries to whom we owe the
excellent Ethnological Dictionary of Navaho and Navaho Vocabulary. It is a
cruelly difficult language that Father Berard treats in this "manual," and if a
number of statements or formulations seem doubtfully correct from the stand-
point of the comparative linguist, it should always be remembered that the
author's point of view is distinctly dialectic and descriptive, and not at all com-
parative or historical. It is very interesting to observe that Father Berard
explicitly recognizes the necessity of noting the pitch of the various syllables of
the Navaho word, though the record in the book is extremely fragmentary on
this point. This statement corroborates Father Legoffs remarks on pitch in
Chipewyan, and E. Sapir's field data on Sarcee and Kutchin pitch.
Editorial Note
Originally published in American Journal of Sociology 32, 511 (1926).
Reprinted by permission of the University of Chicago Press.
Sapir's references are to the Franciscan Fathers (1910, 1912) and to Legoff
(1889). Father Haile later published a considerably more sophisticated intro-
ductory grammar of Navajo (Haile 1941-48) and a stem dictionary (Haile
1950-51), both reflecting Sapir's influence.
A Summary Report of Field Work among the Hupa,
Summer of 1927
Dr. E. Sapir spent the end of June, July, August, and the beginning of Sep-
tember, 1927, in a linguistic investigation of the Hupa Indians, who occupy a
reservation in the valley of the Trinity river in northwestern California. This
work was done under the auspices of the [360] Department of Sociology and
Anthropology of the University of Chicago. A number of good informants was
secured, chief among them being Sam Brown, who served as the most impor-
tant source of material and as interpreter of everything obtained from him and
others. The information gathered was chiefly linguistic in character but a large
body of ethnological material was also obtained, partly in the form of texts and
partly as notes directly communicated in English.
The linguistic material was chiefly obtained as part of a programme for the
comparative study and reconstruction of the Athabaskan languages. A careful
study was made of the grammatical structure of Hupa, which offers many diffi-
culties, and a reasonably complete vocabulary was obtained. The texts, which
number about seventy-five, were so chosen as to duplicate as little as possible
the valuable material already published by Dr. P. E. Goddard. They are pre-
vailingly ethnological in content.
Some of the main results secured from the study of the Hupa language may be
mentioned here. The sound system, as might have been expected, proved to be
much more complex than hitherto represented and more in accordance with
the typical Athabaskan patterns worked out for Sarcee, Kutchin, and Navaho.
The old Athabaskan k-series (intermediate g, aspirated k', and glottalized 'k) is
represented in Hupa by a velar series (g, x, and 'q), but the old prepalatal series
(g, k', 'k) is preserved as such. Curiously enough, Hupa has also developed a
new series of k-sounds which are neither velar nor prepalatal but mid-palatal
(g, k', 'k). These do not represent the old Athabaskan series of k-sounds but are
the diminutive form of the prepalatal set. The sounds s, ts', 'ts represent not
only the old Athabaskan s-sibilants but also the diminutive form of the
Athabaskan c-series (in Hupa terms: voiceless w, tew, "Ic). Vocalic quantity
proved to be of fundamental importance for the understanding of Hupa mor-
phology. This is true also of the use of the glottal stop and of final aspirations.
Many final consonants arc glottalized and there is a characteristic and probably
archaic difference between non-syllabic final consonants and consonants which
have half-syllabic value. This difference is responsible for the parallelism of
"light" and "heavy" syllables, a distinction which had already been worked out
for other dialects.
The old Athabaskan tone system, which can be reconstructed in large part
from striking resemblances in the tone patterning of Sarcee, Kutchin, Navaho,
196 VI American Indian Languages 2
and no doubt by many other northern and [361] southern dialects as well, no
longer appears in Hupa. Mr. Li s researches prove the same loss of tone for
Mattole and Wailaki. Later investigation may disclose the fact that the absence
of tone is one of the distinguishing characteristics of the Pacific (or perhaps only
southern Pacific) group of Athabaskan dialects. There are, however, interesting
tone cadences in the relative forms of Hupa verbs which are most easily
explained as survivals of older forms with a high tone on the final syllable, such
as we actually find in Sarcee and Navaho.
The morphology was completely overhauled. Owing to a faulty phonology
the details and some of the fundamentals of Hupa structure had not been fully
grasped in former works on Hupa. The relative forms of the verb, which are as
important here as elsewhere in Athabaskan, had not been properly kept apart
from the non-relative forms. Owing to this fact the whole verbal system needs
to be presented in a new light. The so-called "past definite" is merely a special
use of the relative form of the perfective (Goddard s "present definite") and is
paralleled by analogous relative forms based on the imperfective ("present
indefinite") and on the continuative. The aspect system of Hupa needs to be
revised also in other respects. The formation of the continuative (as part prefix
and stem form) is entirely analogous to its formation in other dialects. A dis-
tinct permissive paradigm must be recognized not only for the third person but
also for the first. A distinct potential mode was discovered. There are also
special modal forms for the verbal abstract and for the prohibitive. In none of
these cases is reference merely had to the use of certain prefixed or suffixed
particles. As in Navaho, a number of verbs distinguish durative and momen-
taneous stem forms of the imperfective.
Since this report was first written, P. E. Goddard has published a paper
entitled. Pitch Accent in Hupa (Univ. of Calif. Publ. Amer. Arch. Ethn.,
23:333-338, 1928), in which it is shown that a study of Hupa tracings fails to show
that Hupa syllables have inherently high or low tones. This is entirely in accord
with Sapir's auditory record for Hupa and quite opposed to his auditory record
for certain other languages of the Athabaskan groups, particularly Sarcee,
Kutchin, and Navaho. A small amount of independent material obtained on
Anvik (Chapman's "Ten'a," also known as Ingalik) indicates that here too tone
is lacking.
Editorial Note
Originally published in American Anthropologist 30, 359-361 (1928).
Reprinted by permission of the American Anthropological Association.
References to Goddard's earlier work are mainly to his Morphology of the
Hupa Language (1905).
Six: Athabaskan and Na-Dene Languages 197
Sapir's Hupa notebooks and lexical files are now in the Library of the Ameri-
can Philosophical Society (manuscript 497.3 B63c Na20a.4; cf. Kendall 1982:
56, item 4369). The texts, with linguistic and ethnographic notes, have been
edited for publication in Volume XIV. The circumstances of Sapir's and Li's
1927 "expedition" to northwestern California are described informally by Sapir
(1927b, Volume IV).
Golla's Hupa grammar (1970) was based on Sapir's materials, supplemented
by Golla's own field work. Sapir's lexical data are incorporated into a compre-
hensive Hupa dictionary now being prepared by the Hupa tribe.
Excerpt from
The Concept of Phonetic Law as Tested
in Primitive Languages by Leonard Bloomfield
Bloomfield's experience with the Central Algonkian dialects is entirely par-
allel to my own with the Athabaskan languages. These constitute an important
linguistic stock which is irregularly distributed in North America. The north-
ern group occupies a vast territory stretching all the way from near the west
coast of Hudson Bay west into the interior of Alaska. To it belong such lan-
guages as Anvik (in Alaska), Carrier (in British Columbia), Chipewyan, Hare,
Loucheux, Kutchin, Beaver, and Sarcee. We shall take Chipewyan and Sarcee
as representatives of this group. The geographically isolated Pacific division of
Athabaskan consists of a number of languages in southwestern Oregon and
northwestern California. We shall take Hupa as representative. The southern
division of Athabaskan is in New Mexico and Arizona and adjoining regions,
and is represented by Navaho, Apache, and Lipan. We shall take Navaho as
representative of the group. In spite of the tremendous geographical distances
that separate the Athabaskan languages from each other, it is perfectly possible
to set up definite phonetic laws which connect them according to consistent
phonetic patterns. Navaho, Hupa, and Chipewyan are spoken by Indians who
belong to entirely distinct culture horizons, yet the languages themselves are as
easily derivable from a common source on the basis of regular phonetic law as
are German, Dutch, and Swedish.
II.
Ill
TABLE VI
Ath.
Hupa
Chipewyan
Navaho
Sa
1. s
s
e
s
s
2. z
s
8
z
z
3. dz
dz
d8
dz
dz
4. ts
ts
te
ts
ts
5. ts'
ts'
te'
ts'
ts'
1. c
W
s
c
s
2. j
W
z
j
z
3. dj
dj
dz
dj
dz
4. tc
tew
ts
tc
ts
5. tc'
to'
ts'
tc'
ts'
1. X
W
c
s
c
2. y
y
y
y
y
3. gy
gy
dj
dz
dj
4. ky
ky
tc
ts
tc
5. ky' ky' tc' ts' tc'
200 ^f American Indian Languages 2
Table VI shows the distribution in Hupa, Chipewyan, Navaho, and Sarcee of
three initial consonantal sets, each of which consists of five consonants. In other
words, the table summarizes the developments of fifteen originally distinct
Athabaskan initial consonants in four selected dialects. Each of the entries
must be considered as a summary statement applying to a whole class of exam-
ples.^ The table merits study because of its many implications. It will be
observed that no one dialect exactly reproduces the reconstructed Athabaskan
forms given in the first column. Series I is preserved intact in Navaho and
Sarcee and very nearly so in Hupa, but has been shifted to another series in
Chipewyan. Series II is preserved intact in Navaho, but has been shifted in
Sarcee to identity with the series that corresponds to original I, while Hupa has
introduced several peculiar dialectic developments and Chipewyan has shifted
it to the original form of I. Series III is nowhere kept entirely intact but nearly
so in Hupa, while in Chipewyan and in Sarcee it has moved to the original form
of Series II, in Navaho to a form which is identical with the original and the
Navaho form of Series Litis clear from the table that a Sarcee s is ambiguous as
to origin, for it may go back either to Athabaskan s or Athabaskan c. On the
other hand, a Sarcee s which is supported by either Navaho or Hupa s must be
representative of an original Athabaskan s. Sarcee tc is, in the main, unam-
biguous as to origin, for it corresponds to the original Athabaskan ky. It is
curious and instructive to note that, of the four languages given in the table,
Hupa and Chipewyan are the two that most nearly correspond as to pattern but
never as to actual sound except in the one instance of y (III, 2).
Let us take a practical example of prediction on the basis of the table. If we
have a Sarcee form with tc, a corresponding Navaho form with ts, and a
Chipewyan form with tc, what ought to be the Hupa correspondent? According
to the table it ought to be ky.
TABLE VII
Ath. Hupa Chipewyan* Navaho* Sarcee*
*kyan ... tcq n-t-tsq tea
'rain' 'there s a rainfall'
*q represents nasalized a. as in French an. Sarcee a is a peculiar a with velar resonance, regu-
larly developed from Athabaskan a.
Table VII shows the distribution in three dialects of the Athabaskan sound ky
(III, 4) in the word for "rain'. When I first constructed the Athabaskan pro-
totype, I assumed an initial ky, in spite of the absence of the test form in Hupa,
on the basis of the dialectic correspondences. Neither an original ts nor tc could
be assumed in spite of the fact that these sounds were actually illustrated in
4. The apostrophe symbolizes a peculiar type of consonantal articulation, characterized by simul-
taneous closure of the glottis and point of contact in the mouth, with glottal release preceding oral
release. 7 is the French/ of yowr, djh they of English yu^/.x is the c/j of German /c/i; Wis approximately
the wh of English what; e is the th of English thick; 8 is the th of English then.
Six: Athabaskan and Na-Dene Languages 201
known dialects, whereas ky was not. The Hupa column had to remain empty
because the cognate word, if still preserved, was not available in the material
that had been recorded by P. E. Goddard.
In the summer of 1927, however, I carried on independent researches on
Hupa and secured the form kyax^-kyoh,^ meaning 'hailstorm'. The second ele-
ment of the compound means 'big' and the first is obviously the missing Hupa
term corresponding to the old Athabaskan word for 'rain'. In other words, an
old compound meaning 'rain-big' has taken on the special meaning of 'hail-
storm' in Hupa. The Hupa form of the old word for 'rain' is exactly what it
should be according to the correspondences that had been worked out, and the
reconstruction of the primitive Athabaskan form on the basis of the existing
forms was therefore justified by the event.
TABLE VIII
Dialectic Forms for 'Rain'
Anvik (Alaska) rcoA^*
Carrier (B.C.) tcan
Chipewyan tcq
Hare tcq
Loucheux tcien
Kutchin tscin
Beaver tcq
Sarcee tcq
Navaho n-t-tsq
*D represents open o, as in German voll; q is nasalized o. N is voiceless n.
Table VIII gives the chief dialectic forms that were available for the recon-
struction of the Athabaskan word for 'rain'. Observe that not one of these has
the original sound ky which must be assumed as the initial of the word. This is
due to the fact that the old Athabaskan ky and related sounds shifted in most
dialects to sibilants but were preserved in Hupa and a small number of other
dialects, some of which are spoken at a great remove from Hupa. In other
words, in working out linguistic reconstructions we must be guided not merely
by the overt statistical evidence but by the way in which the available material is
patterned.
Editorial Note
Originally published in Rice, Stuart A. (ed.). Methods in Social Science: A
Case Book, pp. 297-306. Chicago: University of Chicago Press (1931).
Reprinted by permission of the University of Chicago Press. Only pp. 302-306
are reprinted here; footnote and table numbers are as in the original.
5. r) is the ng of English sing.
202 y^ American Indian Languages 2
This discussion of comparative Athabaskan phonology forms part of a longer
paper on comparative linguistic methodology (printed in full in Volume I).
Although Sapir s methods are irreproachable, his reconstruction of the Proto-
Athabaskan affricates is almost certainly incorrect, as Krauss has shown (1964).
Sapir gave little weight to forms in his own Kutchin and Anvik notes that are
now generally interpreted as retaining the Proto-Athabaskan contrast between
retroflexed (or flattened) and non-retroflexed palatal affricates (*ts'^ : *ts).
Sapir apparently believed that the retroflex/non-retroflex contrasts in Kutchin
and Anvik (and — had he known — several other Alaskan languages) were of
secondary origin. The retention of the contrast in Tsetsaut with quite different
phonetic realizations, and a corresponding contrast in Eyak, make this an
unlikely explanation. The Tsetsaut data were available to Sapir (Boas and God-
dard 1924), and Krauss has commented that "why or in what way Sapir ignored
that important material remains a dark mystery" (1986: 162).
Two Navaho Puns
It is a well known fact, often stressed by Boas, that the American Indians do
not go in for riddles and proverbs. What few exceptions have been found serve
only to emphasize the rule. If to riddles and proverbs we add puns, as it seems
we have a right to do, it begins to appear likely that the American Indian has a
generalized lack of interest in light verbal fancy. There is plenty of metaphor in
his rituals, there is considerable etymologizing ad hoc in his legends, and his
oratory is famous, but the zest in quick, irresponsible reinterpretation of famil-
iar words or phrases which lies at the bottom of the pleasure that we experience
in the telling of riddles, proverbs, and puns seems strangely un-Indian, what-
ever may be the reason.
Yet puns are not entirely absent, as the two following Navaho examples show.
They were collected in the summer of 1929 at Crystal, New Mexico.
In a council held some time ago the leader of one of the two contending
parties said,^ xdct'e ndhdnd^'d-' 'You people decide on one thing!' A cripple
who was present whispered to one near by, ndcidi-'d-' 'Pick me up!' The latter,
catching on at once, picked up the cripple and, holding him in his arms, asked,
xd-dic ndhdnc'd-' 'Where am I to put him down?' Everybody laughed. This is
said to be a favorite anecdote among the Navaho and depends for its point on
the double meaning of the verb nd-hd-ni- . . .- '«•', which may mean either 'to
decide on the matter' or 'to put him down'.
A close analysis of this pun shows that it is more subtle than appears on the
surface and that to enter fully into its humor requires sensitiveness to no less
than three changes of linguistic front. The fundamental pun is simple and
would hardly be enough by itself to raise a laugh, one suspects. This is the use of
the verb stem -a' 'to handle the "round" object' in the transferred sense of 'to
handle the affair, words, plan, date, decision'. The secondary use oi-'d' (-'d) in
an abstract sense is very common in Navaho, e.g. tc'6-hd- ni. . .-'«•' 'to tell'; nd-
hd-ni- . . .-'d- 'to make the decision'; nd-hd-. . .-'a 'to make plans', - d-nd-hd- . . .
-t'd-' ( = -d-'d-') 'a date is being set for one'. The tingling moment in the anec-
dote comes with the cripple's whispered [218] request; for, in addition to the
very general transfer of meaning already noted, there is the added point that
one does not normally use the verb nd-di- . . .- '«•' 'to pick it up' (and its cor-
relative ni-ni- . . .- '«•' 'to put it down') of an animate being, but only of such
inanimate 'round' objects as a potato, or apple, or watch, or rock. In other
words, the wily cripple, turning his helplessness to humorous account, classifies
his hunched up body as a 'round object', ndcidi- a' substitutes for ndcidi-ftxi-'
and xd-dic ndhohc'd-' substitutes for xddic ndhdhctxi' {-t-txi-' 'to handle the
animate being'). Had a little boy of normal physical health made the request,
1. Grave accent (a) represents low tone, acute (a) high tone, circumflex (a) falling tone.
204 yf American Indian Languages 2
the pun would have seemed a bit far-fetched, for he could not easily be thought
of as lifted up and put down like an inanimate object. The quick understanding
by the second punster of the cripple s use of -'a-' socializes the pun and kindles it
into something like satire of the ponderous doings of important people. Finally,
the climax of the pun, 'Where am I to put him down?', reinterprets the -hd-,
w hich in the first usage has the meaning of something like 'the affair, circum-
stance', while in the second it is a personal pronoun referring to 'this one'.
These two uses of -hd- are historically distinct in Athabaskan. Briefly, then, the
first element of the pun is impersonally contributed, as it were, by the language
itself; the second is the creation of a masochistic cripple; the third is the echoing
understanding of his friend, who equates 'the great business at hand' with 'this
poor chap'.
The second pun is much simpler. It is told as a joke rather than as an anec-
dote. 'So and so has gone over there', one says. 'What for?' xwdjdd-txal,
apparently 'he is going to give one a kick' (future of semelfactive verb yictxat
'I give it a kick'), actually 'he (the medicine-man) will perform a ritual "chant"'
{{uiwTt oi xatx a- 1 '\ perform a "chant"', a denominative verb based on jco/xfl/ 'a
ritual "chant"'). Here again the xwd- (= -hd- above) changes from an imper-
sonal to a personal application. It adds to the flavor of the joke if one remem-
bers that Navaho medicine-men are often hired to come from great distances in
order to direct curing rituals and that the spaces between the scattered Navaho
hogans are wide indeed.
The great number of homonymous elements in Navaho, due largely to the
leveling influence of phonetic laws, and its peculiarly intricate structure, which
derives quite definite meanings from the assembling of elements that are gener-
alized and colorless in themselves, combine to make Navaho a peculiarly
tempting language for the punster.
Editorial Note
Originally published in Language 8, 217-218 (1932). Reprinted by permission
of the Linguistic Society of America.
Problems in Athapaskan Linguistics
The Athapaskan languages, in spite of their essential uniformity in phonetic
type and morphology, are richly ramified and we need many more dialectic
studies than we have yet had in order to understand the group as a whole and to
reconstruct it adequately. The dialects of the Southern division are com-
paratively well known, though less has been published about these dialects than
about the chief languages of the other divisions. Even here, however, much
important dialectic work remains to be done. Manuscript materials on Navaho,
Mescalero Apache, Chiricahua Apache, Jicarilla and Lipan are probably abun-
dant and accurate enough to be called satisfactory. Further work remains to be
done on Kiowa Apache and on San Carlos and other Western Apache dialects.
In the Pacific area the manuscript and published record is probably adequate
for Hupa, Mattole and Wailaki. The languages in this area which need careful
study are Kato, Lassik, Chasta Costa, Galice Creek, and Upper Umpqua.
Kwalhioqua and Tlatskanai, which seem to have belonged to the Pacific group,
are probably extinct. It is particularly important to try to determine if there are
any survivals of the early Athapaskan tone system in the languages of the Pacific
group. That there was such an early archaic system follows from a comparison
of the Southern languages with most of the languages of the Northern group. So
far, however, no Pacific dialect has been found with a tonemic system, though
there is an interesting survival of such a system in certain grammatical forms of
Hupa.
The greatest complexity prevails in the north. Carrier, Chipewyan, Hare and
Kutchin are relatively well known though only a part of the material which has
been obtained is in print. Kutchin, however, probably needs a careful overhaul-
ing in the field because of its great complexity and the necessity of getting fuller
grammatical material. The Kutchin group of dialects is one of the most spe-
cialized and needs careful attention. Carrier may be considered reasonably
well known because of Father Morice's monumental work but the tonal pecu-
liarities of the Carrier group of languages still remain to be worked out. We
have some published material on Beaver and Ingalik but it is altogether
unsatisfactory and needs to be completely overhauled. The Athapaskan
dialects which are either not known at all or only fragmentarily include:
Chilcotin, Tahltan, Tanaina, Tanana, Han, Copper River, Koyukukhotana,
Slave and Dogrib.
Owing to the many intricate problems of phonology set by the task of recon-
structing Athapaskan, we cannot have too much dialectic evidence. The treat-
ment of original final consonants is particularly intricate. Another problem is
the adequate reconstruction of the tone system. Certain languages, like most of
the Pacific languages and Ingalik, seem to have lost the old tone system; others
206 ^f American Indian Languages 2
like Chipewyan and Navaho possess a two-register system; still others, like
Sarcee, have a three-register system, which may either be archaic or, more
likely, a secondary development of an earlier two-register system. In certain
Athapaskan languages, such as Hare and Kutchin, there is a tonal complication
due to the tact that the tonemic structure of a word differs according to whether
it is in final or absolute position on the one hand, or protected by an immedi-
ately following word. All these differentiations suggest the importance of get-
ting as much dialectic material as possible.
An important task in comparative Athapaskan is the recovery of the entire
vocabulary which is basic to the stock as a whole. Because of differentiations of
meaning in different dialects and the partial survival of many words in scattered
dialects, we cannot afford to do without the testimony of all or most of the
recoverable dialects.
Editorial Note
From a typescript with Sapir s ms. revisions, now in the American Philosoph-
ical Society Library, manuscript 497.3 B63c Na.l. Published by permission of
the American Philosophical Society.
The original of this short statement is a short typescript, edited and signed by
Sapir but bearing no date or other indication of its origin. It was perhaps pre-
pared for the American Council of Learned Societies or a similar source of
research funding. Morris Swadesh donated the typescript to the American
Philosophical Society Library in 1946.
Review of
A. G. Morice, The Carrier Language
The Carrier Language (Dene Family), a Grammar and Dictionary Combined.
A. G. Morice. (2 vols., I: xxxv, 660 pp.; II: 691 pp. RM 80. Anthropos Lin-
guistische Bibliothek, IX u. X Band. Modling bei Wien: Verlag der Interna-
tionalen Zeitschrift "Anthropos," 1932.)
This magnificent work, as its sub-title indicates, is not an ordinary grammar
but also a dictionary, the lexical materials being skilfully disposed under appro-
priate grammatical rubrics. The "Vocabulary" at the end of the second volume,
with its page references, helps the student to find his way in these materials.
There is no confusion, only a mutually fertilizing treatment of the complex
grammatical forms of the language and its lexical content. Not only is such an
interweaving of grammar and dictionary allowable for Carrier and its cognate
languages, it is in many respects necessary, at least if the grammatical survey is
to be complete and definite. In all the Athapaskan languages many compli-
cated grammatical rules apply only to single "words" or to small sets of words.
One cannot, therefore, give as adequate a notion of the more intimate structure
of Carrier or Kutchin or Navaho with a schematic statement of processes and
categories as is possible in such languages as Yokuts or Arabic or Jabo (in Lib-
eria), in which grammatical principles, once mastered, [501] can be applied
with a high degree of confidence to the words or word elements. It is not, of
course, a question of the relative complexity of the grammar as of the relation of
the grammar to the vocabulary. Father Morice's "Carrier Language" deserves
to rank as a real contribution to linguistic method. So far as I know, the problem
of handling grammar and vocabulary as a formal unity has never before been
attacked in so original a manner nor with so sure an instinct. Incidentally, this
single attack on a dual problem seems not unsuited to the genius of English,
different as its grammatical contours are from those of Carrier.
The body of the work consists of a preliminary treatment of Phonetics; Part
First, "The normally non-verbal Parts of Speech" (subdivided into "The
Noun," "The Adjectives," "The Pronouns," "The Postpositions, Conjunctions
and Interjections," and "The Adverbs"); Part II, "The Verb morphologically
considered" (subdivided into "The Verbal Stems," "The Verbal Prefixes," and
"The Incorporating Verbs"); Part III, "The Verb grammatically considered"
(subdivided into "General Notions," "The chief irregularities of the Verbs,"
"Grammatical Divisions," "Personal Divisions," "Modal Divisions," "Mor-
phological Divisions," "Temporally incomplete Verbs," "Divisions based on
Endings," and "Verbs with number-indicating Endings"); Part IV, "Syntax and
Linguistic Peculiarities" (subdivided into "Syntactic Notes" and "Linguistic
Peculiarities"); Part V, "Texts" (five texts with interlinear and free translations);
208 VI American Indian Languages 2
and Part VI, "Vocabulary" (an English check list). This list of the main headings
will serve to give some idea of the scope of the work. Details are hardly in place
here.
It should be remembered that the strength of Father Morice's grammar lies in
its wealth of descriptive detail, not so much in its ultimate configurative analy-
sis. A comparative student of Athapaskan may want to reassemble much of the
detail and redefine some of the fundamental outlines but he will always be
grateful to Father Morice for the facts, of which he has an obvious mastery. One
weakness of the work — which is, however, not an important drawback in prac-
tice — is that the orthography is not sufficiently systematized from the
phonemic point of view. Thus, both t and d are used for what is obviously a
single phoneme, a lenis stop, while the corresponding aspirated fortis stop is
written th. One might write these phonemes, with advantage to economy and
phonemic accuracy, either d:t (defining d asa voiceless lenis, varying to the ear
between French d and t, and t as a strongly aspirated stop) or t:t\ Either method
is justifiable and adequate, whereas d:t:th corresponds only vaguely to the true
facts; d.th would be better but uneconomical. It is time that linguists realized
that one of their first and most fundamental problems is to interpret the mass of
purely phonetic data in terms of a phonemic symbolism that is as simple, as
economical, and as powerful as possible. This ideal is rarely attained, yet its
realization is implicit in the phonetic facts themselves.
There are many ethnological remarks and implications scattered in the book.
The anthropologist's attention should be called to the list of ethnological refer-
ences on pages xxix-xxxv of the first volume.
Editorial Note
Originally published in American Anthropologist 37, 500-501 (1935).
Reprinted by permission of the American Anthropological Association.
Internal Linguistic Evidence Suggestive of the Northern
Origin of the Navaho
TNTERNAL linguistic evidence for inferences as to cultural antecedents
-*■ is not in much favor among cultural anthropologists at the present time,
and this for two reasons. Such linguistic evidence is often, if not generally,
tricky as to what of a factual nature can be gathered from it, for words may
change their meanings radically and, furthermore, it is often difficult to
tell whether community of nomenclature rests on early linguistic relation-
ship or on linguistic borrowing attending cultural diffusion. In the second
place linguistic evidence is difficult to handle, full of phonologic pitfalls,
requiring a closeness of knowledge that is often out of proportion to what
little can be obtained from it for tangible cultural inference. Nevertheless,
at its best linguistic evidence, properly controlled, may throw an un-
expected light on remote cultural perspectives. There is reason to think
that as our descriptive and comparative knowledge of unwritten languages
increases, their value for cultural reconstructions and other kinds of in-
ference — not least among which is elimination of theoretically conceivable
possibilities — will grow in importance. It is natural that in the American-
istic field linguistic evidence has as yet yielded but a scanty return to the
historian of culture, but this need not continue to be the case indefinitely.
I shall try to show that there is tangible evidence in Navaho itself for
the secondary origin of apparently fundamental elements of Navaho cul-
ture, such as agriculture, and that such evidence seems to point to an early
association of the culture of these people with a more northern environ-
ment than their present one. It may be said — and with justice — that the
distribution of the Athapaskan languages is such as to make this historical
theory as good as certain, but dialectic distribution is external, rather than
internal, linguistic evidence. It is conceivable, if not plausible, that the
Athapaskan-speaking tribes were originally massed in the Southwest and
gradually rayed out to the north in successive waves of migration. One
might argue that the Navaho and, to a greater degree, the various Apache
tribes present the non-Pueblo aspect they do, not because of their relative
recency in the area of Pueblo cultural development but because, like the
Walapai and other Yuman tribes of Arizona, they represent a simpler and
more archaic Southwestern culture, which proved impervious, aside from
a late Pueblo veneer, to the influence of the more elaborate cultures in their
neighborhood. It is true that the linguistic homogeneity of the Southern
Athapaskan dialects is such and the dialectic cleavages in the nothern
224
210 VI American Indian Languages 2
225
Athapaskan area are so profound that the suggested theory fails to carry
conviction either to the linguist or to the ethnologist, but here again we
are dealing with external linguistic evidence. This external evidence is far
more compelling than can be any evidence derived from details of dialectic
structure or vocabulary, for it is more direct and sweeping. None the less,
the more elusive internal linguistic evidence has its place in giving con-
firmation to a hypothesis based on linguistic distributions.
There is undoubtedly a large amount of relevant cultural evidence
packed away in the vocabularies of Navaho and Apache. For the present
I must content myself with considerations based on the study of four
words or groups of words.
1. The Navaho word for "gourd" is 'ade'.^ The word is used both for
the plant and for the "gourd dipper, ladle. "^ The "gourd rattle," on the
other hand, is otherwise named ('a7al).' But 'ade-' means not only "gourd
ladle" but "dipper, ladle, spoon" in general, the gourd ladle being the ladle
or spoon par excellence. Hence we find the earthen spoon called "mud 'ade'"
or "earth 'ade'," while the modern tablespoon is called "metal (< flint)
'ade'."^ Now the term 'ade' (in form a possessed noun -de*' with indefinite
possessive prefix 'a- "somebody's" or "something's") means not only
"gourd," "gourd ladle," and "ladle, spoon" in general, a natural family of
words, but also "horn" or rather "somebody's, some animal's horn" (d6
"horn" as absolute; 'a-de*' "an animal's horn," parallel to bi-de' "his
[animal's] horn"). In no other Athapaskan dialect does 'ade' or its dialectic
equivalent mean "gourd" or "gourd ladle," while, so far as I can discover,
it is only in Apache that it means not only "horn" but also "ladle" in
general. In Chiricahua Apache* we have possessed -de' "horn (of animal)"
and 'ide' "cup, dish, dipper"; in Mescalero Apache -de" "horn (of animal)"
and 'ide- "cup, dish, dipper." In both Navaho and Apache 'ade*', 'ide',
'ide', in its meaning of "gourd ladle" or "dipper," keeps its indefinite
possessive prefix 'a-, 'i-, when itself possessed, e.g., Nav. be-'ede^' (as-
similated from *bi-'ade') "his gourd ladle," Chiricahua Apache bi-'ide'
"his dipper," Mescalero Apache bi-'ide*. This does not in the least prove
' See, e.g., Franciscan Fathers, A Vocabulary of the Navaho Language (2 vols., St
Michaels, Arizona, 1912), Vol. l,p.99, sub "gourd;" Vol. 2, p. 13, sub ade', where it is defined
as Cucurhita.
' See Franciscan Fathers, An Ethnologic Dictionary of the Navaho Language (St
Michaels, Arizona, 1910).
'/fc/rf, p. 401.
* Vocabulary, Vol. 1, p. 186, sub "spoon."
' My Chiricahua and Mescalero Apache forms are quoted from manuscript material
kindly put at my disposal by Dr Harry Hoijer.
Six: Athabaskan and Na-Dene Languages 211
226
that Navaho 'a-de' "one's horn" and 'ade' "gourd ladle" are unrelated
words, for we have other examples in Southern Athapaskan of double
possessives of type "his-one's . . ."; e.g., Navaho bi-t'a' "his (i.e., bird's)
feather," 'a-t'a' "a (bird's) feather," but be-'et'a' "his-one's-feather," i.e.,
"his (secondarily owned) feather, his plume (used in hair decoration)."
All this suggests that Navaho 'ade*' "gourd ladle" originally meant "ladle"
in general and that this word in turn originally meant "an animal's horn,"
reinterpreted as "horn spoon," very much as our musical instrument, the
"horn," originally a "ram's horn" used for blowing, is now a brass instru-
ment with no obvious relation to an animal's horn. The semantic history
of 'ade'' would, then, be: (1) an animal's horn; (2) ladle made of horn;
(3) any ladle; (4) gourd ladle; (5) the gourd, Cucurhita, of which ladles are
made. Stage 1 would be pro to- Athapaskan; 2, a dialectic Northern and
Pacific, and presumably early Southern, development based on the wide-
spread use of horn for spoons; 3, a Southern Athapaskan transfer of mean-
ing due to the fact that spoons were no longer made of horn; and 4 and 5,
a specific Navaho (in part perhaps also Apache) development. Inasmuch
as stage 2 no longer has validity in Navaho, the meanings of the word
group into two disconnected "sets (1; 3-5), so that 'ade'' is now felt to be
two distinct and unrelated words, the more so as it is tabooed among the
Navaho to use the horn of the deer for the making of spoons. My inter-
preter Albert Sandoval once volunteered surprise that identically the same
Navaho word meant both "a horn" and "gourd, gourd ladle."
If we turn to other than Southern Athapaskan dialects, we find that
the absolute *de "horn," the possessed *-de"' "horn of . . . ," and the form
with indefinite possessive prefix *^e-de'' "an animal's horn" are found in
both of the two other Athapaskan areas. Corresponding to Navaho de we
have, e.g., Kutchin |i "horn" and Hupa -de- (in compounds); correspond-
ing to Navaho -de' we have Carrier -de, Chipewyan -de (Li) (Chipewyan
high tone = Athapaskan [Navaho, Apache, Sarsi, Kutchin] low tone). Hare
-de, Loucheux -|i, Kutchin |i', Beaver -de', Sarsi -da', Hupa -de', Kato
-de', and Mattole -de'; while Navaho 'a-de' has an exact correspondent
in Chipewyan 'i-dk, Hare e-de, Loucheux e-|i, Kutchin c'i-^i', Batard
Loucheux e-d^e, and Hupa ^i-de', all meaning "an animal's horn." The
early use of horn for spoons, which can only be inferred for Southern
Athapaskan, is linguistically reflected in Hupa ^ide-^in', literally "a horn's
handle," whence "spoon," and in Hare ede-k*a "cuiller en come" (Petitot:
k^a "plate, bowl") and Sheep Indian (esbatahot'ine) ede-ka "corne aplatie"
(Petitot), whence "spoon." Obviously, to the Navaho mind 'ade' in its
meaning of "gourd" must be referred to the beginning of things, for the
212 VI American Indian Languages 2
227
term is used in ritual and mythology, for example in the compound term
Gourd Children,® but the feeling of the Navaho is of no more importance
in the historic problem than their conviction that ly' always meant "horse"
(though we can easily prove from comparative evidence that its original
meaning was "dog") and that their ancestors became acquainted with the
horse not too long after the Emergence, as indicated by the origin legend
for the creation of the horse in the four cardinal points out of the four
ritualistically proper materials.
Our linguistic analysis, in short, points unmistakably to two things of
historical interest: that the gourd is not an original element of Southern
Athapaskan culture; and that horn spoons, not directly given by present-
day Navaho culture, must be assumed to have been known to the remoter
Athapaskan-speaking ancestors of the Navaho or, at the least, to early
Southern Athapaskan culture. These inferences go well with a theory of
immigration of the Navaho and Apache from the north (or east) into the
Southwest. Even if one goes no further than to infer the absence of the
gourd and the presence of horn spoons in an early phase of the culture of
the Navaho-Apache tribes, the illumination brought by a close analysis
of Navaho 'ade"' and its Apache cognates is useful for the reconstruction
of the period antedating the massive influence of the Pueblos on the Navaho
and the Apache tribes.
2. The Navaho verb for "seed lies" is -sas, a perfective neuter, e.g.,
sisas "the seed lies," nsas "the seed lies in a row." The original meaning
of these forms is obviously not specifically "the seed lies" but, more gen-
erally, "the mass of finely divided particles (e.g., grain, sand) lies." A cor-
responding active verb, nasas, means, for instance, not merely "I scatter
the seed,"^ but also "I let the mass (of grain, sand) spill (e.g., out of a
bag); I sprinkle it (e.g., sand, water)." I can find no cognate for these
verbs in the material available to me from other Athapaskan dialects, and
. the inference — as so often in analogous Athapaskan cases of apparently
isolated verbs — is that we probably have here a dialectic denominative
formation, i.e., a secondary set of verbs based on a noun.
Now it is perfectly clear from Navaho phonology that all verb stems
beginning in s (after vowels) are contracted products of a "classifier" -1-
and either z or y; in other words, -sas must go back to either -1-zas or
-t-yas. The perfective neuter *si-}-zas or *si-l-yas is analogous in form to
such a perfective neuter verb as silcoz* "the fabric lies." But what is the
« An Ethnologic Dictionary, pp. 351, 353.
' See, e.g.. Vocabulary, sub "broadcast (in sowing)."
•c = ts.
Six: Athabaskan and Na-Dene Languages 213
228
underlying zas or yas? Quite obviously, "snow," Navaho zas, yas (these
two forms, of which the latter is the more archaic, constitute one of the
few cases of dialectic difference within Navaho). Hence the verb sisas
must have meant, originally, "it lies like (flakes of) snow," whence "the
seed lies"; the derived active verb nasas originally meant "I scatter it
about (so that it lies) like snow," whence "I sow the seed broadcast."
As in the preceding case, while the present cultural term is not wide-
spread in Athapaskan but is confined to Navaho (or Southern Athapaskan),
it is not difl&cult to establish a close connection with a universal Atha-
paskan term of differing cultural connotation. Athapaskan *yaxs' "snow"
is found in the majority of Athapaskan dialects: Ingalik yi5, Babine yis
(Jenness), Carrier y3§, Chipewyan ya0 (Li), Slave zah, Hare, Dogrib z^ah,
Loucheux ?iow, Kutchin zah, Kaska zds (Jenness), Beaver yas, Sarsi zas,
Kwalhioqua yaxs, Hupa yahs,^° Kato yas, Mattole yas, Jicarilla Apache
zas, Mescalero, Chiricahua zas. The original meaning of the Athapaskan
word is not "snow" in general but specifically "snow lying on the ground;"^^
another common Athapaskan term, represented by Navaho cil, means
"snowdrift" or "falling snow." This restricted meaning, "snow lying on
the ground," is clearly the prototype of the present Navaho term for "the
seed lies." To summarize, a non-agricultural term ("snow lying on the
ground") takes on a transferred and more general meaning in a classifica-
tory verb ("the finely divided particles lie [snow-like] on the ground")
and, in a secondary, agricultural environment, advances to the technical
meaning of "the seed lies." No other sequence of meanings fits the linguistic
facts.
3. The Navaho word for "corn" is nad4'. The second element, -d4',
occurs in a number of compound nouns referring to plants in which it tends
to be translated "corn" by Navaho interpreters, e.g., hasce'd4'' "box-
thorn," approximately "god-corn" (hasce' is a familiar Navaho god name
and, in slightly abbreviated form, is the first element in the native term
for the Talking God); ci'd4' "buckthorn," perhaps "bitter corn" (cf.
dici' "it is bitter"); ma'id4' "cedar-berries," literally "coyote-corn;"
» My reason for reconstructing to high-toned *y4xs rather than low-toned *yixs is too
technical to give here.
" The Hupa word means not "snow" but "white frost (on trees)."
" See, e.g., Fang-Kuei Li, A List of Chipewyan Stems (International Journal of American
Linguistics, Vol. 7, 1933), p. 146: "y^ snow on the ground;" similarly, for Chipewyan, L. Le
GofiF, Dictionnaire Franfais-Montagnais (Lyons, Marseilles, and Rome, 1916), sub "neige
tomb^e;" for Carrier see A. G. Morice, The Carrier Language (2 vols., Modling bei Wien,
1932), Vol. 1, p. 25, where ye§ (our yas) "snow" contrasts with eel (our Sal) "snow (heavy and
not yet settled)."
214 VI American Indian Languages 2
229
gahcohdi' "winterfat," literally "jackrabbit-corn." The reason why, in
compounds such as these, -di' is translated "corn" rather than "food,"
which is obviously more logical in such terms as "coyote-food" and "jack-
rabbit-food," is probably the use of the abbreviated -d4' for "corn" in
possessed forms (e.g., sid4' "my corn") instead of the fuller na-d4'of the
absolutive. But it is quite easy to prove that -d4' is not, in any true sense,
abbreviated from nad4' "corn" but, on the contrary, is an old term for
"food" which lingers, somewhat disguised, in such compounds as have been
quoted and in possessed forms for "corn" (sidi'' "my food" par excellence,
whence "my corn"). This interpretation, not clear to the Navaho himself
because the word in actual use for "food" is ci-yan and he therefore feels
that the primary meaning of -d4' is, or should be, "corn," is at once made
plausible from within Navaho when we compare -d4' with the medio-
passive imperfective neuter verb -d4 "to be eatable" (e.g., yid4 "it is
eatable"), itself closely related to the durative transitive verb -y4 "to eat
it" (from which ciyan above is independently derived). It looks, there-
fore, as if -d4' originally meant "what is eatable," i.e., "food," secondarily
"corn" in possessed forms.
The na- of nad4' is quite obscure to the Navaho. It seems to follow
no obvious analogy and cannot be equated with the common na.-- "about,
here and there" of continuative verbs. One might venture nad4', originally
"corn is here and there," whence "planted corn, standing corn," finally
generalized to "corn." This is to be taken no more seriously, however, than
an attempt to see our common word tide in the -tide of eventide, whereas
every historical student of English knows that this compounded -tide is
a survival of an old word tide synonymous with time and cognate with
Danish tid and German Zeit. Our problem cannot be considered com-
pletely solved until we have done more than plausibly surmise that -d4''
originally meant "food" and have found a linguistically unforced explana-
tion of na-. The former requirement is met by a consideration of Atha-
paskan cognates, which reconstruct to *dan-6 (itself reduced from *dg-han-6
"that which is eaten, food," relative form in *-g of *d5-han, whence *-dan,
"to be eaten, to be eatable"), possessed form *-dan-e', *-dan-g' "food
of . . .:" Chipewyan d^ne (Li), possessed -d^ne, Sarsi dani, Mescalero
Apache dan, possessed -dan and -d4'', Chiricahua Apache dan, possessed
-dan (also -d4' in nad4' "corn," perhaps borrowed from Navaho), Hupa
possessed -dan' in -dan' sa'an "food of ... is lying" = ". . . is saving with
food," Mattole possessed -dane' "possession, property" (presumably a
meaning enlarged from "food"). These forms enable us to understand the
exact status of Navaho -d4'. It is not the reflex of the primary *dan6
Six: Athabaskan and Na-Dene Languages 215
230
"food" but of its possessed form *-dan6' "food of . . ."; the former (exem-
plified by Chipewyan d^n^, Sarsi dam, and Apache dan [read dan, for an
old Southern Athapaskan *dan, monosyllabic, would have yielded Navaho,
Apache *dan, while an old Athapaskan *dan would have given Navaho,
Apache *d4]) would have resulted in Navaho *dani, *dan. The Apache
possessed forms in -dan are merely generalized from the absolutive dan,
the variant Mescalero -d4-' "food of , . ." being the true reflex of Atha-
paskan -*dan-6' and an exact cognate of Navaho -d4' "corn." We see,
therefore, that on strictly linguistic grounds such Navaho forms as ma'rdi"'
mean, not "coyote-food," but "coyote's food." This makes it doubly im-
possible to interpret nadi' as "corn here and there," which form, if it
ever existed, would have had to yield *na-dan in Navaho. We are driven
to infer that nad4' originally meant "food of na,--," whatever na*- may be.
Once we see that na,'- must have referred to certain beings, human or
animal, whose food was corn, we advance rapidly to a satisfactory linguistic
solution. Many Athapaskan dialects have reflexes of an old word for
"enemy, aliens," occurring in two forms (*na', na,-- in compounds, and,
with indefinite possessive prefix, *]|e-(dg-)na-', *^e-na- in compounds).
These words are frequently used to refer to specific neighboring tribes.
Examples of *^e-na-', *^e-d6-na-' (*de- is collective) and compounded
*^e-na-- are: Carrier 9-d-na, Chipewyan '^-na (Li) "enemy, Cree Indian,"
Slave e-na-kie "Eskimo," Hare e-h-da "enemy," e-na-ke "Eskimo," Dog-
rib e-h-da, Loucheux 9-ne "enemy, Eskimo," Batard Loucheux a-ra-ke
"Eskimo," Kutchin ce--k''6i (contracted from *ca-ne--) "Eskimo," Hupa
]^i-na' "Yurok Indian," Navaho 'a-na-', Mescalero, Chiricahua Apache
'i-nda\ (The -kie, -ke, -k^oi of some of these forms, analogous to Navaho
-ke, is a plural animate sufl5x.) The old compounded form without in-
definite prefix, *na-, is illustrated in Chipewyan na-t'j-i "enemy" (Petitot)
(literally, "the one who acts as an enemy") and, presumably, in Kato
na-cal "orphan" (from "alien" + "child, little"). In Navaho this na,-- is
found in compound nouns, particularly such as refer to foreign peoples, e.g.,
na-lan "Comanche Indians" (from "enemy-many- the"), na'-sfezi "Zuni
Indians" (contracted from na-yist!ezi "enemy"+"the ones who are black-
ened"), nasgali (apparently made over, in accordance with the Navaho
tribal name pattern, from masgali "Mescalero"),^^ natoho "Laguna In-
dians" (apparently also "Isleta Indians"?). The last of these tribal names is
interpreted as "enemies at the water" by the Franciscan Fathers^^ but a
" See Vocabulary, Vol. 1, p. 127, sub "Mescalero Apache."
" Ibid., Vol. 2, p. 135, sub natqoho; better "at the river," for this name is based on
to h, possessed -td -h, "river which does not dry up," rather than on t6 "water."
216 VI American Indian Languages 2
231
more natural interpretation is to take the name as a relative in -i (assimi-
lated to -6) from natoh "enemy-river," presumably an old name for the
San Jose (and Rio Grande?), in contrast to the two normal interpretations
of toh, namely San Juan River and Little Colorado River, the two rivers
in or near the old Navaho habitat which never completely dry up. This is
confirmed by the place name na*t6'(h) siiai' given by the Franciscan Fathers
for Grant, New Mexico (natqo s3,kal)," literally "the enemy river has
its legs distended," "(where) the San Jose turns crotch-wise." The point is
of some importance linguistically as indicating that Navaho compounds in
na"- "enemy" not only mean "... enemies" but also "enemy ..." In
other words, both natoh and nad4' are archaic Navaho words which
qualify basic nouns ("river" and "food") by referring them to the enemy,
in this case the Pueblo Indians.
The Navaho word for "corn," nad4', in summary, which can be an-
alyzed with great probability into an older "food of the enemy," "Pueblo
food," implies that there was a time when the Navaho, an agricultural
people in historic times, were still thinking of corn as an alien food. Later on,
when they had adopted corn as a staple and had built so much of their myth
and ritual around it that it was inconceivable to them that there could be
anything alien about it, they could not possibly feel the na- of their word
for "corn" as akin to the -na' of 'ana*' "enemy" and the na- of tribal
names. The sentiments clustering about the two terms had become irre-
concilable.
4. There is a curious verb stem in Navaho which seems to be used only
in certain quite specialized verbs; this stem has the forms: imperfective
-keh (probably error for -keh), perfective -k\, progressive and future -ket,
usitative and iterative -keh, optative -kel. It is used in an idiomatic verb
referring to sleeplessness, e.g., iterative bit sicanake'h "sleeplessness always
bothers me," perfective bil sicankj "I have been sleepless." The form of the
verb is such (bit "sleep" is subject; -ca- "away from" is preceded by the
indirect pronominal object) as to suggest that the verb stem refers to a
specific type of movement. My interpreter, Albert Sandoval, had no notion
what the underlying metaphor was but said he felt, somehow, that there
was a reference to gliding movement in it: "sleep glides (slips) away from
me." There is no linguistic support for this feeling, which is hardly more
than an ad hoc interpretation to fit the linguistic form. This obscure verb,
as Sandoval pointed out, must have the same stem, in its progressive form
(-kel), as the sacred name of the owl, cahalxel yil nakeli "darkness with-
it the-one-who-comes-gliding(?)-back, the one who comes gliding (?) back
** Ibid., Vol. 1, p. 226.
Six: Athabaskan and Na-Dene Languages
111
232
with darkness." The image of gliding is not so apposite here. The simple
progressive form, which would be *yiket "it glides (?) along," is not in use
in Navaho. There is nothing to be done with these isolated forms except to
see in them survivals of an old set of verbs of movement which perhaps still
occur in other Athapaskan dialects.
Turning to Chiricahua Apache, we find the verb stem: imperfective
momentaneous -ke" (continuative -ke), perfective -kj, progressive and future
-kel, usitative and iterative -ke, optative (evidently transferred from im-
perfective) momentaneous -ke* (continuative -ke). Its meaning is given as
"several run, trot," which is by no means easy to reconcile with the hypo-
thetical "glide" of the Navaho words. If the Navaho and Chiricahua
Apache words are historically related, as is indicated by their strict formal
parallelism, it must be because each dialect has developed specialized mean-
ings that diverge from a third term. Now the distribution of the meanings of
the Northern and Pacific Athapaskan verb stems which are demonstrably
cognate to the Southern Athapaskan stems is such as to leave little doubt
of what this third term must have been. The following table of stem forms
gives a summary of dialectic meanings and of phonetic equivalents for four
selected stems of the set:
Athapaskan
1. Ingalik "to travel by canoe'
2. Loucheux (ditto)
3. Kutchin (ditto)
4. Carrier (ditto)
5. Beaver (ditto)
6. Hare (ditto)
7. Chipewyan (ditto)
8. Sarsi "to travel by canoe;
to go for trade"
9. Ts'ets'aut "to travel by canoe'
10. Chasta Costa (ditto)
11. Hupa "to travel by canoe;
several objects float"
Momen-
Perfec-
Progres-
Continua
taneous
tive
sive
tive
imper-
imper-
fective
fective
*-ke"x
*-ken
*-ke-}
*-ke
-kaix
-kan
-kal
-kai
-k?
-ka
-k*6i
-k-oi
-k^a-
-keh
-kei
-kal
-ke
-kc
■ki
-ktl
(read -keh?)
-kc
-ke
-k9ih
-ki
-kel
-kah
-ki
-kal
(
kin- before
vocalic
suffix)
' -ki-
-ke-
-xe
-XI g
-xil
-xe'
218 yi American Indian Languages 2
233
12.
Mattole "to travel by canoe'
' -kxix
-kxir)
-k?i-l
(relative
form)
13.
Kato "several bathe"
-ke'
(transferred
from cont. pf.,
it., and opt.?)
14.
Chiricahua Apache "several
run, trot"
-kl
-k|
-kel
-ke
15.
Navaho (only as survival in
obscure forms)
-kth
(read -keh?)
-kj
-kel
The history of the meaning of these verb stems is now reasonably clear.
The primary meaning of the Athapaskan verb stems may have been "sev-
eral objects (or persons) move in the water, float" (see Hupa and Kato
above), whence "the group travels on the water, to travel by canoe (as one
of a canoe-party)." Both meanings are preserved in Hupa. The latter mean-
ing, however, may well have been the primary one. The specific meaning of
a group traveling by water seems, under changed environmental conditions,
to have taken on a new meaning in Chiricahua Apache ("several run, trot"),
though the old plural or collective implication is still preserved. In Sarsi
the meaning of "to travel by canoe" is now felt to be rather archaic and to
belong to myth and story. The natural meaning today is "to go to trade,
to go (by foot or horse-back) in order to shop;" this is developed from "to
go by boat (or canoe) in order to trade at a Hudson's Bay Co. trading post,"
itself specialized from the common Athapaskan meaning "to travel by
canoe." Here too the gradual passage to a typical Plains life, with little or
no use of water craft, has brought about a redefinition of a familiar set of
words. The Navaho words seem to stem from an old meaning "to travel by
canoe," naturally entirely eflfaced from tribal memory. A generalized mean-
ing "to float," applying to singular as well as plural subject, cannot be
assumed for early Navaho because there is no evidence anywhere in Atha-
paskan for a reflex of *-ke"x in the sense of "one person floats" and because
all Athapaskan dialects are peculiarly sensitive to the difference between
singular and plural forms of verb stems referring to characteristic types of
movement. "I become sleepy," in other words, seems originally to have
meant "Sleep paddles away from me;" the Owl was ritualistically described
as "he who brings Darkness back in his canoe." Such locutions seem to stem
from a cultural setting in which travel by canoe was so much a matter of
course that it could be transferred to the supernatural world.
Six: Athabaskan and Na-Dene Languages 219
234
The Navaho nakei "he comes 'gliding' home" (of which nakeli in the
sacred name of the owl is the relative form) is contracted from an old
Athapaskan progressive *na-7g-(d6-)ke"l, of which there are exact reflexes
in many of the other dialects, e.g., Sarsi na-71-kal "he's coming back on a
boat, he's returning from shopping;" Beaver na-7a-kil (read -kel?) "he is
paddling back;"'^ Carrier na-s-ket (contracted from *na-7e-s-ke"l) "I am
again navigating, I am returning by boat;'"® Ingalik n3-79-d3-kal "he
paddles again. '"^
The evidence collected in this paper may now be summarized. (1) It is
assumed that there is important external linguistic evidence, distributional
in character, to provide a prima facie probability of the northern origin of
the Navaho and Apache. All the Southern Athapaskan dialects (Navaho,
Western Apache, Mescalero and Chiricahua Apache, Jicarilla Apache,
Lipan, and Kiowa Apache) obviously form a close-knit dialectic unity
which contrasts with the more complex dialectic ramifications of Pacific
and Northern Athapaskan. The geographical center of gravity of these
languages, in short, lies in the north. (2) If we could find internal linguistic
evidence in Navaho, of cultural implications, tending, as it were, to free
Navaho and Navaho culture from their present Southwestern environment,
the initial probability of a northern provenience would be strengthened.
Such supplementary strengthening of an inherently probable hypothesis
is suggested by the linguistic analysis of four Navaho words having cultural
connotations. The cultural inferences that may be derived from this analy-
sis are: that the gourd was not originally an element of Southern Athapas-
kan culture; that spoons in this culture were originally made of horn; that
broadcast sowing of seed was foreign to the culture; that maize, a staple in
historic times, was at one time felt to be an alien food — in other words, that
the Southwestern agricultural complex was originally lacking; and that a
glimpse, faint but not to be lightly argued away, may be had of a time when
the Navaho, or Southern Athapaskans collectively, made use of canoes.
(3) All of these inferences deepen, in a historical sense, the cultural gap be-
tween the Navaho and the Pueblos. This gap is already given, in a descrip-
tive sense, though in lesser degree, by the modern ethnologic evidence. The
first four of the cultural inferences we have listed are theoretically compati-
ble with a non-Pueblo Southwestern cultural setting and, equally, with a
" Pliny Earle Goddard, Beaver Dialect (Anthropological Papers, American Museum of
Natural History, Vol. 10, Pt. 6, 1917), p. 506.
" Morice, The Carrier Language, Vol. 1, p. 279.
" John W. Chapman, Ten 'a Texts and Tales from Anvik, Alaska (Publications, American
Ethnological Society, Vol. 6, 1914), p. 158, 1.1.
220 ^f American Indian Languages 2
235
more northern setting. The last of these inferences, if valid, points more
positively to a northern setting.
"Northern origin" does not in the least imply a direct line of movement
from north to south across the Great Basin. Such a line of migration is
most improbable. It is far more likely that the movement of these peoples
proceeded via the western plains. If this is correct, an analysis of Southern
Athapaskan culture would aim to reveal four strata: a fundamental north-
ern layer, comparable to the culture of the tribes of the Mackenzie basin;
an early western Plains adaptation, more archaic in its outlines than the
specialized culture of the Plains as now defined by ethnologists; a first
Southwestern influence, tending to assimilate these tribes to the relatively
simple non-Pueblo culture of the Southwest; and a second, distinctively
Pueblo, Southwestern influence. To these must, naturally, be added a good
deal of Navaho specialization on the basis of the Pueblo influence. The
disentangling of these various layers is work for the future and, in any
event, is hardly likely to be ever more than fragmentary. Meanwhile, the
geographical sequence: Chipewyan, Sarsi, Kiowa Apache, Jicarilla Apache,
Navaho, may stand as a suggestion of the reality of the historical problem,
though, no doubt, the Plains character of Sarsi and Kiowa Apache culture
is in each instance of a much later type than the hypothetical Plains in-
fluence to be worked out for Navaho cultural antecedents.
Yale University
New Haven, Connecticut
Editorial Note
Originally published in American Anthropologist 38, 224-235 (1936).
A. L. Kroeber called this paper a "matter of virtuosity. . .Sapir just took out of
his head or from scattered literature the particular little bits that he needed and
wove them into this masterpiece" (Kroeber 1984: 133).
Contribution to
Cornelius Osgood, The Distribution
of the Northern Athapaskan Indians
Linguistic Classification within the Northern Athapaskan Area
Linguistic classification within the Northern Athapaskan area is still for the
most part extremely uncertain. I am indebted to Dr. Edward Sapir, whose
authority is preeminent in this field, for the few statements and tentative sug-
gestions which may be presented at this time. First, it appears certain that there
are major linguistic divisions within the Northern Athapaskan area, which are
individually equal in comparative weight to the whole Southern, or Pacific,
Athapaskan division taken collectively. Linguistically, the difference between
Ingalik and Chipewyan, or between Carrier and Sarsi, is as great a contrast as
that between Chipewyan and Navaho. Whereas there is apparently a true lin-
guistic unity in both the Southern and the Pacific Athapaskan divisions, the
Northern Athapaskan area, as such, does not form a linguistic unit. Secondly,
two languages of the Northern Athapaskan group, Kutchin and Tsetsaut, stand
out from any further internal [22] alignment as individually distinct, the former
being probably the most specialized of all Athapaskan languages. Finally, it
seems probable that there are relationships among fifteen other groups which
resolve themselves into six divisions beside the two given above. That these
eight divisions given below are exactly comparable linguistic equivalents is not
to be assumed:
1. Kutchin
2. Tsetsaut
3. Tanaina — Ingalik
4. Carrier — Chilcotin
5. Tahltan — Kaska
6. Sekani — Beaver — Sarsi
7. Chipewyan — Slave — Yellowknife
8. Dogrib — Bear Lake — Hare
Of the remaining eight groups, Koyukon, Tanana, Nabesna, Ahtena, Han,
Tutchone, Mountain, and Nicola, nothing is certain concerning their classifica-
tion. Ahtena may prove to be a distinct division by itself, and it is likely that
most of the others are to be consolidated with the divisions numbered above as
3, 5, and 7.
It is greatly hoped that both linguistic and cultural research may be under-
taken during the coming decade to bring to more definite conclusions, while
still possible, some of the problems which have been mentioned. Such conclu-
sions will have an important and far-reaching effect on the study of the whole of
American culture.
222 Vf American Indian Languages 2
Editorial Note
Excerpt from Osgood, Cornelius, The Distribution of the Northern
Athapaskan Indians, Yale University Publications in Anthropology 7, 21-22
(1936). Reprinted by permission of Human Relations Area Files.
Cornelius Osgood (1905-1985), an anthropologist who did extensive eth-
nographic work among Athabaskan groups of Canada and Alaska during the
1930s, was a junior colleague of Sapir's at Yale during the 1930s and had earlier
studied with Sapir at Chicago.
Section Seven:
Penutian Languages
Introduction
As originally proposed by Dixon and Kroeber (1913), the Penutian lin-
guistic stock encompassed five California groups that had hitherto been
separately classified: Wintun, Maidu, Yokuts, Miwok, and Costanoan. In
modern discussions of North American linguistic classification Dixon and
Kroeber's grouping is referred to as California Penutian, whose status as a
unit within a larger Penutian relationship is still an unresolved question
(Whistler 1977; Shipley 1980; Berman 1983). The recognition of a wider
affiliation of California Penutian, and the extension of the term "Penutian"
to the larger relationship, must largely be credited to Sapir. Between 1915
and 1925 Sapir proposed a series of additions to the California kernel,
beginning with an Oregon group — limited at first to Takelma, Coos, and
Lower Umpqua, but quickly expanded to include Alsea and Kalapuya — then
Chinookan, followed by Tsimshian, a Plateau group (Klamath-Modoc,
Sahaptian, and Molale-Cayuse), and finally by a cluster of Mexican and
Central American outliers. Subsequently, others have used the term Penutian
in even broader senses, but Sapir's proposals have remained the point of
departure in any discussion of the "Penutian Hypothesis" (as for example in
Silverstein 1979 or Greenberg 1987).
Compared with his achievements in the comparative linguistics of Hokan
(see Volume V), Sapir's published record in comparative Penutian is a thin
one. He wrote only one major paper on Penutian ("A Characteristic Penutian
Form of Stem", 1921b), which included no etymological sets or reconstruc-
tions. It is fair to say that, for all his interest in historical questions, Sapir's
major contributions to Penutian linguistics were descriptive rather than
comparative. His Takelma texts and grammar (1909c and 1912h) and his
Wishram Chinook texts (1909d) and grammatical synopses (1907c and 191 Ig
in this volume) are the foundational works in the descriptive study not only
of these two languages but of Penutian languages generally. Moreover, Sapir's
perspective on the diachronic relationship of the Penutian languages was
heavily influenced by his detailed knowledge of the synchronic facts of
Takelma and Wishram. Any general assessment of Sapir's Penutian work
must, then, largely be concerned with his work on these two languages of
Oregon.
It was in fact Wishram on which Sapir, as a 21 -year-old novice field
worker, cut his analytic teeth in the summer of 1905. He had been sent by
his teacher, Franz Boas, on funds provided by the Bureau of American
Ethnology, to gather data on the upriver dialects of Chinook. "Preliminary
Report on the Language and Mythology of the Upper Chinook" (1907c) is
226 ^f American Indian Languages 2
the published version of a paper which Sapir originally delivered shortly after
returning from the field, at the December, 1905, meeting of the American
Anthropological Association. It is a series of comparisons between the Upper
Chinookan (Wasco-Wishram) material that Sapir had collected and the
Lower Chinookan data gathered earlier by Boas. For a student's first pub-
lication of his own field data it is strikingly self-assured, showing many of
the qualities (e.g., meticulous attention to phonetic detail) and concerns (e.g.,
the relation of linguistic to socio-psychological phenomena, as in consonantal
symbolism) of Sapir's mature work. Of particular interest are some historical
suggestions that adumbrate the Penutian hypothesis: that a comparison of
Upper and Lower Chinookan phonetic systems indicates an original Chi-
nookan homeland east of the Coast Range (p. 535), or that the "puzzling"
occurrence in Wishram of locative postpositions reminiscent of case suffixes
can be taken to indicate a connection of some sort with Sahaptian and
Klamath (pp. 541-542).
In the course of his 1905 work on Wishram Sapir collected an extensive
vocabulary, rich grammatical data, and some 30 narrative texts. He published
the texts in 1909, together with 25 Wasco texts collected earlier for the Bureau
of American Ethnology by Jeremiah Curtin (1909d, in Volume VII). The
extent of his material would certainly have allowed him to prepare a full
grammatical description had Boas encouraged him to do so. Instead, Sapir
was asked only to contribute several illustrative sections of Upper Chinook
to the grammar of Lower Chinook that Boas prepared for the first volume
of his Handbook of American Indian Languages (1911). These sections are
reprinted here. Two are expansions of briefer treatments of the same topics
in Sapir's 1907 report: "Diminutive and Augmentative Consonantism in
Wishram" (cf 1907c: 537-538) and "Post-positions in Wishram" (cf 1907c:
541—542). In addition to these relatively self-contained sections, Sapir's
contributions to Boas's Chinook grammar also include Section 17, "Modal
Elements: 2. ni-, 4. ga-, 5. ga-'' (pp. 578 — 579); Section 44, "Demonstrative
Pronouns and Adverbs in Wishram" (pp. 625 — 626); and Section 45, "In-
dependent Personal Pronouns" (the last five paragraphs of which are Sapir's,
paraphrased by Boas; pp. 626 — 627).
Sapir's second field trip, in the summer of 1906, was to the Siletz Reser-
vation in western Oregon for the purpose of gathering linguistic data on
Takelma, a moribund language of unknown affiliation, along with whatever
incidental ethnographic information could be obtained. Sapir wrote a full
grammar of Takelma for his 1907-09 doctoral dissertation at Columbia
University (published as 1912h) and shortly afterwards prepared a volume
of Takelma texts for publicaUon (1909c). Both of these works are reprinted
in Volume VIII. He also published two papers on Takelma ethnography
(1907b, I907d), reprinted in Volume IV.
Seven: Penutian Languages 227
Takelma exhibits many of the phonological and morphosyntactic features
that have come to be recognized as characteristically Penutian. As early as
1907, when Kroeber's study of Yokuts appeared (Kroeber 1907), Sapir had
begun noting structural resemblances between Takelma and certain California
languages. It was not, however, until well after Dixon and Kroeber first
proposed the Penutian and Hokan relationships in 1913 that Sapir began
active exploration of possible genetic links for Takelma. Apparently reading
Leo J. Frachtenberg's grammar of Coos (Frachtenberg 1914a) first stimulated
Sapir's interest in this work. Sapir's correspondence with A. L. Kroeber
reveals that by mid- April, 1915, he had concluded that Coos and Takelma,
and possibly some other Oregon languages, formed a "North Penutian"
group that had been "cut loose" from the California ("South Penutian")
group by the "northern intrusion of Hokan languages" (GoUa 1984: 182).
He compiled a list of 145 potential cognates among Coos, Takelma, and
California Penutian (see Sapir and Swadesh 1953), although his most con-
vincing evidence was grammatical and typological. It was at this time, in all
hkelihood, that Sapir drafted "A Characteristic Penutian Form of Stem"
(1921b), the paper that was to be his only major contribution to comparative
Penutian linguistics.
This paper, however, was not completed until 1919, and its publication
even further delayed until 1921, so that in part it reflects later developments.
Chief among these was the broad expansion of the network of Penutian
connections. By late 1915 Sapir had come to believe that "North Penutian"
should also include Siuslaw and Chinookan, as well as (more tentatively)
Alsea, Kalapuya, and Tsimshian (see GoUa 1984: 201 —203). Leo Frachten-
berg, meanwhile, was independently pursuing a similar line of investigation,
and in 1918 he pubhshed evidence linking Takelma, Kalapuya, and Chi-
nookan, and suggesting (but not claiming) a connection between this group
and the California Penutian languages (Frachtenberg 1918). Sapir recast "A
Characteristic Penutian Form of Stem" as a reply to Frachtenberg, offering
confirmatory evidence of a morphological nature and placing Frachtenberg's
comparisons in the context of his own speculations.
Of particular interest to Sapir was the possibility of a relationship between
Chinookan and Penutian, not least because it seemed so improbable on the
basis of morphological structure. Chinookan is a thoroughly polysynthetic,
head-marking language (Nichols 1986) differing radically in this respect from
most other Penutian languages, which tend to be dependent-marking, with
nominal cases and a rather Indo-European-like distinction between nouns
and verbs. For Chinookan to be Penutian it would either have to be assumed
that all Penutian languages had originally been head-marking, or that Chi-
nookan had originally been dependent-marking. Sapir saw the latter as the
more probable development and believed that certain grammatical differences
228 Vf American Indian Languages 2
between Upper and Lower Chinookan (some of which he had noted in 1907c)
indicated a general drift toward polysynthesis. In "A Chinookan Phonetic
Law" (1926a), pubUshed in 1926 but probably written much earlier, Sapir
explores the processes through which a typically Penutian-like clause syntax
of pre-Chinookan might gradually have "petrified" into the phonologically
close-knit polysynthetic words of attested Chinookan. In this paper — a tour
de force of historical argumentation — Sapir musters internal evidence to
show that the Chinookan ergative prefixes g- (3 sg. masc.) and c- (3 sg. fem.)
plausibly derive from the post-pronominal element -g- that occurs with the
ergative prefixes for other persons. This element, in turn, he sees arising out
of an accentuated demonstrative particle in the original clause syntax.
Sapir's proposal that Tsimshian, spoken in northwestern British Columbia,
belongs in the Penutian family was a bold step that remains controversial.
The idea of such a connection seems to have occurred to Sapir not long after
he had collected a small amount of Tsimshian data from a speaker of the
Nass River (Nisqa) dialect in February, 1915 (see Sapir's letter to Kroeber
of December 9, 1915 in Golla 1984: 201 —202). The nature of Sapir's evidence
remains uncertain, although marginal notes in Sapir's copies (now in the
possession of Professor John Dunn of the University of Oklahoma, who
provided copies to the editors) of Boas's Tsimshian grammar and texts show
that he had noted a number of lexical resemblances (see "Sapir's Comparative
Penutian Glosses" in this volume). In a 1917 letter to Kroeber Sapir cited
several possible kin-term cognates between Tsimshian and Yokuts (Golla
1984: 242 — 244), and he again alluded to this evidence (without presenting
it) in his paper on "Nass River Terms of Relationship" (1920c: 365, note 1).
Sapir began to explore Mesoamerican linkages to Penutian in the mid-
1920s, at the same fime that he was noting the possible Hokan affiliation of
other languages of that area (see 1925b). In 1924 Kroeber suggested to Sapir
that Mixe and closely related Zoque, languages of Oaxaca, were possibly
Hokan (Golla 1984: 409). At about the same time Roland B. Dixon inde-
pendently suggested that Zoque was Penutian (Freeland 1930: 28, note 1).
Sapir evidently found Dixon's evidence more convincing, for he placed Mixe-
Zoque and Huave (whose relationship to Mixe had been postulated by Radin
in 1919) in a Mexican Penutian group in the classificatory table of his
Encyclopaedia Britamiica article on American Indian languages (1929a: 139).
In the same place Sapir also suggested that Xinca (in Guatemala) and Lenca
(in Honduras and El Salvador), as well as perhaps Paya and Jicaque (in
Honduras), "may be remote southern outliers" of Penutian. Here he followed
Lchmann (1920: 767), who had grouped Xinca and Lenca with Mixe-Zoque,
although Sapir rejected Lehmann's further comparisons with Coahuiltecan
and Chumash (i.e., Hokan). In the mid-1920s Jaime de Angulo collected
more field data on Mixe. His wife, Lucy S. Freeland, used these data to
Seven: Penutian Languages 229
prepare a paper on "The Relationship of Mixe to the Penutian Family," a
draft of which she sent to Sapir for "criticism and suggestions." The paper
as published (Freeland 1930) incorporates many of Sapir's comments and
suggestions and for all practical purposes is a coauthored work. In addition
to ten signed footnotes, Sapir is almost certainly responsible for many of the
lexical sets and for the suggested reconstructions. This paper may be consid-
ered Sapir's last public statement on comparative Penutian.
During the 1950s Sapir's student Morris Swadesh was the most active
scholar in the Penutian field. In addition to his own work, which included a
field survey of all surviving Oregon Penutian languages (Swadesh 1954),
Swadesh also edited and published, as "Coos-Takelma-Penutian Compari-
sons" (Sapir and Swadesh 1953), a Hst of 152 comparative Penutian lexical
sets compiled by Sapir around 1915. This is apparently the list mentioned
by Sapir in a letter to Kroeber on April 21, 1915 (see Golla 1984: 184-186)
and also the data Sapir promised to "present in extenso in the future" in "A
Characteristic Penutian Form of Stem" (1921b: 59). In his letter to Kroeber,
Sapir calls particular attention to the sets involving grammatical elem.ents.
"Sapir's Comparative Penutian Glosses" is a collection of Penutian com-
parisons that Sapir jotted down in the margins of works on other Penutian
languages, and which apparently he never thoroughly collated. A few are
contained in the 1915 list (Sapir and Swadesh 1953), but most were probably
made subsequently. The original publication of these marginal glosses (Swa-
desh 1964) contains numerous typographical errors, and the material has
been completely re-edited for publication here. Integrated with these glosses
are the Penutian marginalia in Sapir's copy of Boas's Tsimshian texts (1912).
PRELIMINARY REPORT ON THE LANGUAGE AND
MYTHOLOGY OF THE UPPER CHINOOK »
In the summer of 1905 I was commissioned by the Bureau of
American Ethnology to continue the study of Chinookan Hnguis-
tics and, incidentally, mythology, which had been begun some ten
years ago by Professor Boas, and the results of which, so far as
published, have appeared in "Chinook Texts" and " Kathlamet
texts," both bulletins of the Bureau, and in Dr Swanton's " Mor-
phology of the Chinook Verb " and Professor Boas' " Notes on the
Chinook Vocabulary," both of which articles appear in the Ameri-
can Anthropologist} This published material deals with the dialects
of the Chinookan family spoken at or near the mouth of Columbia
river. It was therefore desirable, in order to gain a somewhat more
comprehensive idea of the peculiarities of Chinookan grammar, to
devote study to the extreme eastern dialects.
The dialect or language to which the following notes refer is
that spoken by the Indians formerly living on the northern shore
of Columbia river, roughly speaking, from White Salmon river to
the Long Narrows. These Indians, who are now on the Yakima
reservation, Washington, called themselves iidxlnit, the ist per.
sing, of which {itcxluit, ' I am an Uaxluit ') is in all probability
the " Echeloot " of Lewis and Clark. They are known by their
Yakima and Klikitat neighbors (tribes of the Sahaptian stock) as
Wticxam, which, in its anglicized form of Wishram, or Wishham, is
their common apellation to-day. The language spoken by them is,
to all intents and purposes, the same as that of the Wasco on the
other side of the river and of the White Salmon and Hood (or Dog)
' Read before the American Anthropological Association at Ithaca, New York, De-
cember, 1905. Published by permission of the Bureau of American Ethnology.
^ Bureau of American Ethnology, Bulletins 20 and 26. American Anthropologist,
N. S., II, 1900, pp. 199-237, and VI, 1904, pp. 1 18-147. i he phonetic system followed
in these works is used in this article. See either Bulletin for the key.
AM. ANTH , N. S., 9-35. 533
232 VI American Indian Languages 2
534
River Indians farther down the stream. More prominent dialectic
differences appear when we get as far down as the cascades ; the
dialect of this locality may be considered transitional between the
Wishram and the Clackamas of the Willamette region.
Viewing the Chinookan dialects as a whole, we find that the
same general morphological characteristics apply to both Upper and
Lower Chinook. In both groups we have the concept of the word
as distinct from the.sentence clearly developed.' Pronominal incor-
poration of subject, object, and indirect object in the verb ; a some-
what elaborate apparatus of pronominal elements and pronouns
(including the dual and an inclusive and exclusive in the first
person dual and plural) ; a peculiar method of expressing the pos-
sessive pronouns (these are prefixed elements related to the pro-
noun subjects of transitive verbs) ; a characteristic use in many cases
of invariable particles accompanied by auxiliary verbs instead of the
use of verb-stems to express the main idea (as though one were to
say in English: "He made the bell ding-dong " instead of "he
rang the bell ") ; a general tendency toward onomatopoesis ; the
extraordinary phonetic weakness of many of the verb-stems (often
consisting of but a consonant or cluster of consonants) ; local or
adverbial prefixes and local and quasi-modal suffixes in the verb ;
and a thoroughly developed system of grammatical sex-gender
(masculine, feminine, and neuter), both in the noun and in the
verb — all of these Jeatures are shared by both the upper and the
lower dialects.
The first important difference between the Wishram and Lower
Chinook is found to be in the phonetic systems of the two. Whereas
the lower dialects affect on the whole a surd articulation (with pre-
* Such a word, for instance, as the Wishram gatctcxcgam,. ' he took them away from
the two (women) ' {,ga- := tense sign indicating remote past ; -tc-^ 3d sing. masc. subj.
of trans, verb; -/- = 3d pi. obj. of undefined gender; -<:-:= 3d dual indirect object of
undefined gender ; -x- = reflexive element indicating that object, -t- is possessed by per-
sons referred to by -c-, here most easily rendered by ' from ' ; -eg- = verb-stem or " root "
meaning ' take ' ; -am ^= verbal suffix generally denoting ' arriving, coming or going to do
something,' but not quite transparent in its application to this verb) must be conceived of
as an indivisible unity in the same sense in which a Latin form like conscripsi is an or-
ganic unit (not merely con -(- icrib -\- s -\- l •&% agglutinated elements intelligible per se^ ;
none of the elements in the given verb-form has any sort of meaning outside of its par-
ticular place in such form. In other words, the word and sentence do not flow into one
another in Chinookan.
Seven: Penutian Languages 233
535
dominance of/, /, and velar surd q over b, d, and velar sonant g), the
Wishram is prevailingly sonant in its use of stops. Thus, where the
Lower Chinook has o'/a, ' yellow-jacket,' and an/oVena, ' I killed
them,' the Wishram has wa^^a and in^/u^ina. Moreover, the short u
and i of Wishram are generally represented in Lower Chinook by long
o and e, as seen in the latter example cited. The peculiar voiceless
palatalized / (written l) of the Pacific coast appears in Wishram with-
out the characteristic stop quality of the Lower Chinook ; thus we
have Lower Chinook Lon ' three,' but Wishram lun. These phonetic
differences, together with a number of local phonetic changes that it
is not necessary to go into here,' would suffice to give the two
groups of dialects a marked acoustic difference. From internal
evidence I am very strongly inclined to believe that the phonetics of
Wishram represents better than that of the lower dialects the orig-
inal condition of Chinookan. Inasmuch as the phonetics of Lower
Chinook is closely allied to that of the neighboring Coast Salish
(such as the Tillamook and Chehalis), the interesting possibility pre-
sents itself that the Chinookan tribes were formerly all located east
of the Coast range and that some of them, proceeding down the
river in their well-built canoes, came to the Pacific coast and there
assimilated the phonetic system of their new neighbors. This,
however, is confessedly mere speculation, and needs confirmatory
evidence.^
Leaving aside these phonetic differences, perhaps the most strik-
ing morphologic difference is in the treatment of the demonstrative
pronouns. Both the upper and lower dialects possess different
forms for the various relations of near the speaker, near the person
' Characteristic Coast features found in Lower Chinook but not in Wishram are be-
sides : the presence of the voiceless palatal spirant x- as in German ich (Wishram employs
instead a voiceless palatal spirant x pronounced far forward, yet quite distinct acoustically
from X', which to Wishram ears sounds like c) ; and the difficulty of distinguishing be-
tween m and b and also n and d, a characteristic Coast Salish phonetic feature.
^ Such a movement of the Chinook down the river would satisfactorily explain also the
severed position of the Salish Tillamook, in Oregon, who are separated from the linguistic-
ally related Chehalis only by Chinookan tribes. Even though all the Salish tribes be of
interior provenience, as generally believed, their occupancy of the Pacific coast, including
the region directly north and south of the Columbia, may have long antedated the coming
to the coast of the Chinook. See A. B. Lewis, "Tribes of the Columbia Valley and the
Coast of Washington and Oregon," Memoirs of the American Anthropological Associa-
tion, 1906, I, pt. 2, p. 198.
234 VI American Indian Languages 2
536
addressed, and near the person spoken of, and both distinguish the
three numbers and the three genders of the singular in the demon-
strative. Whereas, however, the Lower Chinook further distin-
guishes between visibiHty and invisibility of the person or object, no
such difference could be observed in the use of the demonstratives
in Wishram. Moreover, the principle of formation of the demon-
stratives is, in detail, quite dissimilar in the two groups. In Lower
Chinook the demonstrative is built up of three exceedingly weak
phonetic elements : a consonant expressing visibility or invisibility,
a vowel or consonant denoting the number and gender of the person
or object referred to, and a consonant or two vowels defining the
demonstrative relation. In Wishram the principle of formation is
simpler; each demonstrative form is built up of two agglutinated
syllables, one of which is the short form of a 3d pers. pronoun
(defining both gender and number), and the other a characteristic
element indicating the demonstrative relation. Speaking generally,
the demonstratives in W^ishram seem to stand in much closer rela-
tion to the personal pronouns than they do in the lower dialects.'
Reference was made above to the general tendency toward
onomatopoesis in the Chinookan dialects. The impression which
Professor Boas had obtained of such a character in his study of the
lower dialects was in every way confirmed by my own study of the
Wishram. The frequent rhetorical lengthening or shortening of
vowels and consonants, the duplication or quintuplication of imi-
tative elements, and the frequent use of ononiatopoetic particles in
* For convenience of comparison the demonstratives of both Lower Chinook and
Wishram are tabulated below. Those in parentheses are the Chinook correspondents of
the Wishram forms immediately above :
Masc.
Near ist Person f ddnya
hU \{xi'k)
Near 2d Person f ydxdan
iste \ (x'iait)
Near 3d Person j ydxia
ilk \ yx-ix-)
Shortened Pro-
nouns in Wish- yni^x^
ram
The Lower Chinook forms here given are those implying visibility. The correspond-
ing demonstratives used to refer to invisible objects are obtained by changing the initial
X- -Xo q-.
Fern.
Neuter
Dual
Plural
ddiia
dauia
ddtuda
dduda \
{x-uik) r
{x-ak)
{x-lLik)
{x-ictik)
dxdatt
■tdxdan
cddxdaii
ddxdau 1
[xila) \
[x-au')
{x-lLa)
{^x tcta )
dxia
tdxia
cddxia
ddxia "1
{xota) \
{xaX)
{x-oLa)
(xoc/a)
a{x)
ea{x)
cda ( JT )
da {x)
Seven: Penutian Languages 235
537
lieu of verb-stems are not the only phenomena which illustrate this
onomatopoetic tendency. Most characteristic of Wishram, and prob-
ably of the other Chinookan dialects also, is the employment of a
series of changes in the manner and, to some extent, in the place of
articulation of the various consonants, in order to express diminution
and augmentation. This singular rhetorico-grammatical process
works in such a way that all surd and sonant stopped consonants be-
come exploded consonants (better known as " fortis ") to express the
diminutive idea (i. e. b and p become />/, d and / become t!, g and
k become k.'\ while all surd and exploded consonants become
sonant to express the augmentative (i. e. p and pf become b, t and
/.'' become d, k and k! become g, q and q! become ^); in the
case of the velar consonants a possible change to the "fortis" to
denote the diminutive is attended also by a more forward, i. e.
palatal, articulation (i. e. g and q become, not q!, but k!). More-
over, the sibilant consonants c, tc, and tcf on the one hand, and s,
ts, and tsf on the other, are related to each other as augmentative
and diminutive consonants, while dj may sometimes, though rarely,
be employed as the augmentative grade of tc and ts (e. g., idjik-)
djik ' big wagon,' from itsiktsik ' wagon.' The guttural spirant
X becomes x in the diminutive form. Subjoined are a few illus-
trations for the purpose of making the process more easily under-
stood. The normal word in Wishram for ' hip-joints ' is ck.'dlkal.
The diminutive of this word is skfdlkal, in which, it will be noticed,
the c of the first word has been changed to s in consonance with
our rule. The word skldlkal would be appropriately used to
designate a baby's hip-joints, for instance. On the other hand the
augmentative would require a change of the fortis k! to a sonant g
— hence cgdlkal is used to denote ' big hip-joints,' as of a giant.
Similarly, while a^.^^a*/ with velar fortis (^/) is the normal word
for 'knee,' akliixt with palatal fortis (yk!) and guttural spirant
pronounced farther front i^x) is the diminutive, and aodxt^\\\\ sonant
velar (^) the augmentative. Not infrequently there is a slight
change of meaning accompanying the phonetic change. Thus,
while itdVndn (masc.) denotes ' eagle,' itts.'V non (neut.) with
diminutive consonantism means ' bird '; ik.'dlaniat denotes ' stone,'
but igdlainat with augmentative consonantism means ' rock.' It
236 Vf American Indian Languages 2
538
must not be supposed that this characteristic consonant-gradation
is confined to the noun ; it is found just as well in every other part
of speech. An example of its use in the verb will serve to give an
idea of its rhetorical possibilities. InigEltcim is the normal word
for ' I struck him with it.' If the verb-stem -tcim appears, with
diminutive consonantism, as -ishn, it implies that the person struck
is small ; if the verbal prefix -gEl-, which implies in this case
intent to hit, is pronounced -k/sl-, the implication is that the
missile used is a small one. Hence we have four forms : inigAlicim
' I hit him with it,' inigHltsivi ' I hit him (a child perhaps) with it,'
inik.'Eltcim ' I hit him with it (something small),' and inik.'kltsim
' I hit him (a child) with it (something small).' It would seem then
necessary, so far as Chinookan grammar is concerned, to allow as a
regular grammatical process, alongside of reduplication, vowel
change or "ablaut," and pre-, in-, and suffixation, a fourth process
— consonant-gradation or " ablaut."
Turning again to morphology, there was one feature which was
well calculated to arouse a certain degree of surprise. The work
which had been done on Lower Chinook disclosed a paucity of
tenses that is, on the whole, quite in accordance with the general
morphologic character of many American linguistic stocks. In
VVishram, however, I found that it was necessary to distinguish
carefully six tenses : ist, a tense characterized by the prefix ga-
(before consonants) or ga/- (before vowels) in certain cases option-
ally by the prefixed consonant w-), which refers to time long past,
say more than one year ago, and which is used regularly in the re-
cital of myths ; 2d, a tense characterized by the prefix ;/z- (before
consonants) or fiig- (before vowels), used to refer somewhat indefi-
nitely to time past and which is used in speaking of events that
happened say less than a year ago, yet more than a couple of days ;
3d, a tense characterized by prefixed nn- (before consonants) or tia/-
(before vowels) and sufifixed -a, which seems to refer to recent time
exclusive of to-day, more specifically to yesterday ; 4th, a tense
characterized by prefixed t- (before consonants) or ig- (before
vowels), which refers to an action already performed to-day ; 5th,
a tense characterized normally by suffixed -/", referring to an action
now going on but, as it seems, with the implication of its soon being
Seven: Penutian Languages 237
539
completed ; and 6th, a future tense, normally characterized by pre-
fixed a- (before consonants) or al- (before vowels) and suffixed -a}
Besides this series of six positively characterized tenses, I should
not omit to mention that some verbs, when referring to present
time, are morphologically tenseless, and seem to form their immedi-
ate past tense by a verbal prefix -/- which ordinarily denotes action
toward the speaker.^
In this connection I may also mention a group of verb-forms
which are characterized by the consonant / (assimilated in nasal
surroundings to «) suffixed or infixed to the verb-stem, sometimes
by -lal (or -7ian) suffixed to the verb-stem. These forms denote
frequentative or continuative action and, as a rule, do not allow the
verb to be further characterized by a tense element. They may
then, from a certain point of view, be considered as forming a
seventh tense — the present tense with no implication of comple-
tion.^ The most interesting point about these /-frequentatives is
^Examples —
Tbnse:
^a-FoRM
ni-FoRM
«a-FoRM
f gay^y^
J 'he went '
' «/yuya
f wayuya
nalnya.
1 galaya.
\_ 'she went '
nigiiya.
_^rttcigElkEl
Wi'tcigitkEl
W(7tcigElgEla
' he saw him'
galiwi-x.
m'gixaix
«ff/ixuxwa
' he became'
^^'flcgnux
'they two did to me'
wicgnatx
wrtcgniixwa
/-Form
Pres. /-Form
FuT. rt-FoRM
i iyiiya.
fyiii/
\ui/
f ayuya
\ a/uya
?tcigEikEl
(tssiklElutk/)
atcigfelgElrt
r^ixux
/cgnux
'he looks at him
fl/ixiixwa
flcgnuxwrt
cgnux/
^Thus uxt (=a -\-tt -{- xt) means 'she is seated,' but 'she was sitting' is rendered
by dtxt, in which the prefix -«- has been changed to -t-. Cf. , for this interchange,
^wgwat 'they fly (away from me) ' and <Vgat 'they fly toward (me).'
^ Such frequentative forms are :
With Tensb-sign Frequentative
^atksanbnaj'x tksanbnawi'x
' they jumped in the water ' ' they keep jumping in the water '
(verb-stem bna-)
238
VI American Indian Languages 2
540
that certain verb-stems apparently infix the / or ;/. If our EngHsh
word ' look,' e. g., were also a Wishram verb-stem, ' he looked at
it ' would be itciiilook, but ' he keeps looking at it ' would probably
be tcmloolk} I pass over many other verbal peculiarities, such as
the distributive suffix -yit (alxk/wdya ' we shall go home,' but
alxklivdviiiva ' we shall go each to his own home ') or the passive
suffix -ix {itcudxuui ' he ate it up,' hut yulxu mix 'it is eaten up') to
mention the considerable difficulty experienced in analyzing the noun,
apart from its syntactic elements which are transparent enough.
The pronominal elements prefixed to the noun (every noun is
either masculine singular, feminine singular, neuter singular, dual,
or plural) are in Lower Chinook identical with the pronominal
object elements incorporated into the transitive verb, except for the
feminine singular, which in the noun shows o- (from original wa-)
as compared with -a- in the verb. In Wishram, however, the noun
has prefixed a pronominal element differing from the corresponding
element in the verb by an initial zv-{masc. and fem.)or z- (neuter, dual,
and plural). The following table shows the corresponding elements :
Noun in Wishram
Noun in Chinook
Obj. in
Wishram Verb
masc.
wi-, i-
t-
-i-
fern.
wa-, a-
0-
-a-
neut.
ii-
L-
-t-
dual
ic-, {is-)
^-, {S-)
-C-, {-S-)
Pl-
id-
t-
-d-
^atssubKna
tssubEnaMrt«
*he jumped '
'he keeps jumping '
i^atccinqlwSVgunaba
tccinq ! w6'^k°na«p^
* he grasped him wi
th his
claws '
' he scratches him '
_crakdugwiptck
kdiigwi/ptck
' she gathered driftwood '
' she keeps gathering drift
wood
_o-rtyukwa
yugwa/rt/
'he flew'
' he flies about '
' Some examples of th
lis phenomenon are :
With Tensk-sign
Frequentative
,^rt!/ixpcut
ixpcu//'t
' he hid himself
'he is hiding himself
^'-rttciucit'
\.z\\xc\lt
' he used it '
' he keeps using it '
ksKnklitk/'
ksink!idE/k
' she looks at me '
' she keeps looking at me
(verb-stem /k-
Seven: Penutian Languages 239
541
The choice between wi- tva- and i- a- in Wishram is dependent
chiefly upon considerations of syllabic length : wilx ' land ' (cf.
Chinook z'le'e), but ?^nuk ' beaver '; icdtrnX ' marrow ' (cf. Chinook
d'mala), but ^^agilak ' woman.'
It had been hoped that some light would be thrown on the
derivative elements of the noun, but it cannot be claimed that all
desirable success was attained in this direction. Perhaps the most
transparent derivative elements that were found are the suffixes -///
and -mat. The former of these seems to denote a group, particu-
larly a grove of trees. Thus the word ilibuni ' apple ' (borrowed,
of course, from the French /a pomrne) forms the derivative noun
UibumElit ' orchard.' The suffix -?;iai is perhaps best defined as
denoting 'something used for so and so.' For instance, isqxiis
denotes ' the eyes,' isqxusniat means ' something for the eyes,' i. e.
spectacles.' An interesting group of nouns is formed by descriptive
verb phrases, such nouns being in effect pure verb forms. The
loon, e. g., is described as 'he shouts along the river' {tci-iliiviat
tviviat)y and ' telescope ' is rendered by ' people keep looking
through it ' {(jexgElgElini).
The most puzzling linguistic phenomenon found in Wishram,
because it is at complete variance with what we have in the lower
dialects, is the use of a certain number of loosely tagged on postposi-
tions, in some cases optionally prepositions. We have a suffix -ba
denoting ' in ' or ' at,' a suffix -iarnt meaning either ' towards,' or
' from,' a suffixed or prefixed element bdvia meaning ' for,' the post-
er pre-positions dniKiii and knEgi meaning ' with ' or ' made out of,'
and an element -^i"/, meaning ' when,' suffixed to verb forms. ^ The
^ Further examples of this suffix are : igictxmat 'load' (from verb-stem -ctx ' to
carry on one's back') ; itkHcitmat 'tools' (verb-stem -cit 'to use') ; aklixivacamat
' plane ' ( verb-stem -xwac ' to plane ' ) .
^ The following examples illustrate the use of these elements with nouns, pronouns,
and verbs :
ba : wimatba 'in the river; ddiiyaba wilX 'in this country' (lit. ' this-in
country ' ) ; galcigElkElba ' where he saw him ' (gatcigklkEl ' he saw him ' ).
iamt : ivimajiamt ' to or from the river ' ; imigdii naikdyaint ' you are bigger than I '
(lit. 'your bigness [is] me-from, compared with me ') ; dtpXiamd agdiax ' to where she
goes out towards [us] {atpX ' she goes out towards'), the sun,' i. e. 'east.'
bdma : cdn bama ' for whom ?' ; Miilmul bama ' from, belonging to Fort Simcoe.'
dniEni : igdbEnac aniEni ' made out of young oak.'
EnEgi{ngi) : aqti'wiqxi ngi ' with a knife.'
bEf: gayuyabEt 'when he went'; nkldckacbEt ' when I was a child.' In length-
need form bd^^ it means ' as soon as' : gayuyabd't ' as soon as he went.'
240 VI American Indian Languages 2
542
extent of pronominal incorporation of indirect objects and the use
of local or relational prefixes in the verb are such in the Chinookan
dialects that the employment of these local tags (one might be
inclined to call them " cases," if they had less individuality) seems
quite unnecessary. It is of considerable theoretic importance, there-
fore, to note that the neighboring Sahaptian dialects, quite similarly
to the Klamath, make an extended use of such case-suffixes. We
would then have here a good example of the graniniatic, not merely
lexical, influence that dialects of one linguistic stock may exert on
geographically contiguous dialects of a fundamentally distinct stock.^
In conclusion a few words may be devoted to the mythology of
the Upper Chinook. I have not as yet enough texts of myths to
present a really complete description of the mythologic concepts
and elements present in the tales of the Wishram, but some of the
main points seem patent enough. As in other Indian mythologies
it is believed that there was a time antedating the present one when
animals walked about as men, though having approximately the
same mental and, to a large extent, physical characteristics as now.
At that time, when there were no Indians, properly speaking, in the
country, but only anthropomorphic animals, many things were not
as they should be, and, in order to make the country fit for habita-
tion by the Indians destined to hold it, it was necessary for a culture-
hero or transformer to rectify the weak points in creation. This
transformer is, as in the plateau regions to the east, the Coyote.
There is a cycle of myths made up of local tales telling how Coyote
traveled all the way up the Columbia river, transforming monsters
and instructing the people in the various arts of life. This string of
local tales is, if I am not mistaken, continued in unbroken succes-
sion by the Sahaptian tribes living farther up the river, so that we
have here a series of myths, belonging together yet distributed over
a large number of different tribes. Some of the things that Coyote
does are : to stock the Columbia with fish that had been withheld
from the rest of the world by two women ; to transform two women,
who entice wayfarers, into birds ; to provide the people of the
* Of the postpositive elements given above, three, bdma, EnEgi, and dmEni, are
certainly of Sahaptin origin, probably also -ba (cf. Yakima -pa ' in ' ). This explains their
entire absence in Lower Chinook.
Seven: Penutian Languages 241
543
Cascades country with moutlis that had formerly been lacking ; to
instruct men in the art of catching white salmon in basket traps and
of spearing and steaming salmon ; to put an end to the atrocities of
the merman who swallows canoes with men and all, and of the dread
woman, At!at!aha, who steals children- and roasts them on an
island still pointed out at the Long Narrows; and so on. In all
this Coyote is distinctly the benefactor of mankind, but at the same
time he is, as often elsewhere, conceived of as cunning, deceitful,
and gluttonous. In some stories, particularly in such as do not
belong to the cycle of Coyote as Transformer, he is an insufferable
marplot, as when he, contrary to Eagle's injunction, opens a box
containing the souls of his and Eagle's wife and son, thus bringing
death into the world. At the same time he is indescribably obscene ;
some of the deeds of this kind performed by the culture-hero of the
Tillamook, as communicated by Professor Boas, are also told by
the Wishram of him. Although Coyote is the main transformer^ I
think it would be incorrect to speak of him as the hero of the
Wishram. This point comes out clearly when Coyote himself, in
one of the transformation myths, admits that he is no chief, that
title being reserved, among the animals, for the Eagle and the
Salmon. These two may, indeed, be considered the true heroes of
Wishram myth, their deeds being narrated with considerable sym-
pathy and admiration. The Salmon, in particular, may be described
as the local hero of the Chinookan tribes, an elaborate salmon myth
being common to both the Lower Chinook and the Wishram. I
cannot say definitely whether Bluejay, who figures so prominently
as buffoon among the coast tribes, such as the Kathlamet and Qui-
naiult, occupies a corresponding position among the Wishram. So
far as the material collected is concerned, he is quite a subordinate
character, and I suspect that he is almost entirely superseded by
Coyote. The mischievous and spiteful elements of his character, as
of the Mink of more westerly and northerly regions, are embodied
also in the Weasel.
Besides the main type of myth — i. e. the Transformer or Cul-
ture-hero myth, one can discern also a species of nature myth that
is somewhat different in character. This type is represented, e. g.,
by the tale of the contest between the East Wind and the West
242 VI American Indian Languages 2
544
Wind, in which the latter proves successful. Another example of
this type is the struggle of the five East Wind brothers with the five
Thunder brothers, resulting in the death of all but one of the latter,
which exception accounts for the existence of a certain amount of
thunder to-da}\
The single myth motives of Wishram mythology are many,
probably most of them, found distributed over considerable areas
elsewhere. Such well-known incidents as the magic increase of a
small amount of food, the blundering imitation of the host, the life
and death contest at gambling bones, the unsuccessful attempt to
destroy strangers in an overheated sweat-house, the abandonment
and later enrichment of a poor boy while his maltreaters are
starving — all these and many others are common property of the
Northwest Pacific coast and regions to the east and south, though
the setting in which they occur may vary indefinitely. On the
whole, the chief interest of Wishram mythology seems to lie in its
transitional character between the mythologies of the coast and of
the plateau. Although it shares, as we have seen, a local and
specifically Chinookan salmon myth with the Lower Chinook, many
of the myth motives are not duplicated farther down the river,
but are found in other regions, such as the plateaus. Here again
we observe that linguistic and cultural, more specifically mythologic,
distribution areas are by no means necessarily congruent.
Editorial Note
Originally published in American Anthropologist n.s. 9, 533 — 544 (1907).
Reprinted by permission of the American Anthropological Association.
[638]
Contributions to Franz Boas, Chinook
Diminutive and Augmentative Consonantism in Wishram
Very characteristic of Wishram, as also without doubt of ail other Chinookan
dialects, is aseriesof changes in the manner, and to some extent in the place, of
articulation of many of the consonants, in order to express diminutive and aug-
mentative ideas in the words affected. This peculiar process of "consonantal
ablaut," though perhaps most abundantly illustrated in the case of the noun, is
exemplified in all parts of speech, so that it has almost as much of a rhetorical as
of a purely grammatical character. Of the two series of consonantic changes
referred to, that bringing about the addition to the meaning of the word of a
diminutive idea is by far the more common, an actual change to augmentative
consonantism hardly being found outside of the noun. The main facts of conso-
nantic change may be briefly stated thus: To express the diminutive, non-fortis
stopped consonants become fortis, the velars at the same time becoming back-
palatals (the treatment of velar stops, however, seems to be somewhat irreg-
ular); c and its affricative developments tc and tc! become s, ts, and ts! (s seems
sometimes to be still further "diminutivized" to ts, ts to ts!, so that c, s, ts, ts!
may be considered as representing a scale of diminishing values) ; x becomes x,
in analogy to the change of velar stops to back-palatal stops just noted; other
consonants remain unmodified. To express the augmentative, fortis consonants
become non-fortis (generally sonant) stops, no change taking place of back-
palatal to velar; s, ts, and ts! become respectively c, re, and tc! (in some few
cases ts and tc fricatives become dj, pronounced as in English judge, this sound
not being otherwise known to occur in Wishram); other consonants remain
unmodified. [639] The following table of consonantic changes will best make
the matter clear:
Normal Diminutive Augmentative
b, P p! (b)
d, t t! (d)
g> k k! (g)
^, q k!, (g, k) (^)
qx kx (^)
q! k!, (kx) (qx)
p! (p!) h
t! (t!) d
k! (k!) g
c s, ts (c)
tc ts (tc), (?)dj
tc! ts! (tc!), dj
s (s) c
244
VI American Indian Languages 2
Normal
ts
tsf
X
X
Diminutive
(ts), ts!
(ts!)
X
(X)
Augmentative
tc, dj
td, (?)dj
(X)
(?)x
On the whole, there is a distinct tendency to have all the consonants of a word
bear a consistent diminutive or augmentative coloring, though absolute con-
cord in this regard is by no means always observed. In general it may be said
that c and s sounds are most easily varied in accordance with our rule. Final
non-affricative stops seem incapable of change. It often happens that the nor-
mal form of a word is itself partly diminutive in form owing to its meaning; in
such cases the form may be still further "diminutivized" if it is desired to give
the word a more than ordinarily diminutive force. Thus -k!ac- in il-k!a'c-kac
'child' is evidently a semi-diminutive form of the stem-syllable -kac; 'little
child, baby' appears in more pronouncedly diminutive form as itk!a'skas
(Wishram Texts 176.3).
The following table of body-part nouns will serve as a set of examples of
diminutive and augmentative forms. The diminutives would naturally refer to
the body-parts of a tiny child, the augmentatives to those of an abnormally large
being, as a giant.
Normal
Diminutive
Augmentative
i-p!a'qxa 'flat-headedness'
i-ba'qxa
(dim.)
i-gE'tc 'nose' (aug.)
i-k!E'tc
i'i-pc 'foot'
i'l-ps
[640]
i-qxwi't 'leg'
i-khwi't
a-q!o'xl 'knee'
a-k!u'xl
a-go'xl
a-ms'luqtan 'cheek'
a-mE'luk!tan
i-mElExtk!u'lamat 'tongue'
i-mElExtgu'lamat
i-mi'ct 'lips'
i-mi'st
i-k^cxa't 'mouth'
i-k!"sxa't
wa'-kcEn 'finger'
wa'-ksEn
is-qxu's 'eyes (dim.)'
ic-qxu'c
id-mE'qco 'face-hair'
id-mE'kso
i-k!wa'yat 'crown of head'
i-gwa'yat
a-tckE'n 'shoulder'
a-tsk!E'n
wa-qxa'tc 'breast'
wa-kxa'ts^
i-kxa'tc 'tooth'
i-k!a'ts
i-q!a'qctaq 'head'
i-ga'qctaq
ic-k!a'lkal 'hip-joints'
is-k!a'lkal
ic-ga'lkal
is-q!wd'gwdst 'jaws (dim.)'
is-k!wa'gwast
a-mu'q!w>al 'paunch, stomach'
a-mu'gwal
1. Cf. wa-q!a'lc 'thorn', dim. Wa-kxa'ts (Wishram Texts, p. 261).
Seven: Penutian Languages
245
Examples of other than body-
Normal
it-q!u'tcu 'bones'
i-tcfi'au 'snake'
i-tsi'ktsik 'wagon (dim.)'
i-cgi'lukc 'wolf (aug.)'
da-ga'c 'yellow'
i-cga'n 'cedar board'
i-kfa'lamat 'stone'
a-k!a'munaq 'fir'
il-k!a'ckac'c\\M (dim.)'
a-t!u'-gagilak 'good, strong
woman'
[641]
part nouns are:
Diminutive
it-q!utsie'lxlEm 'dog' (literally,
eater of small bones)
i-ts!i'au
is-ts!i'ktsik 'buggy'
il-skli'luks 'new-born wolf
cub (Wishram Texts 56.30)
a-qx-k!i'c 'gold'
wa-ska'n 'box'
wa'-tsk!un 'cup'
Augmentative
i-dji'au
i-dji'kdjik
'heavy truck'
il-k!a'skas
i-ga'lamat
a-ga'munaq
a-du'-gagilak
'strapping big
woman'
In these lists, (dim.) and (aug.) mean that the words so designated are wholly
or partly diminutive or augmentative in consonantism owing to their primary
significance. In i-pfa'qxa, for instance, the diminutive notion implied by p! is
easily understood if we remember that head-flattening is associated with
infancy. In some cases a consonant change involves or is accompanied by a
vocalic change; it seems that the change of a to m or £ has in itself more or less
diminutive force (cf. wa'-tsk!un from wa-ska'n with ila-k!d'its 'very little' [Wish-
ram Texts 176.3] ordinarily -kfaits 'small'). The case of i-cga'n as compared with
wa-ska'n and wa'-tsk!un illustrates the fact that the diminutive form of a noun
often has a specialized meaning of its own. A few more examples are:
Normal
i-tc!i'ndn 'eagle'
i-tc!i'laq 'cricket'
i-q!apca'lwac 'turtle'
a-tca'la 'grindstone'
Diminutive
il-tsfi'non 'bird'
i-ts!i'laq 'grasshopper'
is-k!a'psalwas 'lock (of door)'
a-tsa'la 'file'
It will be observed that several nouns on becoming diminutive in form at the
same time change to a more suitable gender, masculines often becoming femi-
nines (e.g., wa-ska'n), neuters (e.g., ii-sk.'i'luks), or diminutive duals in is-
(e.g., is-k!a'psalwas). Most examples of diminutives and augmentatives hith-
erto given have been formed from nouns that in themselves have no necessary
diminutive or augmentative force. Other examples than those already given of
246 ^f American Indian Languages 2
words with inherent diminutive force, and hence with at least partial dimin-
utive consonantism, follow:
a-k!u'ksk!iiks 'ankle'
a-p!u'xp!ux 'elbow-joint'
i-pfu'xc 'cotton-tailed rabbit'
a-t!antsa 'crow' (contrast i-cka'lax 'raven')
i-skfu'lya 'coyote' (? cf. i-cgilukc 'wolf')
a-gu'sgus 'chipmunk'
a-p!una'tsEktsEk 'mosquito' (? cf. - bEna 'jump')
is-ga'k!aps 'hat'
i-k!a'its 'smallness' (contrast -^ail 'bigness')
i-k!a'stila 'crab'
H-xan '(somebody's) child'
i-sklwo'latsintsin 'swallow'
wa-tsk!E'nLX 'nit'
Particularly instructive as indicating a live feeling for diminutive consonan-
tism are such words as a-lik!u'k 'chicken' and a-lap!u's 'cat' borrowed from
Chinook jargon (p in -pus would not be consistent [642] with diminutive s). It is
perhaps not too far-fetched to recognize augmentative consonantism in the fol-
lowing nouns:
i-ga'nuk 'beaver'
i-gii'nat 'Chinook salmon' (contrast wa-tsu'iha 'blue-black salmon')
i-CE'lqcElq 'porcupine'
ic-ga'kwal 'eel'
i-du'iha 'buffalo'
ic-kcku'ct 'testicles (contrast is-qxu's 'eyes')
i-gu'cax 'sky'
ic-gwo'lala 'gun'
wa'-itc 'tail of mammal'
Ic-li'ct 'fish-tail' (contrast is-p!i'ost 'tail of bulb, dried fish')
It sometimes happens that a change to diminutive consonantism implies not
so much the diminutiveness of the object referred to as a sense of endearment.
This seems particularly true in the case of certain terms of relationship:
N on- diminutive Diminutive
-qcE-n 1 'man's son's -/:.'fl'c-w-c 'paternal grandfather'
ga'c-u (vocative) j child'
-gak-an 1 'man's daughter's -ga'/c.'-w-c 'maternal grandfather'
ga'g-u (vocative) /child'
-gi-an 'woman's son's child' -k!i-c 'paternal grandmother'
Seven: Penutian Languages
247
Interesting as examples of augmentative consonantism are the names of
Coyote's four sons, all of which are derived from words denoting body-parts of
the salmon. The augmentative consonantism implies the lubberliness of
Coyotes sons.
Body-parts of salmon
i-k!la'tcin 'salmon-head gristle'
i-ksa'lk!uts 'backbone of fish'
i-qfwi'nan 'fin'
a-k!a'tk'^tgwax 'adipose fin'
(? better -qfa'tk'^tgwax)
Names of Coyote's sons
Sipa'-glatsin 'Big Gristle' (Wishram
Texts 66.5)
Sipa'-ksalguts 'Big Backbone'
(Wishram Texts 66.6)
Sapa'-gwinan 'Big Fin' (Wishram
Texts 66.7)
Sapag-a'thHgwax 'Big Adipose Fin'
(Wishram Texts 66.8)
As has already been remarked, the noun is not the only part of speech that
illustrates the consonantal play here discussed. Adverbs and particle verbs of
appropriate meaning sometimes show diminutive consonantism: tsfii'nus 'a lit-
tle'; sdk! 'to whistle'; sa'u sau 'to whisper' (contrast Lower Chinook cdu)\
Lower Chinook k!a 'and' may be diminutive to ka. The diminutive form of a
particle verb denotes a less intense state of being or activity than its correlative
form. Sometimes its meaning is considerably specialized: [643]
N on- diminutive
tcic 'cold'
ma'ca 'to spoil'
gut 'to break up (earth) by digging'
Possibly also:
wax 'to pour out'
Iq.'up 'to cut'
Diminutive
(ts/u'nus) a- its d's 'just (a little) cool'
(Wishram Texts 190.15)
ma'sa 'to be ashamed'
kfu'tkfut 'to pluck'
wax 'to set on fire; to bloom'
fkfup 'to shoot'
The dual in is- is not the only example of a diminutive form of a purely gram-
matical element. The diminutive stem -qlwa'lasup 'fast running' occurs with
possessive prefixes showing diminutive consonantism. Thus the normal ele-
ments -tea- 'her' and -cda- 'of them two' appear as -tsa- and -st!a- in i-tsa-qfwa'l-
asup 'she runs fast' (Wishram Texts 66.9) and i-st!a-q!wa'lasup 'they two run
fast' (Wishram Texts 66.13). Similarly, in a song (Wishram Texts 94.23), where
the reference is to is-p!i'ast 'tail of bulb', a noun of diminutive form, the pro-
nominal element cd- and the post-positive local element -ha appear as -st
(? better st!-) and -p!a. Thus:
staimapid' giskip'.i'ast 'it-alone-at the-my-tail'
Finally the verb may show diminutive consonantism, partly in the stem itself,
partly in its local and adverbial prefixes and suffixes, partly and most frequently
248 VI American Indian Languages 2
in its pronominal prefixes. Examples of verb stems in distinctly diminutive form
are not exactly common, but certain cases seem clear enough. Thus gaqiulatla'-
ulx 'he was tossed up' (Wishram Texts 84.26) and gatciutat!a'mElq 'he swal-
lowed him by sucking him in' evidently contain a diminutive form of the verb
stem -lada- 'to throw away'; ^//w'^^wajc 'it trembles' (Wishram Texts 116.10) and
gasi'ximk!na-uk"atsk 'he looked around' (Wishram Texts 30.6) show dimin-
utive consonantism both in their stems {-skw- and -k!na-u-) and in their first
incorporated pronominal objects (dual s-), the latter verb also in its adverbial
suffix -tsk, doubtless the diminutive form of -tck 'up from position of rest';
gats(s)altsgi'ma 'he laid her belly up' (Wishram Texts 56.27) shows diminutive
consonantism in both stem (-tsgi) and incorporated pronominal subject (-ts-)
and first object (dual -s-).
We have already given -tsk as an example of a derivative suffix with dimin-
utive consonantism. Other such suffixes are -p!a 'slightly out (of position)'
(from -ba 'out') in ayulapfa'tcguxwida 'it will tilt up', literally 'it will spon-
taneously move out up from its sitting [644] position' (Wishram Texts 184.10)
and tsu (from -tcu 'down') in ililu'stsu '(water) moved down into the (hollow
place)'. As examples of diminutive forms of local prefixes may be given -klEl-
(from -gEl- 'directed toward') in ga-tssi'k!Elutk 'he looked at him' and its reflex-
ive correlative -xeI- (from -xeI) in gasi'xElutk 'he looked'; -sklEm- 'under' in
iniasklEmla'datcu 'I threw it down under her' is doubtless diminutive to -gEtn-
'next to' (cf. -tcu and -s-tsu above).
The only examples of diminutive consonantism in the pronominal prefixes of
verb forms occur in the case of ts (for tc, third person masculine subject tran-
sitive) and s (for c, third person dual subject intransitive and transitive and
object transitive). Whenever the object of the transitive verb (or the apparent
subject, really first object, of the "half-transitive" verb) is diminutive in form,
the pronominal prefixes tc and c appear as ts and s; the ts by no means implies
the diminutive character of the transitive subject. Examples are: I'wi gatssu'x
isie'nqxoq 'he looked at his fish-line' (Wishram Texts 140.28), where the incor-
porated pronominal dual element -s- of gatssu'x refers to the diminutive dual
object is-ie'-nqxoq 'his fish-line', while the pronominal subject -ts- 'he' agrees
with the object in diminutive consonantism; gatksu'klam {-Iks- always appears
for -skt-) 'the two (women) came home with the (baby)' (Wishram Texts 2.12),
the diminutive dual -s- referring to the grown-up women, not to the baby ; ga^f-
ngatklagwa'x gas ktenaklwd 'st 'it-waves-freely-over-me-my-feathered-cloak'
(Wishram Texts 142.5), where the first object -s- of the half-transitive verb
refers to the diminutive dual noun s-tenakfwd'st '(small) feathered cloak'. Par-
ticularly noteworthy in this connection is the idiomatic use of a diminutive dual
object -.s- referring to an implied, unexpressed noun of diminutive significance;
there need not even exist such a diminutive dual noun to which reference, if
desired, could be explicitly made. A good example is: gaksi'lutk 'she cradled
him', literally, 'she put the-two-small (objects) down to him', where 'the two
small (objects)' refer to an implied word for 'cradle', though the word for 'cra-
dle' in actual use is a masculine (i'-lkau). Similarly, verbs of jumping and somer-
saulting have an incorporated diminutive dual object -s- referring to 'the two
Seven: Penutian Languages 249
small (feet)', though the actual word for 'feet' is plural (i't-pc). Examples are:
gaksu'bEna 'she jumped'; gasixmi' Lgwa 'he turned a somersault' (Wishram
Texts 82.18); and gats(s)altsgi'ma 'he laid her, belly up'. The [645] most trans-
parent example of the use of an incorporated diminutive dual object to refer to
an unexpressed but existing noun is afforded by certain verbs of looking, in
which the -s- has reference to is-qxu's 'the two eyes'. A frequently occurring
example of such a verb is gatssi'klElutk 'he looked at him', literally, 'he put the
two small (eyes) down toward him', the -tc- and -gEl- appearing in their dimin-
utive forms -ts- and -k!El- to agree with the object -s-; gasiximk!na'-uk"atsk 'he
looked around' is another such verb.
As a rule, it will have been observed, a verb form tends to be consistently
diminutive or non-diminutive in its consonantism. It is at least possible, how-
ever, to limit the application of the diminutive idea to some specific element of
the action by "diminutivizing" only some corresponding element of the verb
form. An example already published elsewhere will again do service here. The
normal word for 'I struck him with it' is inigE'ltcim. If the verb stem -tcim
appears, with diminutive consonantism, as -tsim, the implication is that the
missile used is a small one. Hence we have four forms: inigE'ltcim 'I hit him with
it'; inigE'tsim 'I hit him (a child perhaps) with it'; iniklE'ltcim 'I hit him with it
(something small)', and iniklE'ltsim 'I hit him (a child) with it (something
small)'. To be sure, such examples are very uncommon and the one just given is
perhaps little more than a linguistic tour de force. Nevertheless, it shows very
clearly how thoroughly alive is the feeling for the significance of consonantal
play.
Post-positions in Wishram
[650]
Wishram, differing markedly in this respect from Lower Chinook, makes
rather considerable use of a series of post-positive particles [651] defining mate-
rial case relations (chiefly local and instrumental). As most such relations can
be expressed by means of local and adverbial prefixes and suffixes in the verb,
the denominating parts of speech being in apposition to incorporated pro-
nominal elements, this use of postpositions must be considered as un-
Chinookan in origin; the fact that some of the postpositive particles are
phonetically identical with corresponding Sahaptin case suffixes proves the
whole process to be borrowed from the neighboring Sahaptin linguistic stock.
As a rule, such postpositive particles are used with denominating parts of
speech (nouns, pronouns, adjectives), but some of them may also be suffixed to
predicating words (verbs, particle verbs); in the latter case the predicate is to be
considered as substantivized syntactically, though not morphologically, and is
used subordinately to another predicate. Wishram thus utilizes its postposi-
tions to some extent in the building up of subordinate clauses. Where a noun or
250 ^' American Indian Languages 2
other denominating part of speech has been already represented in the verb by
an incorporated pronominal element, its relation to the verb and to other nouns
in the sentence is necessarily already defined, so that no postposition is neces-
sary; even here, however, it not infrequently happens that a postposition is
pleonastically used (compare such English possibilities as "He entered into the
house"). If a noun is modified by a preceding attributive word (demonstrative
pronoun, numeral, noun, or adjective), the postposition is used with the modi-
fying word. The postpositions, with examples illustrating their uses, are listed
in the following paragraphs:
1. -ba (-pa) 'in, at\ With this element should be compared Yakima -pa 'in'.
Examples illustrating its use with nouns and pronouns occur with very
great frequency, so that only a few need here be given.
cikxa'-imdt ci'tfix yak"cxa'tpa 'half of it lies in his mouth' 4.3^
gaklakxa'-ima ilkfa'ckac akni'mba 'she put the child in the canoe' 2.11
atgadi'mama da'uyaba wi'lx 'they will come in this land' 6.17
gayu'yam ixtpd' wilx 'he arrived at one land' 6.28
itcqxE'mEm axqxatcpa 'I am sick in my breast' 12.27
gatci'upmt it.'o'xwatckpa 'he hid it in the bushes' 18.25
gatu'ya yaxka'ba 'he went up to him' 20.10 (one can also say gaiiglu'ya 'he
went to him' with local prefix -gEl-)
[652]
gadiq'.Eltxi'uba icia'gitcba ya'k"cxatpa wamtfu'xiba 'they went out through
him at his nostrils, at his mouth, and at his ears' 28.24
gahi'xuni yaga'iipa wi'mal 'it floated in the great river' 48.7
alxu'ya wa'tcktib' itga'qpuks 'let us go on the tops of the grass' 70.26 (literally,
'the-grass-at its-tops')
Observe that the first two examples illustrate its pleonastic use; the nouns
yak'^cxa't and akni'm have been respectively anticipated in the verb by the
pronominal elements -/- and -a-, while their local relation to the verb is
defined by the prefix -k- 'on' following these elements, -ba is also used with
demonstrative stems to form adverbs of place where: da'ba 'here'; kwo'ba
'there'; ia'xiba 'yonder'.
As subordinating element, -ba denotes 'where'; less frequently it indicates
cause. It is suffixed either to the verb itself, or, similarly to the case of the
modified noun, to an adverb or particle preceding the verb. Examples are:
cta'xya I'nadix qla'tsEuba gatccgE'lgElx 'across yonder (were) the two where
he had first seen them' 8.10 (literally, 'first-at he-saw-them')
galiktcYptck gatcc^ElkE'lxpa 'he came to land where he had seen them' 8.5
e'wi gali'xdx gayaxa'limaixpa 'he looked back to where he had thrown himself
into the water' 8.6
ma'sa gali'xox qfu'mba gagi'iix 'he was ashamed because she had disturbed
him in his sleep' 58.26 (literally, 'disturb-in-sleep at she-made-him')
2. References are to Wishram Texts.
Seven: Penutian Languages 251
2. -/amr (often with palatalized a as -ffl>?i/, -iemt) 'to, from'. This suffix is proba-
bly Chinookan in origin; it may be plausibly analyzed as verb stem -/- 'go'
+ verb suffix -am 'arriving' + tense suffix -t. This analysis would explain
its two apparently contradictory meanings. It tends to draw the accent to
itself. Examples are:
ickte'lgwiptck wimalia'mt 'they collected (driftwood) from the river' 2.2
nigElga'ba iciagitcia'mt 'it flew out of his nostrils' 80.29 (literally, 'out of him
from his nostrils')
gacx^klwa'x txtdqtia'mt 'the two returned to their house' 2.12
gayukfwi'xa ilaxni'miemt 'he swam to the person's canoe' 18.23
mxa'tcktcam wimatia'mt 'go to the river and wash yourself 22.18 (literally,
'go-and-wash-yourself to-the-river')
gatciu'k"{ itq^Udmt ittcqoa 'he took the water to the house' 28.8
As subordinating element it may be translated as 'to where'. An example of its
use after verbs is:
[653]
asEmxElu'tka a'tpxiamd aga'tax 'you shall look towards the east' 188.21 (liter-
ally, 'she-comes-out to-where the-sun')
3. ba'ma 'for, belonging to'. This is evidently the Yakima suffix -pama 'for'.
Examples of its use with denominating words are:
na'ikabam' amtklni' dama itqagi'lak 'for my sake you two will go and get me
the woman' 62.25
ya'xtau laxka'bama IgiubttcEma 'that (fish) he obtains for himself 186.4
gaqxd'gwigax itsH'ndnks wi'lxpama 'animals were taken belonging to the
country' 16.13
ctmo'kct gactu'ix ntca'ikabama 'two of our men (literally, us-for) went on'
216.16
da'nbama qxe'dau mxu'lal 'what for do you speak thus?' 132.24
tga'tqwdm luwa'n qa'xbabama 'he has come I know not where from' 128.17
(literally, 'what-in belonging-to')
k.'a'ya kwo'babama idE'lxam tcduxt 'he had not made people belonging to
there' 44.23
gi'gwalbam' itk'.i'tit 'underclothes' (literally, 'below-for clothes')
Less commonly bama may precede. An example is:
bam ' iiJ'wulx aktugwi" ilk'.a'lamat 'he carries rocks for (i.e. , in order to gain)
strength' 186.17 (cf iLxe'wulx bama 188.2)
When used at the beginning of a predication, bama gives it the meaning of a
clause of purpose. Examples are:
ba'ma ta'-itcka a'lEm' atcludi'na 'in order that he might kill them' 54.2 (liter-
ally, 'for them will he-will-kill-them')
bama capca'p qiuxu'nnit ika'ba 'for chopping up the ice' 188. 19 (literally, 'for
chop-up it-is-always-made the-ice')
When accented {bama'), it is used after predicates to mean 'ever since'. An
example is:
252 ^f American Indian Languages 2
nk'.a'ckacbama' k/d'ya qxantcix itctcgE'ruEin 'ever since I was a child I have
never been sick' 190.9
4. (E)nEgi 'with, by means of, less frequently 'made out of. It seems to be the
Yakima genitive case ending -ngi. Examples are:
axk' E'riEgi amcgiu'xa Iqfo'p 'with it you will cut it off 12.4
Lqfo'p gaigi'ux aqE'tiEkc E'riEgi 'they cut it off with the stone knife' 18.5
gatkld'qi' aiakcE'n EtiEgi 'he counted them with his finger' 18.19
itia'ma ngi gayu'ya 'he went by means of a round-pointed canoe' 38.21
iga'hEnac E'riEgi gatdu'x 'he made them out of young oak' 4.13
[654]
Less frequently ngi may precede. Examples are —
xa'u xau galxu'x ng' ilkcE'n 'they combed themselves with the hand' 78.10
ayakla'lamat ngi wa'nux 'his pipe (was) made out of a stomach' 94.9
5. a'mEni 'made out of, less frequently 'with'. It is perhaps the Yakima -nmi.
Examples are —
sd'q" itk.'a'lamat a'mEni akitxax 'it is entirely out of stones' 82.13
isklu'ly' amEni isga'klaps aqsu'xwa 'a hat is made out of coyote' 182.7
alklwa'dit amEni aqiu'xwa 'it is made of tule' 182.9
itq/u'tc' a'mEni tsE'tsEx gaqtu'x itkfa'munaq 'they split trees by means of
antlers' 182.14
6. -pt 'up to' is used to form adverbs out of demonstrative stems: dapt 'up to
here'; kwopt 'up to there, then, enough'; ya'xpt 'up to yonder'. Probably
etymologically identical with this element is -bEt, frequently added to
verbs or other words in the predicate to form temporal clauses. Examples
are —
gatclE'mquit Iqa'wulqt gagiula' dabit 'he spit blood when she threw him down'
14.11
galikta'tckpEt pla'la igi'xox 'when he had come up out of the water, he
stopped' 22.18
lE'p(b)Et atxu'xwa anigElgd 'ya 'when he dives, I shall take hold of it' 18.20
nk'.a'ckacbEt 'when I was a boy' 188.8
aga'lax alaxu'xwa yaxtadi'wi gali'xux gaixo'qbEt 'the weather will be as it was
when they came together' 130.27
When rhetorically lengthened to -bd't, this post-position has a general
cumulative significance; with verbs it is best translated 'as many as'.
Examples are —
^wE'nEmabd'd Hgwo'mEx antk.'wa'lalaqwida 'I shall be absent as much as five
days' 122.12
kwo'pt natcdup^Enayabd't 'that many (ropes) as he had apportioned' 188.6
qxa'ntcipt alkixa'tgway' atciulxamabd't 'he piles up as many as he tells him to'
186.19
I
Seven: Penutian Languages
253
diwi (emphatic dd'wi) 'like'. This element is very likely of demonstrative
origin, and so does not perhaps belong here. It is freely used, however, as a
post-position, and so may be included. Examples are —
ickla'li diwi datdi'p 'striped like a basket' 166.2
iya'tqx ilgwa'Ulx diwi 'his body (was) like a person's' 166.17
naika dd'wi itcE'lgulit 'exactly like my appearance' 104.10
Demonstrative Pronouns and Adverbs in Wishram
[625]
Near 1st person
Near 2nd person
Near 3rd person
Near 3rd person
(formed from ya'xdau)
Masculine
da'uya(x)
da'ya(x)
[ya'xdau
\ya'xda(x)
ya'xia(x)
yakd'xdau
Feminine
da'ua(x)
da'wa(x)
a'xdau
a'xda(x)
a'xia(x)
akd'xdau
Neuter
da'uia(x)
da'{a(x)
la'xdau
ia'xda(x)
ia'xia(x)
{akd'xdau
Near 1st person
Near 2nd person
Near 3d person
Dual
da'ucda(x)
da'cda(x)
f cda'xdau
\cda'xda(x)
cda'xia(x)
Near 3d person
(formed from ya'xdau) cdakd'xdau
Plural
da'uda(x)
da'da(x)
da'xdau
da'xda(x)
da'xia(x)
dakd'xdau
Plural, persons
da'uda-itc
da'da-itc
da'(u)la-itc
, da'(u)a-itc
da'xdauaitc
la'xdauaitc
a'xdauaitc
da'xiaitc
la'xiaitc
a'xiaitc
dakd'xdauaitc
hikd'xdauaitc
akd'xdauaitc
Note. — It is somewhat doubtful whether ya'xdau should be so read or as
ya'xdau. (x) in personal and demonstrative pronouns is deictic in value.
-ka may be added to demonstratives in -/7c.
Elements -tfa and -tHkc are perhaps "diminutive" forms of demonstrative
pronominal stem da 'this' and personal plural dike.
254 VI American Indian Languages 2
Following is a list of the demonstrative adverbs of the Wishram dialect:
Locative
up to
towards, on . . . side
Stem da(u)
da'ha 'here'
dapt
dabd't 'little ways further
on'
kwobd't
Stem kwo
kwo'ba 'there'
kwopt
iyax da'ba ^^A^y
Stem iaxi
ia'xiba'yondef
ia'xi 'away, off
ya'xpt
iaxd't 'further on'
Stem di
di'ka 'here'
(pt also in
di'gat (18.17)
(dika data' 92. \\)
qa'n tcipt
'how long?')
[626]
Stem gi
g/>/ (18.17)
{i'wat 'to you (place)'
i'wa 'thus, there' (106.22)
[iwa'tka (158.24)
Note. — Compounded with gi are also da'ngi 'something'; qa'tgi 'somehow';
qxa'matgi 'somewhere' (96.11).
Related to di'ka and di'gad is perhaps digu'tcix 'perhaps' (96.17); also di'wi
'like'.
In -xi we have, besides ya'xi, also (aga) du'xi 'oh, well'! (60.4).
Note. — Ya'xa 'indeed ' (also in quct i'axa 'as it turned out'); au (perhaps =
aw', a'wa, and related to Chinook _vflwfl) in da'n au ayamlu'da 'what, pray, shall I
give you?' (154.6); yaxa'wa 'however'.
Note also kwo'bixix 'right there, not very far'.
-a'dix forms: a'ngadix 'long ago'; ixtka'dix (192.2); ina'tkadix (192.5).
With stem dau: kwo'dau 'and'; da'ukwa 'just as before'; qxi'dau 'thus'.
Independent Personal Pronoun
These correspond to Wishram forms recorded by Sapir:
na-ima 'I alone'
Uiinuidikc, dci-iniadikc , dimadikc 'they alone'
mci-ima "thou alone'
Ixa-imadikc 'we (incl.) alone'
[627]
Besides these, Doctor Sapir has recorded in Wishram the following:
Shortest form:
nafx) T ya(x) "he da'-ifc 'they'
la'-itc 'they' (Wishram Texts 48.4)
a'-itc 'they'
inclusive:
nai't'.a 'I too' ya'xt'.a 'he too' la'-it'.ikc 'they too'
Ixai't'.ikc 'we too' da'-it'.ikc 'they too'
a'-it!ikc 'they too'
3. References are to Wishram Texts.
Seven: Penutian Languages 255
He remarks that the demonstratives of the third person (ya'xia) seem mor-
phologically parallel to first and second personal emphatic pronouns (na'ya);
that the demonstrative element -/- is characteristic of the first and second per-
sons, -X- of the third; as in
na-i-ka T ya-x-ka 'he'
na'-i-tfa 'I too' ya-x-t!a 'he too'
na'-(i)-ya T ya'-x-ia 'he'
These elements -/- and -x- are probably identical with Chinook -/"- and -x-, -x
in x-i 'lu and x-ix-, xax.
Modal Elements
[578-579]
2. ni-. This prefix is confined to the dialects east of the Kathlamet. It takes the
form nig- before vowels, like the preceding a-, prefix indicating a change
from one state into another, translated 'to become' in intransitives and as an
aorist in transitives. It occurs in transitive and intransitive verbs. It expresses
a somewhat indefinite time past, and is used in speaking of events that hap-
pened less than a year or so ago, yet more than a couple of days ago.
ni-y-u'ya 'he went' {ni- 'past'; -y- 'he'; -uya 'to go')
nig-u'ya 'she went' (the same before vocalic element; -a- 'she', being con-
tracted with -u- into -u)
ni-tc-i-gil-kErhe saw him' {ni- 'past'; -tc- 'he'; -/- 'him'; -g// verbal prefix;
-knl 'to see')
[579]
3. a-. [Prefix indicating the future in intransitive verbs.] In the dialects east of the
Kathlamet it is used also with transitive verbs.
a-tc-i-gE'l-kEl-a 'he will see her' {a- 'future'; -re- 'he'; -/- 'him'; -^eI- verbal
prefix; -kEl 'to see'; -a 'future')
4. ga-, before vowels gal-. This prefix is confined to the dialects cast of the
Kathlamet. It expresses time long past, and is always used in the recital of
myths
ga-y-u'ya 'he went' (see analysis under 2)
gal-u'ya 'she went' (see analysis under 2)
ga-tc-igE'l-kEl 'he saw him' (see analysis under 2)
n- may be used in place of this prefix.
5. na-, before vowels nal-. This prefix is confined to the dialects east of the
Kathlamet. It refers to recent time exclusive of today, more specifically to
yesterday. Its use is analogous to that of the preceding.
256 VI American Indian Languages 2
WISHRAM TEXT'
By Edward Sapir
Coyote and Itc!e'xyan
673 Aga' kwo'pt^ gayu'ya* isklu'lya'^ wi'tlax." Na'2wit' gayu'yam;*
Now tlu'ii he went Coyote again. Straightway he arrived
going;
galixE'ltcrna(i® isklu'lya gwa'nisini'*' qtiilatla'iiiElqt" idE'lxam"
he heard Coyote always they findef.) are always the people
Mwallowing them down
674 itclE'xyan.'^ Qxa'damt" gayu'y"^ ikni'm** na'wit gatcigE'lga"
Merman. Whither it went the canoe straightway he got hold of it
itclE'xyan; gatciutatla'mElq^'* ka'nawi'® dan.^° "NaitI'" a'g'"
Merman; he always swallowed it down every thing. "Me too now
atcnuiatla'niElEqEma,"" isklu'lya galixtu'xwa-it." Aga kwo'pt
he will swallow me down," Coyote he thought. Now then
gayu'y' isklu'lja; gatcigE'lga yag'ail" ikla'munaq.^^ Aga kwo'pt
he went Coyote; he got hold of it its bigne.ss the tree. Now then
675 La'x" gali'xox.'* GatcigE'lga itclE'xyan, gaqiulat!a'raElEq."
insight he made him- He got hold of merman, they (indef.) swallowed
self. him him down.
Na'wit ittcqo'ba^" gi'gwaP' isklu'lya galixi'maxitam^'' wi'lxba.*'
straightway in the water below Coyote he arrived falling on the ground
Aga kwo'pt gatcugi'kfil^* Igabla'd'* idE'lxam; Jgabla'd'^ akni'ni"
Now then he saw them their multitude the people; their multitude the canoes
axu'xf* kwo'ba^® gi'gwal iltcqo'ba. Aga kwo'pt gatcigE'lkEl*"
they are there below in the water. Now then he saw it
piled together
isklu'lya itclE'xyan yago'mEnii^' qxwoL^'' iki'ax." Aga kwo'pt
Coyote Merman his heart hanging it is. Now then
676 gaqiu'lxam"* isklu'lya: "Ya'xdau"'' itclE'x3'an yago'mEnii.'' Aga
they (indef.) told Coyote: "That Merman his heart." Now
him
kwo'pt Lq!6'p"'= gatci'ux;"^ Lqlo'p"*^ gali'xox*^^ itclE'xyan yago'niEnii.
then cut he made it; cut it made itself Merman his heart.
Aga kwo'pt ka'nawi gatkxEni'yutck" sJi'q"''^ akni'm kwo'dau"
Now then all they each floated up out entirely the canoes and
of water
idE'lxam kwo'dau isklu'lya.
the people and Coyote.
Aga kwo'pt gali'kim" isklu'lya: "Lga^^ pu" qii'ma^" ma'ima"
Now then he said Coyote: " Perchance would how you alone
itclE'xyan qxi'dau" amdu'xwa" idE'lxam? Da'uya^^ wi'gwa" aga
Merman thus you will do to them the people? This . day now
677 kwo'pt^® qxi'dau amdu'xwa idE'lxam. Na'ika" isklu'lya yamu'lxam.^*
that far thus you will do to the people. I Coyote I have told you.
them
Kwa'ic*" da'uyaba®" wi'lx atgadi'mama" idE'lxam. Kwo'pt
Soon in this land they will arrive coming the people. Then
alugwagi'ma,®^ 'Qxi'dau ^ex«^ gatci'ux^* isklu'lya itclE'xyan.'
they will say, 'Thus exercising he did to him Coyote Merman.'
supernatural
power
Kwo'pt a'ga itclE'xyan pla'l'^^ amxu'xwa.""
Then now Merman being quiet you will make yourself."
Seven: Penutian Languages 257
1 A connected English translation of this text will he found in Sapir's \\ ishram Texts, Puhlicatlons of 673
the American Ethnological Society, ii, 4 1 , 43. The Indian text as here given has been very slightly normal-
ized from its form as there published (pp. 40, 42).
« Used partly with weak temporal force, partly as mere connective in narrative. It is frequently prac-
tically untranslatable into English.
3 kw6pt, THEN, AT THAT TIME, is regularly used with preceding aga to mark new step In narrative. It
can be analyzed into demonstrative stem kwo- (or ktoa-) that (= Chinook go there) and local suffix
-pt UP TO (so AND so) FAR. Neither of these elements occurs freely, kwo- is not used to form demon-
strative pronouns, only occurring stereotyped in several adverbs; l)esides kwopt we have kwo'ba there
(note 39), and kwo'dau and (note 46). -pt also hardly seems to occur except stereotyped in adverbs;
cf. dapt AS FAR AS THIS (related to da'ba, this-in=here, as kwopt is to kwo'bc), and yaxpt, as far A3
THAT YONDER, from ya'zi off yonder). See also note 56.
* ga- (gal- before vowels) = tense prefix denoting remote past, regularly used In myth narrative. - y-^
3d per. masc. subj. intr., referring to isk.'u'lya. before consonants it would appear as -i-, while gal- would
then appear as tense prefix (ga-y- = gal-i-: see notes 9, 28, 32, 47). -u- = directive prefix away from
SPEAKER, -ya = verb stem to go.
* i- = masc. noun prefix with which -y- in gayu'ya is in agreement, -sk.'u'lya = noun stem coyote,
apparently not capable of analysis; perhaps loan-word from Klickitat spi'lya. Chinook has another stem,
-t.'d'lapas.
« Composed of wi'tfa again and deictic particle -i: cf. da'uya (note 54) and da'uyaz this. wi'Ha
is most plausibly explained as stereotyped adverb from wi-, masc. noun prefix (originally independent
masc. pronoun? See notes 19 and 33), and -t!a, emphatic particle added to pronouns, too, also (see note
21). According to this analysis wi't!a(x) was originally formed from *wi as ya'xtfa(i) he too from ya-X'
HE. Originally it must have meant that (masc.) too, but was later generalized in meaning.
' Rhetorically lengthened form of nu'it immediately, right away. When thus lengthened to na'wit,
it seems to imply direct, unswerving motion without interference of other action; it may then be rendered
as STRAIGHT ON or ON AND ON.
8 As in note 4, except that instead of verb stem -ya we have its shorter form -y. -i- (as in yu'it he goes;
cf. also note 61). To this is suffixed verb suffix -am arrive while — ing, go (or come) to do . Several
verb stems have two forms,— one in -a, and one without this -a (e. g., -pa and -p to go out; cf. gali'pa
SHE WENT OUT With atpx SHE COMES OUT).
9 gal- = tense prefix ga- before vowels, -i- = 3d per. masc. obj. before reflexive element (reflexive verbs
have, morphologically speaking, no subject). -xEl- = indirect reflexive composed of reflexive element -r-
and local verb prefix -I- to, into, -tcmaq = verb stem to hear. galixE'ltcmaq means literally to him-
self HEARD. TO HEAR SOME ONE is expressed by -x-tcmaq with prefixed transitive subject and object
pronominal elements.
10 Adverb not capable of analysis.
" q- = indefinite transitive subj. -t- = 3d per. pi. obj. tr., referring to idE'lxam. -u- =directlve prefix
(very many verbs have this "directive" -u- even when no definite idea of direction away from speaker
seems to be implied). -tatfaniElq- is exampleof rarely occurring compound verbs. -Za^'a- is "diminutive"
form of verb stem -lada- to throw down, away (in this case its meaning seems to correspond somewhat
more closely to that of its Chinook cognate -Lata to pull back); -mElq- is best explained as verb stem
-mEq. (or -mq-)TO vomit with infixed -I- of frequentative or continuative significance (that -Ms not really
part of stem is shown by form itciulatfa'maq he swallowed him down); pull back -i- vomit may
be construed as meaning vomit backward, draw to one's self and swallow, -t = tpnse suffix of
present time. Observe peculiar sequence of tense, he heard . . . they sn- allow them down. Verbs
that are dependent on other verbs, chiefly of saying or perception, are always present in tense, no matter
what tense is logically implied; cf. below gatcigE'lkEl . . . iki'az (note 43) he Syi if it ... it /5.
'2 id- = 3d per. pi. noun prefix, in concord with -t- in preceding verb, -ham (-E- is inorganic) = noun
stem village (wi'liam village is formally masc. sing, of idE'lxam people); -liam is evidently
related to -li (see note 33).
" j- as in note 5. -tc.'Eiyan = noun stem merman, protector of fishermen (see Wishram Texts, p. 40,
note 2; p. 42, note 2; p. 256, note 2); no etymology suggests itself. Syntactically itc.'E'iyan is subject
Implied, but not grammatically referred to, by q- of preceding verb. This clause can hardly be considered
as quite correct; properly speaking, itc.'E'xyan should go with tclulat.'a'mElqt.
258 Vf American Indian Languages 2
675
•< From interrogative stem qia- (or qa-), seen also in qa'xba what-in? = where? qa'xia of what
KIND? and qa'ngi what-with? = how? -damt = local suffix toward found suffixed to several
adverbs (cf. ca'ialadamt toward above, gigwa'ladamt toward below). This -damt is evidently
related to local noun sufTix -iamt to, from. Qxa'damt here introduces indirect question, and may best
be translated as no matter where.
li = gayu'ya. Final vowels are regiilarly elided when following word begins with vowel. For analysis
of form, see note 4.
•« i- as in note 5. -knim = noun stem canoe. This stem can be only secondarily monosyllabic, for
otherwise we should have * wiknim (see note 3.3); its Chinook cognate -kanlm shows original dissyllabic
form. See also note 37.
1' ga- = tense prefix as in note 4. -tc-.= 3d per. masc. tr. subj., referring to following itc.'E'xyan as sub-
ject, -i- = 3d per. masc. tr. obj., referring to ikni'm of preceding clause as object. -gEl- = verb prefix of
adverbial force, toward (with purpose, intent to reach); it here replaces directive -u- of most transitive
verbs, -ga = verlj stem TO get hold of, seize; it is possibly to be identified with verb stem -ga stick to,
its particular active significance being gained by use of transitive pronominal prefixes and verb prefix -gE 1-.
18 ga-tc-i- as in note 17, -i- here referring to following dan. -u-lat.'a'-mElq as in note 11.
" ka'nawi all, every is most probably compounded ofkana- all together (found in such niuneral
fonns as ka'nactmokct all-the-two = both and, with unexplained -m-, in kanEmlu'nikc all three
PEOPLE) and old 3d per. masc. demonstrative pronoun *wi (cf. note 6) now no longer preserved as such
(except in such petrified words as wi't.'a and ka'nawi), but specialized, like its corresponding fem. wa-,
as 3d per. noun prefix (see note 33). These old pronouns *wi and *wa are best explained as substantivized
from pronominal elements-!- (masc.) and -a- (fem.) by means of demonstrative element w- (or «-); this
latter element is probably identical witli -u- in demonstrative stem da'u- this (found also as da-; see
note 54), and with ( hinook -o- in demonstratives near 3d per. {I'OLa, locta, iota), ka'nawi must origi-
nally have meant something like all (of) that (masc), but, like wi't.'a, was later generahzed in signifi-
cance, ka'nawi is here, as often, rhetorically lengthened to ka'nawi to emphasize its meaning of totality.
""> Interrogative and indefinite pronoun referring to things, what, anything, something. Though not
provided with any sign of gender, it is always construed as masculine, hence -i- in gatciulat.'a'mElq. Its
correlative can (Kathlamet ban) referring to persons, who, anybody, somebody, is always neuter in
gender; he swallowed everybody down would he gatciulat.'a'mElq ka'nawi can.
21 Elided from na'it.'a (see note 15). Composed of 1st per. sing, pronominal stem nai- (seen also in na'-
ika I) and emphatic suffix -t.'a too, also (see note 6). All independent pronouns in -fco can be changed
to emphatic pronouns by merely replacing -ka by -t.'a (e. g., ya'ika he becomes ya'it.'a he too).
Syntactically na'it.'a here anticipates -n- in following verb (see note 23) as 1st per. sing. obj.
M = a'ga (see note 15). This particle is very frequently used before future verb forms in conversation.
23 a- = tense prefix of future time, -tc- = 3d per. masc. tr. subj. -n- = 1st per. sing. tr. obj. -u-lat.'a'-
mElEq- as in note 1 1 {-E- before -q- is inorganic). -Em- = connective before future suffix -a: verbs that are
continuative or frequentative in form regularly use this connective -Em- before certain suffixes (such as
future -a, cessative -tck, usitative -nil), -a = tense suflix of future time; in \V ishram verbs regularly form
their future by prefixing a- or al- (before vowels) and suffixing -a. It is somewhat difficult to see why this
form should be frequentative; one would rather except atcnulat.'a'mEgwa.
2< gal-i- as in note 9. -x- = reflexive element; literal translation of verb would be (to) himself thought.
•lux^w)- = verb stem to think, -a-it = verb suffix of rather uncertain significance here; it is found in all
tenses of verb but present, where it is replaced by -an {iilu'iwan he thlnks).
26 ya- = i-ya-. i- = masc. noun prefix, determining gender of noun stem -gail. -ya- = 3d per. masc.
possessive pronominal prefix, referring to masculine noun ik.'a'munaq. -gail = abstract noun stem big-
ness, yagailik.'a'munaq the tree's bigness may, like all other possessive constructions, be construed
either attributively (the big tree) or predicatively (the tree is big). Its attributive character is here
determined by presence of true verb (gatcigE'lga) as predicate.
•6 J- as in note 5. -k.'a'munaq = noun stem tree, stick, wood. This word is difficult of etymologic
analysis, yet can be no simple stem; -k.'a- is undoubtedly to be regarded as noun prefix (cf. ik.'a'lamat
ROCK, perhaps from verb stem -la to move), -k.'a- is most plausibly considered as "diminutive" form
of verb stem -ga- to fly '- r in air (as first elenient in comi)ound verbs); cf. itciuk!wa'la he whetted it
with itci'ula HE filed it, and miugwala' da-ulx i threw it up on top (of something) with iniula'da-uLf
I threw it up.
2' Particle verb. Though verbal in force, it is purely adverbial morphologically, having no grammatical
form of its own. In regard to tense and person it is defined by following verb, which serves as its form-
giving auxiliary.
» gal-i- as in note 9. -i- = reflexive clement. -6- (modified from -«- because of preceding and following
velar consonants) = directive prefix; ordinarily reflexive -i- replaces directive -u-, but there are several
Seven: Penutian Languages 259
676
verbs that retain it even when reflexive in form, -i = verb stem to do, make, -x-u-x to do to one's
SELF, MAKE ONE'S SELF, is regularly used to mean become. For other forms of verb stem -i see notes
43, 53, 64, and 66.
^ ga- as in note 4. -q- = indefinite tr. subj. -i- = 3d per. masc. tr. obj. -u-lat/a'-mElEq as in note U.
FOTms with indefinite -q- subject are very commonly used in Wishram in lieu of passives.
'• il- = 3d per. neut. noun prefi.x. -t- = inorganic consonant, serving as glide between I and c. <q6-
(= <qa-; a is velarized to 6 by preceding q) = noun stem water ; its shorter form <q- is seen in kta'cq
THE WATER OF THE TWO (Wishram Texts 190.14). -ba = local noun suffix in, at (see also notes 33,
39, and 60).
31 Adverb; -al is probably not part of stem, for it is found also in correlative ca'i-al above.
32 gal- as in note 9. -i- = 3d per. masc. intr. subj., referring to preceding i-sk.'u'lya. -lima- = verb stem
to put down, put on ground, lay down (as tr.); lie down (as intr.); probably composed of -x- on
GROUND(?) and -ima- put (cf. ga-ya-i-a'l-ima-lx HE put himself into the water [Wishram Texts 2.5));
whenever indirect object with -k- on is introduced, -x-ima- becomes -la-ima- (e. g., ga-k-l-a-k-ia'-ima
she laid it down on it [Wishram Texts 2.11]). -lit = quasi-passive suffix; -x-ima-xit- = be laid down,
LAY one's .SELF DOWN, FALL DOWN TO GROUND, -am = verb SUffix ARRIVE ING (cf. note 8).
33 wi- = 3d per. masc. noun prefix; masc. noun stems that are non-syllabic or monosyllabic require wi-
(cf. note 55); those that have more than one syllable have i- (see notes 5, 13, 16, 26); for probable origin of
wi- see note 19. In Chinook wi- has entirely given way to i-, except as archaism in some place-names
and in songs, -li- = noun stem land; seen also in wi'lxam village, idE'lxam people (see note 12);
probably also in wa'lxi FisHittG station and icE'lxlx staging for fishing, -ba as in note 30.
3< ga- as in note 4. -tc- = 3d per. masc. tr. subj. -u- = 3d per. pi. obj., referring to following idE'lxam
(before verb prefix -gEl-M per. plural obj. -Ms replaced by-w-, -gEl-ih^n becoming -^(w)'-; in other words,
-t- before gEl- is treated analogously to when it comes before -gel-), -gi- = plural form of -gEl-{see note,
40)^ouT FROM enclosed SPACE (cf. gi-l-a-g E' l-ba it flowed out of her [Wishram Texts 94.4]); analo-
gously to -</ir/-( see note 17) directive -u- is here replaced hy -gEl-. -kEl= verb stem to know (cf. l-k-d-u'-
kulHE knows them [Wishram Texts 176.10)); -gEl-ksl = to know from out one's (eyes), hence to
SEE, GET sight OF.
34 ^ = 3d per. neut. noun prefix, defining gender of abstract noun stem -bla4. -ga-= 3d per. pi. pos-
sessive pronominal prefix, referring to idE'lxam. -blad = noun stem multitude, great number. Igabla'd
idE'lxam is construed like ya'gnil ikfa'munaq (see note 25).
36 As in note 3.'>, except that -ga- = .3d per. fem. possessive pron. prefix (merely homonymous with -ga-
of note 35), referring to akni'm.
iT a- = 3d per. fem. noun prefix; though many fem. dissyllabic stems have wa- (e. g., wala'la pond),
it is here replaced by analogy of iknt'm (see note 16), as in related nouns i- and a-, wi- and wa- generally
pair off respectively, -knim as in note 16. logically akni'm canoes is plural, morphologically it is
fem., being so referred to in aiu'xt, (note 38); another example of fem. as plural is wa'mwa maggots, masc.
wi'mwa maggot.
38 a- = 3d per. fem. intr. subj., referring to akni'm. -i- = verb prefix on ground, on bottom (?) -u- =
directive prefix, -xt = verb stem to lie, sit, be placed, corresponding in use to Chinook -c. This verb
stem allows of no formal modification by means of tense affixes.
3" Composed of denionstrativestem kwo- (see note 3) and local suffix-6a Csee note 33): that-in = there.
♦0 As in note 34, except that incorporated obj. is -i- = 3d per. masc, referring to yago'mEnil, and that
-gEl- is unmodified.
<i ya- = t-j/a-asinnote25, i- defining heart as masc. in gender, while -ya- refers to itc.'E'iyan. -g6mEnil
heart seems to be verbal in form, -Enil being usitative suffix; yagO'mEnil may also be used predicatively
to mean he is alive.
« Particle verb, for which iki'ax serves as auxiliary.
<3i-= 3d per. masc. intr. subj., referring to yagd'mEnil. -kiax to be is another tenseless verb (cf. note 38).
It is best, though somewhat doubtfully, explained as composed of verb prefijc -ki-, which shows lack of
object of ordinarily trans, verb, and verb stem -i to do (cf. Kng. he oo£S well, i. e., gol^ along well)-
-a- would then have to be explained as inorganic glide vowel (cf. Chinook i-ke'-i he is and Wishram
i-ki'-i-ai he is, has become). For syntactic construction, as subordinated to gatcigs'lkEl, see note 11.
«B ga- as in note 4. -q- = indef. tr. subj. (cf. note 29). -i- = 3d per. masc. tr. obj., referring to Lik.'u'lya.
•u- = directive prefix, -liam = verb stem to say to with personal object. This verb form is logically
passive.
260 VI American Indian Languages 2
677
43b Donionstrative pronoun, showing location near 2d person, composed of simple form of independent
3d personal pronoun + demonstrative element -z- (of. also ordinary forms of independent 3d personal
pronoun ya'x-ka and similarly for other genders) + demonstrative stem -dau (= -da + -u), for vifhich see
note 54. Syntactically ya'zdau, here used suijstantively, agrees in gender with yago'mEnil, to which it
refers. There is no expressed predicate in this sentence, yago'menil (it is) his heart being so used.
«<: Particle verb, to which following verbs gatci'ux and gali'iox, both from verb stem .-i to do, serve
as auxiliaries. Lqfdp doubtless has onomatopoetic force.
«d See note 64.
<*» As in note 28. cut it-m.\de-itself = it became cut.
<< ga- as in note 4. -t- = 3d per. pi. intr. subj., referring to akni'm, idE'lxam, and isk.'u'lya as combined
plural subject, -k- = regular replacement of directive -u- whenever intr. subj. -t- would theoretically be
expected to stand before it. -xEni- (or -luni-) = verb stem to float, drift, -j/m- = distributive suffix
EACH separately (gatkxEni'tck would mean they floated up in one body), -tck = local verb suffix
UP to surface, up from position of rest (cf. also gal-i-x-lE'-tck he moved himself up from sitting
POSITION, he arose [Wishram Texts 4.6]; gal-i'-kta-tck he rose (sticking his head) out of water
[op. cit., 10.5]); combined with -6a out of interior, -tck appears as -ptck from water out to land
{gatkxEni'yuptck they each floated on to land; for change of -ba to-pcf. galags'lba with lagE'lpt
[Wishram Texts 94.7]). This -tck should be distinguished from -tck of cossative significance, whose
function it is to deprive verbs that are continuative or frequentative in form of their continuative
force (e. g., yuwi'lal he is dancing, gayuwi'lalEmtck he was dancing (but is no longer doing so).
« Adverbial in force. Logically sa'g"- (rhetorically lengthened to sd'qu to emphasize idea of totality)
often seems to be used attributively with nouns (translated as all), but grammatically it is best con-
sidered as adverbial, even when there is no expressed predicate.
« Composed of demonstrative stems kw6- (see note 3) and dau- (see note 54). Its original significance
was evidently that (which precedes) and this (which follows).
<7 gal-i- as in note 32. -kim = verb stem -to say (without personal object; cf. note 58).
« Adverb of modal significance, serving to give doubtful coloring to verb.
« Adverb of potential and conditional significance; in formal conditions introduced by cma'nii if, it
often has contrary-to-fact implication. This use of modal particles in lieu of verb modes is characteristic
of Chinookan.
60 Evidently contains interrogative stem qa- what, seen also in qxa'damt (note 14). -ma can not be
explained. This word has been found only in such passages as here, and is very likely felt to be archaic.
Iga pu qa'ma occurs as stereotyped myth-phrase in transfonner incidents (cf. Wishram Texts 6.13,
38.6, for similar passages).
51 Forms in -aima alone may be formed from simplest forms of personal pronoims (subject intr.
incorporated); e. g., na'ima i alone, ma'ima you alone, ya'ima he alone. It is doubtful, however,
whether these forms should be considered as intransitive verbs from verb stem -aima. Since personal
plurals in -dike (e. g., la'imadikc they alone) occur, it seems preferable to consider them as formed by
suffixed -ma alone? (cf. qa'ma note 50) from independent pronoun stems in -ai- (as in na'ika, note 57,
and na'it.'a, note 21); this -ai- is in these forms found also in 3d persons (e. g., la'ima it alone, as con-
trasted with la'xka and la'xt.'a). Chinook nd'mka i alone, analyzed by Boas as intr. subj. pronoun +
verb stem -nmka, Is probably best explained as simple independent pronoun in -a- {na, ma, and corre-
sponding fonns for other persons occur not rarely in Wishram) + -m{a) + -ka just, only (cf. lu'nka just
three).
'"- Adverb composed of.relative particle qxi- (cf. qzi as relative pronoim in Wishram Texts, 188.1) and
demonstrative stem dau- this (cf. note 54). qii'dau thus means literally as, like this.
"a-= tense prefix of future time, -m- = 2d per. sing. tr. subj. -d- = 3d per. pi. tr. ohj., referring
to idE'lxam. -u- = directive prefix, -x- = verb stem to do (to), -w- = inorganic consonant induced
by -u- preceding k- sound, -a = future suffix.
M Demonstrative pronoun, showing location near 1st person, composed of demonstrative stem daw
(= da-, as in da'6a here + -u-, see note 19) and simple form of 3d per. independent pronoun in -a (masc.
ya, fem. a, neut. la, du. cda, pi. da). Forms without -u- (e. g., da'ya) occur, though much less frequently;
deictic -i may be added without material change in meaning (e. g., da'uyax or da'yax). -dau also occurs
as second element in demonstrative pronouns showing location near 2d person (e.g., ya'idau that masc.,
note 43b). da'uya is here masculine because in agreement with masc. noun wj'j/u'a. Chinook seems to
Seven: Penutian Languages 261
preserve da- only in isolated adverbs like <a'fc£ then (= da'ka just this or that [cf. Wishram da'uka
JUST so]).
is wi- = masc. noun prefix, with w- because noun stem is monosyllabic, -gwa = noun stem day.
da'uya wi'gwa this day is regularly used as stereotyped phrase for to-day; dau' aga'lax this sun is
also so used.
56 Analysis given in note 3. Here kwo'pt, with well-marked stress accent, preserves its literal meaning
of that far, thus much, aga kwo'pt being regularly used, outside of narrative, to mean enough. Chi-
nook kapc't enough is doubtless related, but fco- can not be directly equated with kw6-, which corre-
sponds rather to Chinook go (see note 3).
5' Ordinary form of independent personal pronoun, composed of stems in -ai- (for 1st and 2d persons) or
-a-x- (for 3d persons) and suffi.xed particle -ka just, only, found also suffixed to numerals, na'ika is here
grammatically unnecessary, but is used to emphasize subject of following verb form.
58 = iyamu'lxam. i- = tense prefix of immediate past time, -yam- = combination of 1st per. sing. subj.
and 2d per. sing. obj. -u- = directive prefix, -ham = verb stem to say to with expressed personal object.
59 Temporal adverb referring to action just past or about to occur, either just now, recently, or soon.
Seems to be Klickitat loan-word.
6" da'tij/a as in note 54; masc. because in agreement with masc. noun will, -ba = local noun suffix in
regularly suffixed to demonstrative pronoim preceding noun instead of to noun itself.
81 a- as in note 53. -t- = 3d per. pi. Intr. subj., referring to idE'lxam. -ga- = element regularly Intro-
duced after 3d per. pi. Intr. -t- before -d-i- to come and, before verb stems beginning with k- sounds,
after 3d per. pi. intr. -u- (cf. note 62). -d-i- to come consists of -d- = directive prefix hither, toward
SPEAKER, correlative to directive -«-, and -i- = verb stem to go. -mam- = form of -am- (see notes 8 and
32) used after vowels, -a as In note 53.
62 al- = tense prefix of future time employed before vowels (al- and a- used analogously to gal- and ga-).
-u- = 3d per. pi. Intr. subj.. used, Instead of -t-, before verb stems beginning with k- sounds (as here
-gim-). -gwa- = -ga- as In note 61, -w- being Inorganic, due to Infiuence of -u- preceding k- sound (cf.
note 53). -gim- = verb stem to say; -kim (as in note 47) is used when accent immediately precedes, -gim-
when suffix (here -a) is added and accent Is pushed forward, -a as In note 53. In Chinook -ugwa- appears
as -ogo- (gwa regularly becomes go); alugwagi'ma is paralleled in Chinook by ogogoe'ma.
63 Particle verb to use supernatural power, transform, to which following gatci'ux serves as aux-
iliary. It is one of those very few Wishram words in which glottal catch Is found (other words are -tci^
OR, H'c^ic BLUEJAY, daWa'x perhaps).
6* ga- as In note 4. -<c-= 3dper.masc.tr. subj., referring to jsfc/u'^ya. -i- = 3dper.masc.tr. obj., referring
to UdE'xyan. Observe that subject noun regularly precedes object noun, their order being thus analogous
to that of Incorporated pronouns with which they stand In apposition, -u- = directive prefix, -x = verb
stem to do (TO).
* = p.'a'la. Particle verb, with which following amxu'xwa is used as auxiliary. pfa'V amiu'xwa QUiET
You-wiLL-BECOME (I. e., you will stop, desist).
6«a- as In note 53. -m- = 2d per. sing. obj. with following reflexive element (see -i- in notes 9 and 28).
-z- as in note 28. -u-x-w-a as in note 53.
262 ^f American Indian Languages 2
Editorial Note
Excerpts originally published in Franz Boas, "Chinook," in Handbook of
American Indian Languages, Part 1 (Bureau of American Ethnology, Bulletin
40), 578-579, 625-627, 638-645, 650-654, 673-677. Washington, D.C.: Smith-
sonian Institution (1911).
Sapir s extensive grammatical files on Wishram were later used by Walter
Dyk, a student of Sapir s at Yale, in preparing a Wishram grammar for his 1933
doctoral dissertation. This grammar, which remains unpublished, also drew on
new material collected both in the field and at Yale, where Philip Kahclamat,
a young Wishram speaker, was in residence for a few months in 1932. Sapir's
and Dyk's manuscript materials on Wishram are now in the Boas Collection,
Library of the American Philosophical Society.
Sapir s description of Wishram diminutive-augmentative consonant "grada-
tion," here and in the previous paper, is the first detailed treatment of this
common feature of western North American languages. Sapir's exhaustive
description of the elaborate phonetic symbolism of Nootka (1915a) is also
reprinted in this volume. For a thorough survey of the literature on American
Indian consonant symbolism see Nichols (1971). See also Haas (1970).
A CHARACTERISTIC PENUTIAN FORM OF STEM
In spite of its somewhat top-heavy intro-
duction Dr. Leo J. Frachtenberg's " Com-
parative Studies in Takehnan, Kalapuyan and
Chinookan Lexicography " (International
Journal of American Linguistics, vol. I, n° 2,
pp. 175-1S2) is a decidedly valuable and wel-
come contribution to our gradually progressing
knowledge of the relations between the seve-
ral linguistic " stocks " of the Pacific Coast. He
does not go so far as to deduce genetic rela-
tionship between Takelma, Kalapuya, and
Chinook on the basis of the lexical evidence
presented in his paper, but he does__" make
bold to predict that additional data will be
produced in the near future " tending to
confirm such a conclusion. Elsewhere in the
paper he states : " I have collected a mass of
material establishing a probable conmion ori-
gin for the Kusan, Siuslawan, Yakonan, and
(perhaps) Kalapuyan languages which will be
presented in the near future. " Though he
chides Dixon and Kroeber for their hasty
annouqcement of the Hokan and Penutian
groupings, he remarks, somewhat unexpec-
tedly : " The absence of conclusive evidence
concerning Penutian and Hokan is the more
unfortunate, as there exist strong reasons to
believe that the Takelman, Kusan, Siuslawan,
Yakonan, Kalapuyan, and (perhaps) Chinookan
languages spoken in Oregon may be proven to
be Penutian sister-tongues.
All this is very interesting to me, as it
chimes with conclusions or hypotheses I had
arrived at independently. On the appearance
of Frachtenberg's Coos grammar ' it soon became
clear to me that the morphological and lexical
resemblances between Takelma and Coos were
too numerous and fundamental to he explain-
ed away by accident or plausibly accounted
for by borrowing. This in spite of the very
great differences of phonetics and structure
that separate the two languages. The appea-
rance of Frachtenberg's" Siuslaw material ^ has
only tended to confirm this impression, further,
to make it perfectly obvious that Coos and
Siuslaw, as Frachtenberg announces, are
divergent representatives of a single linguistic
stock. Meanwhile comparison of Takelma, Coos,
and Siuslaw with Dixon and Kroeber's Penutian
group of California (Costanoan, Miwok,
Yokuts, Wintun, and Maidu) disclosed an
astonishing number of both lexical and mor-
phological correspondences, correspondences
which were first dimly brought to my con-
sciousness years ago by certain morphological
resemblances between Takelma and Yokuts,
later and more vividly by the decidedly Penu-
tian " feel " of Coos grammar. In spite of our
slight knowledge of most of the Californian
languages involved, I succeeded, in getting to-
gether what I believe to be a quite respectable
mass of evidence tending to unite the southern
1. Coos, an IlUisirative Skctcli (Handbook of Ame-
rican Indian Languages, BBAE 40, part 2, 1914,
pp. 297-429)-
2. Lower Umpqua Texts (CU, 4, 1914) , Siuslawan
(Lower Umpqua), an Illustrative Sketch (Handbook of
American Indian Languages, BBAE 40, part 2, 191 7,
pp. 431-629).
264
VI American Indian Languages 2
59
languages with those of Oregon into a large
and highly differentiated " stock ".
Unfamiliarity with Alsea (Yakonan) and
Kalapuya made it impossible for me to follow
the chain of evidence geographically. Neverthe-
less, Chinookan points of contact soon mani-
fested themselves too persistently to be brushed
aside. After hesitating for a long while to take
up seriously the possibility of affiliating
Chinook, one of the most isolated and mor-
phologically specialized languages in America,
with the Penutian languages of Oregon, I now
find myself forced by the evidence to admit
such an affiliation as not only possible bat
decidedly probable. In view of the clear points
of lexical contact and of the phonetic shifts
that Frachtenberg has established, even if only
in a preliminary way, between Takelma and
Kalapuya, his further hypothesis of a funda-
mental connection between Kalapuya and
Chinook was, forme, to be looked for a priori.
I believe it only fair to add that the manuscript
evidence that I possess of the relation between
Chinook and various languages to the south is
much stronger than the comparatively scanty
lexical data presented by Frachtenberg '.
1. Even of the nineteen cognates given by Frachten-
berg no less than nine or ten, it seems to me, are almost
certainly borrowings. This is particularly obvious in the
case of Kalapuya U skan cup and u'lxayu seal. Both of
these nouns have taken over bodily the Lower Chinook
feminine prefix u- of u'-cgan cup and u'-lxaiu seal.
Moreover, the Kalapuya, as an inland people, could
hardly be expected to possess a native term for the seal.
As for u'-cgan, that is merely a secondary feminine for-
mation from Chinook i'-cgan (masc.) CEDAR. This
alone would prove Kalapuya u'skan cup to be a borrow-
ing from Chinook.
The fifty-three Takelma-Kalapuya correspondences are
of a very different sort and bear all the ear-marks ot
genuineness. I take this opportunity of correcting a few
errors. For Takelma Ja*'- to finish (p. i8o)read t\ap'g-;
for t'pdW SNAIL (pp. i8o, i8i) Ttzd t'billt' ; {or dip'
CAMASS (p. i8o) read dip' ; for t]ewex flea (p. i8i)read
tUwix. Frachtenberg's statement that « Takelma ts
remains unchanged >i in Kalapuya (p. i8o) is incorrect, as
However, further evidence serving to hnk
Chinook both with Kalapuya and with Coos-
Siuslaw-Alsea will no doubt have been discov-
ered by Frachtenberg by the time this paper is
published.
The greatest surprise was still awaiting me.
Tsimshian occupies a peculiarly isolated posi-
tion. In its morphological aspects it offers
hardly any specific points of resemblance to
the neighboring Nadene languages (Haida,
Tlingit, Athabaskan). With the Wakashan-
Salish-Chimakuan group to the south it shares,
e. g., initial reduplication to express distribu-
tive and plural ideas and a series of classifying
suffixes appended to numeral stems. In its
deeper morphological features, however,
Tsimshian stands quite apart from these lan-
guages. Lexically too, aside from a number of
more or less patent loan-words, there seem to
be no stems that Tsimshian shares with any of
the neighboring languages. A tentative com-
parison with the Penutian (or, if one prefers,
hypothetically Penutian) languages of Oregon
revealed a considerable number of correspon-
dences both in the lexical material and in some
of the more intimate and fundamental features
of the morphology. Should it be possible to
demonstrate (and I am fairly sanguine that it
can be demonstrated) thai Tsimshian is a de-
tached northern offshoot of Penutian, we would
be compelled to face a most interesting fact in
linguistic differentiation and in the distribu-
tion of American tribes.
The data for the various assertions I have
made in this paper I expect to present /;/
extenso in the future. The fundamental type of
Penutian language seems to be a predomi-
nantly inflective one, as exemplified say in
Miwok, Yokuts and, in more specialized form,
in Takelma. The tremendous morphological
Takelma possesses no ts ; al-ts-M red (pp. i8o, i8i)
should be corrected to al-ts-\il.
Seven: Penutian Languages
265
60
divergences exhibited by the various languages
here grouped together are due, it v»'ould seem,
to two main types of change : the gradual
breaking down of inflective forms and the
consequent growth of more analytic forms (e.
g., in Costanoan and, in a less extreme way, in
Coos) ; and the evolution of special synthetic
developments, as in Chinook, often, no doubt,
under the influence of alien languages, as in
Maidu and Tsimshian. It is, of course, entirely
premature to group the Penutian languages
genetically, but the following provisional
scheme may be of some s'ight value as an aid
to convenient visualization.
A. Californian group
[ Miwok
\ Costanoan
2. Yokuts
3. Wintun
4. Maidu
B. Oregon group
1 . Takelma
2. Oregon Coast group
Coos
Siuslaw
Yakonan
3. Kalapuya
C. Chi nook an
D. Tsimshian
It may well be that Frachtenberg's as yet
unpublished data on Kalapuya and Yakonan
will force a readjustment of the languages
under B ; as for group A, possibly Wintun and
Maidu form a closer unit, analogous to that
undoubtedly comprised by Miwok and Costa-
noan '.
I. See A. L. Kroeber, The Chumash and Costanoan
Languages, UCP 9: 237-271, particularly 259-263(1910);
also The Languages of the Coast of California north of
San Francisco (/oc. cit., 191 1, pp. 273-455), pp. 306-508.
One of the most striking of the more inti-
mate structural peculiarities of Takelma is the
presence of a large number of stems of type
cv, cv(c2)-, i.e. of stems with repeated vowel.
In the verb, this method of stem formation has,
to a large extent, taken over a grammatical
significance, in some cases characterizing the
aorist stem, in others, less numerous, the verb-
stem proper. It is, however, frequently illus-
trated in the noun^ adjective, and adverb as
well ^. Examples are :
yana ' acorn
viahai large
waga- TO BRING (non-aorist)
baxam- to come (aorist)
herein- KIDNEY
o
yeuue^ perhaps
sehe- TO roast (non-aorist)
lehei- to drift dead to land (aorist)
yiwin speech
-xiivi- TO ROT (non-aorist)
ginig- TO GO SOMEWHERE (aotist)
nwt'op' SEED-BEATER
Olom FORMERLY
odo- TO HUNT FOR (nou-aorist)
Jop'od- TO RAIN (aorist)
Silli'V'k' CRICKET
ilyftfs- TO LAUGH (aorist)
Turning to Coos, we find that this same type
of stem formation, though apparently not pro-
vided with formal significance, is very common.
It is so very characteristic of the noun, indeed,
that Frachtenberg's failure to mention it in his
Coos grammar must be set down as a curious
oversight. By actual count I find in his Coos
2. For details see The Takelma Language of South-
western Oregon (BBAE 40, part 2, 191 2, pp. 1-296),
§§40, 86, 109.
3. lam omitting all stress and pitch accents in this
paper, also normalizing all orthographies to correspond
to the' recommendations of the Phonetic Committee of
the American Anthropological Association.
266
VI American Indian Languages 2
6i
vocabulary ' no less than 140 examples of such
nouns (and adjectives), 35 of verb stems, and
9 of adverbs. These figures show conclusively
that we are dealing with a real formative
feature of the language, not with an arbitra-
rily selected congeries of words. A few of the
examples are :
yabas maggots
yalaq gray (of hair)
fiianat crowd
Cola't'iS, CRACK
(jaiva CHEEK
maha- to scare
ak'auak- to stick out
kayal- to decay
qala"- to hurt
//rt/a'^'- to bark
hakwai as if
ehe GONE
luebel stomach
tsehes alive
^el'eths corner
tlpene wings
en'ek- to stick out
eqe to die (plur.)
xeyetl to twist to one side
lek shouts
luekx- TO stoop
kweek to live
ehenlc far away
Cccd MAT
kzvims feather
W'.t'.n blood
f'./"u: razor-clam
hkiuKt feathers
i'tud- TO RECOGNIZE
h'.tl- TO CARRY
hmiS- TO HOLD BACK INFORMATION
^l//-^- TO OVERTAKE
IV'.xkm- TO STEAL
I. See pp. 191-215 of Coos Texts, CU i, 1913.
ydki' LONG AGO
tfyW RAINBOW
nwytis anus
tso'no' BOTH ways
It does not seem plausible that these and
numerous analogous examples are merely
secondary phonetic developments due to pro-
gressive or regressive vocalic harmony ^. To at
least some extent Coos possesses forms with
repeated vowel that are of grammatical sig-
nificance, notably certain irregular plurals of
nouns and adjectives, e.g. :
tSi'y-UX^' SMALL
knes HUNCHBACK
plur. tsi'yt'-ne
kene'yes-e
The very irregularity of such forms points to
their archaic character. They are obviously
analogous to such Takelma adjectival plurals
as :
-t'gem BLACK
-t'gcy-ap-x ROUND
ba'Hs i,ONG
plur. -fgeme-t'it'
-t'geye-p'-it'
ba-Has-iV
Analogous forms seem to be found in Lower
Umpqua (Siuslaw), though apparently not so
characteristically. I do not feel as clear about
them as yet as about the corresponding Coos
formations. Examples are :
yafqa-a"' hole
hayci'tit WIDOW
halca-t LONG
pa'iiqa medicine-man
taha-mk quiver
ca'ya penis
t'saxan stomach
k\iyak EAGLE
qa'xal fish-net
XlVil'ka HEAD
iazvat- to gamble
2. Sec Frachtenberg's Coos, BBAE 40, 2 : § 7.
Seven: Penutian Languages
267
62
amha- to be willing
atas ONLY
ivi'imtCKKT
prs<.p FISH-HAWK
misi'-a" ELDER SISTER
t\hmS CHILD
CriUVSt SUNSET
tSimd'z' MUSKRAT
tsi'kwi LEGGINGS
hmilta MATCHES
pmd-t SHARP
aiorhin- to lose
zui'tsm- TO CAMP
mU'l'U BREECH -CLOTH BELT
tlxmu'hv BOW-STRING
Many more forms of this type, though not as
many as in Coos, could be cited, but, on the
whole, one gets the feeling that it had ceased
to be a live and productive method of stem
formation. An apparent tendency to loss of
vowels and consequent formation of consonan-
tal clusters would seem to have counteracted
the persistence of the type. Comparison with
Coos seems to confirm this, e.g. :
L. Umpqua
(Siuslaw)
Coos
nia''q"li crow
tipa'^'n, ipz't'v wings
ma'qati
t'ipenc
In both Coos and Siuslaw, it should be noted,
forms with repeated vowels are sometimes
related, as in Takelma, to monovocalic forms,
e.cr. :
Coos manat crowd
t'ipenc wings, feathers
xalaw-is heat
L.U. valqa-a" ■ hole
piurcja medicine-man
nianta-^- to keep compan\-
tlpe wings
xahu-is hot
yafq-' i-lq- to dig
p:iiq- to dance a mtdicinc
dance
Whether the type of stem with repeated
vowel is found also in Alsea (Yakonan) must
wait for an answer until Frachtenberg's mate-
rial is available. This applies also to Kalapuya.
The Kalapuya forms that he gives in his com-
parative study number several of the type,
e.g. :
thanaq fly
qalavi silverside salmon
qauwan Chinook salmon
palam drunk (Chinook Jargon)
qoloq swan (Chinook)
diigulhu' owl
niu'Iukwa cow
but, suggestive as such examples are, not much
can be done with them when torn apart from
their grammatical and statistical context. In
Chinook and Tsimshian the type cannot be
demonstrated to exist. It is evident from inter-
nal evidence and from its numerous and often
complex clusters of consonants that vowels
have dropped out very frequently in Chinook,
so that an originally present type of stem for-
mation (c)vc,v(c2)- would be expected, under
the destructive influence of phonetic processes,
to have disappeared as such. Possibly, however,
compar,ative evidence and closer study of
Chinookan material may eventually reveal the
former existence of the type. Such forms as
-kala MAN (cf. also Wishram -gi-kal husband),
-ka-holf EEL < *-ka'kwale' (cf. Wishram
-gakwal\ -saJa fresh-water clam, /av.'//; six (cf.
Wishram tx.^ni), and ca'y.m grizzly bear (cf.
plural caya'm-ukc) are, for the present, sugges-
tive but too isolated to constitute convincing
evidence.
As soon as we turn to the Penutian languages
of California, we are at once struck by the
large number of nouns, adjectives, and verb
stems thatareformed on the pattern(c)vc,v(c2)-.
Examples from Wintun ' are :
I. Taken from S. A. Barrett's vocabulary, pp. 81-87
of The Ethno-oeography of the Povio and jicighhoring
Indians (University of California Publications in Ameri-
268
VI American Indian Languages 2
63
apti- mother's brother
tahal tongue
daka belly
anak knee
kama bed
taral mush-paddle
kada deer-snare
aka bitter
k'afta to sleep
hara to go (D) '
elct infant
k'me rib
keicel house
wene medicine
ii'nik nose (Southerly dialect)
iniil milk
tkvi-l rattlesnake
hi'ii' shell-beads
t'iki sand
luini to see (D)
ko'so'I lungs
poto intestines
mrko' arrow
no'Io' smoke
do'ko' knife
tco'nr fish-net
cotok pestle
SH'iio' nose (Xortherly dialect)
pi fill' heart
irdii'i rabbit-skin robe
tii-nfl magnesite beads
tii'nii'k cradle
swrii't burden net
yu'rw rain
ku'pirm fingers
ti'tcw- younger sister
lu'iiiw dead
iinrhii' to sin"
can Archaeology and Ethnology, voL 6, 1908, pp. i-
332).
I. Forms marked (D) are quoted from R. B. Dixon's
Outlines of Wintun Grammar (Putnam Anniversary
Volume 1909, pp. 461-476).
It is evident at a glance that the type is very
much in evidence in Wimun. By actual count
I find that considerably more than a third of
the nouns and adjectives listed by Barrett for the
« Southerly » dialect belong to it.
Equally evident is the type in Maidu. From
Dixon's grammar ^ I may quote :
luaka' meat
tsaka- pitch
tsa'zva teeth
paka sinew
ma' tun hip
yaha' good
ivasa' bad
-tala- to crush
heme gopher
sede blood
benek tomorrow
/;/;// eye
hini net
ki'ivi' back
;///;//■ nipples
yiini arm
pili' dung
si mi month
tsibi' nails
-piuili- to roll
-tibil- to wind around
U'dhol heavy
koJo- to rotate
orw- head
ko'zuo' armpit
yoso' field-mouse
kono' baby
nolo' pack-basket
bono' ear
poko' sun
koyo' valley
WO' no'- to die
2. Maidu, an illustrative Sketch (Handbook of Ameri-
can Indian Languages, BBAE : 40, part i, 19 10, pp. 679-
734).
Seven: Penutian Languages
269
64
buhl tail
butu fur
suhu' smoke
ku'lw evening
viu'su' face
It is clear that the type is as much alive in
Maidu as in Wintun, and, as in Wintun, it
seems, further, to be rather more characteristic
of the noun than of the verb. This may be
only apparent, however, as our knowledge of
the verb morphology of these languages is far
from complete.
Turning to Yokuts, we find the type of stem
formation still very much in evidence. I select
from Kroeber's lists '. Examples of nouns
are :
V'apnd shoulder
ma ,ad tears
kabad wings
tabak deer-rick
capan coal
wadak head-net
k'twz't back
tthzty head louse
tczhvr, fog
bidik faeces
dimik prairie falcon
tyipin sky
idik water
cikid arrow
ip'i7i ground acorns
P'ir,ik nose
mi'kJc throat
lir,it earth
wdho'cit panther
iirta youth
oty^ hair
p:>tJ penis
I. See The Yokuts Lau^uas^e of South Central Califoinin,
UCP 2 : 165-577(1907). For this paper I use only
Yaudanchi forms (see pp. 240-254 for vocabularv).
comot lungs
■rpho grizzly bear
coyod antelope
topo■r^ buckeye
tcoxJtc soaproot
so'hn tobacco
cohd hole
odJt ball
luocok belt
putuc acorn
The list might be greatly extended. A large
number of verbs also belong to this type, e.g. :
awaty to dislike
tawac to be thirsty
paxat to mourn for
tcadax to turn
dixid to make a basket
pinity to ask
pitid to tell
pitciiu to catch
icitid to pull hair
luinis to be ready
xityiw n be angry
tdxitc ■ be sick
i'dik to sing
iv:)d:> to dodge
od:^y to be on
ciitnx to skin
dukud to bury
dumuk to sweat
k'uyuk to scratch
nuhuk to kneel
hupnc to select
Before certain suffixes these stems, like other
disyllabic stems, lose their second vowel, e.g. :
diikttil (imperative) )
dxkud-ji (past) )
( duhd-un (present-future)
( duhd-ut (passive)
According to Kroeber this stem alternation is
purely phonetic, not morphologic, in charac-
ter, so that direct comparison with the stri-
270
VI American Indian Languages 2
65
kingly similar stem alternations of Takelma
(^e-g. zonst p'eleg- TO go to war : non-aorist
p'elg-) is not feasible. However, this matter
certainly needs further investigation, particu-
larly as not all of Kroeber's forms seem to
conform to his phonetic rule (loss of second
vowel before a suffix beginning with a vowel).
It may well be, on the other hand, that what
was in remote origin a purely phonetic alter-
nation in Takelma gradually assumed a mor-
phological function. There is nothing to in-
dicate this at present, however, for Takelma.
In Miwok (Moquelumnan) the presence of
the type (c)vc,v(c),- is as easy to demonstrate
as in Wintun, Maidu, and Yokuts. For examples
of nouns and adjectives I quote from Barrett's
vocabularies of Coast Miwok ' :
nawa old man
ata elder brother
saka cradle
panak red-headed woodpecker
patca poison
luala Indian potatoe
heleke neck
ewe milk
wene medicine
killi antler
ki'ti'lak butterfly
pi'tci' nails
ko'no' bow
loko tule
olok ocean
posol lungs
mo'M condor
koto grasshopper
tso'to'i short
pu'tw infant
pu'lu'k belly
ku'lu'm bone
pu'tcw wild onion
I . See The Etbno-geography of the Porno and neighboring
Indians, UCP 6 : 68-80.
u'kw hand
tirnu'k cradle
su'ku'i sand
mu'lu'-ta black
n'lu'-ta red
hu'hf stinking
From Kroeber's material ^ the following verb
stems are cited :
yar^a to sleep
kata to shut
ete to see
etepo to lie on one's stomach
kelpe to swallow
nete to count
nipito to sit with folded leg
hili to pinch
oiuo to eat
koyok to see
toloye to hear
lutsu to ascend
uhii to drink
iinii to come
ynnu to kill
kusH to sit with stretched leg
nuxii to undress
uku to enter
iitcii to stay
yiiku to swing
Costanoan, finally, is perfectly analogous in
this respect, as in so many others, to Miwok.
Examples of nouns and adjectives are again
quoted from Kroeber > :
ama person
wara body
patcan blood
2. See his Miwok sketch (pp. 278-3 I9)in The Languages
of the Coast of California north of San Francisco, UCP 9 :
273-435 (191 1). The Northern Sierra dialect is quoted.
3. See vocabularies in The Chiimash and Costanoan
Languages (UCP 9: 237-271 [1910]).
Seven: Penutian Languages
271
6G
lawan bow
wasar wind
lanai neck
hatac foot
tcara sky
wahan snow
hesexem old woman
£)'Ci beard
/c^jyw jackrabbit
t^d;'(?u rabbit
wetel large
j/;n' child
kinir fish
^/7m belly
wznzf heart
liti bow
tt/fx/ fish
worox hair
/co/^o boy
o/co ear
koro foot, leg
rono^ bow
moto cloud
yokovi ice
/o/oi knee
co/o^ moccasin
to/o fire
locko-, laskci- white
^o/^o- black
xntii belly
kunuc pipe
«;/n<x wolf
pMJM/ small
It is worth noting that disyllabic stems ot this
type are in some dialects, particularly in that
of Monterey, sometimes reduced to mono-
syllables by the loss of the second vowel, e.g. :
Monterey
xurks neck
tols knee
tuis jackrabbit
S. Juan Bautista
xorkos
tolos
tceycs
Some examples, furthermore, of vocalic repeti
tion may be only the secondary result of as-
similation, e.g. :
S. Clara oroc bear ores (four other dialects)
S. Cruz lasa tongue lase- (four other dialects)
tcipi knife S. J. B. Pipe
S. Francisco tipa
For examples of Costanoan verb stems of type
(c)wc'y{c2y I quote from De laCuesta's Mutsun
vocabulary ' ;
atiia to eat
ara to give
axa to comb
ata to examine
icara to cut
xaiva to call
saya to shout
ene to write
ele to raise
luexe to shield
nietc to hide
sepe to cut hair
tere to cut hair
pele to stick
ipili to lie down
witi to fall
wixi to fish
niini to strike
xiri to make dried meat
olo to become blind
yoko to become ashes
vioko to be born
roto to drown
xopo to give water
tcorok to become sad
iipu to pay
utu to guard
Inxun to stick in mud
I. See J. A. Mason, The Mutsun Dialect of Costanoan
based on the Vocabulury of De la Cuesta (in UCP 1 1 :
399-472 [1916]).
272
VI American Indian Languages 2
67
rusu to spit
siimti to decay
ttipun to finish
tculti to jump
This list might be vastly extended. That the
second vowel is felt as an integral part of the
stem is shown by its persistence after infixed
frequentative -s-, e.g. :
ele-pu to go
epe to pass
else -pti many go
epse many pass
The final impression left on one's mind is
that the stem type(c)vc,v(c2)-is quite unmista-
kably, not to say exuberantly, represented in
all the Penutian languages of California. It is,
further, just as unmistakably in evidence in
Takelma and Coos. Further north it appears
to die out, in all probibility owing to the ope-
ration of destructive phonetic laws. Its persis-
tence to a minor extent in Lower Umpqua
(Siuslaw), however, is highly probable, while
further data may eventually indicate its
presence also in Kalapuya. In other words, we
are not far from having demonstrated its charac-
teristic presence in the Penutian languages as
a whole.
Whether any conceptual significance origi-
nally attached to this type of stem formation
it is quite impossible to say at present ; quite
probably it will always remain impossible. It
will be interesting to determine later whether
the Takelma alternation betwen the aorist verb
stem with repeated vowel and non-aorist verb
stem with single vowel is an archaic Penutian
feature or a specialized development peculiar
to Takelma. It is quite likely, indeed, to prove
related to the alternation in Coos and Siuslaw
between the short form and the « amplified »
form of stem (generally with a-vowel), e.g. :
Coos tkwi'tl- to follow tliwi'yali
tC'.Jiti- to reach tanatl
Siuslaw tku'nt- to close tkwani-
hi'ts- to put on h'yats-
i'iq- to dig yaiq-
hau'- to be ready haiua-
wi'in'- to agree lui'hca-
The primary function of the amplified stem is
the indication, apparently, of durative activity.
Assuming the « amplified » stem of these lan-
guages to be related to the aorist (present-past)
stem of Takelma, it remains to be determined
whether their amplifying rt-vowel is genera-
lized from stems with radical a {e.g. Siuslaw
hail'-) or, on the contrary, the repeating vowel
of Takelma is a secondary assimilation from an
older -a- (e.g. p'eleg from *p'elag-'). More pene-
trating study of the Californian Penutian lan-
guages mav help to solve this and other simi-
lar problems.
Seven: Penutian Languages 273
Editorial Note
Originally published in International Journal of American Linguistics 2,
58-67 (1921).
Sapir's suggestion of the "characteristic presence in the Penutian languages
as a whole" of disyllabic stems with a repeated vowel should be compared to his
identification of a similarly diagnostic stem-type (with deletable initial vowel)
for Hokan (see 1917e, 1920b, 1921o, and 1925b).
There is evidence that Sapir had second thoughts about Penutian stem shapes
after the completion of this paper in early 1919. In November of that year, fol-
lowing the publication of Dixon and Kroeber s full lexical evidence for Califor-
nia Penutian (Dixon and Kroeber 1919: 56-61), Sapir wrote to Kroeber, "I feel
the typical Penutian stem is long, often 3 or 4 syllables in original form," sug-
gesting such reconstructions as *ilapita 'fish' and *yikati 'one' (Golla 1984: 315).
A characteristic stem-type is not included by Sapir as a diagnostic trait of Penu-
tian languages in his final classificatory statement (1929a: 140). Later work in
comparative Penutian morphology has tended to emphasize the centrality of
stem vowel ablaut, frequently coupled with suffixation as a "morphological pro-
cess," rather than focusing on stem shape per ^e (Hymes 1957; Silverstein 1979:
663-667).
A CHINOOKAN PHONETIC LAW
It is the purpose of this paper to show how
the operation of a phonetic law, hitherto unno-
ticed, brought about a number of irregularities
in the use of pronominal elements in Chinoo-
kan. Certain incidental inferences on more
fundamental points of Chinookan linguistic
history also suggest themselves. These will be
briefly referred to at the end of the paper.
If we examine the Chinookan system of
transitive and intransitive pronominal prefixes
of the verb and corresponding possessive pre-
fixes of the noun ', we shall note three appa-
rently unrelated irregular features which
involve an alternation of g (which may be
modified to k or kx ^) and the palatal sibilant
aff'ricate tc. These are as follows :
I. The possessive prefix for the third person
singular feminine (" her ") is -ga- when the
noun itself is feminine, neuter, dual, or plural,
i.e. is preceded by the gender-number pre-
fixes :
sing. fem.
neut.
du.
plur.
Lower Chinook Wishram '
6- (w) a-
L- ii-
C-, s- ic-, is-
t- id- (it-)
but \s-tca- when the noun itself is masculine,
1. See F. Boas, Chinook (in Handbook of American
Indian Languages, Bureau of American Hthnology, Bul-
letin 40, part I, igii.pp. 559-677), particularly pages
580, 581, and 585.
2. For consistency's sake I am preserving Boas' Chin-
ook and my own Wishram orthography without modi-
fication.
3. An Upper Chinookan dialect. I quote from my MS
data.
i.e. is preceded by the gender-number prefix :
masc. sing. e-, i- (jio)i-
Examples (Wishram dialect) are :
Absolute Possessive : " her '
wa-ska'n cup a-ga'-skan
ii-tcqiua' water il-ga'-cq
is-qxu's eyes is-ga'-xus
i't-qHi house it-ga'-qH
{-ho'-qH)
but
wi'-lxam village i-tca'-lxatn
1. The possessive prefix for the first person
singular (" my ") is -^e- (Wishram -^-, -k- ;
-.y- before /e-stops) when the noun is femin-
ine, neuter, dual, or plural, but -/^e-, -ki-
(Wishram -tc-') when the noun is masculine.
Lower Chinook (C.) and Wishram (W.)
examples are :
Absolute Possessive : " my "
C. o-pO.ike bow d-gu'-pL\ike {-gu- labra-
lized from -^e- because
of preceding 0-)
W, a-kni'm canoes a-x-hii'm
C. L-gE'-qacqac my grand-
father
W. is-qxu's eyes is-k-xu's
W. i'f-pc feet i't-k-pc
but :
C. i-tslENiEnd wooden
spoon
W. wi'-iq body
i-tcE'-ts\EmEno
i'-ic-iq
276
VI American Indian Languages 2
io6
3. Aside from certain secondary irregularities
in the third person dual and third person plur-
al which do not concern us here, the pronom-
inal subject of the transitive verb differs from
the pronominal subject of the intransitive verb
(and pronominal object of the transitive verb)
only in the case of the third person singular
masculine and third person singular feminine,
the difference between the two sets of forrns
being for the most part indicated by position
(the subjective pronominal prefix preceding the
objective pronominal prefix) and, in part, by
the use of a " postpronominal " particle -g-
which indicates that the preceding pronominal
element is used as the subject of a transitive
verb. For " he " (and " him ") and " she "
(and " her "), however, the following distinc-
tive forms are used :
Intransitive Transitive
3d person singular, masc. tc- he z'-he, him
fem. ^- she a- she, her
The forms will be better undestood from the
following Wishram examples :
i-tc-i'-uwaq he killed him (/- is temporal ; tc-
" he " ; -i- " him ")
i'-tpx he comes out of the house (/- " he")
i-g-i'-uwaq she killed him (-^- *' she ")
a'-lpx she comes out of the house (a- " she ")
Contrast :
i-m-i'-uwaq you (sing.) killed him (-///-
" thou ")
a-ni-tbay-a you will come out of the house
(a- is temporal)
i-tc-m-u'woq he killed you,
in which -m- " thou, thee " is used both as
transitive subject and as intransitive subject and
transitive object.
How are wc to explain these irregularities ?
The distribution of the forms in question is
such as to make it probable that we are dealing
with a phonetic factor rather than a morphol-
ogical one in the first instance. Cases i and 2
are parallel phonetically :
1. a-ga- " her " (fem. noun) : i-lca- " her"
(masc. noun)
2. a-g- " my " (fem. noun) : i-tc- " my "
(masc. noun)
and suggest at once that the masculine prefix
/- palatalized the older -ga- " her ", -^(e)-
" my ", to -tea-, -/^(e)-, perhaps via palatali-
zed k- sounds (*-§a-, *-^^-). But how account
for the forms in case 3 (fem. g- : masc. tc-\
and why should only these forms be exclusive-
ly characteristic of the subjective transitive
verbal paradigm ? If we venture to reconstruct
them in accordance with cases i and 2, we get :
*ag- " she" (transitive subject)
*itc- " he " (transitive subject)
< V
The phonetic parallelism would then be per-
fect in the three cases. If we compare the theo-
retical forms *ag- " she " and *itc- " he " with
the remaining subjective forms of the transi-
tive verb, we obtain at once a perfectly regu-
lar and intelligible set of forms. Including the
" post-pronominal " -g-, the system is as fol-
lows :
1st pers. sing: n-
exclusive dual nt-g-
exclusive plural nc-g- (also heard as
ntc-g-, with f-glide)
inclusive dual /a:-^- (simplified in Wish-
ram to t-g-^
inclusive plural Ix-g (simplified in Wish-
hram to /-f-)
2nd pers. sing, m-
dual mt-g-
plural mc-g-
3d pers. sing. masc. *i-tc- <i*i-g-
sing. fem. *a-g-
sing. neut. i-g-
dual. c-g-
plural t-g-
Seven: Penutian Languages
277
107
Compare these pronominal prefixes with
the corresponding intransitive subjects (and
transitive objects) :
1st pers. sing. n-
exclusive dual nt- (nd-)
exclusive plural nc- (nlc-^
exclusive dual tx-
exclusive plural Ix-
2nd pers. sing. tn-
dual mt- (^md-)
plural mc-
3rd pers. sing. masc. i- (y- before vowels)
sing. fern. n-
sing. neuter i-
dual c-(in certain cases c/-, erf-)
plural t- (in certain cases u-
gwa-, Lower C. o-go-)
Aside from the irregular intransitive subjec-
tive (not objective) forms in the third person
dual and plural (ct- ; ti-gwa-), whose use is
limited to certain cases, the transitive par-
adigm obviously derives from the intransitive
by the addition of a transitivizing particle -g-
to the pronominal element, except in the
first person singular («-) and second person
singular (w-), in which cases position alone
differentiates the transitive and intransitive
subjective uses. If our analysis is correct, the
actual transitive subjects for the third person
singular, masc. Ic- and fem. g-, are not true
pronominal forms in origin; the older pronomi-
nal elements, i- and a- respectively, still in
evidence in the intransitive praradigm, have
disappeared as such but have left their tVace in
the different treatment of the old transitivizing
-f-, which now appears in twofold form and with
transferred function as tc- " he " and o^- " she. ''
We may therefore reasonably infer that in
all three cases what now appears as a peculiar
morphological alternation of^^: tc is really a
survival of an old phonetic law, according to
which g (k) was palataUzed by immediately
preceding i to anterior palatal ^ Q), which in
turn shifted at an early period of Chinook his-
tory to tc. Presumably an old *ik\ shifted to
itc\, but I have no evidence of this. The law
is no longer operative as such. It had run its
course long before Chinookan split up into its
present dialects, its consequences are now of
a strictly functional character, and its operation
was probably checked at an early period by
analogical leveling. There may at one time
have been such alternations as *i-tca'la " man " :
it-ka'lukc " men " or *n-i-tcim " he said " :
n-a'-kim "she said", which were then leveled
out to the forms i-ka'la : it-ka'lukc, n-i'-kim :
n-a'-kim that we now possess. But there is
nothing to prove this and it is more probable
that the phonetic law had ceased to operate
before the welding of noun and verb stems
with pronominal and with gender-number
class prefixes. It is not at all unlikely that such
elements as /-, wi- of masculine nouns and
n-i-, n-a- of verb forms were independent elem-
ents or assemblages of elements (e.g. *w-i
" he " ; *n-i " then-he ", *n-a " then-she "),
which became attached to noun and verb stems
at a comparatively recent date. On the other
hand, we must assume that such assemblages
as *i-tca < *i-ga " hers " (masc), *i-tc3 <
*i-gd " mine " (masc), and *i-tc3 < *i-g3 " he "
(transitive subject), *a-gd *' she " (transitive
subject) and *c-g3 " they two " (transitive sub-
ject) formed firm units at a much earlier date.
Within such units the phonetic law could
operate but not outside of them.
Comparative evidence, making use of data
outside of Chinookan, may some day succeed
in confirming our phonetic law by showing
that certain cases of tc in stems go back to g
(/^) after ;■ (say -i/c, fem., "tail of quadruped "),
but at present we cannot do this. So far I
know of only one other case of tc which may
be presumed, with some plausibility, to derive
from palatalized k. This is Upper Chinook
278
VI American Indian Languages 2
1 08
-i-tc{i), which forms personal plurals of demons-
trative and personal pronouns, e.g. Kathlamet
La-i-tci " those " (indef.), ta-i-tci " those ''
(def.), based on pronominal stems La- " it ",
ta- " they " ; Wishram da'-i-tc, fa'-i-tc, a'-i-tc
" they ", based on pronominal stems da-
" they ", in- " it ", a- " she"; Wishram
da'uda-i-tc " these people ", h'xia-i-tc " yon
(indef.) people ", based on demonstrative
da'iida, la'xia '.This suffix corresponds morphol-
ogically to -kc-, -di-kc, -i-kc of other forms
(e.g. nominal plural -^^in cases like W. it-ka'-
lu-kc " men " ; W. ia'it\i-kc '' they too ", cf.
fa'itla " it too " ; da'ima-di-kc " they alone ",
cf. da'ima ; Kathlamet tata-ihc " these people ",
cf. tala-x " these "). Presumably Upper Chin-
ook -i-kc is umlautcd from older *-a-kc -C
*-a-ki-c (cf. ia'ii\a : fa'it\i-kc ; -di-kc < *da-ki-c,
personal plural in *-ki-c oi da- " they "). In
Lower Chinook plural forms in-kc,-i}ic,nnd -tike
occur plentifully with nouns, both animate and
in animate (e.g. L-q\eLxa'pu-kc " coats ", /-w'-
gala-i-kc " his firrs ", id'wux-ti-kc "his younger
brothers "), but not with demonstrative or
personal pronouns. The Lower Chinook suffix
for personal plurals in the demonstrative is -c
(t.o.x'ltac " those people", visible, qota-c "yon
people", invis.), probably the same element as
the -c- of prefixed nc- " we " (excl.) and mc-
" ye " (cf. n- " I ", m- " thou "). The best
way to explain the various plural suffixes in
Chinookan seems to be to assume an old ele-
ment *ki, preserved in palatalized form in Upper
Chinook -i-tc{i) ; a plural element -c \ and a
double plural -kc <C *-ki-c.
Another survival of the old *-ki plural may
be Lower C. L-a-tcl " mothers " •< *i-a-ki-t :
L-aa " mother ".
The data presented in this paper suggest a
number of further problems, which we can
hardly do more than touch upon.
I. The disappearance of */- and *a- in the old
I. See F. Boas, op. cit., pp. 623, 62 5, '^627.
iorms*itc(py "he"and*a^^- "she" is merely an '
early phase of a phenomenon that seems to have
been characteristic of Chinookan at all times,
the loss of short unaccented vowels. The accent
of Chinook is a strongly expiratory one, seems
to have been regulated by morphological consid-
erations (contrast W. gahi'pa " she went out of
the house " whith aluba'ya " she will go out of
the house " ; future -a, as shown also by Lower
Chinook evidence, shifts the accent forward),
and has left in its train a number of phonetic
consequences, both early and dialectic, the
chief of which are the disappearance of short
vowels (cf. vowelless stems like -iq " body ",
-tcktc " to wash"; alternations like ti- " I",
as verb prefix, with independent na'-, W. -0^-
" my " with Lower C. -^e'- ; loss of final
vowels which reappear in protected forms, e.g.
Lower C. i-saniEF " lid " <; *-sa'm^lga : L-ia-
SEniElqa-ks '' their lids " -< *-savid'lqa-), the
shortening of unaccented and the lengthening
of accented vowels (e.g. Lower C. *i-cayiin
" grizzly bear " : plur. L-caya'm-u-kc <C early
Chinook *cayam : *caya'm-)., and the weaken-
ing of consonants after unaccented vowels
(cf. above examples of W.'/): /»'; Lower C. q: 'g').
2. As regards the old unaccented */- and *a-
which we must suppose to have disappeared
before transitivizing *-gy-, we may note that
it is characteristic of all transitive forms that 1
the pronominal subject is unaccented, while in
many instances the intransitive subject receives
the accent. Lower Chinook seems to preserve |
the old accentual conditions better than the
upper 'dialects, which have undergone further
shifts of accent with resulting loss of reduced ||
vowels (e.g. Lower C. atcE'tax " he made
them " : W. gatclux^. A good ex^ample of such
accentual alternation is Lower C. le'- " it "
(intr. subj.) : L- " it " (tr. subj.) in aLE'nkatka
" it comes flying above me " : aLgigE'ltcxEm
" it sings for him ' ". This is not the place to
I. Op. cit., p. 588.
Seven: Penutian Languages
279
109
pursue the matter further, but we may at least
point out that the transitivizing *-gi- is in all
probability identical with the -gE- of the
** adverbial prefixes " gE-l- "■ for, on account
of" and -gE-tn " with, near ". It is remarkable
that the pronominal element (indirect object)
to which -gEl- and -ghn- are suffixed never
receives the accent, which either strikes the -^e-
or some syllable following it. In other words,
the treatment of pronominal element + tran-
sitivizing *-^<?-and thiU of pronominal element
-f- indirective -^e/-, -gEm- are parallel. But
note that the pronominal element (indirect
object) is frequently, even typically, accented
before the " adverbial prefix " (really postpo-
sition) -/- " to, for ", with which -gEl-'\s com-
pounded; e.g. W. inia'lut *' I gave it (masc.)
to her, " Lower C. Ld'Joc " it was to her ".
If -/- and -gEl- were strictly parallel elem-
ents, it should be possible to have such paral-
lel forms as a'-l- " to her " and a'-gEl- " for
her ", whereas we consistently have a'-/- " to
her" but a-gE'l- or a-gEl-' " for her ". This
can only mean that -/- and -gvJ- are not mor-
phologically parallel, but that -gE- is an element
which somehow displaces the pronoun and
draws the accent to itself. Its power to take the
accent away from preceding elements is further
indicated by the fact that it regularly occurs
with voiced 0-, not with voiceless k or affricative
kx (cf, remarks in i.).
What is this old element *-^^'-, which now
appears as transitivizing -^(e)-, as third person
singular masculine and feminine transitive
subject (/c[eJ- " he ", g^E)- " she '"), and as
first component of the verb prefixes -gEl-
" for" and -gEm " with " ? It seems likely that
it is an old demonstrative or deictic stem which
is either predicatively related to the preceding
pronominal element or which serves to emphas-
ize or displace the pronominal element and
to which the postposition (-/-, *-ni-) is at
tachedas an enclitic. A cluster of elements like
i-gEl- " for him " originally meant i-gE'-l-
" him -that one-to, for him ". Similarly, at a
far earlier stage, a transitive cluster like t-gE-n-
(W. t-g-n-') " they (subj.) -me (obj.) " or
*i-gE'-m- > *i-tcE-m- >> tcE-m- (W. tc-m-')
" he (subj.) -thee (obj.) " really indicated
" ihey-that (it is) -me ", " he -that (it is)
-thee ". It this is correct, the original difference
between the intransitive and transitive phrase
must have been one of sentence idiom. " He
goes " was expressed as " he goes ", but *' he
kills her " as " he that (is) (who) kills her "'.
This deictic or demonstrative *o-p'- can only
be a reduced form of post-accentual -ka, which
occurs freely in Chinookan numerals, pro-
nouns, and adverbs as deictic element (" only,
just "). Examples are W. i'xt-ka " just one ",
na'i-ka " I ", a'x-ka " she ", Lower C. nd'm-
hi " I alone ", e'-ka " thus ", kaiua't-ka " soon ",
nau'it-ka " indeed ", W. iiua't-ka " to yon
(place) ". This deictic '-ka, in turn, is obvious-
ly merely an enclitic use of an old demonstra-
tive stem ka " that " which is no longer in
free Chinookan use but which survives in Lower
C. ka, c-ka " and " and as petrified temporal
ka- " that (time) " in Lower C. ka-iua't-ka
" soon " (< " to just that [time] ", parallel
in form to W. i-wa'l-ka "to just yon [place] ")
and ka-wi'x' " early " {c{. luiix't '' tomorrow ",
W. wax "dawning") ; cf. also W. tense pre-
fix ga-, ga-l- of remote past time. All these
Chinookan elements (ka, ka-, -ka ; ga-; -gE-),
finally, are reflexes of a wide-spread demon-
strative stem *ka " that ", often used as a gener-
al term of reference, found in other Penutian
languages (e.g. Coos -kd in xd-ka " he " ;
Takelma j^^ " that"; Yokuts/fert " that "[vis.] ;
Miwok i-ka " that " ; Tsimshian -_^e absent
connective, -ga absent demonstrative added to
final noun in sentence).
3. It is fairly clear that the two fundamental
factors in the development of the somewhat
irregular morphology of Chinookan were a
280
VI American Indian Languages 2
I 10
strong and movable stress accent and, as a
result of this, the tendency for vowels to drop
out and for originally independent elements to
melt together into complex assemblages. Thus
the old sentence, which seems to have been
constructed on rather simple, analytical, lines,
tended more and more to petrify into a highly
synthetic sentence-word. We have already hint-
ed at the probability that the phonetic change
of g to tc antedated the inclusion of certain
elements in the verb. Internal evidence makes
it practically certain that at least the tense pre-
fixes were late in coming into the verb com-
plex. In the first place, the tense prefixes of
Lower Chinook differ considerably from those
of the upper dialects ', so that it looks at
though an old set of temporal particles or
adverbs (Lower C. a ; «-; W. a, a-l; i, i-g-^
na, na-l ; ni, ni-g ; ga, ga-l \ ;/-) had coalesced
with the following pronominal prefixes of the
verb in the independent life of the various dia-
lects. Moreover, these elements do not behave
as though they had ever coalesced into a
phonetic group with the early Chinookan
forms of the transitive forms for " he" (^iicc-)
and " she " (^ags-'). Thus, in Wishram we
have forms like i-g-i'-ux " she made him ", in
which the tense prefix /- does not palatalize
the following -g- to -Ic-, no doubt because it
did not enter into the verb complex until long
after the palatalizing effect of an /- had spent
its force. If the /- had been prefixed at the time
that the pronominal element^^- " she "still exis-
ted in the fuller form *agj-, it would have re-
quired an intervocalic -g~ and the form *ig-ag3'-
would have arisen (cf. modern forms like ig-a-
-tpa "■ she came out of the house "). Similarly,
a form like Lower Chinook atcE tax " he made
them "evidently arose before the tense prefix a-
was part of the verb complex, for a- could not
have palatalized an original *-_^?- to*- to- while the
I. Op. cit., pp. 577-79.
older pronominal form */to- would have requi-
red as tense prefix the prc-vocalic «-, hence
*n-iicy-.
4. It is a well known linguistic phenomenon
that similar or identical sounds, groupings of
sounds into phonetic patterns, or phonetic pro-
cesses may characterize a number of indepen-
dent languages or even linguistic stocks within
a continuous area. Such examples are sugges-
tive of phonetic interfluences between distinct
languages presumably through the medium
of bilinguals. The change oi g or k to dj
or tc because of the palataUzing influence of a
preceding or following front vowel (/ or e) is
perhaps too general a process to warrant our
attaching much importance to its occurrence in
a number of contiguous languages. Neverthe-
less it is of some interest, and, it may be, of
historical significance, to point out that the
change of ^- sounds or of palatalized k- sounds
to Ic- sounds is found in a continuous or nearly
continuous area from a northern point on the
west coast of Vancouver Island south to the
mouth of the Columbia. All the Nootka dia-
lects, both NootKa proper and Nitinat-Makah,
have altered the original Wakashan anterior
palatal Ic- sounds, preserved in Kwakiutl, to
corresponding palatal sibilant affricatives; Kwa-
kiutlj g' (^'•)and t (^)appear as Nootka tc; t \
(^') as tc' ; and x (.v) as c. A large number of
Salish dialects, furthermore, have altered the
original unlabialized /.'-sounds to /^-sounds. I
am quoting Dr. Boas' personal statement on
this point and am unable to give the geograph-
ical distribution of the Salish /^-dialects. I
should perhaps add that the Lower Chinook k'-
Q-) sounds, which correspond to ordinary k-
sounds in Wishram, are a comparatively recent
dialectic development before /-vowels and that
they have nothing whatever to do with the
old, general Chinookan, change of 0^ to tc after
/-vowels which is the subject of this paper.
Seven: Penutian Languages
281
Editorial Note
Originally published in International Journal of American Linguistics 4,
105-110(1926).
Silverstein (1977), working with richer comparative data, has reaffirmed
Sapir s analysis of the Chinookan ergative prefixes, although he posits a basic
zero marker, rather than a-, for the third person masculine subject/agent pre-
fix. Like Sapir, Silverstein sees "heavy categorial influence from the languages
surrounding [Chinookan] on the coast," resulting in a rich elaboration of
ergative inflectional morphology out of an earlier "nominative-accusative phra-
sal-enclitic syntax" (1977: 154).
Sapir's ms. corrigenda on his copy are as follows:
Original
For:
p. 106,
col.
1,
first table
p. 107,
col.
2,
last para.
ta-
p. 108,
col.
1,
1. 11
-kc-
p. 108,
col.
1,
1. 22
in animate
p. 108,
col.
1,
fn. 1
See,
p. 108,
col.
2,
1. 1
age-
p. 108,
col.
2,
1. 3
charasteristic
p. 108,
col.
2,
1. 8
whith
p. 108,
col.
2,
1. 26
q:'8'
p. 109,
col.
1. 4
gE-L-
p. 109,
col.
1. 5
-gE-M
p. 109,
col.
1. 18
to her,"
p. 109,
col.
last 1.
tachedas
p. 110,
col.
1. 25
itce-
p. 110,
col.
2,
1. 29
Kwakiutl, g-
Read:
Reverse headings "intran-
sitive" and "transitive"
ta
-kc
inanimate
See
ag^-
characteristic
with
-GE-L-
-GE-M-
to her"
tached as
/7C9-
Kwakiutl g-
THE RELATIONSHIP OF MIXE TO THE PENUTIAN FAMILY ^
By L. S. Freeland.
Morphological
THE structure of Mixe is rather bare and
scanty, at least when compared with the
rich and fairly intricate grammatical patterns
found in some of the Penutian tongues of
California and Oregon. One gets the im-
pression that in Mixe the morphology has
worn thin. 2 There is a strong tendency to
fall back on word-order to express relational
ideas. New morphemas appear which are
but one step removed from concrete seman-
temas. In some cases these "empty words"
still retain their full concrete meaning in
other connections, as for instance the word
mid which is the verb "to have". We find it
in Mixe used to express the concept "with,
of" very much in the same manner as the
"belong" of Beche-de-mer. In Miwok or in
Maidu this would be expressed by a comita-
tive suffix.
^ The Mixe material for this study was obtained
by J . de Angulo in Oaxaca during the course of a
linguistic survey of that region for the Department
of Anthropology of Mexico under the Direction of
Dr. Manuel Gamio.
After writing this article I sent it to Dr. Edward
Sapir for criticism and suggestions. Dr. Sapir was kind
enough to send me a Ust of further cognates from
Takelma, Coos, and other languages from the northern
Penutian group. Dr. Sapir was also kind enough to
write some marginal notes which I have taken the
liberty to reproduce as footnotes with the initials K. S.
In his communication to me Dr. Sapir says that Prof.
Roland B. Dixon had already been struck by the Penu-
tian character of Zoque, and had sent him a list of
Penutian cognates in this language. Zoque is very
close to Mixe. The credit for the discovery of a Penu-
tian language in southern Mexico belongs therefore to
Dr. Dixon, by priority.
^ Cf. analytic wearing down in Mukne and Coos as
contrasted with Miwok and Takelma. — E. S.
One somehow gets the impression that the
Penutian morphology having reached too
great a degree of intricacy in its usual
methods for expressing the relational, has
thrown the whole baggage overboard and
started on a new tack. This may be due in
some measure to the influence of other
Central American tongues which seems to
be in the direction of sparseness of linguistic
expression. Nevertheless it is true that a
very similar drift is at work in some places in
California, for instance in the western dialects
of Miwok as compared with the eastern
dialects.^
Whatever may be the interpretation, the
fact of the morphological sparseness of Mixe
remains, and naturally precludes extensive
morphological comparisons. The evidence
for classing Mixe in the Penutian family must
therefore necessarily^ be largely lexical. But
in spite of its limited grammatical apparatus
Mixe possesses morphological traits that
have a strong Penutian flavour. We will
consider these briefly:
(l) INTERNAL MODIFICATIONS OF THE RADICAL.
As in Maidu, in Miwok and in Yokuts, so
in Mixe the radical often appears in several
forms so closely allied that it is sometimes
difficult to say which is the primary one.
^ This is just what happened in Chinookan, where
older Penutian features are only survivals and a new
autonomous and rather complex morphology was built
on top of the older system by means of coalescences of
sequences. — E. S.
284
VI American Indian Languages 2
2q
In Maidu these symbolic changes express
derivational concepts, in Miwok and Yokuts
temporal aspects, in Mixe differences of
person.^
Examples (a hyphen indicates the omission
of irrelevant affixes) :
Maidu :
a) tot to drag ; tttt to touch ; tat to stroke ; tet to reach
for.
b) tea to eat grass; tco to eat meat.
c) yopom-ia to throw; yapam to slap.
d) witcap to tear; witcep to tear a little; witcohit to
wring; witcup to tear off.
e) ivitcot to skin a hare or small animal; iviicut to
scrape hides; witcat to split with the hand;
ivitcet to split a small thuig.
Miwok: 2
to play
to die
to smoke
to run
Yokuts :
cadik, cadak to awake; tcadxin tcadax to turn;
ipe, epi to get water; hiwet, hcut to walk, go, move;
tawidc, taudj to overcome ; waid, waadi to break up ;
diikdu, dukud to bury; tui, toy to shoot.
iESENT
TiME Removed
Freque
FROM Present.
TATIVE
awi-n-
awin'-
dwni
tcdmci-
tcam'ic-
tcdnric-
hdksi-
hakis-
hdk'is-
huwd't-
huwdt--
huwta
Mixe:
to get up
to fall
to like
to sleep
I
pedelk-as
ri'gunac-els
n-zoik-bes
may
thou
pedetk
m-gunac-el
n-z6ik-he
may
he
pedtg
kiivac-el
coik-be
mey
we
pedelk-matc
n-gunac-edom
dsok-medj
mau-medj
ye
pedtk-tetc
ni-gunac-tel
m-dsok-taba
mau-detc
they
pedik-nedj
kunac-tel
tsok-teba
mey-dj
(2) THE INCORPORATED PRONOUN.
Yokuts and Mukne (Costanoan) are appar-
ently lacking in incorporated pronouns.
Southern Maidu uses independent forms in
the ordinary statement, but possesses v^hat
seems to be a depleted series of suffixed
pronouns (possibly old objective forms), used
Examples :
in subordinated expressions of time and in
hortatory and optative modes. Miwok is
rich in pronominal series, specialized in their
use for present, past, future, subjunctive
and the Hke. Although the}'' are much less
numerous the Mixe forms come closest to
those of Miwok. 3
Maidu
Present
Miwok
Past and
Future
Mixe
I
-5, -se
-m
-/
-is
n-
thou
-ne, -no, -men
-s
-n
—
m-
he
-i
—
—
—
y-, IV
we
-es, -nes, -hese
-mas
-m
-ma,
-mats
n-
ye
-maw, -mem
-tos
-ton
-te
m-
they
-i
-P
-k
-te
y-, w
Both Miwok and Mixe show a considerable
amount of fusion and obliteration of elements
^ Cf. internal vocalic changes of verb themes in
Takelma, Coos, Lower Umpqua. — E. S.
- These alternations are amazingly like the Takelma
in putting together their verb forms. This
gives often quite a similar impression:
ones, e. g. ydan- "to go": fut. yana-. — E. S.
^ Takelma like Miwok and Mixe. — E. S.
Seven: Penutian Languages
285
30
Miwok
Mixe
I have eaten
i'wicak I
eat
hi' kcebes
thou
■i'wicas
hi'kceb
he
i'wica
heykc
we
i' w'icakmas
hi'kcemets
you
i'wicak tos
hi'kcteh
they
i'wicap
hi'kcteb
(3) THE VERBAL PREFIX.
Rather a large number of the Mixe verbs
in the list are bi-syllabic, and many can be
analyzed into two elements. They suggest
very strongly the Maidu verbs with prefixes
or first-position-stems such as he- action with
the foot, ka- action with the hand, and the
like. Mixe probably has nothing like the
number of these prefixes that there are in
Maidu, but three or four are quite clearly
discernible and very suggestive:
Maidu: heye creep; bedokoi run; bedoi
kick; helom step over; hewet hurry, kadokok
push away; kapadum roll up; katut touch;
katcaduk grab; pai foot, trail; -pat after.
Mixe: petc-h to climb; pegig to run; pedig
to rise; pasl-h to go out; pa- along, by;
pa-otsm to go along; pa-witsm to creep;
pa-nas to pass by {nas "the ground") ; pa-boy
to chase; pa-son to follow; pa-wep to kick
[wep to strike) . kuhath to pay ; kudigx to put ;
kutuk to command; kuyat to play; kunas to
fall {nas ground); ni-pan to cover; ni-kes-b
to be on top of something high [kos, kes
above, on, upon); ni-wits to close; neksm to
go, ni-neksm to go for the benefit of some
one else; kutuk to order, ni-kutuk to order
through the agency of another; kes-wits to
place; kes-ets to place to boil {ets-b to boil);
niwis-dud to open {niwits to close) ; nipan-dud
to uncover [nipan to cover) ; rswits to take
off [kes-wits to place, ni-wits to close) ;
rspit-n a skirt [pit to make thread) [rspit-m
also means "to tangle"); rsmats to let go
[mats-b to take, pats-b to find) ; rsmats
trotisers ;
II. Semasioi,ogicax^
(l) NATURAI, EIvEMENTS:
1. sun: si Cf. Maid, sa fire; Miw. hi, hit sun, day;
Muk. icnien, hismen sun; Wint. sun sun.
2. moon: po Cf. Maid, pombok moon (also poko
sun, poho night, poketcuk star); Yok. opodo,
upic moon; Wint. po fire; Tak. be svm; Kal.
pyd-'n sun, day.
3. shine : tuks Cf . Miw. lulep- shine ; Muk. tuxts day.
4. night: su, us Cf. Miw. sien- be night; Wint.
sinol night; Tak. xii'n night (Tak. x regularly
comes from 5).
5. cloud: yots Cf. Maid, ya; Tak. hat; L. Um. hiai ;
Pen. * h-yai.
6. sky: tsap Cf. Muk. tcarak.
7. star: musa Cf. Muk. mur night.
8. fire: hen Cf. Miw. wik-, wik-, wul-, wel-; Yok.
licit, oxit; Muk. xii, xihi, xute go for a fire,
light a fire ; Tak. -'/cm/- to set fire to, to catch
fire; Coos cutl to set fire, Hcil- to burn, 'tcwel-
fire; Pen. * swilihe, *'lswil- to bum (cf. "to
burn" next semantema).^
9. bum isai Cf. Maid, tco; Miw. tsup.
10. salt kan Cf. Muk. akes, aks.
11. dry tes Cf. Miw. tsutul.
12. stone: tsa Cf. Miw. cawa; Muk. isin.
13. earthquake: pumimb Cf. Wint. pomoko.
14. hole: Aw/ Cf. Maid, tuke; Muk. kutui.
1 In making the following list of semantemas I
found it necessary to alter somewhat the orthography
of the different sources in order to obtain a common
basis of comparison. The system of transcription we
have followed is that of the American Anthropological
Association. It may be well to point out that in Mixe,
s and c are interchangeable ; i' and f'are the "unrounded"
forms of u and o; in Mixe the dynamic stress accent
is invariably on the semantema, and this in turn is so
seldom of more than one syllable that a di-syllabic stem
may be looked upon confidently as resolvable under
further scrutiny.
For Miwok, Wintun and Maidu, I have made use of
my own unpublished material and of that of Mr. J. do
Angulo, also of the grammar of Maidu by R. B. Dixon
in the first volume of the Handbook of American Indian
Languages.
For Pomo I have made use of my own and J. de
Angulo's unpublished material, also of The Ethno-
geography of the Pomo Indians by S. A. Barrett. For
Mukne I have used the Mutsun Dialect of Costanoan
by J. A. Mason.
For Yokuts I have used The Yokuts Language of
Southern California by A. L. Kroeber.
Many of the kinship terms are to be found in E. W.
Gif ford's exhaustive work "California Kinship Termi-
nologies".
* Pen. *swil- "fire" ? : Pen. *'tswil- "to burn" ? —
E. S.
286
VI American Indian Languages 2
15. water, river; ni Cf. Maid, dim wet; Miw. nuni-
to be submerged; Wint. hm boat.
16. wet: 50^ Cf. Miw. cumek to soak up.
17. rain: tub, tsuy Cf. Miw. upa; Yok. ciwek drizzle.
iS. wash: iL'ibux Cf. Maid, dp to dive; Miw. upux
bathe, swim; Yok. ep-, ip- swim; Muk. upxi
to sip; Tak. 'pag- to bathe.
19. wind: poh Cf. Maid, biye; Miw. pwc to blow;
Yok. piuate whistle; Muk. put blow, pusninyis
whirlwind; Wint. pul whistle; Tak. -p'ouw-
to blow; Chinook po to blow; Kal. ptU to blow;
Coos pjix- to spout; Tsim. bmi to blow.
20. red: (saps Cf. Muk. lasas.
21. green: stisA Cf. Maid, kotcis; Muk. icutsu.
22. then: we/ Cf. Miw. ne this; Muk. «e this.
31
23. Temporal affixes: Past te- Cf. Miw. -/it; Muk.
-te, -kie. "essential" : -b, -n Cf. Muk. -n infinitive;
Tak. -n noun suffix, sometimes used to make
verbal nouns, exactly as in Mixe; Maid, -n
infinitive ending, "contmuative" -its, -ats
(forms of the verb to be) ; Cf . Miw. -ntcu to live,
also a continuative ; Mukne tcira always,
continually.
24. big: mi, mik, mix (V. also old) Cf. Yok. nioxodo
old; Muk. muk adult; Tak. mahai.
25. little: mutsk. Cf. Muk. pusnt.
26. round: pik Cf. Miw. pol; Yok. huk-\ Wint.
bakak.
27. Numbers:'
Mixe
Maidu
Miwok
Yokuts
Mukne
Wint.
I
tig
wite, suii
kerje
yet
emet-
ete
2
matsk
pen
otto
punoi
utxin
pampe
3
tugig
sapm
toloko
soopin
kapxan
ponol
4
niaktask
tcii
oyisa
hatepanai
usit, utit
emus
5
mogosk
mawik [ma
hand) macoka
yitecinil
parues
ete-sem
6
tuduk
timbo
temoka
tcolipi
nakitci
serpol
7
ustug
penimbo
kenekaki
7iomtcin
takitci
serpote
8
tuktug
pentcii
kawinta
monoc
taitimin
pan-emus
9
tastug
tciinbo
wo^e
nonip
watsu, paki
pan-emute
10
mak
matcam
na^aica
waditc
matstt
pampa-sem (two hands)
20
ipc
inaiduk
na'a
kay
28. more: maak Cf. Miw. manik.
29. no, not: ka-, kedi. Cf. Miw. ke, ken, ket; Yok.
k'a>nu; Muk. ekwe, kwe.
30. yes, positive, good: oy Cf. Maid, he; Miw. hi;
Yok. houu; Muk, he; Wint. 0.
31. thus: sa Cf. Miw. saka like, as; Muk. sata.
32. if: pen. Cf. Muk. pini.
(2) animals:
33. dog: uk. Cf. Maid, suku; Miw. tcuku; Yok.
tce-jej; Muk. hutce; Wint. cukut; Tak. 'tsixi.
34. bird: hon Cf. Maid, hu to fly, horn nest; Miw.
hotju egg; Yok. hoy egg.
35. eagle: wits Cf. Miw. wipayak; Yok. witisnl,
eagle, witc, condor.
36. snake: tsan Cf. Muk. lisana.
37. frog: ink Cf. Miw. wataksay, kotola; Wint,
watak.
38. fish; ak Cf. Maid. mako.
39. flea; picg Cf. Muk. wipsur.
^ Note to "two" Chin, mokct., Mixe matsk cannot be
directly compared to Mixe us- in us-tug "7" (= 2 -|- 5)
and to Miwok otto and Mukne utxi-n but may be related
if we assume m- as archaic durative intran.sitive prefix,
such as we have far more transparently in Hokan and
Muskogian. — E. S.
40. louse: ag Cf. Muk. rax.
41. bedbug: /// Cf. Maid, dt^ louse.
(3) plants:
42. wood, tree: kep, kip Cf. Miw. kapiim, bark;
Yok. yapkin ; Muk. xipur, tree.
43. brush, bush, forest: yuk Cf. Miw. yomyum,
brush; Yok. yawud; Maid, yo, flower.
44.. tobacco: xuyge Cf. Miw. hutia.
45. moss: sarjk Cf. Muk. sasuk.
46. fruit: uts Cf. Muk. owos.
47. flower; pux Cf. Maid, pu, to blossom.
(4) man:
48. person, people: hay Cf. Maid, maidik; Miw.
miw; Muk. muwe; Coos md; Yok. mai.
49. man: yayek, yadyek Cf. Miw. naya-yak, Yok.
muk-yamk. (also Coos -iyag plural suffix in
terms of relationship).
50. woman ; tosdyek Cf . Miw. occayak, Yok.
latc-yamk.
51. old man: mix (also "big") Cf. Maid, muk, big;
Yok. moxodo; Muk. muk, adult.
52. foe, enemy: was Cf. Mukne wayas.
Seven: Penutian Languages
287
3^
53. youth, boy: nakn Cf. Yokuts notco.
54. father: tyedj, tat Cf. Maid. te\ Yok. -atet; Muk.
ete, grandfather; Win. tata.
35. mother: tagh, nana Cf. Maid, na, ne; Miw. ita;
Yok. -ajaj; Mukne ti/a, parents: Tak. ni-, -hin;
Kal. 7ii; Low. Chinook -wafl; Coos e'n-dtc {-ate
kinship ending); Coos nik/a "mother!" (voc);
Tsim. (Nass R. dialect) nc-^i "my mother"
(-H "my"), nd'd (voc.).
56. grandfather: ap Cf. Maid, opa, pa; Miw. apa,
father; Mukne apa, father; Win. apa; Yok.
bapa, bap', father's mother; Tsim. (Nass
R. dial.) no-beb-i', my mother's brother, bip'
(voc).
57. elder brother: ats Cf. Miw. ata, tatci; Mukne
taka.
58. yoimger brother: its Cf. Maid, tu, t'i; Miw.
tcale; Mukne tare; Miw. iDi'' (Sapir MS.).
59. younger sister: vis Cf. Wint. utcii; Miw. Wa'*
(Sapir MS.).
60. uncle: haym Cf. Maid, yam, kam; Miw. kaw\
Yok. kawa, komoy.
61. elder sister: isyo Cf. Maid, cti; Miw. De'De*
(Sapir). 1
62. aunt: tsugu Cf. Yok. guiha, nusus.
63. child: tt»;g Cf. Miw. aijst, son; Mukne rwis, son.
64. son: wza»; Cf. Mukne mos.
65. daughter: ^e/, wyi's.
06. diminutive or endearing suffix (suggested by
Sapir) -s, -c, -j (e. g. wi'-y, daughter; Cf. ur)-g,
chUd. Pen. stem perhaps *ariu, child; dimin.
*a'yiu-si > Mixe tji-s, Miw. ajj-si, son, Mukne
ini-s, son). With stems ending in -t, -d this old
diminutive combines in Mixe to form -ts, -ic,
-dj, hence: tyed-j, father (probably dim. form
of tat) ; at-s, elder brother : Miw. ata ; it-s,
younger brother: Maid, ti, Miw. iDi'^; ut-s,
yoimger sister: Miw. Wa'\ This dimin.
*-s! survives also in other Pen. dialects: Miw.
ay-si, son; S. Coast Miwok (Barrett) taiyi-s,
man; kuleyi-s, woman; hena-s, boy; oyi-s, old
man; potci-s, old woman( ?) ; Mukne ini-s, mo-s,
son; Wintun utcu, younger sister (diminuti-
vized < *utu < *utaP); Tak. -xi, -x < *-si,
*-s (in hap-xi, child, haap-x, one's children,
cf . haap'-, child in other combinations) ;
Chinookan (Wishram dialect) -c in -k!acu-c,
paternal grandfather, -gak ! u-c, maternal grand-
father, -k/i-c, paternal grandmother (these
stems are etymologically related to respective
reciprocals: -qcE-n, -gaka-n, -gia-n), further in
wi-n-am-c, my father (stem -am) and wa-n-aq-c,
my mother (stem -aq-) ; Coos -ca, endearing suffix
(e. g. huumik-ca "dear old woman"; umd-ca-
^ c- of Maidu cti and -sy- of Mixe tsyo may be
diminutivizing.
tc, uma-c "grandmother": vocative uma. This
Penutian *-si diminutive is characteristic, it
would seem, as contrasted with its vmdoubtedly
cognate Hokan *-tsi {*-'tst). — (E. 8.)^
67. I: es, -s, -es Cf. Maid. -5, -mus; Tak. -xi "me",
from < *-si.
68. I: n- Cf. Miw. kan; Yok. na; Muk. kan; Wint.
nanu "my"; Coos n; I/Ow. Ump. na "1", -n
"I, me"; Tak. -'n, -n (trans, subj.); Chin, -n-
"I, me" (Wishram dial, also indep, «a.)
69. thou: migs, m- Cf. Maid, mi, -mam, -mem;
Miw. mi; Yok. ma; Mukne me; Wint. mi; Tak.
ma; Chin, -m-, -mi-; Tsim. — n (thee); m-.
70. he: yt Cf. Maid, -i; Miw. /.
71. we: -m,, -em, -ma Cf. Miw. -m, -me; Yok. mai;
Mixkne makse, fnak; Tak. -aw; Coos -ami;
Tsim. -Em.
72. ye: -t, -te, -ta Cf. Miw. -tos, -ton, -tok, -tc.
73. which: wud Cf. Mukne watt "someone"; Coos
wit; Low. Ump. watc.
(5) BODY PARTS AND ACTIVITIES:
74. eye: win, is "to see" Cf. Maid, hin; Miw. c'in-ti,
eye, c'iy, to see ; Yok. oil, to see ; Mukne xin ;
Wint. wini, to see, ca, eye; Tak. al-xi-g-, -xi-k-,
(Tak. -xi- < *-si-), also -xanw- ("to look")
< *-san-w-.-
75. mouth: aux Cf. Miw. awo; Yok. ata, ca; Mukne
xai.
76. tongue: _va»e Cf. Maid, eni; Miw. He/??/, leiip;
Yok. tadxat, palat "project tongue"; ^lukne
lase; Wint. tahal; Tak. ela; Coos Ae//a ; Low.
Ump. i^al.
77. to drink: ug Cf. Miw. wcczt; Yok. uk; Mukne
ukis; Tak. iigw-.
78. to speak: mo^ Cf. Maid, mo-, ma-, action with
mouth; Miw. mo-, answer; Mukne }>!ons.
79. to answer: azoy Cf. Miw. has-, ask; Mukne
osehe.
80. word: kops Cf. Maid, ka, say; Miw. /fa-, say;
Mukne ko, say.
81. to call: mo Cf. Maid, wo; Miw. «'o; Low. Ump.
waa, say, speak, tell; Chin, wawa, talk; Tak.
yaway-, talk (dissimilated from *waway- ?).
82. nose : hikb Cf. Maid, hikii; Miw. A«/j ; Mukne hus.
^ Nominalizmg suffix -c = Mukne -s verbalizing
suffix (e. g. yitc-s "language" <; rite "to speak";
= Tak. -{a)x infinitive suffix of intransitive verbs
< *-(a)s; = Low. Ump. -s verbalizing suffix (waas
"language" << waa "to speak").
^ In other words we have two old Penutian stems:
*(i)sa or *(i)say (> Miw. ciy-, Wint. ca, Mixe is, Tak.
-xi-) and *{i)san, *(i)sat > Miw. cln-ti, Yok. cil, Mukne
xin. Maid, hin, Tak. -xan-w-, Tsim, saal "to notice",
perhaps also Coos xil-, "to look around". — E. S.
288
VI American Indian Languages 2
33
83. ear: tatsk Cf. Miw. tokoc; Yok. luk; Mukne
liiksiis.
64. hand: ki Cf. Maid. ka-. with the hand; Miw.
eku. uka « *esu. Miw. k = Mukne s. regularly.
— E. S.); Yok. ko, to hit with hand; Muk.
isu; Win. kupum.
85. left hand: ava Cf. Mukne an-si.
86. to take: mats Cf. Mukne nm;tr, get; Low. Urap.
mate-, to lay; Tak. mats lag-, put, place; (Pen.
base *mats- > Ta^. mas-g-); Chin. (Wishr.
dial.) -A--n»a, to lay, -a-l-ima-lx, to put into
water; Tsim. mag-, to put dowTi. ; Maidu me, to
catch.
87. to give: mo, to have: mid Cf. Maid, me; Miw.
a>«-; Muk. ;trMm, cumi.
88. to do, make, work: dun Cf. Miw. taiiha; Yok.
/nc; Muk. /w« finish.
89. to put, place: -wits Cf. Maid. wi-. pull, press;
Miw. wtk-; Tak. -ttyhA, to spread out.
go. foot, leg: t'ik Cf. Maid. /^A, jump; Miw. tek,
kick, Aa/f, foot, tirji, hip; Mukne /^A^, walk on
toes.
01. knee: koc, Cf. Tak. gouk; Low. Ump. qoq;
Chin- -q!oxl.
92. to go: ois Cf. Maid, u-; Miw. «-.
93. to walk: wag Cf. Miw. iwi-n, weeta; Yok. wadix,
pass by; Mukne twa//, come, ze/a/^, go; Tak. wi-,
to go about.
94. to go: y- Cf. Maid, i-ye; Miw. yoc-; Chin, -ya,
-i-\ Tsim. yaa; Tak. yaw-; (Cf. Hokan *iya,
to go).
95. action with the foot, walk, trail, etc.: pa-
(pason, to follow, paboy, to chase, pesamb, to
go out) Cf. Maid, be-, with the foot, puiyi, hunt.
track; Miw. puy, move, depart; Mukne paya,
run, payta, himt.
96. to stay : wi (also in : tti^ic, a bed)^ Cf. Tak. teiay-,
to sleep ; Kal. ivai, to sleep ; Tsim. woq, to sleep ;
Coos, haya-ti, to lie down (Coos A often comes
from iv) ; Yok. woi-, to sleep.
97. to sit: uy Cf. Maid, in-; Yok. uy, to lean;
Mukne one.
98. bone: pak Cf. Maid, -/'o^, -puk; Wint. /jaA.
99. skin: po Cf. Maid, ^tt, posa.
100. neck; yok-yi Cf. Yok. ogun; Tak. gwen; Coos
kwin-ts; Low. Ump. kwininicu.
loi. rib: /la/c Cf. Maid, /cz ; Miw. Atf/c, bone; Mukne
/ca/c, bone.
102. phallus: tsii-k Cf. Maid, utcu, piss; Yok.
icwyo ; Mukne tcoxo, pudenda.
103. testicles: istca Cf. Miw. caii-tal; Mukne tcatia.
104. anus: pxut Cf. Miw. poii, navel; Yok. poiodo,
intestines; Mukne puttus, belly; Wint. pot.
105. to urinate: tatsp Cf. Miw. ota, otso; Mukne
tcala (Cf. Wint. tcunu-s; Tak. xan. — E. S.).
106. heart, belly, thought: hot Maid. ho7i; Yok. hut,
hon, to know, horj, heart; Mukne xtitu, belly;
Coos haw-, to imagine, think; Low. Ump. hai,
opinion; Tak. heivhaw-, to think. ^
107. fear : tsok Cf . Miw. ceki ; Yok. dote : Mukne susu.
108. love, like: tsok Cf. Maid, -sak-; Yok. icik;
Mukne otciko.
^ < "Sleeping instrument". See -i nominalizing
suffix. — E. S.
' Maidu ho-n and Yokuts ho-r] probably nouns in -n,
cf. "infinitive" and noun-forming -n, from base *ho-
< Pen. *haw-, "hau-; Mixe ho-t probably factitives in
-t from same base. — E. vS
Editorial Note
Originally published in International Journal of American Linguistics 6,
28-33 (1930).
Ancestral Mixe-Zoque is now supposed by most authorities to have been the
language of the Olmecs, the oldest of the archeologically known high cultures
of Mesoamerica (Campbell and Kaufman 1976).
Sapir's "Mexican Pcnutian" (Mixe-Zoque and Huave) was expanded by
Whorf ( 1935) to include Mayan, Totonac, and Uto-Aztecan. Greenberg (1956)
accepted the addition of Mayan and Totonac but excluded Uto-Aztecan.
Seven: Penutian Languages 289
Greenberg has recently (1987) reiterated his version of Mexican Penutian as
part of a hemisphere-wide classificatory scheme bolstered by extensive lexical
comparisons. Neither Whorfs nor Greenberg's claims have met with favorable
reaction from specialists in Mesoamerican languages (see Campbell 1979: 964).
The connection of Huave to Mixe-Zoque, proposed by Radin (1916) and
accepted uncritically by Sapir, is now considered improbable. The most likely
affiliation of Huave is with Otomanguean (Campbell 1979: 964).
COOS-TAKELMA-PENUTIAN COMPARISONS
Edward SAPiRf and Morris Swadesh
1. Introduction
1.1. Date of manuscript
1.2. Description
1.3. Phonology
1.4. Why withheld
1.5. Symbols and arrangement
2. Sapir's comparisons
1. The Coos-Takelma-Penutian compari-
sons presented here were made by Edward
Sapir. Morris Swadesh has prepared them
for pubhcation and has provided the intro-
ductory information and discussion.
1.1. Sapir's manuscript, which now forms
part of the Franz Boas Collection of the
American Philosophical Society, evidently
dates back to about 1914, the publication
year of Leo J. Frachtenberg's Coos, An
Illustrative Sketch, Extract from Handbook
of American Indian Languages (BAE-B
40.2.297-429). It surely represents an im-
portant part, though far from all, of the
lexical evidence Sapir refers to in A Charac-
teristic Penutian Form of Stem (UAL
2.58-67), pubHshed in 192L In this article
he concurs with Frachtenberg's suggestion
of Penutian affinities of certain Oregon
languages, stating:
All this is very interesting to me, as it chimes
with conclusions or hypotheses I had arrived at
independently. On the appearance of Frachten-
berg's Coos grammar it soon became clear to me
that the morphological and lexical resemblances
between Takelma and Coos were too numerous and
fundamental to be explained away by accident
or plausibly accounted for by borrowing. The
appearance of Frachtenberg's Siuslaw material
has only tended to confirm this impression,
further, to make it perfectly obvious that Coos
and Siuslaw, as Frachtenberg announces, are
divergent representatives of a single linguistic
stock. Meanwhile comparisons of Takelma, Coos,
and Siuslaw with Dixon and Kroeber's Penutian
group of California (Costanoan, Miwok, Yokuts,
Wintun, and Maidu disclosed an astonishing num-
ber of both lexical and morphological correspond-
ences. ... In spite of our slight knowledge of
most of the Californian languages involved, I
succeeded in getting together what I believe to
be a quite respectable mass of evidence tending to
unite the southern languages with those of Oregon
into a large and highly differentiated "stock".
In the same article Sapir refers to evidence
for still further Penutian relationships. It is
possible that other manuscripts will yet be
discovered containing this material. Much of
it may have been in the form of marginal
annotations made by Sapir in books and re-
prints, like those compiled in Comparative
Penutian Glosses by Edward Sapir, in the
Boas Collection.
1.2. The present manuscript consists of
13 pages 8x11, handwritten in Sapir's usual
small writing. There are 152 numbered
groups of cognates, with columns for Coos,
Takelma, Penutian. There is no other title.
At least two shades of ink are distinguishable,
a dark blue for all but set 152, and a lighter
blue for the latter and for additions to the
first 151 sets. Since the additions include
items in the Coos column marked "Mil."
(that is Miluk dialect of Coos) and four
annotations of 'see L.U.' (that is, Lower
Umpqua or Siuslaw), one may infer that the
main list was made before and the added
notes after the appearance of Frachtenberg's
Coos and Siuslawan sketches. The original
Coos items must then have been based on
Frachtenberg's Coos Texts, which appeared
in 1913. This assumption is corroborated by
the fact that Sapir's copy of Coos Texts has
marginal glosses of Takelma comparisons in
what seems to be the same ink as the manu-
script list, and there are 'L.U.' annotations
in the lighter ink. All the foregoing observ^a-
tions on the ink used in the manuscript were
made by inspection only. The identifications
seem for the most part quite obvious,
except for those parts of the manuscript
which have been most affected by light.
132
292
VI American Indian Languages 2
133
Besides the entry numbers the manuscript
has a set of check marks varying with a
diagonal criss-cross at a few points. Sapir
used such marks to register a later critical
reexamination of his work. That is, at some
time after he had made the list, he must
have gone through it again, with constant
reference to the sources, checking the entries
which seemed plausible and putting the
criss-cross by the ones that on reflection
seem improbable. During this process he
would often insert additions to the original
material. In the present manuscript, we find
that the criss-cross sometimes confirms and
sometimes overrides an earlier hesitancy re-
flected by parenthesized question mark or
annotations (e.g. in 57 after the Yokuts
form: "perhaps better to Tak. -ha-n,
no. 55"). Three asterisks between entries 55
and 56 may be a pause mark, showing where
Sapir had temporarily stopped in his
check-up. A short horizontal line between 86
and 87 is evidently another pause mark.
1.3. The manuscript does not include any
full analysis of phonological relationships.
However, one finds phonological notes at
various points. Some of these merit general
comment. The occurrence of Yokuts d for
earlier 1 is indicated or implied in a number
of places (8, 9, 19, 50, 54, 71, 72, 74, 78, 79,
83, 84, 87, 94, 95); Newman's work' has
since shown that this is a regular dialectal
development in Wikchamni, while the other
dialects preserve the lateral. In two cases (10,
23) we have Wintun X, Takelma s recon-
structed to t; item 7, disregarding the first
Coos form (Hanis dialect), suggests that
I or X might be a better reconstruction. The
development of Takelma x from earlier s is
posited in 11, 35, 41, 61, 75, 99, 121, 123. The
origin of Takelma s from earlier c is suggested
in 129. This fits in with the fact that Takelma
has no c even though it has s and the com-
bination s?, the latter being pronounced as a
glottalized affricate. Sapir's idea was evi-
dently that pre-Takelma s changed to x,
' See Stanley S. Newman, Yokuts, Viking Fund
Publications in Anthropology 2 (1944), p. 16.
coinciding with original x, and that new
instances of s developed out of older c and ch.
The evidence in general seems to bear out
Sapir's theory, but the proof is complicated
by the presence of more than one s-sound in
Yokuts, Coos and perhaps other Penutian
languages, and by the existence of two
x-sounds in Coos and Siuslaw. Robert Shafer
has attempted a clarification of Penutian k
and x-sounds,^ but he does not fully suc-
ceed; his material is insufficient in quantity
and does not properly distinguish between
probable cognates and similarities which
might have resulted from borrowing. How-
ever, the problem may be expected to yield
to patient research in the future.
Where Sapir gives two or more spellings
for a word, they may be variant phonetic
recordings of one and the same form, dialect
variants (as in Yokuts), or functional vari-
ants (e.g. verb stem and aorist stem in
Takelma) . The first kind of variation is here
eliminated to the extent possible by using
phonemic spellings. The other two types
have been retained because of their value in
tracing phonological relationships.
For Yokuts it was desirable and easily
possible, by consulting Newman's mono-
graph and with additional data kindly pro-
vided by Newman, to specify the dialect of
most of the cited forms. Where Sapir gives a
single form and it coincides with the Yawel-
mani dialect, it has been labeled Yawelmani
(Yy). Where Sapir's form is from another
dialect or includes more than one dialect, the
Yawelmani has been added after Sapir's
Yokuts entry.
1.4. Why did Sapir not publish his Coos-
Takelma-Penutian evidence?' The answer is
perhaps suggested by the following remarks
(UAL 2.59, 1921):
Unfamiliarity with Alsea (Yakonan) and
Kalapuya made it impossible for me to follow
* Penutian, UAL, 13.205-19 (1947).
' Some of Sapir's comparisons and structural
observations are included in L. S. Freelander,
The Relationship of Mixe to the Penutian Family,
UAL 6.28-33 (1930).
Seven: Penutian Languages
293
134
the chain of evidence geographically. Neverthe-
less, Chinookan points of contact soon manifested
themselves too persistently to be brushed aside.
After hesitating for a long while to take up seri-
ously the possibility of affiliating Chinook, one
of the most isolated and morphologically
specialized languages in America, with the
Penutian languages of Oregon, I now find myself
forced by the evidence to admit such an affiliation
as not only possible but decidedly probable. . .
The greatest surprise was still awaiting me.
Tsimshian . . . Should it be possible to demon-
strate (and I am fairly sanguine that it can be
demonstrated) that Tsimshian is a detached
northern offshoot of Penutian, we would be com-
pelled to face a most interesting fact in linguistic
diflferentiation and in the distribution of American
tribes.
These comments show that Sapir con-
sidered the Coos-Takelma-Penutian rela-
tionship to be part of a larger grouping which
he felt could be demonstrated. Evidently he
was awaiting the appearance of Kalapuya
data and was hoping to work out in detail
the relationship of Chinookan and Tsim-
shian. He felt sufficiently sure of himself to
publicly state his theory, but perhaps he
thought it well to wait with the specific
lexical data until he could present the evi-
dence for the fuller picture. The case would
then be comparable to that of Sapir's theory
relating Na-dene and Sino-Tibetan.'* When
students asked Sapir why he did not publish
his evidence for this connection, he said it
might be better to hold off until he had
worked out and published the full details of
Athapaskan phonology. However, in retro-
spect, one can only regret that this material
was not published long ago.
1.5. Sapir's 1914 orthography is here
modernized and spellings are given in
phonemicized form to the extent possible. In
the case of Takelma, we take the aspirates
to be clusters of stop with h and the glot-
talized to be clusters with glottal stop;
Takelma accent is here reduced, with reason-
able justification inferred from Sapir's de-
* See discussion (by Morris Swadesh) of
Athapaskan and Sino-Tibetan by Robert Shafer,
UAL 18.178-81 (1952).
scription,^ to two types, a normal accent (')
with tonal variations according to position
in the word, and a rising or sustained long
accent (') occurring on long vowels or on
vowel-sonant groups in contrast to falling
tone for the normal accent in this position.
Points of phonetic usage: x is for back
velar spirant, c is for sibilant affricate (ts),
H for shibilant affricate, double letters are
used for long vowels, r is for voiced velar
spirant (in Coos). The languages are abbrevi-
ated by using the first letter for Takelma,
Coos, Siuslawan, Yokuts, Wintun, the first
two letters for Chinook, Miwok, Maidu,
Mukne (Costanoan — Freelander's term). Yy
is used for the Yawelmani dialect of Yokuts.
For convenience in printing, Sapir's
columnar arrangement of the manuscript
has been dropped. Sapir's diagonal cross for
improbable comparisons is retained, his
check marks are omitted. His parenthetic
question marks, also indicating doubt, are
retained but are placed after rather than
before the entry. Other than in such matters
of arrangement, punctuation and phonetic
symbols, the materials are given as fisted by
Sapir.
2. 1. T kuu-x-m/e (?) (see S); Mi kule,
kulei wife, woman. Ma kiila woman. 2. C
k'^aaxaX how; T gaP bow; Mi kono, W kul-sak
how. 3. C walwal knife; T wiih-A;m/e; Mi
hulaya knife. 4. tuu good (aor. tuu\\^iu-k- to
be good); Mi towi-s good. ^5. C asoo again,
asoow-is second; Mi os(s)a, otta two. 6. C
e'^n-ee^ mother, niika mother!; T hin-, ni-
mother; W nen-^m, nenin. Ma ne, Mu ana
mother. 7. C huuI nose, Miluk fin-nuuq; T
sin-, sinii-x- nose; W Xinik, suno, Y tiiDiik
nose, Yy tinik. 8. C helta tongue; T ela
tongue; Ma eni, W tahal. Mi le-mtep, le-tip,
Mu las, lase, lasa, Y talxat, '''aladis, m-ada-t
tongue, Yy talxat tongue, "^aalat- to lick.
9. C kxla leg; T k«el- leg; W koli leg
(<*k*ili ?), Mu kolo, koro foot. Mi kollo
* Takelma, Handbook of American Indian Lan-
guages, 40.2.17-18 (1922).
294
VI American Indian Languages 2
135
foot, Y kada-Sa lower leg, Yy kalassa'i*. 10. C
Miluk Hil-li leg; T sal- foot (<*tal-?); W
Xel-ma (<*tel-?).
11. T xuu''*n night (for x:s, cf. xi water
below); W sinol night. 12. T b§e sun, day.
poo now, today; Ma poko sun, W po day,
Y 'i'opo-do sun, "i^op-di day, Yy "^op moon.
13. T n6ox rain-, W luha rain. 14. T p'^ii^re;
W po fire. 15. T s'i'aaw lake; W nahi lake.
16. T som mountain (<*tom?); W toX, ho\
mountain. 17. T tan rock; W Son rock. 18. C
Sximl black bear; T x§,mk grizzly bear; W
Silai bear, grizzly bear. 19. T -s'i'il red; W
tulu-ka, tede-kit red. 20. C bni ^ood; W
laiyok good.
21. T ma thou; W mi thou, [see 47]. 22. T
pam- up, pam-ls sky; W panti-be up, panti
on, upon. 23. tqanX- (?) [to hit, to strike with
instrument] ; T saansan-, sana- to fight with,
spear; W Mia to shoot (<*tana?). 24. T -t
present participle, adjective suffix; W -u-t
future participle, -t adjective suffix with sub-
jective noun, Ma -do- present participle.
25. T -khi?, -ki? if; W -kila conditional.
26. C -uu interrogative suffix; W -wi inter-
rogative. 27. T -de?, the? / (intr.), future
-tee, -thee; W -da /, Mi -t, -te /, me, my.
28. C iin not; T ?anii? not; W eleu
(<*ene-u?), Y am, Yy ?ohom, Mi (E.) ela.
29. T yaan-, yana- to go;W hara to go. 30. T
p^nx hunger; W bira to be hungry.
31. C tk"*- to kick; T tkuuntkan-, tkuun-
to kick; W qow(a) to kick. 32. -eenii reciprocal
verb suffix, -inii (relatives) to one another; T
-an- reciprocal verb suffix; W -pu-ra re-
ciprocal verb suffix (?). 33. T ?al- to (?),
ka-nan in; W el in, into. 34. T pay- out of
house (vb. prefix); W pat out. 35. C yipsn
three (?) (see S [Siinx]); T xipini three; Ma
sapwi three, Yy soopin, Mu kapan, kaphan,
kapxan three. 36. C SuuX- to set fire to, m\- to
burn, Hweel^re; T -s^iiluii-, -s^iil- to set fire
to, to catch fire; Ma sa, Y ?oSit, Yy ?osit,
Mu Soto fire. 37. C mee human being; Mu
ama person, Ma mai-dii Indian, mai-ki boy,
Y may person. 38. T yap?a person; W
yapaitu person. Ma yepi man, husband. 39. T
tak-ax- head; Mu taxa-S head. 40. C x'^iHux'^ ^
head (?); T ?ulu-k?-i- hair of head; Mu ut, ^|
uri, uli hair, Yy ?otow head. Ma ono head.
41. C '))[ to drink; T xl water; Mu si water.
Ma sewi river. Mi kik water. 42. C Xipi^
arrow; Mu tep-s, temo-x, nemo arrow. 43. C
Iqawe to die; T lohoy-, loho- to die; Mu
laku- dead. 44. C cimsimt to sleep (plur.,
redupl. <*tim-t or reformed from *ir)tim >
*t-ir)tim); Mu eten, enen, etin-i to sleep. Mi
eM, Y ?er)tim, Yy ?entam a sleep-inducing
substance. (Is Esselen acin- after all merely
borrowed from Costanoan?). 45. T ti inter-
rogative enchtic; Mi mau-ti whof hi-ti
whatf Ma -de interrogative, Y ti enclitic,
Mu -sa, -s (< *-ti). 46. C n- /; T -?n I—him,
-n (fut.) I— him; Ma ni /, W ni, Yy na?.
47. T ma thou; Mu me, men, mene. Mi mi,
Yy ma?, W mi, Ma mi thou. 48. C Xtaa
earth (?) (see S [Xa?ay]); T tkaa earth; Ma
kau ground. 49. Ch -k-, -x- my; T ki /, -te-k,
-t-k my, koom we; Mu ka, kan, kana /, Mi
kanni 7, -ka, -ka-n my. 50. C -en numeral
suffix so and so many times; T -n, -(t)an j
adverbial numeral suffix; Mu -na adverbial '
numeral suffix, Ma -nini, Y -d, -1, Yy -? . . . il'.
51. C puux"^- to spout; T phoophaw-,
phoow- to blow; Ma bo to blow. 52. T mena
bear; Y moloy bear. Ma mode brown bear.
53. T maan- to count; Ma -mak- to know,
count, measure. 54. C -enii to do, to make
something; T -(a)n(a)- causative verb suffix;
Mi -ne causative, Y -la, -da causative (for
1 < n, cf. no. 50), Yy -aala. 55. C -n, -ni-,
-eene plural element found in few nouns (e.g.
mee-n human beings, ceey-eene stnall pi.,
qe-ni-xet short pi.) — its irregularity and
rarity show it to be survival from remote
past; T -ha-n, -an noun plural (not very
frequent); Y -n, -in plural of pronouns, -in
numeral collective, W -1 dual of pronouns,
-li-t dual obj., -le-t plur. obj., -li-n dual pos.,
-le-n plur. pos., Ma -nono many (?).
56. C -e, -ii petrified noun plural (e.g.
tummeeX-e old men, timil-ii men, keneeyes-e |
hunchbacks); -a petrified noun plural (e.g. f
aXimaq-a big pi.); Y -i, -a noun plural (for
Seven: Penutian Languages
295
136
Y ablaut, e.g. 'i'onmid: ?onemad-i, cf. C
toomiX: tummeeX-e, knes: keneeyes-e), Yy
-i, -a subject case with plural stems. ^57, C
hinna so and so many each; Y -hin collective
(e.g. yapkan-hin many trees) (perhaps better
to T -ha-n, no. 55). 58. T ^uuk'^- to drink;
Yy "i^ugun- to drink, Mu ukis, Mi uhu. 59. T
ka that (indef.); Y ka that (vis.), Mi i-ka that.
60. C t that there, ta so, such, te- article with
pos. pronoun; T ii-ta (ka) that yonder; Y ta
that (inv.), Yy ta that.
61. C -is noun suffix (often suffixed to
verb stems: Xeey-is language Xee-c he spoke),
-s, -t-s, -en-is nouns of quaUty; T -(a)x
infinitive (often used as verbal noun: S
?ipn-ax speech); Mu -s noun-forming from
verb stems (e.g. ni^i-s language < nm to
speak), W -s verbal noun suf., Mi -s noun
ending (?), Y -0$ noun suf. 62. C -ne it is (e.g.
n-ne it is I); T -(a)n-, -in- petrified intransi-
tive suffix; Y -in- intrans. (?), Mi -ne, -fie in-
transitive verbifying. 63. C ii-ta emphatic
particle; ii-ta(ka) that; Mi i-, i-ni, i-sa, i-mo,
i-ka, i-ti that. 64. T -n noun ending (e.g.
yiwi-n speech < yiw to speak); Ma -n in-
finitive ending. 65. T -a'i'n person, people of;
Yy -? . . . inin people of. 66. C axaax maternal
uncle.; T has- maternal uncle; Y 'i*aga$
mother's brother, Yy "i^aagas. 67. C ^inl^in
eyebrow; Y nimejid eyebrows and lashes.
68. C cneex beard; Y jamo^ beard, Yy
daamut-. 69. C k'*'in-c throat, neck, k'^'in- to
swallow (Coos kw: T k^, cf. k^'ees-is wind, T
k^alt); T k*en- neck; Yy ?oogun, Ma kuyi
neck. ^70. Xpene wings, feathers; Y ^a'i'ada
fur feathers (?).
71. T telk-an buttocks; Y teda anus, Yy
iooto*^. 72. T nihwlk'' black bear; Y du?uxun
bear (<*lu?uxu-). 73. teec lice (see S); T
fi^eld louse (?); Y tehet head louse, Yy ti'i^it.
74. T laap leaf; Y dapdap leaf (<*lap-).
75. T bix-dl moon (?) (x < 5); Y 'i'upiS moon
(cf . sun), Yy "^op. ^76. C t(^een soot; Y neheo
fog. 77. T tk^'d thunder; Y taka'i'a thunder.
78. C lahX-is earth; Y diioiit earth (< *luout).
79. T hok--ai, -h6k-h-al hole (<*-Sok-al);
Y Sogod hole (<*Sogol), Yy sogol. 80. T
tkism- green; Y nii'i'imat green.
81. T -tkSm black; Y qimgutan black.
82. T mii now, then (particle) ; Yy 'i'ama'i* and
then, then. 83. C liixUi to pass out, to pass by,
landing place; Y day (< *lay), lay to step, to
kick. 84. C al-$ toy; T loo-l-, loo- to play,
loo-sf toy; Y do to play (<*lo). 85. C tooh-,
tous- to hit, strike; T t^omom-, toom- to hit,
kill; Y do battle, (?), do5 to beat, overcome
(?), Ma -tul- to break flat thing (?), tup-, tus-
to break (cf. 117). 86. C Xou- to eat; Y duy to
eat (<*\uy). 87. T -(a)l-, -Iha- continuative,
frequentative; Y -ad (<*-al) continuative,
-li frequentative. 88. C maux- to chew (up)
(cf. Ch); Y mok to swallow, Yy meeki-.
^89. C nix(t)- to touch; Yy niti- to squeeze
with hand. 90. C tou- to fall; Y ?ot- to fall.
^91. C t^- to come in; Y tax- to come. 92. C
$anx- to shake (see S); T s'i^elel-, s?eel- to
rattle (?) ; Y ta^it to shake. 93. C ^au- to come
apart, to pull apart; Yy tatay- to break, Y taw
to overcome. 94. T helel-, heel- to sing; Y
'i'uduk-, Yy ^ilik- to sing. 95. C welex- to
stoop, to lie down; Y wodo to dodge ( < *wolo),
Yy wolooyee-. 96. C w^kin- to touch with
stick; Yy woto- to hit with a stick. 97. C
yak"-, yak'^-t- to pick, to gather (see S); Y
yam- to gather seeds, yitw- to gather. 98. T
-koyo-k*^-, -koy-k*!*- to touch, nudge; Yy koyo-
to butt. 99. C xil- to look around (?); T
-xanaw- to look (out) (?); Y Sil'i- to see, Yy
sil'i-. ^100. C xint- to go fast, to run (?); Y
Silit- to jump, Yy siUt-.
101. T wayaan-, way- to sleep; Y woy- to
sleep, Yy woo'^uy-. 102. T mahay large; Y
met, mat, may- large, Yy mayaahay. 103. T
khay- woman, girl; Yy gaa'i'ina woman, girl.
^104. T tola hollow tree; Y ton, toga-S digger
pine. 105. C k6s shell used for ornament; Y
gu^i^ beads. 106. T s^fxi dog; Y neSe^, ^exa,
Ma sii, Mu huhu dog. 107. T hSiiy cloud;
Ma yaa cloud (?). 108. T nf, nii- teats; Ma
mini nipples (dissimilated < *nini?). 109. T
-k?otk'i*at-, -k'i*oot- to break in two; Ma -kot-,
-kut- to divide. 1 10. T kele-k, kel-k- to drill for
fire; Ma -kel- to perforate.
111. C qal- to dig (?); T k'^olol-, kool- to
dig; Ma -kol- to bore (?). 112. T kala-p-,
kal-p- to twist (thread) by rolling ; Ma -kol- to
296
VI American Indian Languages 2
137
roll. 113. T k'^ataay-, k'^'aat- to pick, pluck;
Ma -ket- to graze. 114. T lopap-, loop- to
pound (acorns, seeds); Ma -lop- to move with
friction (?). ^115. C six- to shake off; Ma -sil-
to shake (?) (see no. 92 [in different ink,
showing that comparison with 92 was prob-
ably made after the original entry and after
the entry had been marked with criss-
cross]).
^116. T tama-k?-, tam-k*^- to choke; Ma
-tap- to squeeze (?). 117. T -tk^eltk^al-,
tk'^eel- to break in two; Ma -tala- to crush (?).
118. T t?alal-, taal- to crack; Ma -tala- to
crush (?). 119. C -aami /, we 2, or we-thee,
you 2, or you; T -(i)k-am we (fut. subj. intr.),
-(a)nak-am we (fut. subj. trans.), -am us,
-t-am our; Mi -ma(si) our, -m, -me us, me
(subj. 1 [sic]). 120. T -nk he (fut. trans,
subj.); Mi -k, -ko him, he (subj. 1, including
future), W -k he (past).
121 . T -xi me, I (subj . of passive) ( < *-si) ;
Ma -s(i) /, Mi mu-su, mo-s, mu-$u, $i-ma
I — thee. 122. C -aa'^is thou, ye 2, or ye — me,
u^ 2, or us; W -s-ka, -s-ke-n thou, ye 2, ye
(-ka, -ke- as in Ma -nka-no thou, ye 2, ye,
-nka-s we 2, -nke-s we?). Mi -s thou, -tok-su
ye (subj. 3: present and perfect). 123. T -x
non-agentive (quasi-passive or reflexive),
-al-x intransitive (-x < *-s) ; Ma -us reflexive,
Mu -s-tap, -s-tapse impersonal, passive (?),
Y -wi-5, -wi-s reflexive. Mi -si passive. 124. C
C xeen-is sick; T xil-am sick (< *xin-an) ; Mi
hali sick. 125. T -molo'i'mal-, mol^mal- to
turn things over, .stir food in basket-bucket; Mi
mole to spill.
126. T phili-phal-, philphal- to squash
(insects), to whip (children); Mi pilapa to
pinch (?). 127. T -hit, -thit plural of ad-
jectives; Mi -ti plural of adjectives and
verbs, W -te subjective pronoun plur. 128. T
-k inferential past; Mi -ke, -ka past. 129. T
-sii, -saa agentive (-s- < *-c-); Y -i^, -ic
agent, Yy -ic. 130. C -a-'^eiwat causative
frequentative (-eiwat frequentative), -iya-t
causative, -e-'^et (-a-'i*at) causative passive
(-et, -at passive); Y -i, -u, -a causative.
131. T -(a)pa'^ let us'.; Ma -p6, -pe, -pe'^'e
let us I ^132. C -qm, -xm to be in a condition,
to be in the act of; Mi -imi continuative (?).
133. C -H instrumental noun suffix; Mu -cu
comitative (?). 134. C -et (-at) past passive;
Y -t passive, -n-it future passive. 135. -iiyas
plur. of nouns of relationship, -iiye, -eeye
adjective plur.; Mi -ya noun plur. 136. C han
about to, hanX shall, will; Y hi future particle.
137. C k^'a it seems, as if, hak'*'a-i as if, kind
of; Y akam perhaps, it seems. 138. C hi'i
particle indicating surprise; T his almost,
trying to but in vain; Y hiqa perhaps. 139. C
hei emphasizing particle, hi enclitic em-
phasizer indeed; Ma -hehe only, just. 140. C
qa-1 down, below, under (<qa -f- at, qa- is
local adverbial prefix), qa-lin from under
(<qa- -f- ahn); Y 'i'adid down, low, below
(<*'^alil).
141. C yuu very, very much; Yy yow ami,
also, again. 142. T -ta*^ subordinating suffix
of intransitive aorist; Ma -we-te after having
(for -we- cf. -wea and -weu temporal
suffixes), -ce-te when, while, -ya-tan past
participle (for -ya- cf. -yak temporal suffix).
143. T -na*^ subordinating suffix of inferen-
tial past and transitive aorist; Ma -wono
past participle (-wo- assimilated to -no,
<-we- as above). 144. T -mai* subordinating
suffix of aorist passives, -tat from, to, in —
direction, -t locative (occurs in very few iso-
lated adverbs: k'^cn-t in back, behiml <
k^en- neck, nape, back), -(a)ta local adverbial
suffix; Ma -di in, on, at (?), Mi -ta for, Mu
-ta, -ta-k locative, -ta-5 in, on, at. 146. T
-k'a- comitative verb suffix; M -ko having.
^147. C witin blood; Mi weteti, wetete red.
^148. C -ume where . . . is; Mi -m, -mo loca-
tive, Mu -me with, at the house of, -m, -mo.
149. C -a imperative with object of third
person; Mu -i imperative with object of
third person. 150. C -is, -inis place of; Y -lis
habitual place of (-1- frequentative).
151. T -(a)n aorist passive; Y -han pas-
sive in dependent clauses. 152. See S [-uus
locative]; Y -w locative.
Seven: Penutian Languages 297
Editorial Note
Originally published in InternationalJournal of American Linguistics 19,
132-137 (1953).
Comparative Penutian Glosses
Edited by Victor Golla
Based on "Comparative Penutian Glosses of Sapir,"
edited by Morris Swadesh (1964)
Introduction
It was Sapir's custom to enter linguistic notes, both comparative and descrip-
tive, in the margins of his books and offprints, especially dictionaries and gram-
mars of American Indian languages. In this way, he frequently began various
researches which he pursued throughout his lifetime. One of the most impor-
tant of these was the gathering of comparative data substantiating his theory of
a Penutian stock enlarged to include languages of Oregon, the Plateau, British
Columbia, and Mexico. The broad outline of this theory was first presented in
the introductory paragraphs of "A Characteristic Penutian Form of Stem"
(Sapir 1921b, this volume), but with no comparative lexical data. That Sapir had
what was to him a satisfying amount of such data is clear from a few published
allusions (see especially Sapir 1920c: 265, note 1) and miscellaneous citations in
his letters to Kroeber (see Golla 1984: 201-204, 242-244, 314-315) and other
colleagues. Only a few fragments of Sapir's Penutian data reached print during
his lifetime, the most substantial being the comparisons incorporated into L. S.
Freeland's "The Relationship of Mixe to the Penutian Family" (Freeland 1930,
this volume). Unfortunately, most of Sapir's comparative Penutian working
files seem not to have survived. What these may have consisted of is hinted at in
a passage in a letter from Sapir to Robert Lowie, dated July 6, 1918:
Just at the moment I am carding some of my Penutian-Takelma-Coos-Siuslaw-Chinookan-
Tsimshian correspondences. It is technical work, of course, but quite interesting, as many
lines of historical linguistic research are opened up. Yes, my boy, Tsimshian. Not a bit
isolated. Very specialized in development, but showing clear threads, in my humble and
heterodox opinion, binding it to Oregonian "stocks" (Lowie 1965: 27).
No collation of the sort Sapir refers to in this passage is known to exist among his
surviving papers. A 14-page handwritten list containing 315 sets, later published by
Morris Swadesh as ''Coos-Takelma-Penutian Comparisons" (Sapir and Swadesh
1953, this volume), seems to have been compiled quite early in Sapirs Penutian
research (probably in 1915); it includes no Chinookan, fsimshian. Plateau Penu-
tian, or Mexican Penutian material. Under these circumstances, it is useful to
publish as many as can be retrieved of the marginal jottings that were Sapir's raw
material. Unsystematic and abbreviated as they arc, they nevertheless indicate the
breadth of Sapir's knowledge of the data and the general dimensions of his theory
of Penutian comparative phonology and grammar.
300 VI American Indian Languages 2
The comparisons in sections 1 through 5 below were culled by Morris Swadesh
from several books left to him by Sapir. The work was done in 1953, and Swadesh s
original manuscript is in the Boas Collection of the American Philosophical Soci-
ety (ms. 2U[P1.4]). A decade later, Swadesh published these lists under the title
"Comparative Penutian Glosses of Sapir" (Swadesh 1964). For the present work,
Swadesh s lists were thoroughly checked against the publications annotated
(although the copies with Sapir s notes could not be found), and against the sources
of most of Sapir s comparative glosses. A number of errors in transcription have
been silently corrected. Translations of the glosses — for the most part omitted by
Sapir (or Swadesh) — have been supplied where they are not obvious from the
context. In a few (but far from all) instances where Sapir cited a language without
giving a specific form, it has been possible to identify the form Sapir had in mind.
These and all other editorial interpolations appear in brackets.
Swadesh, both in his 1953 manuscript and in his 1964 publication, cited all forms
in a normalized orthography. To minimize future philological difficulties, the origi-
nal orthographies have been restored wherever possible. This includes all of the
forms in Wintu, Tsimshian, Coos, and Lower Umpqua in the publications that
Sapir annotated. Lacking access to Sapirs original marginalia, I have retained
Swadesh's orthography for these, but in many instances his re transcriptions have
been supplemented (in brackets) by the orthography of Sapir's probable source
(most frequently Sapirs own published work on Takelma). This, unfortunately,
could not be done with Sapir s citations of his still unpublished Chinookan field
data. Swadesh described his normalization conventions as follows (1964: 182-183):
Back velar voiceless spirant is shown by /X/ as against /x/ for the front type , /G/ is a velar voiced
stop, /y is a velar voiced fricative; /i/ is central vowel. /«/ is front-rounded; double letters are
used for long vowels, but length is disregarded in Chinook, where it is non-distinctive. The Ta-
kelma stops are given in the system adopted for Coos-Takelma-Penutian comparisons (Sapir
and Swadesh, 1953): /p t/ and so forth for Sapir s/bd/ and so forth, and also for his p /orp' t' in
final or preconsonantal position. Aspiration, in positions where it is distinctive, is repre-
sented by /h/; glottalization by a separate P/ after the character; /s^/ is phonetic glottalized
(ts). The Takelma tonal accents are interpreted as normal and rising, the latter found only
on long vowels and on vowel-sonant sequences; it is assumed, on the basis of Sapirs treat-
ment of tone, that the sustained (on sequences of "inorganic" vowel plus sonant) and the
rising may be phonemically equivalent. The acute sign is used for the normal accent in its
different forms, and the inverted circumflex is used for rising or sustained long accent.
Comparable phonemicization has not been attempted for the other languages.
The comparisons in the last section (6) are marginalia from Sapirs copy of Franz
Boas's Tsimshian Texts, New Series (1912), which was apparently not available to
Swadesh in 1953. It is now owned by Professor John Dunn of the University of
Oklahoma, who graciously provided the editor with a photocopy of the book
and a list of all annotations in Sapir's hand. This material is presented in the
same format as the comparisons in sections 1-5.
Abbreviations for the languages cited by Sapir, and the sources from which
most of his data were obtained, are as follows:
Tak. = Takelma (Sapir 1909c, 1912h; section references to the latter are given
where appropriate).
Seven: Penutian Languages 301
Coos = Hanis and Miluk Coos (Frachtenberg 1913b, 1914a).
L.U. = Lower Umpqua-Siuslaw (Frachtenberg 1914b, 1917).
Ts. = Tsimshian (Boas 1911, 1912).
Chin. = Chinook (Boas 1911; Lower Chinook unless otherwise indicated).
Wish. = Wishram Chinook (Sapir's fieldnotes).
Yok. = Yokuts (Kroeber 1907).
Miw. = Miwok (Barrett 1908).
L Wintun
The following are comparative glosses found in Sapir's copy of Roland B.
Dixon's "Outline of Wintun Grammar" (1909). Dixon's frequently inaccurate
forms are here supplemented by phonemic forms from Alice Schlichter's Wintu
Dictionary (1981) and Harvey Pitkin's Wintu Grammar (1984). Section refer-
ences to the latter are given in parentheses. Swadesh's normalizations of the
Miwok and Yokuts forms cited by Sapir have not been checked against the
original sources (Barrett 1908 and Kroeber 1907).
-ibi, -be, present [-be-, visual evidential (262.11)]: Miw. -b«s, bi-s,
continuative.
-liba, -libo, future [-le-, inevitable future (241.61) + -ba-, -boh, durative auxil-
iary (262.12)]: Miw. -bo, fut. imperative,
-kila, conditional [-kele, hearsay evidential (243.12)]: cf. Tak. -khi^, -ki^ [-
k'i^, -gi* conditional suffix, §71].
-wu, -u, imperative [-U-, imperative stem formant (222)]: Miw. -pa (?).
-ut, future particle [= ?]: cf. Tak. -tu [?].
-s, forms nouns from verbal stems [-s, general aspect (322)]: cf. Tak. agentive -"^s
[-^s §80] (?).
-wi, interrogative [-(w)u-, 1st person interrogative suffix (241.75)]: Coos -uu [-u
interrogative, Frachtenberg 1914a: 372].
-t, used with adjectives, where the noun is in the subjective case [-t, particular
aspect (320)]: cf. Tak. -t [-f adjectival suffix, §108.1].
-da, first person [-da. 1st person subject (243.21)]: Tak. -the^, -te^ [-t'e^, -de^
§60].
eleu-ibida tconmina T am not dancing' [^elew, negative auxiliary (262.422)]: cf.
Tak. anii^ [an! negative particle, §113.3].
ni hara-kila 'if I go' [har- 'to go, move away from speaker']: cf. Tiik. yana- ('to
go'],
-s, verbal noun [-s, generic aspect (322)]: cf. Coos -is [-is nominalizing suffix,
Frachtenberg 1914a: 365].
bira'ibe-wi 'is he hungry' [bira-- 'to be hungry']: Tak. panx [banx 'hunger'],
tcona-bem 'he kicks' [c'on-, c'una- 'to kick']: cf. Tak. tkuuntkan- [t'gu"nt'gan-
'to kick off'].
302 VI American Indian Languages 2
tcon-pura-ibi-da 'we kick each other' [-p'ure-, reciprocal (241.41)]: Tak. -an-
[reciprocal, §55, being compared here with -ra-].
ya'paitO 'person' [ya-paytu- 'person, white man, guardian spirit']: Tak. yap^a
[yap!a' 'person, people'],
el 'in, into' [^el]: cf. Tak. al- ['face, with eye, to, at', §36.15].
pat 'out' [pat]: Tak. p^ay- [p!a-i- 'down', §37.13].
xan 'away from' [xan]: Tak. hee- [he*^^- 'off, away', §37.3].
panti 'on, upon' [pan 'on top' + -ti, locative suffix (500)]: Tak. pam- 'up' [bam-
'up into air' §37.15].
"Sometimes the locative particle is repeated after the verb, and in some cases two
particles may be used together": cf. Tak. for use before verbs and in postposi-
tion with nouns [§37.93].
kiilun' 'edge's' [kulu 'edge (of basket)' + -n, genitive case]: Tak.-Coos-Maidu-
Yok. type of stem [i.e.,disyllabic, with repeated vowel, the "characteristic Penu-
tian type of stem" discussed in Sapir 1921b].
2. Tsimshian (Boas 1911)
The following are comparative glosses found in Sapir's copy of Franz Boas's
Tsimshian (1911). The page on which a gloss is found is given in brackets. Forms
not marked "Tsimshian" are in the Nass River dialect.
bax- 'up along the ground' [300]: Tak. paa- [ba''- 'up', §37].
y!a§a- 'down along the ground' [300]: Coos qa-yaa-c [qaya^tc 'down-stream'].
ts'Elem- 'into, from the side' [301]: cf. L.U. q-cii [q-tsi 'inside'] (?).
na- 'out of the woods in rear of the houses to the houses' [303]: cf. Tak. noo
[no" 'downriver'], hinaw [hina'" 'upriver']; Coos nooc [?], L.U. tnu 'outside,
outdoors'],
lagauk- 'from the side of the house to the fire' [303]: for-k-cf. L.U. adverbs in
-k.
spl- 'out of water' [304]: Tak. pay- [ba-i- 'out (of house, water)'] (?).
logol- 'under' [305]: Coos qat [qal 'below, under'],
lax- 'to and fro, at both ends' (Nass), lagax- (Tsimshian) [309]: cf. Chin. laX
sa- 'off [309]: Tak. xam- [xam- 'into the water'] (?).
his- 'to do apparently, to pretend to' (Nass), sis- (Tsimshian) [316]: Tak. his
[his, hl's 'nearly, almost, trying', §114.7] (?).
wadi- 'like' (Tsimshian) [317]: cf. Wish. tiwi.
xpl- 'partly' [326]: Coos x- [? discriminative prefix, Frachtenberg 1914a: 324].
hwin- 'innermost part' (Nass), wun- (Tsimshian) [329]: Tak., Chin. [ = ?].
qa- 'location' (Nass), g-i- (Tsimshian) [331]: cf. Coos, L.U. [ = ?].
an-, used to transform verbs into nouns, and express abstract terms, local
terms, and even instruments (Nass), n-, nE- (Tsimshian) [333]: cf. Coos n-
[n- 'in, at, to, on, with'], prob. = a-n- 'it-on', cf. Chin. -n-.
Seven: Penutian Languages 303
yu-..-k" 'one who has' [334]: cf. Tak. comitative -k"^- (-(a)gw- §46].
ha- instrument' [334]: cf. Tak. wa- [wa- instrumental, §38] (?).
iagai 'already, however, rather' (Nass), y!a^ai (Tsimshian) [339]: cf. Coos,
tak. [ = ?].
-En, causative [344]: Tak. -an- [-an- §45].
-sk", expresses primarily the elimination of the object of the transitive verb
(Nass), -sk (Tsimshian) [344]: Tak. -xa-, -x- [-xa- intransitivizing suffix,
§53].
-A, instead of -sk with words ending in p, t, s, ts, q, x, L, and sometimes in 1
(Tsimshian) [346]: Not at all. Evidently suffix has different meaning from -
k^. Cf. Coos -ani (?).
-s, used in Nass and in Tsimshian in place of -k and -tk after k, x, k^, q, and x
[346]: cf. Tak. -x.
-ES, a few times after p in place (?) of -sk, denotes back of object (Tsimshian)
[347]: Tak. -xa-.
-d, indicative of many transitive verbs [347]: cf. Coos [-t, transitive, cf.
Frachtenberg 1914a: 328].
ga" 'to take' (Tsimshian) [348]: Chin. [ = ?].
dzak 'dead', dzak-d-u 'I kill' (Tsimshian) [348]: cf. Coos ck'^i-t-s [tskwits 'he
speared him'],
reduplication of words beginning in hw (Nass), w (Tsimshian) [372]: cf. Coos
(see vocabulary under w-) [Frachtenberg 1913b: 194-195].
smax- 'meat' (Nass), cf. Tsimshian sami 'bear' [377]: Tak. xamk [xam'k'
'grizzly bear'],
ie' 'to go (sg.)' [381]: cf. Tak. [?yana- 'to go']. Chin. [ = ?].
L6 'to go (pi.)' [381]: cf. Coos [ = ?].
gaksk 'to wake up (sg.)' (Tsimshian) [381]: Tak. [ = ?] (?).
-Em 1st pi. objective (Tsimshian) [381, 384]: cf. Tak. [-am, §62].
awa'" 'proximity', awa'"t 'near him (his proximity)' (Tsimshian) [393]: Tak.
waa-ta [wa'^-(da) 'to, at-him'].
gwa" 'that' (Tsimshian) [394]: cf. Chin. [ = ?].
-1, demonstrative element (Tsimshian) [395, 406]: cf. Tak., Chin., Coos,
nda 'where' [406]: Chin, ta-n, sa-n, Tak. ne'k-ti [nek'di 'who?', §105].
-L, imperative of transitive, indef. obj. (Nass), -\ (Tsimshian) [407]: cf.
Coos 1.
dzE, tsE, weakens statements [408]: Wish, -ci^, Coos na-cii [natsi, Frachten-
berg 1914a: 387], ci [tsi, restrictive particle, Frachtenberg 1914a: 394].
optsE 'else, lest' [408]: Wish. pu.
ami 'if (event assumed as not likely to happen) (Tsimshian) [409]: Tak.
mii^wa [ml'^wa 'probably, perhaps'].
3. Coos (Frachtenberg Texts)
The following are comparative glosses found in the vocabulary section of
Sapir's copy of Leo Frachtenberg's Coos Texts (1913b).
304 VI American Indian Languages 2
a'lEC 'toy': Tak. lou-si [lo"-s-i].
a'lqas *fear": Tak. hin^x [hin^x] (?).
alts- 'to be in the wrong place, to be mistaken': Tak. iils^-ak^ [i'lts!-ak'^ 'bad'].
ai^- 'to kill (pi. object), to take away': Tak. ey-yi- [ei[y]-i-].
a"q- 'to take off: L.U. aaq- [aq-].
e« 'thou': Tak. [see Section 4 below].
e^natc 'mother': Tak. ni- [ni-].
is 'we two (inclusive)': L.U. -ns [-ns].
ix- 'canoe': Tak. ey [ei].
ilx- 'to look': Tak. liw- [liw-].
ilxa'ka 'they': Tak. [see Section 4 below].
itE, emphatic particle: Tak. [cf. hi (§114)].
in, negation: Tak. [see Section 4 below].
I'n-ta 'not so, bad': Tak. [cf. hit (hiit) 'no' (§113)].
I'l- 'to tell, to say, to send': Tak. hiim-t- [hi'm-d- 'to talk to'] (?).
yEq- 'to go away': Tak. yew- [yeu- 'to go back, return'] (?).
yat- 'to coax, to persuade': Tak. yamat- [yamat- 'to ask (tr.)'] (?).
yi'pSEn 'three': Tak. xipini [xib'ni].
yi'qa 'nevertheless, right away': Tak. [ = ?].
yux"- 'to rub': Tak. yuluyal- [yuluyal-].
yuwil- 'to divide': Tak. yowo-, yows*^- [yowo-, yo"^s- 'to start (when startled)'
(?)•
yut- 'to tear off: Tak. yonk^- [yonk!- 'to pull forcibly'] (?).
waha'^tcas 'sickness': Tak. hawax [hawax 'rottenness, pus, foul odor'] (?).
wa'lwal 'knife': Tak. wiil-ii [wi'li' '(stone) knife'], waya [waya 'knife'],
wa'wa 'little girl': Tak. waiwii [waiwi' 'girl, female'],
weste'n 'so many times': Tak. [ = ?].
wint 'forehead': Tak. «l«k^-i- [ii'liikl-i- 'head-hair'] (?).
wispa'ya 'arrow': Tak. wilaw [wilau] (?).
wif 'to look far, to search': Tak. wow-lt- [wo"-ld- 'to go far, go to get'].
weL-, wIL- 'to twist': Tak. wiik*^- [wi'k!- 'to put around (neck, head)'],
wl'n- 'to cheat': Tak. wiim-at- [wi'm-ad- 'to exercise supernatural power upon']
ha'yati 'to lie down (pi.)': Tak. wayaan-, way- [waya^n-, wai- 'to sleep'],
ha^- 'to think, to imagine': Tak. hewe-haw [hewe-haw-].
han, temporal particle: Tak. [ = ?].
han- 'to wrestle': Tak. hemem- [hemem-].
hant- 'to pick out, to choose': Tak. henee-t- [hene'^-d- 'to wait for'] (?).
hak- 'to crawl': Tak. waka-xa- [wage-xa- 'to climb up'].
hala'qi:s,hali'qas 'relatives of husband': Tak. wayaw- [wayau- 'daughter-in-law']
halt! 'now': Chin. alta.
halk"- 'to take off: Tak. weet-k- [we^t'-g- 'to take away from, deprive of] (?).
halq-, helaq 'to climb up': Tak. hilw- [hilw-].
halq 'tir-tree': Tak. x6 [x6] (?).
he, temporal particle: Tak. [see Section 4 below].
Seven: Penutian Languages 305
heilta 'tongue': Tak. ela- [ela-].
hen, particle denoting hearsay: Tak. [ = ?].
henl- 'a while, a long time': Tak. [see Section 4 below].
he'lmi 'tomorrow': Tak. [see Section 4 below].
helq-, he'laq 'to arrive': Tak. wowk- [wo"g-].
hitc, particle denoting surprise: Tak. [see Section 4 below].
hrk!am 'monthly courses': Tak. wtHilh-am- [wii^lk-am- 'to have first menstrual
courses'] (?).
hu"'mik- 'old woman': Tak. wuun- [wu"n- 'to be, grow old'] (?).
pEnL- 'to tear off, to come off: Tak. potpat- [bot'bad- 'to pull out (somebody's)
hair'],
pa^w- 'to smoke (a pipe)': Tak. phoy-amt- [p'oy-amd- 'to smoke out (wasps)']
pa^- 'to fill': Tak. ptwk^- [bii'^k!-]. Chin. paal-.
pils- 'to tear, to smash to pieces': Tak. philiphal- [p'ilip'al- 'to squash, whip'].
pilx"- 'to break, to crush': Tak. pilw- [bilw- 'to jump (at), fight with'] (?).
pidj- 'to come up (from water), to appear': Tak. piis- [bi's- 'to look up, lift up
one's head' (used only in myths)].
pux"- 'to spout': Tak. phowphaw-i- [p'oyp'aw-i- 'to blow'].
mEanl'yas 'parents, grown-up people': Tak. ma- [ma- 'father'].
maha- 'to watch, to look after': cf. Hokan.
ma'qaL 'crow': Tak. mel [mel].
ma'luk" '(Indian) red paint': Tak. maanx [manx 'white paint'] (?).
ma^L! 'flood': cf. Chin. maX-ni.
ma'x" 'vulva': Tak. min- [min- 'vagina'] (?).
mitsl'le 'pregnant': Tak. mix^ii- [mAhwI-].
mi^la 'liver, waist': Tak. p^an [p!an 'liver'].
mi'iaq 'arrow': Tak. smelaw'^x [smela"^x].
ml'naqas 'pole': Tak. mal [mal 'salmon-spear shaft'].
da'mil 'strong, male being, husband': Tak. tomxaw [domxau 'big crooked-nosed
salmon'].
drl 'something': Tak. al-til [al-dil 'all'].
tE 'that there': Tak. [ = ?].
tEW- 'to be in upright position': Tak. t^epe-, teep- [t!ebe-,de<-"b- 'to get up'].
ta, conjunction: Tak. [ = ?].
tat'n- 'to fix, to divide': Tak. t'-^ilii-, tiln- [t!ili-, dlln- 'to distribute to'].
te'lex- 'crosspiece': Tak. tele-p- [dele-b-] 'to stick into', taia-k- [dala-g- to pierce
nose, ears'].
ti'k- 'to stand': Tak. tiiktak- [di'k'dag- 'to erect, cause to stand up'j.
ti'lpl 'gopher': Tak. thiis |t'i's].
tlyet- 'to store up food': t««lt^al- [dult!al- 'to stuff (basket) with'].
tl^- 'to coil': Tak. t^uukuuy-, ttmk^- [tlugui-, dii^gw- 'to wear (garment)'].
ti'ntc 'remnants of meal': Tak. t'-^ayay-, taay- [t!ayai-, dai- 'to go to get some-
thing to eat'] (?).
toh- 'to hit, to strike': Tak. toy-k^ [dui-k!- 'to push'] (?).
306 VI American Indian Languages 2
to"s- "to hit, to strike': Tak. t^omom-, toom- [tlomom-, do"m- 'to kill'],
tkwii'tuk" 'shoulder', t^kwa'tkwis 'elbow': Tak. tk^iintk^- [t'gwi'nt'gw- 'upper
arm'].
tEk-elm- 'to dive, to sink': Tak. tkeP [t'gel^ 'drop, fall'].
tqanL- 'to hit, to strike (with instrument)': Tak. tk^el-tk^al- [t'gwelt'gwal- 'to
break in two'].
tqa'L- 'to put on a belt': Tak. tkenc*?- [t'gents!- 'to put about one's middle'].
tEq!e''^n 'soot': Tak. sin [s-In 'wood-coals'] (?).
t!c- 'to shove, to push': Tak. t^osot^as [t!os-ot!as- 'to walk around'].
t!kw- 'to kick': Tak. tkuuntkan- [t'gu"nt'gan-].
t!kwa 'roof: Tak. tkuup- (t'gu^b- 'to cover over, put lid on'],
nl'kla 'mother (vocative)': Tak. ni- ['mother'].
s^aL! 'pitch': Tak. seel [se'^l 'black paint, writing'], k^^al [klwal' 'pitch'] (?).
sin! 'hair on penis': Tak. seen- [sen- 'hair'],
sil- 'to drop (of liquids)': Tak. tk^il-i- [t'gwili-] (?).
so^t- 'to trade, to exchange': Tak. -xoot- [-xo"d-] (?).
sto"q- 'to stand, to be in an upright position': cf. Tak. [ = ?] (?).
sq- 'to seize, to take': Chin, -sk- (?).
sla '(male) friend, cousin': Tak. snaa [sna 'momma!' (vocative)] (?).
ci- 'to drink': cf. Yana si-,
canx- 'to shake': Tak. s'^^el- [ts!el- 'to rattle'].
cima- 'to draw by means of breath': Tak. s^usm- [ts-!usm- 'to chirp'],
cl'tla 'pet': Tak. c^ixi [ts-!u'xi] 'dog'.
cuL- 'to set fire to': Tak. s^«l-k^- [ts-!ul-k!-] (?).
tSE7i- 'to crush': Tak. c'^atat-, saat- [ts-!adad-, sa^d- 'to mash'] (?).
tSEL- 'to stand, to lie side by side': Tak. xoxok^- [-xoxogw-'to string salmon'] (?).
tsa^ntc 'on the back' (borrowed from the Siuslaw tca^'n- 'to lean backwards, to
recline'): Tak. xaah-am- [xa-^h-am- 'back, waist'] (?).
tsn- 'to stretch': Tak. tin-k^- [din-k!-] (?).
tsk"- 'to hit against, to strike against, to spear': Tak. saak^- [sa^gw- 'to shoot
(arrow), paddle (canoe)'],
tsx- 'to take': Chin, -sk- (?).
tsqal- 'to defecate': Tak. xalaxam- [xalaxam- 'to urinate'] (?).
tshm 'summer': Tak. sama- [sama-].
tci'ntsan 'friend': Tak. k^hinax- [k'winax- 'kinsman, relative']. Upper Tak.
k^htHtnakst [k'ii'^naks-t' 'his relatives'],
tcils 'penis': Tak. khal [k'al].
tco"-, tco^" 'to jump': soo-'^k- [s-6"^k'-].
tcut 'nose': Tak. sin- [s-in-].
tcH 'raft': Tak. sil [sil 'canoe' (in verbs)].
tc!a- 'to walk': Tak. s'^aak- [ts!a^g- 'to step'].
tc!ici'lis 'sweet', tc!ical- 'to be sweet': Chin. [ = ?].
tc!il- 'to burn': Tak. -s^«l-k^- [-ts!ul-k!- 'to set fire to'].
tc!iltc! 'hammer': Tak. s'^al-s'^al- [ts!alts!al- 'to chew'].
tc!6"- 'to lie down, to go to bed': Tak. s'^oot- [ts!6"d- 'to touch, strike against'].
Seven: Penutian Languages 307
kwE'li7Es 'intestines, a person created from the intestines': Tak. k^aas [gwas
'entrails'].
kwe'ik- 'girl', kwe's 'young woman, girl': Tak. khay^- [k'ai^-].
kwil-, k!"hil- 'to burn': Tak. k«l-k^- [giil-k!- 'to blaze, glow'], of. Tak. k'winax
: k'ii'^nax [for phonological detail].
kwit- 'to leave': Tak. k*?otot-, koot- [k!odod-, go"d- 'to bury'].
kwiL- 'to dig (clams)': Tak. k^olol-, kool- [k!olol-, go"l- 'to dig'].
kxla 'foot, leg': Tak. sal- [s-al- 'foot'], k^el- [gwel- 'leg'] (?).
k!a 'rope': Tak. k^ank^an- [k!ank!an- 'to twist (hazel switch)'].
k!a' 'to listen, to keep quiet': Tak. skek^ii- [sgek!i'- 'to listen'].
k!walxa'ya 'butter-ball' [duck sp.]: Wish. [ = ?].
kiwa'sis 'wind': Tak. k^alt [gwaPt'].
k!wehe 'willow': Tak. k^^aay [k!wal 'grass'].
k!wints 'throat, neck', k!win- 'to swallow': Tak. k^en- [gwen- 'neck, nape'].
g-a'we 'sea otter': Tak. tkam [t'gam 'elk (hide, hide armor)'] (?).
k-i'nwis 'lazy', k-i'ria" 'tired': Tak. kenaw- [genaw- 'lie curled up dog-
fashion'].
k-ow- 'to munch, to pick and eat': Tak. k^owoo-, kuuw- [k!owo"-, gu"w- 'to
throw a mass of small objects'].
k-!em- 'to practice': Tak. k^emn- [k!emn- 'to make, treat as'].
k-!si'lis 'green': Tak. tkism- [t'gis-m-] (?).
k-ele 'shouts': Tak. skelw- [sgelw- 'to shout'].
qano'tc 'outside': Tak. noo^s [n6"^s 'next door'].
qal- 'to dig': Tak. k^olol-, kool- [k!olol-, go"l- 'to dig'], -holohal- [-holohal- 'to
dig into (ashes)'] (?).
qal- 'to cry (sing.)': cf. Chin. [ = ?].
qa'ya 'breath': Tak. hekehak- [hegehag- 'to breathe'] (?).
qa'yis 'sky, day, world': Tak. haay [hai 'cloud'] (?).
qai'na 'cold': Tak. tkuun-p- [t'gu"n-p'-] (?).
qa"'wa 'evening, night', qa"'m'(tc) 'evening': Tak. hoo-xa [ho"xa 'yesterday',
§112.2].
qEto" 'to be in suspended position, to hang': Tak. xataxat-na- [xadaxat'-na- 'to
hang up in a row'].
q!al- 'to take out': Tak. k^^alk^- [kiwalgw- 'to let alone'].
q!a'na 'young': Tak. k^^al-thaa [k!wal-t'a'' 'youngest (of two or more)'].
q!e}e 'pitchwood': Tak. k^^al [k!war].
q!m- 'to eat, to cook': Tak. k^uum-an- [k!u^'m-an- 'to fix, prepare'] (?).
xa'yusLatc 'relatives by marriage after death of person causing that
kinship': Tak. ximni- [ximni-].
xan- 'to be sick, to be sorry', xii'nis 'sick': Tak. xilam [xil-am 'sick, dead per-
son, ghost'].
xa'ka 'he, she, it': Tak. haa^ka [ha'^^ga 'that yonder', §104] [see Section 4
below].
xo'xwel 'frog': cf. L. Chin. [ = ?].
x"kwi'natc 'maternal aunt': Tak. xaka- [xaga-].
308 VI American Indian Languages 2
xwelap 'lungs': cf. L. Chin. [ = ?].
xwi'nLis 'snot': Tak. xin [xin].
xwi'lux" 'head': Tak. xuul-i- [xu"l-i- 'brains'] (?). [See different comparison in
Section 4 below.]
xqas 'white': Tak. tk«y^s [-t'gu'^s-].
xlls 'slime': Tak. xla^px [xle^p-x 'roundish dough-like cake of deer-fat or
camass'].
X il- 'to look around': Tak. xanan- [xanan- 'to look out (pi.)'] (?).
Eqa"'we 'to die (sing.)': Tak. loho- [loho- 'to die'].
E'71 'good, nice': Tak. tuu [dii] (?).
aa- 'to put around': Tak. laat- [la^d- 'to put (belt) around waist'],
atclya- 'to call by name': Tak. laalaw- [la^law- 'to name, call'],
aix"- 'to jab': Tak. laaw-t-an- [lau-d-an- 'to hurt'],
ewi 'it is, that is': Tak. laaHi- [la-^'l-i'- 'to become'],
em- 'to be in upright position, to stand (of inanimate objects)': Tak. lemk^-
[lemk!- '(people) move, go, to take along (pi. obj.)'] [see lim- below],
e'xalx 'string': Tak. laaw- [la^w- 'to twine (basket)'],
e'xum 'buzzard': Tak. moxo [moxo].
ic- 'to shake, move (intransitive)': Chin. -la.
icla7- 'to swing': Tak. lewe^law- [lewe^iaw- 'to swing (shells) in one's ear'],
smilismal- [- smilismal- 'to swing'] (?).
i'kwit 'feathers': Tak. xliwi [xliwi 'feathers worn in war-dance'] (?).
^x-il- 'to like, to love': Tak. miil- [mlli'-d- (?).
, abbreviated form of dl4 ['something']: Chin. t-.
a- 'to go': Tak. yaan- [ya'^n-] (?).
im- 'to put inside (pi. object)': Tak. lemek^- [lemek!- '(people) move, go, to
take along (pi. object)'] (?) [see lem- above],
^yuwil- 'to move, to wiggle': cf. Ts. [ = ?].
h- 'to get weir, the- (the-) 'to rest': Tak. likii-n- [ligi'-n- 'to rest'],
n'nas 'name': cf. L.U. hin [tin],
k"- 'to sew': Tak. lep-t- [lep'-d-] (?).
kwilt'red': Tak. al-s^iil [al-ts!il] (?).
qalk"- 'to bite': Tak. lek^ei- [legwel- 'to suck'].
Lala^- 'to bark, to growl, to shout, to wail': Tak. khewekhaw-al- [k'ewet'aw-al-
'to bark at'] (?).
L'^va 'son (vocative)': Tak. khapa- [k'aba- 'son'] (?).
Loc 'clam': Tak. kos [gos- 'clamshell'] (?).
Lkwa^- 'to cut off: Tak. skoot- [sg6"d- 'to cut'] (?).
Lx-- 'to drift (away)': Tak. thiyii- [t'iyi'- 'to float'].
L!ta 'land, earth, country, ground, place': Tak. tkaa [t'ga].
L!ka- 'to string': Tak. tpaak- [t'ba^g- 'to tie up (hair, sinew)'].
L!kw- 'to cover up': Tak. tpook-tpak- [t'bo"k't'bag- 'to put away'].
Seven: Penutian Languages 309
L!k-- 'to pour, to spilK: Tak. tkiiy-al-x- (t'gi'y-al-x- 'tears roil down ones face'].
L!le- 'to come out (from water)': Tak. theek- [t'e'^g-].
4. Coos (Frachtenberg Grammar)
The following are the comparative glosses found in Sapir's copy of Leo
Frachtenberg's Coos (1914a), omitting comparisons duplicating those in his
copy of Coos Texts (Section 3, above). The page on which a gloss is found is
given in brackets.
he'niye 'a while', he'nihen 'many times' [313]: Tak. he^ne [he'^ne' 'then',
§112.3], hem-ti [hemdi 'when', §113.2].
In, negative particle [314]: Tak. anii^ [anP §113.3].
"Hiatus .... Broadly speaking, it may be said that the coming-together of two
vowels of like quantities and qualities is avoided by means of infixing a weak h
between them" [314]: cf. Tak. ["inorganic"] -h- [§24].
helmi'his 'next day' [314]: Tak. te-wenxa [dewenxa 'tomorrow', §112.2].
xwi'lux" 'head' [316]: Tak. «l«k^i- [ii'liikl-i- 'head-hair'] [see different com-
parison in Section 3].
n-, 1st person [321]: Tak. -a^n [-a^n 1st person sg. transitive subject (aorist)],
-an [-an (future), §63].
e*-, 2nd person [321]: Tak. -te'^, -'^t [-de^, -'^t', 2nd person sg. possessive, §90].
nkwa'xLa 'with bows' [322]: Tak. kaP [gal^ 'bow, gun'].
mitsT'ltfye 'become pregnant' [323]: Tak. max^ii^ [mAhwi^] ? = mix^ii^ [see
also Section 3].
ntcla'ha '[equipped] with walkers [i.e., legs, feet]' [323]: Tak. s^aaks^ak-
[ts!a^k'ts!ag- 'to step'],
-t, transitive [327]: Tak. -s-, from -tx- [indirective, §47].
-ts, transitive [327, 329]: Tak. -s-, from -t-s.
-I, e', neutral [ = mediopassive] suffix [327, 334]: Tak. -ii [-1, suffix forming verbs
of position, §57].
-s, general nominal suffix [328]: Tak. -s [§87.8] (?).
-aai, intransitive [332]: cf. Tak. -xa- [§53].
yoyo^waai 'is stopping' [332, 364]: cf. Tak. yo-, yowo- ['be sitting'] (?).
-me", reciprocal [332]: cf. Tak. -an- [§55].
-ts-XEm, -t-XEm, reflexive [333]: cf. Tak. -k^i-, -k^i- [-gwi-, -k'wi- §54].
laatsxEm 'put herself in' [333]: cf. Tak. las^ak-, lask- [lats!ag-, lasg- 'to touch'].
-e'wa(t), -o"wa(t), frequentative [336]: cf. Tak. -eeha, -iiha [-eha, -Iha §43.6].
nhaml'yat 'I brought it out' [340]: Tak. hemek- [hemeg- 'to take out, off'],
-ani, transitive [341]: Tak. -an- [causative, §45].
310 VI American Indian Languages 2
tsak"kw-, -tskw- 'to spear' [341]: Tak. saak^- [sa^gw- 'to shoot (arrow), paddle
(canoe)'] [see Section 3].
tqa'nL- 'to strike' [341]: Tak. saansan- [sa^nsan- 'to fight'] [see Section 3 for dif-
ferent comparison].
\\ tqanLLii'nl 'they mutually strike one another' [341]: Tak. sana-x-in-iyan-
[sana-x-in-iyan-].
wil- 'to fight' [342]: Tak. te-wiliw-alt- [de-wiliw-ald- 'to fight with, "go" for'
(< wiliw- 'go, proceed, run')] (?).
-ayu, -e'yu, -lyu, (past) passive [327, 344]: Tak. -yaw- [-iau-, impersonal, §58].
-aya", past participle [347]: Tak. -lya [§83] (?).
-a, in formation of past participle [347]: from-ii, [-1] cf. L.U. and umlauting force
in Coos,
k-ela 'hand' [348]: Chin, -ksi, -ksn (?).
te*L 'this' [348]: Chin, ti-ka.
-It, suffixed to verbs that are transitivized by means of the transitive suffix -aya, 3
pers. obj. [348]: cf. Tak. [ = ?].
-enl, verbalizing suffix, 'to do, make something' [349]: Tak. causative -an- [§45].
-a, transitive verbs [354]: cf. Tak. [§44-51] (?).
-is, nominalizing suffix [360]: Tak. -(a)x, [intransitive] infinitive [§74].
-onis, -si, verbal noun formants [363]: cf. Tak. passive [(aorist)] -an [§66].
L!x-T'nt 'he examined it' [364]: Tak. xiik- [xi'g 'to see'] (?).
e^niitc 'mother', ni'kla 'mother!', vocative [366]: Tak. ni- [ni-(xa) 'his mother'], -
hin [(wi-)hln '(my) mother'],
teka'^tsi 'granddaughter!', vocative [366]: cf. Chin. [ = ?].
a'tatc 'paternal aunt' [366]: Tak. that- [t'ad-].
mi'nkatc 'son-in-law' [366]: Tak. mot-, moo- [mot', mo"-] (?).
-ex, -lyEX, -lyetEX 'pertaining to . . .' [367]: Tak. -^iixi [-^r'xi 'belonging to',
§108.10].
yiqantcimex mii 'the last [ = previous] generation' (< yiqantc 'behind')
[367]: Tak. -imik^i [-imik!i, suffix forming adjectives from temporal adverbs,
§108.7].
slaatc 'cousin' [371]: cf. Tak. snaa [s-na 'momma!' vocative, §91] (?).
-en, multiphcative suffix, 'times' (in numerals) [373]: Tak. -an, -tan [-an, -dan
§111].
hl''me 'children' [374]: Tak. haap- [ha'-'p'- 'small child'],
ma 'human being', men, plural [374]: Tak. -(h)an [§99].
L!ta'yas 'village', derived from L!ta 'earth, ground, country' [375]: ? from
X'taay- [L!tay-] (cf. L.U. X'ay [L!a'»'] 'place, world, people', X'ayuus [L!ayu's],
locative form) + -as. [See Section 3 for Takelma comparison],
hu^mi'k-ca 'dear old woman' [375]: cf. Chin, wi-na-m-s wa-n-aq-s.
-I, found suffixed to the article, expressing instrumentality [377]: cf. Ts. -ii (-1)
(?)•
final reduplication [380]: cf. Tak. [§30].
tco'xtcox 'rabbit' [381]: Tak. hoow [hou] (?).
he 'usually, frequently, habitually' [384-5]: Tak. -wi^ [-wi'^ 'every, §101] (?).
Seven: Penutian Languages 311
c*^, expresses slight surprise at a state of affairs that has come into existence con-
trary to one's expectations [389]: cf. Chin. s-.
hitc, indicates surprise [391]: Tak. his [his 'nearly, almost, trying', §114.7].
L 'must, necessarily' (exhortative particle) [392]: Chin. XX (?).
hamiL, mIL, iL 'let me, I should like to, better, you may, please, a while' (exhor-
tative particle) [392]: Ts. amii [ami], Tak. mii^wa [ml'^wa 'probably, perhaps',
§114.6].
n'ne, 1st person (independent pronoun) [396]: L.U. -n.
e^ne, 2nd person [396]: L.U. -nX [-nx].
i'sne, dual inclusive [396]: L.U. -ns.
i'cne, 2nd person dual [396]: L.U. -c [-ts].
xwin'ne, dual exclusive [396]: L.U. -Xun [-xiin].
lin'ne, 1st person plural [396]: L.U. -nl, inclusive, -nXan [-nxan], exclusive.
cin'ne, 2nd person plural [396]: L.U. -ci [-tci].
itxa, 3rd person plural [396]: cf. Chin. 1- (?).
ntet, 1st person singular reflexive (cf. tet 'body') [400]: Tak. -ta^x [-da^x
restrictive suffix, "denoting the isolation of the person," e.g. 'only I', §103].
la", ha" 'he, it is' [401]: Tak. haa^- [ha'^^ demonstrative stem, §104].
lewi 'it is' [402]: Tak. laalii- [la^l-i'- 'to become'].
-tc, adverbial suffix of modality [405]: Tak. -xa, -x [adverbial suffix, §112.2].
i 'when, as, since, while' [409]: Tak. isi^ [T's-i^ 'despite, although, even if,
§114.4].
a'watu 'whether or not' [411]: Wish, awa-ci^ 'or'.
nouns as qualifiers [412]: cf. Tak. [§88].
k!al- 'to shout' [413, 416]: Tak. skelw-, skelew- [sgelw-, sgelew-].
cx-Iml 'bear' [415]: Tak. xamk [xam'k' 'grizzly bear'].
5. Siuslaw (Frachtenberg Grammar)
The following are comparative glosses found in Sapir s copy of Leo Frachten-
berg's Siuslawan (Lower Umpqua) (1917). The page on which a gloss is found is
given in brackets.
nx-l-k becomes nak [446]: cf. Wish. -tX-k becomes -t-k.
winx- 'to be afraid' [446]: cf. Tak. hin^x [hin^x 'fear'].
-Imts, indirect object of first person imperative [501]: cf. Tak. -xi [?].
Jan- 'to call by name' [514]: Tak. laalaw- [la''law-].
wihc- 'to send' [520]: cf. Coos will- [wil- 'to look for, to search'] (?).
-t, present tense [527]: cf. Wish. [ = ?].
mate- 'to be in a horizontal position', ma-'tc- 'to lay' [536]: cf. Tak. mas'^ak-,
mask- [mats!ag-, masg- 'to put'].
-a, nominal object of an action, also the local idea of rest [ = accusative case-end-
ing] [541]: cf. Yokuts -a.
312 yi American Indian Languages 2
-a, modal adverbs [557]: cf. Tsim. adv. -a.
qa'wl 'blood' [560]: cf. Chin. [ = ?].
-a^mu, nominalizing suffix indicating place [563]: cf. Coos -ame.
-t, adjectival suffix [564]: cf. Tak. -t [-(i)t' §108.1].
tsami'tsEm 'chin' [565]: Coos cneex [ts'na'x] 'beard'.
-wi, suffix found in a small number of nouns [565]: cf. Coos -eyeewe [-eyawe,
noun of agency, Frachtenberg 1914a: 364].
tsxan- 'to comb one's hair' [565]: cf. Chin. [ = ?].
duplication of final consonants [567]: cf. Tak. [§30].
tcaq- 'to spear' [568]: Tak. saak^- [sa^gw- 'to shoot (arrow), paddle (canoe)'],
hlq" 'wildcat' [570]: Tak. yaak^ [yak'^].
stem amplification [573]: equivalent to Tak. aorist verb stem [§39-40].
tqut- 'to shout' [575]: Tak. skelel- [sgelew- 'shout', sgelel- 'keep shouting'] (?).
xa'ts!un 'four' [586]: = 2 x (2), -n cf. Tak. -an, -tan [- an, -dan multiplicative
suffix, §111].
qo'x"m 'offshore, out in the water' [589]: Tak. xam- ['in river, into water', §37.16]
stim, stimk 'there' [589]: cf. Ts. [ = ?].
aldu 'Hkewise' [592]: Tak. altii [aldl 'all'].
"I 'and' [592]: Coos il [il 'surely, indeed'], Tak. altii-1, -tiil [aldI-1, -dil 'all,
everything'].
aL 'now' [593]: L. Chin, alta, alqa.
hanhan 'indeed, to be sure' [594]: Coos han [temporal particle],
na, particle of interrogation [599]: L. Chin. na.
6. Tsimshian (Boas Texts)
The following are comparative glosses found in the vocabulary section (pp.
256-284) of Sapir's copy of Franz Boas's Tsimshian Texts, New Series (1912).
aya- 'successfully': Chin. aiaq.
awa' 'proximity': Tak. wa- ['to, together with'].
ap 'bee': Wish, wa-ba.
am- 'only': Wish, nai-ma, L. Ch. nam-ka.
ami 'if: cf. Tak. mT''wa ['probably, perhaps'] (?).
amuks 'to listen': (cf. mu 'ear'?).
al 'but': L.U. "1 ['then, so, and', introductory particle] (?), Coos il ['surely,
indeed'].
a'Mks 'servant': cf. Kwakiutl Elk",
at 'not' (in interrogative sentences): cf. Na-dene.
emx 'beard': Chin, -miqcu.
Seven: Penutian Languages 313
ya° 'to go, sing.': Chin, -ya, Tak. yana-.
y!an 'excrement': Tak. xa^n- 'urine' (?).
yaltk, pi. yilyaltk 'to return': Tak. yew-.
yei, pi. yikyei 'fat': cf. L.U. [= ?].
y!u'«ta 'man': L.U. hit-c (?).
wak- 'brother', waik- 'elder brother': Tak. wa^-, Chin. awi.
walp, pi. huwalp 'house': Tak. wih.
wa 'to invite': Tak. wo"- ['to go for, go to get'] (?).
woq 'to sleep': Tak. wai-, Yok. woi.
wun- 'innermost part', wun-^aus 'brain' (cf. gaus 'head'): Chin, -wan, Tak.
ha-^win-i-(de) ['inside of (me)'],
wul 'being': Tak. wala' ['really, come to find out'] (?).
wuir^l, pi. wulwuir^l 'to rub': Tak. yul-, final red[upIication].
hau 'to say': ? < *waw-.
haowi'n 'before': Tak. ha'wi- ['still, yet', §113.2] (?).
hats! 'hardly, now': demon, ha- (cf. Coos [he 'the', ha 'his, her, its']),
ha^x 'goose': cf. Tak. [ha^k'a^].
haldEm- 'upward': cf. lak. hilw- ['to climb'].
hi- 'beginning': L.U. hlq!- ['to begin'],
ha 'to shout': L.U. hal-.
han 'salmon' . . . SE-wI-ha'n 'twins ( = making plentiful)': cf. Nootka.
hum 'to smell something': cf. Chin.
hu^t 'to escape, pi.': Tak. ho"gw- ['to run (without expressed goal of motion)'].
hu°tk, pi. hukhu'ot'^ 'to call, summon': L.U. hant'- .
huk- 'expert' (par.): Tak. yok'y- ['to know'] (?).
bEn, pi. §a-bE'n 'belly': Chin, -wan (?).
bEla'x 'moss': Tak. blls.
balax, pi. bilba'lax 'ghost': Tak. bilam ['having nothing, empty'] (?).
pe 'liver': cf. Tak. p!an.
bu", pi. gabu'o 'to blow': cf. Tak., Coos, Chin,
pha'r 'to relate': Tak. p!al-g- ['to tell a myth to'] (?), L.U. pulk-na- ['to speak']
p!El-mu' 'ear-ornament': cf. p!al 'button'.
p!ia'n 'smoke', x-p!ia'n 'to smoke ( = to enjoy smoke)': cf. Southern Paiute.
m 'thou (transitive subject)': cf. Chin., Tak.
mES- 'reddish' . . . mEsi'^n 'copper': cf. Hokan, Algonquian.
mEsx 'breast': Chin, -mxtc 'heart'.
m!an- 'up through the air': cf. Hokan 'up, sky'.
mag 'to put down one object', mak-sk 'to put down several objects': cf. Tak.
[masg- 'to put'].
ma'Mk 'to burn': cf. Tak. [me^l- 'to blaze'],
mat, pi. mElmal 'to tell': cf. Tak. [malg- 'to tell, speak to'],
mu 'ear': Hokan?
du'la, pi. ga-du'la 'tongue': cf. Wintun.
tgi- 'down through the air': Tak. t'gi'y- ['tears roll down one's face'].
314 Vf American Indian Languages 2
tgin 'to drill fire, [t]gu- 'around': Tak. t'gei- ['to roll, put around'].
txa 'skin': Tak. t'gwa^x-an- ['to tattoo'] (?).
txal- 'against': Tak. =t'gwa"l- ['to run about, whirl past'] (?).
t!a« 'to slap': cf. L. Chin.
t!alp 'to cover with hot ashes or fire': Coos t!al 'roast'.
t!ii" 'to sweep, tran.', t!ii°sk, intrans.: Tak. -xa- [intransitivizing suffix].
t!u"s 'to push, to beat with fist': cf. Coos t6"s ['to hit, to strike'].
SEksu' 'urine': cf. Tak. xa^n (?).
sa°l 'to notice', only in negative form: Tak. -xanaw ['to look out'] (?).
su"nt 'summer': Tak. sama- (?).
sta 'half, one side of a long thing': Chin. cit!-ix, Tak. -xdil'- ['notch'].
stslal 'beaver': Coos tEtci'na (?).
sg-an 'gum, pitch': Coos s'^aL! (?).
dzab, pi. dzEbdzab 'to make': Tak. xe^b- ['to do (so)'] (?).
dzai, pi. dzEldzat 'to eat up, to beat in game': Tak. -xeml- ['to desire to eat'] (?).
tslE- 'inside of a thing' (only in the form ts!E-wa'lb 'inside of house'): L.U. [qtsi
'inside'].
ts!ats!a 'hail': cf. Tak. [ts!ele-m- 'to hail'].
ts!al, pi. ^-ts!Elts!al 'face, eye': Tak. tc!elei- ['eye'].
ts!i"p 'to tie, to close eyes': Tak. -ts-!ibib- ['to shut (doorway, hole)'],
n-, UE-, nominal prefix: Chin, [illegible].
na'ya 'mother!' (said by girl): Tak. [s--na 'momma!' (vocative)].
ne'tSEks 'fish-tail': Chin. -lict.
n!axn!6 'to hear': Uto-Aztecan.
gao 'to take sing, obj.': Chin. -ga.
gu'p!El 'two round objects': Tak. ga"m ['two'].
gun- 'to cause' (par.): Haida (?).
gwa" 'that': cf. Chin.
gwanks 'cooked, done', SE-gwanar 'to cook': disyllabic stem,
ksax 'to go out, pi.', ksEr 'to go out, sing.': ? = *ksaxwax"-.
gP-tc 'to swell, to flood', g-I"n 'to cause to swell': cf. Tak. [-(a)n-, causative],
qa 'rabbit' (?): Athabaskan.
gal-, qal- (par.) 'empty': cf. Tak. k!wal- ['let alone'] (?).
gala'r 'cedar': Chin, -cgan (?).
ga"d 'heart, mind': Tak. giixw- [ = ?] (?):
gox, qox 'to peck', q!ax, pi. q!alx 'to pluck': note infix -1-.
la- 'not quite in the right manner' (par.): Chin, la'ax.
lEt 'excrement' (?): Tak. la'.
lamdzEX 'to enter, pi.': Coos tim- ['to put inside (pi. object)'].
laq 'to take name' (?): Tak. la^law- ['to name, call'],
lo'e 'neck-ring of cedar-bark': Chin. -Ixwa (?).
lu- 'in': Chin. -1-.
lu"p-k 'to sew': Tak. [lep'd-, lebe-].
tEb 'smooth', Ie'Iep 'to smoothen': infixed -I-.
lantk, pl. hitantk 'to move', la^'l, pl. h:»a '^'1 'to move something': Chin. -la.
Seven: Penutian Languages 315
ta" 'to slide': L.U. sLox" ['to descend, slide down'].
latk 'to move': Chin.
toga 'to go, pi.': Coos la- ['to go'].
Ika'ak 'sister' (said by man): Chin, -ixt (?).
Section Eight:
Wakashan and Salishan Languages
Introduction
Within a few weeks of assuming his duties as Chief of the Anthropological
Division, Geological Survey of Canada, in September 1910, Sapir set out for
British Columbia for linguistic and ethnographic field work on the Nootka of
Vancouver Island. A genetic connection between Nootka and Kwakiutl had
been proposed by Boas (1891:678-679), and although this was accepted by
Powell (1891), who provided the family with the name "Wakashan," the details
of the relationship remained to be worked out. This was the sort of task Sapir
relished, and his first report on his Nootka work (1911e) was a preliminary sur-
vey of Nootka-Kwakiutl comparative linguistics. He apparently continued to
work on the project intermittently for several years. The notes he accumulated,
now in the Franz Boas Collection in the Library of the American Philosophical
Society (manuscripts 497.3 B63c Wla.2 and W1.3; cf. Freeman 1966: 216, item
1954, 380, item 3836), formed the basis of an extensive "Wakashan Com-
parative Vocabulary" (American Philosophical Society Library, manuscripts
497.3 B63c Wla.26 and Wl.l; cf. Freeman 1966: 380, items 3837-3838) pre-
pared by Morris Swadesh in 1950-51, which is published for the first time in
Volume XII of The Collected Works.
Sapir visited the Nootka twice, from September to December, 1910, and from
September, 1913, through February, 1914. His principal informant was Tom
(Sayach'apis), an elderly blind man (19181, 1921e, and 1922y). From Sayach'apis
and others Sapir obtained an extraordinarily detailed record of Nootka lan-
guage and culture, including 71 texts, several of considerable length. During
the second trip Sapir taught two of his young interpreters, Frank Williams and
Alex Thomas (the latter a grandson of Sayach'apis), to take phonetic dictation ;
he arranged for them to continue to transcribe texts and collect other data after
his departure (1915j: 172). This collaboration was explicitly modeled on the
relationship Franz Boas had developed with his Kwakiutl interpreter, George
Hunt, and proved just as successful, particularly in the case of Thomas (Golla
1984: 133-134). In addition to providing documentation of many aspects of
Nootka culture from a native" point of view, Thomas's contributions eventually
more than doubled Sapir's collection of Nootka texts. Although Sapir worked
on these materials fairly steadily during his Ottawa years, he was able to see
only a small fraction of it into print until very late in his career. The three papers
(1911e, 1924g, and 1929e), the short monograph (1915a), and the set of annota-
tions (1916c) that are reprinted here represent nearly all of Sapir's published
work on Nootka linguistics before 1939 (for his publications on Nootka eth-
nography see 1911e, 1912a, 1913b, 1914b, 1915h, 1919e, 1921e, 1922y, 1922aa, and
1925c, all in Volume IV).
320 VI American Indian Languages 2
"Some Aspects of Nootka Language and Culture" (1911e), subtitled "The
Linguistic Relationship of Kwakiutl and Nootka," is one of two studies pub-
lished together as the first fruits of Sapir s 1910 field work. (The other part, "The
Nootka Wolf Ritual," appears in the full article reprinted in Volume IV.)
Sapir s aim here is to provide Wakashan comparative linguistics with a firm
structural basis. He outlines the basic sound correspondences between Nootka
and Kwakiutl; he sketches the parallels in morphology and provides a number
of cognate affixes; he discusses related morphological processes, particularly
reduplication; and he broaches comparative syntax in a treatment of pro-
nominal systems and conjunctions. He concludes with a list of 40 apparently
cognate stems and affixes.
The short monograph on "Abnormal Types of Speech in Nootka" (1915a)
belongs together with Sapir s description of consonant symbolism in Wishram
(in 1911g), his paper on male and female speech in Yana (1929d), and his experi-
mental study of the psychology of sound symbolism ( 1929m) as an expression of
his enduring interest in the points of contact between formal linguistic struc-
tures, social categories, and psychological universals. This paper is sometimes
cited in the sociolinguistic literature as a pioneering study of variation corre-
lated with such factors as age, sex, class, and ethnicity. In fact, Sapir's subject
here is not sociolinguistic variation as such, but a set of literary-rhetorical
devices used in both formal and informal Nootka discourse to index the social
categorization of the topic (having thus a clear link to Sapir's discussion of
"recitative" in Southern Paiute myths, 1910d). Typically, these are regular pat-
terns of phonetic alteration used in traditional narratives to identify the speech
or actions of salient mythic characters or of socially stereotyped groups such as
foreigners, deformed individuals, or outcasts. At least some of these devices
are regularly used in conversational interaction for a variety of interpersonal
ends, including affection, mockery, and social exclusiveness.
"The Rival Whalers, A Nitinat Story" (1924g), a short text obtained from
Sayach'apis in 1913, is provided with a word-by-word analysis in order to pres-
ent a "serviceable" introduction to Nootka linguistic structure. Publishing an
analyzed text in lieu of a full grammar was a device Sapir had earlier resorted to
for Wishram ( 191 Ig: 673-677) and for Yana ( 1923m). In Sapir's hands what could
easily be a mere pastiche of disconnected information is crafted into a tightly
organized descriptive statement. Most major inflectional and derivational pro-
cesses are explained (with several paradigms given in full), a wide range of
morphophonemic alternations are exemplified, and the complex morphosyntax
of Nootka words is thoroughly explored. Although superseded as a formal pre-
sentation by the grammatical sketch in Nootka Texts (Sapir and Swadesh 1939)
— the orthography of which is also fully phonemic — "The Rival Whalers"
remains the best pedagogical introduction to Nootka linguistic structure.
"Nootka Baby Words" (1929e) is a brief report on the Nootka baby-talk regis-
ter — a small set of lexical substitutions, phonological alterations, special mor-
phological devices, and other special features used by or in speaking to small
children — and should be read in conjunction with the paragraph on the use of
Eight: Wakashan and Salishan Languages 32 1
the diminutive in baby talk in "Abnormal Types of Speech in Nootka" (1915a:
3-4).
During the 1930s two of Sapir's students, Morris Swadesh and Mary Haas,
carried out further Nootkan field work under Sapir's direction (cf. Swadesh and
Swadesh 1933). In the summer of 1934 arrangements were made for Alex
Thomas to come to Yale for intensive work with Swadesh, with a view to prepar-
ing some Nootka materials for publication. The eventual result, published in
the year of Sapir's death, was a major volume on Nootka linguistics containing a
selection of Sapir's (and Thomas's) texts, a grammatical sketch, and a list of
lexical elements (Sapir and Swadesh 1939, Volume XI). A second volume of
texts, largely Alex Thomas's, was published by Swadesh after Sapir's death
(Sapir and Swadesh 1955). Even so, over half of the texts, many of considerable
ethnographic interest, have remained unpublished, as have virtually all of
Sapir's Nootka ethnographic notes. (See Volumes XI-XII.)
Besides those specifically concerned with Nootka or Wakashan linguistics,
the papers reprinted in this section include two that deal wholly or in part with
Salishan languages. "Noun Reduplication in Comox" (1915f), a Straits Salish
language, is based on data that Sapir collected during his first visit to the
Nootka in 1910 in typically incidental fashion, from a Nootka man whose
mother was Comox. The short list of "Salishan-Wakashan Comparisons," pub-
lished posthumously (as part of Swadesh 1949), was apparently compiled by
Sapir at the time of his Comox work or shortly afterward (see Golla 1984: 108).
It is all that survives of Sapir's own lexical evidence for the "Mosan" hypothesis,
the genetic linkage of Salishan, Wakashan, and Chimakuan originally pro-
posed by Frachtenberg (1920: 295; see also Swadesh 1953: 26-28) and incorpo-
rated by Sapir into his general classification of North American languages
(1921a and 1929a).
Sapir also prepared a phonetic retranscription of, and notes to, the anony-
mous 18th century Nootka vocabulary published by Boas in "Vocabularies from
the Northwest Coast of America" (1916c); this contribution, which appears in
Volume XII, is a minor work, but not without interest. The vocabulary was
taken at Nootka Sound, in the northern part of the Nootka dialect chain,
whereas the dialect with which Sapir was familiar, that of Barkley Sound and
Alberni Canal, was near the southern end of the chain. Sapir's notes include
numerous observations on the differences between the two dialects and may be
read as a preliminary essay on Nootkan internal diversity.
THE RIVAL CHIEFS.
A KWAKIUTL STORY
RECORDED BY GEORGE HUNT."
Synopsis.
Fast-Runner and Throw-away are chiefs of the Kwakiutl, and
close friends. The latter makes up his mind to give a feast of salmon-
berries, and sends his four attendants to invite his own clan and his friend
Fast-Runner. After Fast-Runner has been assigned his seat and the
feasting-songs have been sung, four canoes that Throw-away has been
careless enough not to have cleaned out are brought in, filled with berries
and boxes of grease, and put before Fast-Runner and his clan and three
other chiefs. Fast-runner, instead of proceeding to eat of the food, lies
down on his back and covers his face with a blanket, while his fellow-
clansmen, observing that he is displeased, follow his example. After a
long silence, Fast-Runner's attendant arises and tells his fellowmen of the
dirty treatment accorded their chief, in that Throw-away has not washed
out the canoes before using them as food-receptacles. Throw-away
rebukes his friend for his haughtiness ; Fast-Runner expresses unwilling-
ness to eat of the dirty food, and, to show his superiority in wealth, sends
for his copper "Sea-Monster" and puts it into the fire. Throw-away
retaliates by putting in his own copper, " Looked -at-askance," in
order to " keep the fire burning." Fast-Runner meets this by sending
for a second copper, " Crane," and putting this also into the fire so as to
"smother it." Throw-away, not possessing another copper, tries to
borrow one from his fellow-clansmen, but is unable to do so. Hence he
virtually confesses himself beaten by his rival, and his fire " has died
out." The guests thereupon leave.
The following day Fast-Runner returns the feast, and sends his
attendants to invite his friend Throw-away. The latter is assigned his
seat, and the feasting-songs are sung. Then four feasting-dishes are filled
' [The text as originally written down by Mr. Hunt has been kindly revised
by Mr. Edward Sapir, whom I have to thank for having changed the system
of spelling to that now adopted by Professor Boas in his Kwakiutl Texts of
the Jesup Expedition Publications. For the explanation of letters and sounds
see Vol. Ill of that series, p. 5. — Editor.]
108
324 VI American Indian Languages 2
109
with crab -apples, wild cherries, and grease, and put before Throw-away
and his fellow-clansmen. Throw-away does not eat, but returns his
friend's insult by saying that he will not taste the dirty food offered, and
then sends his four attendants home. They soon come back, however,
with the copper "Day-Face," which Throw-away puts into his rival's
fire. Fast-Runner arises and says that his fire has been " extinguished."
He then puts on the recklessness of the Fool-Dancer, and, with his
father-in-law's leave, sends his attendants to the latter's house to break
four of his new canoes. They return with the fragments. Fast-Runner
puts these on the fire so as to " build it up " again, and wishes by the
heat to drive away his friend, who is lying on the ground near the fire.
The intense heat causes Throw-away great physical pain, but he does not
flinch and holds his ground. After the blaze has begun to die out, he
gets up and eats of the crab-apples, thus showing that his rival's deeds
have not in the least affected him. Then the guests of Fast-Runner are
directed by his attendants to leave, and they do so.
Some time later Throw-away secretly informs his fellow-clansmen of
his intention to give a winter-dance in order to outdo his rival. Fast-
Runner, however, hears of this, and determines to do likewise. As pre-
paratory to the winter ceremonial. Throw-away has his son and daughter
"disappear," whereupon Fast-Runner, not to be behindhand, causes his
two sons and two daughters also to ' ' disappear." At the approach of the
ceremonial period, Throw-away has a Sea-Monster mask carved out for
his daughter, who is to be a war-dancer, and a Grisly-Bear mask for his
son. The "disappeared" children of Throw-away are "caught," songs
are sung, the Sea-Monster and Grisly Bear perform the proper cere-
monies, and a canoe is given to Fast-Runner.
Fast-Runner then begins with his winter-dance and begs his people
to stand by him in his attempt to shame Throw-away. The following
day all are invited into his house to eat, whence they repair to the singing-
house. Here they sing and dance continuously through the night. The
next day the " disappeared " Grisly Bears and warriors of Fast-Runner are
sent for and brought in four canoes. A slave is scalped, pursued, and
butchered by the Fool-Dancers, Cannibals, and Grisly Bears, and his
body eaten up by the Cannibals. Fast-Runner's recklessness in sacri-
ficing a slave shows him to be a richer man than Throw-away, who is now
clearly outdone. Fast-Runner gives Throw-away his slave's scalp. In
the evening the Grisly Bears are ' ' tamed ' ' and the two war-dancers sing
their secret songs. The attendant asks one of the war-dancers what she
wishes done to her, and she requests that they be put into the fire. So
the war- dancers are tied down on boards, a high wall of fire -wood is put
Eight: Wakashan and Salishan Languages
325
no
about the fire, and the daughters of Fast-Runner are ostensibly to be put
inside to be burned. Two slaves are adorned just like the true war-
dancers and tied down on boards. These are to be burned as substitutes
for the war-dancers. They are told not to scream when put into the fire,
as they will be struck dead if they do so, but otherwise will come back
to life after four days. The slaves acquiesce and are burned to death,
while the real war-dancers are hidden, being supposed to have been
put into the fire. After four days they pretend to revive, the ashes of the
two slaves having been preserved in boxes. Now Throw-away is com-
pletely worsted.
Throw-away goes off to fight against the Nootka ; but he and all his
men are slain by them, only one man surviving to tell the tale. " This,"
says Mr. George Hunt, " is the true story of the two chiefs who were true
friends in the beginning, and turned out to be worst enemies at the end."
Go'kula^laeda g-a'lasa Kwa'g-ule lax Qa'logwise. Wa,
They lived, it is first of Kwakiutl, at Crooked-Beach. Well,
said, the the
laE'm^awIse ^na'mok-aleda g-i'gama^yasa SE'riLlEmexa
now it is said they were friends to each chief of the Sun tribe
TsEx^wi'dEx-Luia
named Fast-Runner
(long ago)
other, the
and
g-i gama'yasa
the chief of the
Lo'yalalawaxa Ts!ex-
Hair-turned-up-in- named
Front tribe
^i'dEX'Lula. Wa,
Throw-away Well,
(long ago).
laE'm^lawIse
then it is said
a'laEl
really
laE'm^lawIse
then it is said
q!a'^lap!xes
they watched (knew)
each other's
la ga'la la ^na'mok-ala,
now long time
they were friends
to each other.
^ne^na'qa^e.
thoughts
(secrets).
la'xulapla.
they loved each
other.
Wa,
Well,
Wa,
Well,
'na'xwaEm lae
all, it is said,
Wa, laE'm^lawiseda ga'lasa Kwa'gule xu'lsaxa ^riE'mxsa
Well, then it is said the first of the Kwakiutl were downcast one
^na'la. Wa, laE'm'^lawise TslEx^i'de ha'nak-axes ^nEmo'kwe,
day. Well, then it is said Throw-away asked leave of his friend
lax TsEx^wl'de. Wa,
of Fast- Runner. Well,
laE'm ne'la.ses
then he told of his
k!we'lats!eneLasa
going to be a place of
giving a feast with
326 VI American Indian Languages 2
III
qle'nEme qIa'mdzEkvva la'xa ga'la Kwa'kug-ula. Wa,
many salmon-berries to the first Kwakiutl (clans). Well,
he'x-^ldaEm^la'wTse TsEx^wi'de ^ne'ka : " ^ya, qast, la^mo'x
right away, it is said. Fast- Runner said, " O friend ! Now that
e'k'os wa'ldEmaqos qa e'k'lex^IdesEns g'o'kulotex qaxs
good your word, so that may be happy tribe, for
your our
xE'nLElaex xu'Isa," ^ne'x-^lae. Wa, he'x-^idaEm^la'wise
very they are he said, Well, right away, it is said,
downcast," it is said.
5 TslEx^i'de axk-a'laxes a^yi'lkwe qa e'x%itsE^weses g'o'kwe.
Throw-away asked his attendants that they sweep out his house.
Wa, g i'FEm^la'wise gwal e'kwasE^wa g'o'kwaxs lae'da
Well, as soon as, it is said, finished was swept house when then the
out the
a^yi'lkwe mo'kwa yae'ltsEmtsa dEnE'me. Wa, la'xda^x"-
attendants four hung about their cedar-bark Well, then they,
waists the (belts).
^lae Le'^lalaxa Kwa'kug-ule qaes gi'gama^e TslEx^i'de. Wa,
it is said, to invite the Kwakiutl for their chief Throw-away. Well,
went
g i'l^mese gax %e'laeLexs la'e axk'la^le TslEx^i'daxes
as soon as came all in, when then asked Throw-away his
they
lO a^yi'lkwe qa las e'tse^stasE%es ^nEmo'kwe TsEx%i'de. Wa,
attendants that they go call again his friend Fast-Runner. Well,
he'x-^ida^mesa mo'kwe a^yi'lkwa e'tse^staq. Wa, g-i'Fmese g-a'xe
right away the four attendants again called Well, as soon as came
him.
TsEx^wi'de g'a'xeLa, la'e TslEx^i'de axkMa'Iaq qa las
Fast-Runner came into then Throw-away asked him that he go
house,
k!wa'ga^lil la'xa ^nEqe'walelases g'o'kwe. Wa, g-i'l^mese
sit down at the rear of his house. Well, as soon as
in the house
k!wa'g-a^ll}Exs la'asa kiwe'le klwe'^'lala dE'nx^ida, yi'sa
he sat down in the then the feasters sang feasting- began to sing with
house, when songs
15 mo'sgEme klwe'la^layo q!a'mq!EmdEma. Wa, g-i'l^mese gwal
four means of feasting songs. Well, as soon as they finished
Eight: Wakashan and Salishan Languages 327
112
klwe'^laleda k!we'laxs la'e LEle'LEmeda mo'ts!aqe xwa'xu-
to sing the feasters, then were brought four (long) small
when they into house
xwaguma. Wa, laE'm hawe'xaEm ts!o'xwEgEntsE^waxs la'e
canoes. Well, then never they were washed inside then
when they
qupIa'laxsElayowa qle'nEme q!a'mdzEk" laq. Wa, la-
were used for pouring in many salmon-berries into Well,
them.
^me'se ^na'xwaEm ^naE'ngoyaleda mo'tslaqe xwa'xuxwagumaxa
then all half full the four (long) little canoes
qIa'mdzEkwe. Wa, la'^lae ax^e'tsE%a mo'sgEme dE'ngwats!e 5
salmon-berries. Well, then it is were taken four (round) grease-boxes
said
L!e"^na qa^s k!u'nq!EgEmaeda ^na'FnEmsgEme la'xa
olachen-oil so that were put in the each one (box) into
^na'l^nEmtsIaqe xwa'xwaguma. Wa, g-i'PEm^la'wise gwal
each (long) small canoe. Well, as soon as, it is finished
said, they
k!u'nqasoxs la'e LE'lLElbEntsE^a xwa'xwagume qa^s le
were put in, then it was lifted up at small canoe, so that went
when each end the they
ha'nx-d/.amolelEm lax TsEx^wi'de LE%e's ^nE^me'mote. Wa,
was put down at Fast-Runner and his clan. Well,
in front
la'^laeda ^nE'mtslaqe ha'nxdzamolelEm lax O'dze^stalise 10
it is said one (canoe) was put down in front at Wrong-around-
the World
LE%e's %E^me'mota Maa'mtagila, yixs SE'nLlEmae ^nE^me'-
and his clan Maa'mtag-ila, while Sun tribe the
motas TsEx^wlMe. Wa, la'^laeda ^nE'mtslaqe xwa'xwagum
clan of Fast-Runner. Well, it is said one (long) small canoe
the
ha'nx-dzamolelEm la'xa Ge'xsEme lax gi'gama^yase Ki'm-
was put down in front at the Chiefs (clan) to their chief Met-by-
k-EqEwede. Wa, la'^'laeda E'lxLa^e ha'nx-dzamolelEm lax
(Chiefs). Well, it is said last one was put down in front at
the
Lla'qwagila yixs gi'gama^yasa La'alaxsEnt!ayo. Wa, 15
Copper-Maker when he was chief of the Breakers-(of-all- Tribes). Well,
328 VI American Indian Languages 2
"3
g-i'l^Emla'^wise ^wi'lga^lllExs la'e TsEx^wi'de a'Em
as soon as, it is said, all had been put then Fast-Runner only
on floor,
tIe'xbEtalel qa^s ^naxu'mdeses ^nEx^una'^e Lla'^ya. Wa,
lay down on his and he covered his face blanket (of) black Well,
back in his seat over with his bear.
laE'm'^lawise TslEx^i'de axk'la'laxes a^yi'lkwe qa wa'xesexa
then it is said Throw-away asked his attendants that they tell to go
ahead the
k!we'ie qa wa'g*es yo's^itses k'a'k'EtslEnaqe, qaxs
feasters that they should eat with spoons spoons, for
with their
5 ^na'xwamaeda kiwe'ie qe'qEplEpElaxes k'a'k'EtslEnaqe. Wa,
all the feasters carried in the folds of spoons. Well,
their blankets their
a'Em'^la'wisa ^na'xwa klvvei do'qwalax TsEx^wi'de, qaxs
only, it is said, all the feasters kept looking Fast-Runner, for they
at
^ma'lt!allmaaqexs ya'x-sE^mae na'qa^e la'xes k!wa^e'lase.
recognized that bad the heart at his place of sitting
on floor.
Wa, a'Em^lawisa klvve'ie la ^na'xwaEm la tle'x-alila o'gwaqa.
Well, only, it is said, feasters then all now lay down on also,
the their backs
Wa, he'lat!a^la gae'iExs la'e La'xulile a^yi'lkwas
Well, that, however, long on floor, then stood up the attendant of
it is said,
lO TsEx^vvi'de. Wa, la'^lae ya'q!eg-a^}a. Wa, laE'm*la'wise
Fast-Runner. Well, it is said he spoke. Well, then it is said he
^ne'k-a, la'xes klwe'l^wEote : " V^, g'o'lg'Ekulot, we'g-a
said to his fellow-feasters, "O fellow-tribesmen! goon
do'qulaxg-in g-i'gamek-, la'xg-a gvvae'lasg-as la'xox
see this our chief here at this his manner at this
on floor
g-o'kwasos ^nEmo'kwox TslEx^i'dex, yixs k-!e'sek- ^nex-
house of his friend here. Throw-away, for he not says
qa^s ha'^mx-^idexwa ^mo'^mxselayoxsa ^mo'xulax. Wa'lax'de
that he eat these dirty things of this dirty (chief). Could not he
15 la'Lox tslo'xulExsaxes le'loqlulela xwa'xuxwagumaxs
have gone he washed out his feasting-dishes small canoes when
Eight: Wakashan and Salishan Languages
329
114
k-!e's^mex-dex gu'xalESElasa qla'mdzEkwe la'qwaq, qaxs
not he began to pour the salmon-berries into these, for
he'mEnala^ma'axEl t!e'lats!exa t.'e'tlElema (he'Em gwo^yo'seda
they were all the time soaking-recep- what was (that is what he
tacles for that
soaked
meant the
p!E'Lasde LE'^vva ma'leqasde LE'^wa mo'qwasde). Wa, yo'^mets
1 hali- and the dried hali- and the (?) ). Well, this is
your
Dut-fins
but-heads
gwo^yo's qa yo'sasoso'xda q.'wa'iobEse la'xg-exs la'xos
wish that be eaten with spoons the soot now inside in
lelo'qulelaqos TsIex^iM," ^ne'x-^lae. Wa, la'^lae he'x-^lda^me 5
your feasting-dishes. Throw-away," he said, it is Well, it is said immediately
said.
TslEx^i'de
Throw-away
La' xu Ilia.
got up in the
house.
Wa,
Well,
laE'm^lae
then, it is
said,
k-!es
not
wa'ldEma la'xes a^yi'lkwe. Wa, la'^lae ^ne'k-a :
word to his attendants. Well, it is said he said.
^ne'x'tslots
he said
^ya, qast,
"O friend!
gwa'ldzas xe'iileI LE'mlEmqIa'loL. Ha'aqos gwex-s q!e'q!ade
don't you very talk proudly. That is you as if you much having
qa'es wa'ldEmos. Wa'g-adza a'Em laxs gwa'exsdaasaos,"
on account word. Well, go on just to your wish,"
of your
^ne'x'^lae. Wa, he'x-^IdaEm^la'wise TsEx^wi'de La'xuliia.
he said, it is Well, immediately, it is said, Fast-Runner got up in the
said. house.
la'^lae ^ne'ka
it is said he said,
^ya, qast, Ts!Ex*^i'd, k-!e'sEn ^ne'x-
' O friend Throw-away ! not I say
Wa, 10
Well,
qEn
that I
ha'^mapexa
eat the
^mo'xul.
dirty man !
^ma'^moxselaqwe ha^mgi^la'yo g'a'xEn,
dirty things given for food to me.
ya
o
Wa, la^me'sEn ^ma'^moxwalaLOL. A'la^meg'in
Well, and now I shall meet your dirty Truly I am
deed.
q!e'q!adaga'%ayos,"
much having among you,"
a^yi'lkwe qa
attendants
las
that they go
and
'^ne'x'^laexs
he said, as
la'e
then he
'^ya'Iaqases
sent his
ax^e'dEx Lla'qwase Tsle'gese. Wa, 1 5
get his copper Sea-Monster. Well,
330
VI American Indian Languages 2
115
g-i'l^Em^la'wise g'ax ho'gwcLcda mo'kwe a^yi'lx"sexs la'e
as soon as, it is said, came went into house four attendants, then
they
ts!as lax TsEx%i'de.
gave it to Fast- Runner.
the
Wa,
Well,
la'^lae TsEx^wi'de da'x'^idxes
then it
is said
Fast-Runner
took his
Lla'qwa.
copper.
Well,
la'^lae
then it is
said he
qa'sElli
walked in
house.
la La'yabots lax
and he went pushed it under to the
qa^s
lEqa'%alilas g'o'kwases ^nEmo'kwe. Wa, laE'm k'li'lxas la'xa
fire in middle of house of his friend. Well, then he put out to
the floor of the fire with it
5 kiwe'lase. Wa, he'x-^idaEm^la'wise TslEX^i'de o'gwaqa ax^e'dxes
his feast. Well, immediately, it is said. Throw-away also took his
Lla'qwe LlEsaxEla'yo. Wa, laE'm^laxaa'wise La'yabots la'xes
copper Looked-at-askance. Well, then it is said also he pushed it under to his
10
15
kIwe'lasdEma
feast-place
kIwe'lasdEma
feast-place
lEgwe'ia.
fire.
lEgwe'ia
fire
Wa,
Well,
qa
in order
that
laE'm^lae x'a'x"iq!as
then it is
said he
k-le'ses
it should
not
tried to burn it
kMi'lx^eda.
die out.
la'xes
in his
Wa,
Well,
laE'm
then
Wa,
Well,
^nEma'x-Es l6^' lEqwi'las lEqwa', yi'ses Lla'qwax'de.
just as if with he made fuel, with his past copper.
ith
la'^lae
then it
is said
TsEx^wi'de
Fast-Runner
e'tled
again
^ya'laqases mo'kwe
sent his four
a^yi'lkwa qa las ax^e'd ^nE'msgEme Lla'qwase AdE'mgule.
attendants that they go take one (other) his copper Crane,
and (round)
Wa, g i'l^Em^laxaa'wise g'a'xda^x" ho'gweLElaxs la'e tslas
Well,
as soon as again, it is
said, they
went into house,
when
then
they
gave
it
la'xes gl'gama^c TsEx%i'de. Wa, a'Em^laxaa'wise TsEx^wi'de
to their chief Fast- Runner. Well, only, it is said, Fast- Runner
again
la La'yabolesas lax kIwe'lasdEma lEgwe'iases %Em6'kwe.
went pushed it under to the feast-place fire of his friend.
Wa, laE'mxaa kli'lxas lax lEgwe'lases ^nEmo'kwe.
Well, then again he put out to the fire of his friend,
with it
Eight: Wakashan and Salishan Languages
331
ii6
Wa, laE'm %Ema'x-Es lo^ dzn'mas lelao'xwasa Lla'LlEqwa
Well,
then
just as if
with he covered
with
cost of the
coppers
lax klwe'lasdEtna lEgwe'lts go'kvvases ^nEmo'kwe, qa
to the feast-place fire of the house of his friend, so that
k-!eo'ses xi'xsE^watsa lEgwe'le. Wa, he'^mes la'g-ilas
it be nothing being burned iire. Well, that is reason of
by the
Le'gadEs kli'lxaxa lEgwe'lasa k!we'lase. Wa, la'La TslEx^i'de
having name putting out fire of the feast. Well, but then Throw-away
of the
^nEma'xEs lo^ mo'xLalases Lla'qwa. Wa, laE'm ^mEx°stE%e'sa 5
just as if
Lle'^na
olachen-
grease
Lla'qwa,
copper
g'o'kwas.
his house.
Lla'qwa
copper.
with he lighted his
fire with his
LE'^wa lEqwa'.
and the fuel.
copper. Well, then it is in place of the
La'giias o'gwaqa LasLa'lases
That is also pushing on
reason of the fire his
qa k-!e'ses k-!i'lx^ede k.'we'lasdEmas lEgwe'las
for its not dying out the feast-place fire of
Wi,
Well,
la'a^lase
then it is
said
g-i'^Em^lawise TsEx^wi'de La'sLEntses
Fast-Runner pushed on fire his
as soon as, it
is said,
TslEx'^'i'de
Throw-away
dak -la' lax
asked for a
L.'a'qwa
copper
la'xes
to his
*nE^me'mota Lo'yalalawa.
clan Hair-turned-up-
in-Front.
Wa, laE'l k'leo's Lla'qwas. Wa, 10
Well,
it is there was no his copper. Well,
said
la^me' ya'k-awe Ts.'Ex^i'de la'xeq. Wa, la^me' k'li'lxEkwe
then he was beaten Throw-away in that way. Well, then was extinguished
k!we'lasdEmax-de lEgwe'lts g-o'kwas. Wa, a'Em^la'wise la
the past feasting-place fire of his house. Well, just, it is said, now
ho'qawElseda kiwel la'xsde. Wa, la^me' hawe'xa yo's^Id la'xa
went out the feasters the passed. Well, then they never ate wiih at the
spoons
kIwe'ladzEm qla'mdzEkwa. Wa, la'^lae TsEx^wi'de ^ncx- qa^s
given in feast salmon-berries. Well, it is said, Fast- Runner wished that he
q!a'FaLElex na'qa^yases ^nEmo'kwe TslEx^i'de. Wa, laE'm'la'wise ic
find out the thought of his friend Throw away. Well, then it is said he
332 VI American Indian Languages 2
117
axk!a'laxes a^yi'lkwe qa las Le'^lalax Le'lanEmx-dases
asked his attendants that they go invite the ones that had been
and invited by his
^nEmo'kwaxa lE'nsde. Wa, laE'm^la'wise a^yi'lkwas Le'^'lalaq.
friend the past Well, then it is said his attendants invited
day. them.
Wa, he'x-^idaEm^la'wise g-a'xeda Le'lanEme ho'gweLa. Wa,
Well, immediately, it is said, came the invited ones came into Well,
house.
le'x'aEtn^la'wise TslEx^i'de k'!es g'a'xa. Wa, laE'm^^lawise
only, it is said, Throw-away not came. Well, then it is said
5 T.sEx'vvi'de '^ya'laqaxcs a^yi'lkwe qa las e'tse^staq. Wa, k!e's-
Fast- Runner sent out his attendants to go and call him Well, not,
again.
^lat!a ge'x"^ideda a^yi'lkwaxs g'a'xae la'k'Elax TslEx^i'de. Wa,
however, were long attendants when came following Throw-away. Well,
it is said, the
he'x'^idaEm^la'wise TsEx^l'de q!a'x"sldzeq. Wa, laE'm^lae
immediately, it is said, Fast-Runner led his feet. Well, then it is
said he
%ex- qa las klwa'g-a'^lila Lla'salilases ^nE^me'mote. Wa,
wished that he go sit down outside of his clan. Well,
and
laE'm ma^e'ma^lokweda a^yi'lkvvas lax wax'sano'LEma^yas la'xa
then there were two on his attendants at both sides of him at the
each side the
10 nExwa'Lala la'xa lEgwe'lases g'o'kwe. Wa, g-i'l'-Em^la'wise
neighborhood to the fire of his house. Well, as soon as, it is
said, he
klwag-a^ll'laxs, la'as k!we'lg"a^'leda kiwe'lasa klwe'laya^layu
sat down, then they began to sing feast-giver there were sung
feast-songs the
mo'sgEm q!a'mq!EmdEma. Wa, g i'FEm^la'wise
four songs. Well, as soon as, it is
said,
qlwe'l^ededa k!we'lala dE'nxElaxs la'e ax'^e'tsE^a tse-
stopped the feasters singing then were taken out crab-
tsE'lwats!e mo'sgEm k!ek!i'myaxLa, qa^s ga'xe mExa'lelEm
apple boxes four (round) boxes, and they came were put down
on ground
15 lax ma^sta'yasa a^we'LElasa klvve'layats! g'o'kvva. Wa, la'^lae
at the near by to the inside of door feasting- house. Well, it is
of the receptacle said
Eight: Wakashan and Salishan Languages 333
ii8
e'tled ax^e'tsE^vveda mo'sgEme dE'ngwatsle Lle'^na. Wa, la'^lae
again were taken out four (round) grease-boxes (of) olachen- Well, it is said
grease.
ax^e'tsE%eda ma'^lEx-La si'siul lo'quleia. Wa, he'Em^la'vvisa
were taken out two on it double-headed feasting- Well, that, it is said,
serpent dishes.
^nEme'xxa na'na lo'qulela. Wa, he'Em^la'wisa ^nEme'x'La
one on it grisly- feasting-dish. Well, also, it is said, one
bear
na'ne lo'qulela. Wa, he'Em^la'wisa ^nEme'x-La
grisly- feasting Well, also, it is said, one
bear dish.
a'LanEm lo'qulela. HEm%ex-Lago la'e lo'Elqulelas 5
wolf feasting-dish. That is all then the feasting-dishes of
TsEx%l'de. Wa, laE'm^lawise guxtsla'laso^sa t.'E'lse. Wa,
Fast-Runner. Well, then it is said were poured into wild Well,
they with cherries.
laE'm^lawise naE'nguya'leda lo'Elquleiaxa tsElx",
then it is said were half full the feasting-dishes of crab-apples,
la'e k!u'nq!Eqas6^sa Lle'^na. Wa, laE'm q6'qut!as.
then were poured into olachen- Well, then they were full
they with grease. with it.
Wa, laE'm^laxaa'wise he'Em g"il k'a'x-^itso'^ses ^nEmo'kwe
Well, then also, it is said, that first it was set before his friend
TslEx^l'de. Wa, g-i'l^Em^a'wise ^wi'^la k-a'x"^Itsa 10
Throw-away. Well, as soon as, it is said, all were set before
the
io'Elquleiaxs la'e a^yi'lkwas wa'xaxa k.'we'le qa
feasting-dishes, when then his attendants urged the feasters that
wa'g'es yo's^eda. Wa, he'x'^idaEm^la'wise TslEx'^i'de La'xullia.
they go eat with Well, immediately, it is said, Throw-away stood up in
ahead spoons. the house.
Wa, la'^lae ya'q!eg-a^la. Wa, la'^lae %e'k-a : "^ya, qast,
Well, it is said spoke. Well, it is said said, "O friend!
he he
k-!e'seg-in he g-a'xile qEn plax^aLE'le la'xos kIwe'ladzEmaqos,
not I that reason of that I should taste of your your feasling-objects,
coming
yO'Laxs a'laaqos mo'xula bEgwa'nEma, qfist, qaxs hawe'xaaqos 15
you that really are dirty man, friend, since you never
334
VI American Indian Languages 2
15
119
tsIo'xug-indxEn
washed inside my
ha^maa'tslex, yi'xos lelo'qulelaqos,"
food-receptacle your feasting-dishes,"
^ne'x'^laexs la'e ^ya'laqases mo'kwe a^yi'lkwa qa las dadag'ill'lEla
he said, it is then sent out his four attendants that they take out (some-
said, when
he
lax g-o'kwas. Wa,
in his house. Well,
k-!e's«lat!a
not, it is said,
however,
ga'laxs
long when
they
he'x-'^idaEm^la'wise
immediately, it is said,
they
g'a'xae e't!ed
came again
go and thing)
la'x-da^xwa. Wa,
went out. Well,
ho'gwcLa.
they went into
house.
5 laE'm^lae dag-i'lqElaxa Lla'qwa '^nE'lgEmala.
then it is carried in hand copper Day-Face,
said they the
Lla'qwa
^idaEm^la'wise TslEx^l'de da'x-^idxa
mediately, it is said. Throw-away took the copper
boleses lax lEgwe'lasa klwe'lase. Wa,
it under to the fire of the feast-giver. Well,
house of the
k'li'lxax lEgwe'lasa gi'gama^e TsEx^wi'de
put out the fire of the chief Fast-Runner.
Wa,
Well,
qa^s
and he
Wa,
Well,
he'x -
im-
La'ya-
shoved
laE'm
then
6 gwaqa
also he
Wa, la'^laeda
Well, it is said the
g i'gama^e
chief
house of the
TsEx^wi'de
Fast- Runner
La'xulila
got up in
the house
qa^s
ya'qlega^e. Wa,
and he spoke. Well,
10 la'^lae ^ne'ka : "^ya, qast, la^mo'x k-!i'lx^edEn kIwe'lasdEmaqEn
it is said
he
said.
"O friend!
now extinguished my
this is
feasting-place
this my
lEgwe'la. We'g-il la e'tla^lilEL k!wa'g-a^lllLOL, qEn do'qwale
fire in house. Go on now will again will vou sit down so that I look
in house
will you sit down so that I
in house
qEn wa'ldEm o'gwaqa,'
for my word also,"
^ne'x"^laexs la'e
he said, it is said, as he
xwa'sa la'xes
was excited in his
nulEmalaena^e. Wa, la'^lae ^ya'laqases mo'kwe a^yi'lkwa qa
fool-dance. Well, it is said he sent his four attendants that
las o'gwaqa d6'x%edEx axe'ilaxa lax go'kwases nEgu'mpe
they go also look for what they might in the house of his father-in-law
take
Wa, k-!e's^Em^lawise ho'qawElseda mo'kwe
Well, not then, it is said, went out of four
house the
and
^mo'^nakula.
Moving-Load.
Eight: Wakashan and Salishan Languages
335
I20
a^yi'lkwaxs
attendants when
lase la'xes
place for he
in house
la'e
then
La'xullle
stood up in
house
^mo'^akula
Moving- Lx)ad
g'a'xene^me
had also come
k!we'}a.
as feaster.
Wa,
Well,
la'xes
at his
la'^ae
it is said he
k!wae'-
sitting-
^ne'k-a:
said,
"O
nEgu'mp,
son-in-law !
^ya'laqalas
send
qa
that
lase
they go
and
ax^e'tsE^wa
be taken
mo'tslaqa t!e't!Egu'na
four (long) flat-bowed canoes
qas
for
lEgwe'los," ^ne'x'^lae.
your fire
in house,"
he said, it is
said.
Wa,
Well,
he'x-^ldaEm^la'wise TsEx^wl'de ^ya'laqases mo'kwe a^yi'lkwa 5
immediately, it is said, Fast-Runner sent out his
Lo^me's ha^ya'Fa qa les
young men that they
and also
his
laq
t!e't!Egu'na.
flat-bowed canoes.
Wa,
Well,
tso'tsoxsEndxa
break to pieces the
k-!e's4at!a
not, it is said,
however,
four
mo'tslaqe
four
ga'laxs .
long when
attendants
alo'-
new
^we'g-iLEla'yoweda tso'gukwe t!e't!Egu'na. Wa,
carried into house on broken flat-bowed canoes. Well,
shoulders the
ga'xae
la'^lae,
then it is
said they
^mo'x''Lalayo la'xa kIwe'lasdEma lEgwe'la g-o'kwas Tsex-
were piled up at the feasting-place fire in house the house of Fast-
^wl'de.
Wa, la^me'
a'lax-^id
la x'ix^e'de
lEgwe'las.
10
Runner.
Well, then
truly
now was burning
the
fire in the
house with it.
Wi la^
me' TsEx%I
'de wa'lac
elax TslEx^i'
de he'ltsasa
Well, then Fast-Runner wished that Throw-away might run
away from the
Lle'sala.
Wa, la
e'x-ax*Em
L!6'p!eda,
qo k!esL
heat.
Well, then
he thought it
good that
he be roasted
if not will
he'ltsaLEs
Wa,
a'Em^lawise
Ts!Ex^i'de
t!e'x-^al!i
he will run
away from it
Well,
just, it is said,
Throw-away
lay down on
his back
la'xes
k'.wae'lase.
Wa, a'
Em^lawise
^nEx^una'^yas
at his
sitting-place
Well, just, it is said,
his blanket
in house.
L!a'^ya
la tlE'mg
^u^nakula.
Wa hewa
'xaEm^la'wise
15
(of) black
bear
now became scorched.
Well, then never, it is said,
336
VI American Indian Languages 2
121
qlwe'nalElile TslEx^i'daxs wa'x-'^mae la pEne'^nakule Lle'sas
moved in house Throw-away although became covered with skin of
blisters the
pp.'e'g-a^yas,
his knee
la'xes
at his
a ene me
just being
^nExuma'lases
face being covered
with his
L.'E'ntsEmx-de
past bear-skin
I
^nEx^una"^ya
Wa, g-i'FEm^la'wise k'li'lx-a^nakuleda lEgwe'laxs
blanket. Well, as soon as, it began to die out the fire when
is said,
la'e TslEx^l'de klwa'g-a^lll qa^s yo's^ldexes lo'quleda tsElx".
then Throw-away sat down and ate with spoon dish of crab-
he out of his apples.
Wa,
Well,
laE'm
then
^nEma'x'Es l6^ ne'laxs
this is just if
he showed
that
k-!e'sae tslE'x-^aLEle
not he became sick,
gwe'x-^ldaasaq wa'x'^mae la lEgu'la. G i'Pmese gwal yo's^ededa
what had been although he now he was As soon as finished eating with
done him burnt. spoons the
kiwe'laxa tsE'lxwe la'e tslE'lwaqaso^s a'^yi'lkwas TsEx%l'de.
feasters the crab-apples then were praised by attendants of Fast- Runner.
they the
Wa, g'i'Fmese qlwe'Fideda a^yi'lkwaxs la'e ho'qawElseda k!we'le.
Well, as soon as stopped speaking attendants then went out of feasters.
the house the
Wa, ga'laEm^a'wise k-leo's e'tled gwe'g'ilasa. Wa, laEm^la'-
Well, it was long, it is said, nothing again way of doing Well, then it is
lo Wise
said
TslEx^l'de
Throw-away
Le'^alaxes
invited his
thus.
^nE^me'moteda ^ Lo'yalalawaxa
clan (were) the Hair-turnedup-
in-Front at
ga'nuLe. Wa, laEm^la'wise ne'lases ya'^wix-ilaexsda^e.
night.
Well,
then it is said he told that he desired to give a winter-
dance.
he'x-^ldaEm^Ia'wise
immediately, it is said.
^nE^me'motas ^mo'^las
his clan
were grate-
ful for
wa'ldEmas.
his word.
Wa,
Well,
Wa,
Well,
laE'm^ae
then it is
said he
wax"
tried
sEna'nEmaq
to find out
qa^s
that he
la'g-iias ^nex* qa'^s he gwe'x-'^lde.
for that said that he thus did.
reason he
1 5 qla'^le TsEx'
Wl
dax
watched
Fast-Runner
wa'idEmas.
his word.
yak-a'masex TsEx%i'de ;
might beat Fast- Runner ;
Wa, he'x'^idaEm^la'wise
Well, immediately, it is said,
Wa, laE^m^Iaxaa'wise
Well, then it is said, also he
Eight: Wakashan and Salishan Languages
337
122
o'gwaqa Le'^lalaxes ^nE^me'motaxa ga'nuLe. Wa, laEm'^la'wise
also invited his clans at night. Well, then it is said he
ne'iaxes ^'nE^me'motaxs lE^ma'e o'gwaqal ya'wix-ilalxa tslawu'nxe.
told his clan that he will also will give winter- winter.
dance in the
Wa, laEm^la'wise ^nex* qa^s a'lagawa^eses hayo'te TslEx^l'de.
Well, then it is said he wished that he be always equal rival Throw-away.
to his
Wa,
Well,
Wi,
Well,
laEm^a'wise gwa'ie wa'ldEmasexs la'e ho'qawElsa.
then it is said were their words when then they went out of
finished
house.
k-!e's«lat!a
not, it is said,
however.
ga'iaxs
long
la'e
when
x'is^l'deda
disappeared the
ba'bagume 5
boy the
xuno'x"s TslEx^i'de LE^we's ts!a'ts!EdagEme xuno'x"s.
child of Throw-away and his girl his child.
Wa, he'x-'^'idaEm^laxaa'wise x-is^i'de bEgwa'nEme xuno'x"s
Well, immediately, it is said, also disappeared man the child of
the
ma^o'kwe ts!eMaqe sa'sEmaxa la'e
two women children at then
the
TsEx^wl'de LE^we's
Fast-Runner and his
e'tled
again
[next]
la'e
then
ga'nuFeda.
night.
Wa,
Well,
la'4ae
it is said
e'tled
again
ga'nul^edExs
it was night when
x'is^l'deda ^nEmo'kwe bEgwa'nEm xuno'x°s.
disappeared one man
laE'm^ae
then it is
said
laE'm^lawise
then it is said
g-i'ta qae'.
carve for him.
ts!e'ts!aqeda
danced winter-
dance the
TslEx^i'de
Throw-away
ga'la
first
he'laxa
hired the
his child.
Kwa'kug'ula.
Kwakiutl clans.
g'ltle'noxwe
wood-carver
qa
that
Wa,
Well,
Wa,
Well,
las
he go
and
Wa, laE'm^ae tsle'gese g-ita"-'yasa gitle'noxwe
Well,
then it is
said a
sea monster carving of the
the
qa Lo'gweltsa ts!a'ts!EdagEme xuno'x"s
that it be the supernatural girl the child of
treasure in house of the
to'x%ldeLe
was to be war-
dancer
qo
if she
ga'xL
should come
ne'l^'ldELo.
should show
lierself.
Wa,
Well,
wood-carver
TslEx^i'daxs
Throw-away
when she
laEm^la'-
then it
10
15
338
VI American Indian Languages 2
123
wise
na'neda
ba'bagume
qo
ne'i^ldELo.
Wa,
is said
a
grisly bear
was to be the
boy
if
he should show
himself.
Well,
a'Em^a
'wise TsEx^w
i'de ho'Lelax
wa'ldEmas. Wa,
laEm-
only, it is
said, Fast-Runner was listen-
ing to
their words. Well,
then
^la'wise
la'^stod
lax k I'm^'yaEnxLaq.
Wa, laE'm^'la-
it is said
it
near
to the time of being about
to meet them.
Well,
then it
wise
k!wa'i^ideda
Kwa'kug-ule
la'xa
q!a'mdase.
Wi,
is said
sat down the
Kwakiutl clans
at the
singing-place.
Well,
5 laEm^la'wise TsEx^wi'de k!wa'gElasEq, qaxs ^a'xwa^mae la'da
then it is said Fast-Runner sat with them for all they went the
be'bEgwanEme la'xa q!a'mdase. Wa, laE'm^lawise gwa'lExs
men to the singing-place. Well, then it is said finished when
la'e na'^nakwa. Wa, la'^lae ga'nuFlda, la'e k-ik-i'lnala ^wi'^leda
then went home. Well, it is said night came, then tried to all the
they bring back
(the novice)
q'.e'nEmula le'lqwalaLa^ya. Wa, laE'm^lawise ^na'x'^ldxa gaa'laxs
many past tribes. Well, then it is said daylight came early mom-
in the ing when
la'e gwa'la. Wa, he'x-^idaEm4a'wise la da'da ga'la Kwa'g'ulxa
then finished. Well, immediately, it is said, went took first Kwakiutl the j
they the
10 ma^o'kwe x'isa'ias sa'sEmas TslEx^i'de.
two disappeared children of Throw-away,
ones
ga'xalesa da'xdaxa x'isa'laxde la'xa
came on the who had taken having disap- on the
Wa, g'i'l^Em^Ia'wise
Well, as soon as, it is
said.
LlEma'esaxs
beach when
la'e
then
beach those
qa'kasE^wa
had her head cut
off the
1
the peared ones
to'x^wide. Wa, la'^laeda
war-dancer. Well,
na'ne
LE qases
it is said the grisly bear struck with
his
15
ge'ts!Eme la'xa Sva'lase tiEgu'na. Wa, g i'FEm^la'wise
claws at the large flat-bowed Well, as soon as, it is said,
canoe.
gwa'lExs la'e g-e'x-^edayoweda t!Egu'ne lax TsEx^i'de.
finished when then was given the flat-bowed canoe to Fast-Runner.
Wa, laE'm^lawise gwal q!a'mtasE%a to'x%idaxs la'e
Well, then it is said finished was sung for the war-dancer then
when they
Eight: Wakashan and Salishan Languages 339
124
ho'x^wEsdesa qa^s la ho'gweL la'xa lo'bEkvve. Wa, la'^lae
went up beach and went went inside to the (emptied one) Well, it is said
they winter-dance
house.
ga'nuFldExs la'e nana'qamasE%eda da'nEme. Wa, laEm^la'wise
night came when were brought to their caught ones. Well, then, it is said,
senses by singing, the went the
tsle'gese Lo'gwa^yasa to'x^wlde. Wa, laE'm^lae g-a'xustalileda
sea-monster the supernatural war-dancer. Well, then it is said came up from floor
treasure of the of house the
tsle'gese la'xa ogwe'waliiasa g-o'kwe. Wa, he'Em^la'wise
sea-monster at the rear of the house. Well, that, it is said,
wa'lox%ede. Wa, laE'm gwa'le TslEx^i'de ya'vvixila. 5
was all that Well, then finished Throw-away giving winter-
was done. dance.
Wa, la'4ae TsEx^wi'de Lla'yogulsa.
Well, it is said, Fast-Runner changed with him
on the ground.
Wa, laE'mxa^e TsEx^wiMe ^ya'laqases a^yi'Ikwe qa
Well, then also Fast- Runner sent his attendants that
las Le'^lala Hvu^na'laxa la ga'ia ga'nuLa la'xes
they go invite secretly in the now long night to his
and
^nE^me'mote. Wa, gi'FEm^la'wise g*ax %!'^laeLExs
clans. Well, as soon as, it is said, they came all into the
house
la'e LEne'x'^ItsE^we t!ex-i'lasa ts!a'gats!e g'6x"s 10
then was barred the door of the winter-ceremonial the house
receptacle of
TsEx^wi'de. Wa, laE'm4awise ne'le E'lkwas TsEx^wi'daxes
Fast-Runner. Well, then it is said told the attendant Fast- Runner his
of
^nE^me'motaxs lE^ma'e lal k!wa'iaixa hi'La dza'qwaltsa la'La
clan that they go sit in the future evening of the future
e'tledEl ^na'x'^IdEL. Wa, he'^mesexs lE^ma'e ^ne'k'es gi'gama^e,
again morning. Well, that is when he said his chief
yix TsEx%i'de qa^s we'g'iL e't!edEi mo'mas^idElxes ^nEmo'kwe
that Fast-Runner that he will go will again will do harm to his friend
on
TslEx^i'de qo lal ne'l^idELes x-e'xisala. " Wa, laE'm^la'wisEns 15
Throw-away if will go will show disappeared " Well, then we are told to
his ones.
340
VI American Indian Languages 2
125
gwa'laial qEns g-o'xHvedeLEnsaqEk- qa^go ha^ala'lasoLo,"
be ready that we shall help hitn if he will be faced by
his rival,"
^ne'x-^ae. ^na'xwaEm^'la'wise ^nE^me'motas ^ne'kExs lE^ma'e %i'^la
All, it is said, his clan said they all
he said, it is
said.
said
when
gwa'laia. Wa, laE'm^lawise gwa'le wa'ldEmasexs la'e ho'qavvElsa.
are ready. Well, then it is said finished their words they went out of
when the house.
Wa, laE'm^awise ^na'xwa ga'x'staeiaxs ga'lae ho'qawElsa.
Well, then it is said all went to bed first they went out of
when (as soon as) house.
5 Wa, laE'm^awise a'l^Em tslEx-^i'dxa
Well, then it is said they soon awoke at
Wa, he'x-^idaEm^a'wise
Well, immediately, it is said,
they
la lE'nsa.
now next day.
gwe'gudza
winter-dancers
(sparrows)
Wa, g-i'l^Em^a'wise gwa'lExs la'e Le'^lale TsExVi'daxa
LE'we's
and their
tsle'daqe
a la
ga'la '^na'laxa
now
1
long day on the
[late in)
^na'xwa
all'
la kwa'seda
went washed them-
selves the
Lo'^ma
g i'ng inanEme.
and the
children.
Well, as soon as, it is said.
finished
when
then
invited
Fast- Runner the
. gwe'gudza LE'%a tsle'daqe Lo'^ma
winter-dancers and the women and the
10 las he'yasEla la'xa ts!a'gats!as
they go breakfast at the winter-ceremonial
and house of
grng-inanEme
children
TsExHvi'de.
Fast-Runner.
qa
that
Wa,
Well,
g'i'FEm^la'wise gwal ha^ma'pExs lae'da a^yi'lkwe o'paiaxa
as soon as, it is finished eating when then the attendants whispered to
said, they
*na' xwa
all the
gwe'gudza qa la.s la'xa
winter-dancers that they go to the
qia'mdase la'xa a'Lle. Wa,
singing-place in the inland. Well,
bEgwanEme ^wi'^la q!wa'g"a4ii
kiwa'iaase
sitting-place
he'x-^ldaEm^la'wisa
immediately, it is said, the
le a'Le^sta.
15 g-i'FEm^la'wise %i'^la
as soon as, it is said, all
they
qa^s
and they went around
inland.
la klusE'lsExs la'eda
went to sit down on then the
the ground, when
la'xa
to the
be'-
Wa,
Well,
f
nema
song-
Eight: Wakashan and Salishan Languages 341
126
gade dE'nx^Itsa ma^ltsE'me q!a'mq!EmdEmasa to'x^wlde.
masters began to sing the two songs of the war-dancers.
Wa, he'^mesa ma%sE'me qIa'mqiEmdEmasa na'ne. Wa,
Well, also the two songs of the grisly bears. Well,
g'l'PEm^la'wise gwa'la ne^na'gade dE'nxElaxs la'e
as soon as, it is finished song-masters singing when then
said, the
TsEx%i'de ya'qleg'a^a. La'^lae ^ne'k'a: " '^ya, ^ne'^nEmokwa',
Fast- Runner began to speak. Then it is said, "O friends!
said he
laE'ms %i'^lal qlwa'lax'alxwa ga'nuLex, qas e'k-!eqEleL6s 5
now you will all will dress this night, that may be happy
you
qaE'n wa'idEmLex,"
'^ne'x-^lae.
Wi,
la'^lae
%i'^la^ma
on account word,"
he said,
Well,
it is said
all the
of my
it is said.
gwe'gudza na'^nax^meq. Wa, la'^lae ^ne'k-a : " He'LEnu^x"
winter-dancers answered him. Well, it is said said, " This we shall
they
gwe'laLe " ^ne'x-^lae. Wa, la'^lae he'x-^ida^ma mo'kwe a^yi'lk"
shall do they said. Well, it is said, immediately the four attendants
thus," it is said.
qa's^id qa^s la xa'set!alaxa le'ladEnokwe. Wa, g-i'FEm^la'wise
walked and went asked to wash those who had Well, as soon as, it is
they their bodies dances. said,
la'da mo'kwe a^yi'lkwaxs la'e ho'x^wult'.eda qIa'mtlEsde, 10
went four attendants then went out of the song-experts,
the woods the
Wa, he'x-^idaEm^a'wise xwa'nal^ededa , %a'xwa gwe'gudza
Well, immediately, it is said, got ready all the winter-dancers
LE%e's ts!e'daqe LE%a' ^'na'xwa g-i'nginanEme. Wa,
and their women and all the children. Well,
laE'm^lae k-iki'lnala. Wa, a'FEm^la'wisc gwa'lExs la'e
then it is said they tried to bring Well, just, it is finished when
they them back. said, they
^na"^nakulaxa gaa'la. Wa, he'x'^idaEm^la'wise la'da mo'kwe
daylight came early Well, immediately, it is said, went the four
in the morning.
a'^yi'lk" ax^e'dxa mo'tslaqe awa' tIe't.'Eguna qa LE'nkwesesa 15
attendants took the four (long) great flat-bowed and tied them together
canoes with the
342
VI American Indian Languages 2
127
mo'ts.'aqe g-i'lsgilt!a dzo'xuma.
four (long) long poles.
sao'kwc. Wa, g"i'FEm^la'wisc
boards. Well, as soon as, it is said,
they
^na'xwa gwe'gudza laq
all the winter-dancers
on
them.
Wa,
Well,
LE'wa'
and the
no'EniEmala
fool-dancers
Lo^ '^na'xweda le'laenenokwe
and all those having dances
5 LaE'm^lae a'Em kludzl'l la'xa lo'bEkwe. Wa,
Then it is said just sat in house in the winter-dance
they
ga'leda
long the
awl'lbes
point of
Wa, la'^lae pak-EE'ntso*^sa
Well, it is said were covered over
they with
gwa'lExs la'e ho'x^aLExseda
were finished then went into canoes
when
laE'm^lae k-!es la'da nena'ne
not went the grisly
bears
la'xa da.
to the taking.
k-!e's'^lat!a
not, however,
it is said,
then it is
said
daxs
taking
when
ga'xae
house (emptied
one).
Eyo'x^ed lax
back at the
Well,
Qa'logwise.
Crooked- Beach.
LlEma'esasa lo'bEkwe.
beach of the winter-dance house
(emptied one).
Wi,
Well,
Wi,
Well,
g-a'x'^lae
came, it is said,
laE'm'^lawise
then it is said
gwa'klodilba^e
opposite northern
end of point
ha'ng-a^les lax
being in front at the
of beach
^nE'me na'ne
one grisly bear
hax"ts!a'gvvewe lax
lay down inside in the
a gwiwa^yasa
bow of
of bov
^nE'mtslaqe
one (long)
tiEgu'na.
flat-bowed
canoe.
10 Wa, la'^'laxaeda ^nE'me na'ne he'xat! gwa'la lax a'gwlwa'^yasa
Well, also, it is said, one grisly that also thus in the bow of the
(other) bear
the
^nE'mtslaqe.
one (long)
Wi,
Well,
Wa.
Well,
Wa, laE'm^lae k'!es ya'wix'a'lag"iiExsa
Well, then it is said not moved about in canoe,
(mother canoe). they
laE'm'lawise q!a'mt!etsE^weda ma^lo'kwe teto'x^wlda.
then it is said were sung for the two war-dancers.
g-i'l'Em^la'wise gvval q!a'mtasoxs la'e
as soon as, it is said, finished were sung for then
they when
^walatsa'yokwe q!a'k*6 bEgwa'nEma qa^s le q!o'dEg-ivvaletsa
stout slave man that he 'go keep bow with a
15 dzo'xume la'xa ^wa'lega^yasa t!et!Egu'ne. Wa, g i'PEm^la'wise
pole in the large one among flat-bowed Well, as soon as, it is
^ya'lagEma
was sent a
the
said, he
Eight: Wakashan and Salishan Languages 343
128
La'x^walExs
la'eda
ba'baklwe
Le'gadEs
QE'nx%ida'yo
stood up in
the canoe,
then a
warrior
named
Frowned -upon
haxsE'mlExsaq. Wa, la'^lae ^nEmo'kwase '^na'xulalese, ax^e'dxes
fell uf>on him Well, it is said his friend Warrior-all-round- took his
in canoe. the-World
k'le'LEnxe qa^s k-!e'ltse^stales mak'inxE'ndalaxox awE'nxa^yaxsEns
sharp-edged and cut around with it close to this edge of our
(knife) he
SE^ya'x. Wa, g i'FEm^la'wise Ifi'lgEwe k!e'La^yasexs
hair. Well, as soon as, it is said, met where it was cut
when
la'e ne'xodEx Lle'tsEma^yas xo'msdasa q!a'k-owe. Wa, 5
then pulled off the scalp of the head of the slave. Well,
he
la'^lae LaxSvuito'dxa qla'k'o. Wa, g'i'FEm^la'wise la^sta'xs
it is said they pushed out of slave. Well, as soon as, it is said, he went into
canoe the he the water,
when
ga'xaeda ha'Emats! LE%a' ne^na'ne LE^wa' no'EniEmala ho'qawEls
came the cannibals and the grisly bears and the fool-dancers went out
la'xa lo'bEkwe qa^s la qa'qayaxa qia'kaxs la'e dzExwae'sEla
from the winter-dance and went pursued the slave then ran along
house they when he
lax LlEma'esas Qa'logwise. Wa, he'^mes la qaqaya'atsa
at the beach of Crooked- Beach. Well, and that is then place of pur-
suing of the
ne^na'naq LE^vva' no'EnlEmala qa^s sa'k'apleq. Wa, la'^'lae 10
grisly bears and the fool-dancers that they tried to Well, it is said,
him spear him.
na'papleda wao'kwaq. Wa, la'^laeda ne^na'ne Ui'qaplEses
tried to throw several at him. Well, it is said the grisly bears tried to strike
stones with their
xexElya'yowe laq. Wa, we'gaa^latla lax ^na'lanegwesas
claws at him. Well, not he reached, to the beach on south side
however, of
Qa'IogwIsaxs la'e ya'x%alesa. Wii, laE'm 1e4 la'xeq. Wa,
Crooked-Beach when he fell down dead Well, then dead there. Well,
on beach. he is
he'x'^ldaEm^'la'wisa no'EnlEmala sEsox"sE'ndxa q!a'kox"de,
immediately, it is said, the fool-dancers cut up in pieces the former slave.
Wa, g-i'l^Em^'la'wise gwal sa'kwasoxs la'eda ha'Emats!a jr
Well, as soon as, it is said, he finished was cut up when then the cannibals
344
VI American Indian Languages 2
129
GEORGE HUNT
ho'qunts!es la'xa LlEma'ese qa^s la ha^mx-^l'dEq. Wa,
went down to to the beach that they go eat him. Well,
beach
la'^lae o'gwaqa'^ma ne^na'ne LE%a' no'EnlEmala ha^mek'la'la
it is said, also the grisly bears and the fool-dancers asked to eat
laq.
of it.
Wa, wel?.x"dze'^lae ge'x-^idExs la'e '■"wi'^laq qaxs
Well not it took, it is said, long when
they
(ate) him
all up
for
ma^itso'gugiya^eda bEgwa'nEme ha'^ma'pxa q!a'k'oxMe. Wa,
120 were the men eating the former slave. Well,
la'^laeda ne^na'ne, yi'xa
it is said the grisly bears, that is the
ma^le'
two
da'nEma,
novices
(taken ones).
he'x-^idaEm
immediately
le'nEmax SE^ya'x-das
took away his past hair
ha^mg'i'^layo laq. Wa,
was given to Well,
food them.
qa^s le'nEmapleq.
and took it from
they each other.
Wa,
Well,
laE'm
then
he'^mesexs
and that is
when
lE^ma'e
he
k*!e'lakas6^s
was killed by
TsEx%l'de qae's hayo'te TslEx^i'de. Wa, la^me' ya'k-aweda
Fast- Runner on account rival Throw-away. Well, then was beaten
of his
the
10
g-i'gama^e Ts!ex^
chief Throw-
"I'de la'xes ^nEmo'kwe
away by his friend
TsEx^'wl'de.
Fast- Runner.
Wa, g-i'l^Em^la'wise gwal a'mleda ma'^le'
Well, as soon as, it is said, finished playing the two
ne%a'nesa
grisly bears
with the
sa'bEkwe SE^ya'sa
skinned hair of the
q!a'k-oxMaxs la'e TsEx%l'de
former slave, then Fast- Runner
da'x-^'ldxa
took the
sa'bEkwe SE^^ya'
skinned hair
qa^s la ts!as lax
and he went gave it to
TslEx^i'de.
Throw-away.
Wa, la'^lae
Well, it is said he
^ne'k-a : " Wa, qa'sta,
said, " Well, friend.
la^mo'x
now this
qo'sL," ^ne'x-^lae.
will be he said,
yours," it is said.
Wa, g-i'P'mese gwa'lExs la'e
Well, as soon as he finished then
ho'x^wuita
went out of
canoe
15 ^wl"leda k i'm'yaxdaxa x i'sa'lax-de. Wa, g-i'FEm^a'wise
all those who had surrounded former disap- Well, as soon as, it
the peared ones. is said.
Eight: Wakashan and Salishan Languages
345
130
dza'qwaxs la'e a^'yi'lkwas TsExHvi'de qa's^Ida. Wa, k-!e's'lat!a
evening came, then attendants of Fast-Runner walked Well, not, however,
the (to call). it is said,
g-a'laxs g-a'xae %i'^laeLeda ga'la Kwa'kugula. Wa,
long when came all into house the first Kwakiutl clans. Well,
he'x'^idaEm^la'wise ya'lasE^weda ma^e'
immediately, it is said, were tamed the two
ne'^na'na.
Wa,
Well,
g-i'FEm^a'wise gwa'iExs,
as soon as, it is said, finished
they with it,
la'aEl
then it
is said
ya'laqweda
sang their sacred
songs the
grisly bears.
ma'lo'kwe
teto'x^wlda. Wa, g-i'pEm^la'wise qlwe'FedExs la'e dE'nx^ededa 5
war-dancers. Well, as soon as, it is stopped speaking, then began to sing
said, they the
ne^na'gade. Wa, g-a'xeda ma^lo'kwe teto'x^wid ^yEx%uit!a'-
song-
masters.
qa^s
lllEl
came out and
of house, they
Well, came the
la^stali'lEle
went around
in house
two
la'xa
at the
war-dancers dancing as they
lEqawa'lllasa
fire in middle
of the
•o'kwe.
house.
Wa,
Well,
dEma.
laE'm^la'wise
then it is said
they
gwal
finished
dE'nxElasa
singing with
'nE'msgEme
one (round)
q!a'm-
song.
Wa,
Well,
he'x'^ldaEm^la'wisa
immediately, it is said,
the
ya'yaq!antemiie la lax
speaker of the house went to the
qlwae'lasasa
standing-place
of the
ma^lo'kwe
two
teto'x^wlda.
war-dancers.
Wa,
Well,
laE'm^lawise
then it is said
10
wuLa'sE%eda ^nEmo'kwe to'x^wid la'xes axe'xsdESE%a. Wa,
was asked one war-dancer for her desired (thing). Well,
he'x-^ldaEm^lawise to'x^wide %e'ka: " Wa'laqeleg'anu^x"
immediately, it is said,
the
war-dancer
said,
qEnu^x" lEqwi'lasE^e la'xwa lEgwe'lex,"
that we
be made fire
[put into fire]
into that house-fire,"
la'^Iaeda ya'yaq!antemile e'talas la'xcs
it is said speaker of the house repeated it to his
the
" We desire
^ne'x-^lae.
she said, it
is said.
^nE^me'mote
clan
Wa,
Well,
yi.s
the
wa'ldEmasa teto'x^wide. Wa, he'x-^idaEm^^lawisa mo'kwe a^^yi'l.x^'.s 15
word of the war-dancers. Well, immediately, it is said, the four attendants of
346
VI American Indian Languages 2
131
TsExSvi'de ya'q!eg-a^ia. Wa, laE'm4awise ^ne'k-a: "^ya, pepExala',
Fast-Runner spoke. Well, then it is said they said, "O shamans
(dancers),
wc'gadzaxins ^nE'mplEna na'nageg-exg-axg-a wa'ldEmg-asg ins
let us for one time obey this desire of our
^ne^nEmo'kuk-.
friends here.
Wa,
Well,
we'g'a
go on
ax^e'dEx ma^Exsa' ts!a'ts!ax"-
take two flat short roof-
SEtna qEns we'gl yiLEdzoda'lasEk- laq," ^ne'x-^lae. Wa,
boards that we go on tie them on to them," they said. Well,
it is said.
5 laE'm^lawise Le'^IalasE%a ma^lo'kwe ba'bEbaklwa, yix
then it is said were invited two warriors, that is
QEnx%ida'yowe LE^e's ^nEmo'kwe
Frowned-upon and his friend
^na'xulalese qa
Warrior-all-round- that
the-World,
ga'xes yiLEdzo'ts la'xa ts!a'ts!a6x"sEme. Wa,
they come
and
tie them on
to the
daEm^la'wise
ately, it is said,
10
qa^s
and
they
Wa,
Well,
qa^s
and
they
Wa,
Well,
g"a xe
came
ax^e'tsE%a
were taken
pa'xalelEm
put them down
short roof-boards.
ma^lExsa'
two flat
Well,
he'x-^I-
immedi-
ts!a'ts!Ets!a6x°sEma
short roof-boards
la'xa ogwewali'iasa g-o'kwe.
in the rear of the house.
laE'm^awise
then it is said
da'x'^ItsE%a
were taken the
ma^lo'kwe
two
teto'x'wid
war-dancers
naEriLEdzoda'yowe la'xa ^na'FnEmxsa tsIa'tslaox^sEma.
were laid on their backs on the
each flat
laE'm^lawisa
then it is said a
g-i'Itla dEnE'm
long cedar-bark rope
yaLa'4ayosa ts'.e'daqe to'xw^ld
instrument of women war-dancers
tying of the
Wa, laE'm^laxaa'wise he' Em
Well, then, it is said, also, that
LE'^wa
and the
short roof-board.
ax^etsE^wa qa
was taken for
ts!a'ts!aox"sEme.
short roof-boards.
gwe'x-^ItsE%a
thus was done to the
1 5 Wa, laE'm^lawise gwa'lExs la'e
Well, then it is said they finished when then
ax^etsE^wa
were taken
''nEmo'kwe.
other one.
LESLEkwe'
thick
Eight: Wakashan and Salishan Languages 347
132
tE'mg-ik° lEqwa' qa^s qElxasusta'lasE'^we lax awi'^s-
blocks fire-wood and they were piled up at the around
tasa lEgwi'le. Wa, a'^mese la nExts!E%e'da
the fire. Well, and only then was in centre the
lEgwl'le laq. Wa, laE'm^lawise ha'lsElaEm^la heltsla'pElcda
fire in it. Well, then, it is said, hardly could look over a
g'i'ltlExsde bEgwa'nEm lax xixtsla'laq. Wa, laE'm gwa'lala
tall man in the putting head Well, then it is ready
out into it.
qae'da ma^lo'kwe teto'x^ld qo lal latslo'^yoLo laq. 5
on account two war-dancers if will will be put in into it.
of the then middle
Wa, he'^maaxs la'e ^ne'k-a teto'x^wide qa^s laxLa'nowe
Well, that when then they said the war-dancers that should be put
they on top
la'xa lEgwi'le. Wa, la ax^e'tsE%a ma^lExsa' ts!a'ts!aox"sEme
on the fire. Well, then were taken two flat short roof-boards
la ^na'^nEmax iyaak" LE%a' la nELEdza'yaatsa teto'x^wlde, Wa,
made just like also the now place of lying on war-dancers. Well,
back of the
la'^lae ax'^e'tsE^a q!wa'xe, Wa, la'^laxae ^na'^nEmax-iyaak"
it is said were taken hemlock- Well, then also, it were made just like
branches. is said, they
LO^ qeqExIma^yasa a'laklala to'x^wlda. Wa, laE'm- 10
also hemlock head-rings true war-dancer. Well, then
the of the
^lawise qEximda'yo la'xa q!a'q!Ek*o ma^lo'k" tsle'daqa.
it is said were tied around to the slaves two women.
they
Wa, laE'm^lawise la ne'nELEdzodayo la'xa ts!a'ts!Ets!aox"sEme.
Well, then it is said now were laid down on on the short roof-boards,
they their backs
Wa, laE'm^lawise yii^eda'yoweda gi'ltle dEnE'm laq lax
Well, then it is said were tied the long cedar-bark to in
ropes them the
gwa'laasasa ma^lo'kwe teto'x'^wida. Wa, laE'm^lawiseda
same manner as two war-dancers. Well, then it is said
the the
ma^o'kwe ba'bEbaklwa ^ne'x'xa ma^lo'kwe ts!e'daq q!a'q!Eka : 15
two warriors said to the two women slaves.
348
Vf American Indian Languages 2
133
" ^ya, sa'sEm, gu'no gwaLlEXLa'lalaxo qa'so lal lax'La'not
"O children! do not scream if you will you will be
I
put on top
axga'da
lEgVVi'iek'.
Wa,
he'^maaxs
k-!e'seLaqos
on this
■fire.
Well,
that when
you not
gwaLlEXLa'laLOL, wa, la'LEs mo'plEnxwa^s^Eml k"!es g'ax
you will scream, well, then you four (times) day not come
will,
q!ula'x'^idELOL. Wa, g-i'pEml^its gwaLlEXLa'laLoL, wa,
you will come to life. Well, as soon as you will you will scream, well,
5 la^me'sEnu^x" kwe'xaplELOL qa^s lelE'^'laos. Wa, laE'ms XEk'la'l
then we shall strike the so that you die. Well, then you will stay
nape of your necks away
IcIe'^U la'xaq," ^ne'x-^lae. Wa, laE'm^lawise ^na'xwa ya'qleg'a^e-
will be by this," they said, it Well, then it is said all spoke
dead is said.
da ^we'^woselaga ts!e'daq q!a'q!Ek-a. Wa, la'^lae ^ne'k-a:
poor women women slaves. Well, it is said said,
they
"We'g-a a'Em ha'^lilalax laE'mx- iak!we'masg-anu^x" ne^na'qek*
" Go on only do it quickly this we are strong in our hearts
qEnu'^x" k*!e'se gwa'LlEXLa'la qEnu'^x" k-!e'sel ga'ial qEnu'^x"
for we not scream so that we shall not shall be that we
long
10 ga'xei e'tledEl q!ula'x-^ida," ^ne'x'da^x^^Iae. Wa, laE'm^lae
shall come shall again come to life," they said, it is said. Well, then it is
said they
gwa'llla. Wa, la'^laeda no'Enlamala Le'^lalaso^s TsEx^wi'de
were ready. Well, it is said the fool-dancers were invited by Fast- Runner
qa^s
that
they
la
go
la
go
and
La'g ililaxa la ya'gudzayaatsa teto'xSvlde, qa*s
war-dancers that
they
carry up the now place of lying tied on
board of the
Lax"LE'ndEs la'xa lEgwi'le.
put it on top on the fire.
ga'xeda no'EnlEmalaxs
came the fool-dancers when
1 5 gwe'gudza. Wa,
winter-dancers. Well,
la'e
then
laE'm^lae
Wa,
Well,
^na'xwa
all
g-i'FEm^la'wise
as soon as, it is said,
qlwa'ga^lila
stood up in
house the
xo'lexulila. Wa,
then it is said they were all confused Well,
(running about).
Eight: Wakashan and Salishan Languages
349
134
laE'm^lawisa
it is said the
x'i'lplede
turned
around
no'EnlEmala
fool -dancers
La'g-a^lllaxa
took up in
teto'x^ide
war-dancers
qa^s
the
la'xa
in the
ogwewali'iaxs
rear of house
when
la'e
then
they
la'xa go'kwe.
in the house.
Wa,
Well.
la'^lae
it is said
they
X'i'lpled
turned around
la'xa
at the
and
they
la^stali'lElas
went around in
house with them
o^stall'Jasa
door of the
g'o'kwe.
house.
Wa,
Well,
la'^lae
it is said
they
e't!ed
again
lag'iyoll'lElaxs
went to rear
when
la'^lae
it is said
they
la'e
then
they
axa'Hlaxa
put down the
heyak ili'lEla la'xa helk*!ote'waliie. Wa,
went back to at the right-hand side. Well,
inner room
a'lak-!ala teto'x^id qa^s q!ula'Fedeq. Wa, la'^lae La'g-a^lllaxa
true war-dancers and hid them. Well, it is said took up the
they they
ma^lo'kwe q!a'q!Ek*o qa^s le laltla'lii la'xa gEmxote'walile.
two
Wa,
Well,
laxs
when
la'^lae
it is said
they
slaves and went
they
le^stall'iElas
went around with
them
went out at the
left-hand side.
qa^s
and
they
le
went
la'e
then
la'^laeda
it is said
the
g-a^ll'lEm
stopped
pled
around
%Elg'a^li'lEma
was stopped
^nEmo'kwe he^stali'lElaya
^nEmo'kwe
one
one
[other]
la'xa
at the
LE'*wa
and the
was taken around
qa's
and
he
la'xa
to the
q!a'k"o.
slave.
le'El
went
Ogwewali'iaxs
rear of house
o^stali'lElaxs
door when
la'e
when
they
la'e
then
they
^nE^ma'x-^it
at same time
o^sta'li-
door
Wi,
Well,
^weI-
was
xi'l-
turned
waxsano'dxa
were on each side
of the
qElxase'lakwe
piled-up
lEgwi'la.
fire.
Wa,
Well,
la"^lae La^na'kulamatsE^wa
it is said were placed upright
one after another the
ma%'x"de qIa'qiEkoxs la'e la'xxana. Wa, laE'm'^lae havve'xa
two former slaves when then were put on Well, then it is never
they top (of pile). said they
gwa'LlEXLalaxs la'e we'qumaxa. Wa, laE'm^lae lelE'^la. Wa,
screamed when were shoved Well, then it is were dead. Weil,
they down. said they
10
15
350 Vf American Indian Languages 2
135
gi'FEm^la'wise q!u'lx-^idExs la'e ax^e'tsE%a ma^tsE'me
as soon as, it is said, burned to ashes then were taken two
they
xa'xExatsEma qa^s g-a'xe ha'nEmg-a^lllEm la'xa ogwewali'le.
small boxes and came were put down on floor at the rear of house,
they
Wa, he'^latla le'da ^nE'msgEmeda o^sta'lile. Wa, la'^lae
Well, that, however, was one at the door. Well, it is said
it is said,
ax'-e'tsE^a gi'lt!a klipLa'lae qa^s k!ip!e'tsE^we xa'lxEqasa
were taken long tongs and they were picked up the bones of
5 ^naFnEmo'kwe q!a'k-o qa^s le k-!ipts!a'layo la'xa ^nal^nE'msgEme
each (person) slave and then were put in into the one to each (round)
they with tongs
xa'xatsEma. Wa, g-i'FEm^la'wise ^wl'^lts!axs la'e yikuyE'ntsE^wa
box. Well, as soon as, it is were all in then they were covered
said, they
yi'ses ye'yik"ya^e. Wa, la'^ae ha'nga^lllEm la'xa ogwewali'lasa
with their covers. Well, it is said were placed on at the rear of the
they floor
ts!a'gats!e. Wa, la'^lae mo'plEnxwa^se ^na'las he gwae'lExs
winter-dance Well, it is said, four times day day of that being thus in
house. house
la'e ya'laqweda teto'x^wide. Wa, laE'm qlula'x-^ldbola. Wa,
then sang their war-dancers. Well, then they pretended to Well,
sacred songs the become alive.
10 laE'm^Iae a'lax-^ld la ^ya'k'owe TslEx^i'de la'xeq. Wa, la'^lae
then it is truly then was beaten Throw-away after that. Well, it is said
said
TslEx^i'de ^nex- qa'^s le wl'naxa Mo'tsladxwe g*a'sa la'xa
Throw-away said that he would make war Nootka going at the
go upon the through
dzE^la'lasa ^nE'mgese. Wa, la'^lae te'nox%ed la'xa wa. Wa,
lake of the ^nE''mges. Well, it is said poled canoes on the river. Well,
they
g'i'FEm^la'wise la'gaa la'xa t!ex i'las TiE'se la'e yu'dux"sEnd
as soon as, it is went to the road of Nootka then in three pieces
said, they Inlet they
LE'mx'^idxes ^ya'^yats!e qa^s la ^we'k'ilk-ilaqexs la'e e'k'!e^sta
split their canoe and went carried it on their they went up
they shoulders when
15 la'xa nEg-a', Wa, g i'FEm^la'wise la'g'aa lax was TIe'sc
to the mountain. Well, as soon as, it is came to the river of Nootka
said, they Inlet
Eight: Wakashan and Salishan Languages
351
la'e
then
they
Wa,
Well,
tlE'mx-^ldxes ^a'^yatsle qa^s yo'Ix-^ide la'xa
sewed together their canoe and they drifted down at the
g'i'PEm^la'wise
la'x'SE^yod la'xa vvaxs
as soon as, it is arrived at at the river
said, they mouth
sex"sale'sEla la'xa tlo'kwaxs la'e a'Em
paddled through at a narrow passage then only
when they
la'EnaLlEme. Wa laE'm^lae
arrows. Well,
136
wa.
river.
la'e
then
they
q!a'yaxaxs
were startled
when
la'e
then
they
ha'nialExsElaso^sa
were shot at by
lE'^le
then it is was
said killed
TslEx^i'de
Throw-away
qlu'laxa
was alive
that
LE'we's
and his
g'a'xe
le'Elote.
crew.
ae'daaqa.
back.
Wa,
Well,
Wa,
Well,
la'^lae
it is said
he'^me
that is the
^nEmo'kwa 5
one
q!a'lag-iltsa
reason of know-
ing of the
ga'la
first
Kwa'g'ul
Kwakiutl
qexs
that
lE'^mae
they
lelE'^la.
were killed.
Wa,
Well.
la^me^
then
^ya'k-owe Ts!Ex*l'de la'xeq. Wa,
was beaten Throw-away after that. Well,
laE'm la'ba.
then end.
Editorial Note
Originally published in Boas Anniversary Volume. New York: G. E.
Stechert, 108-136(1906).
This paper, a contribution to the Festschrift honoring Boas on the 25th anni-
versary of his doctorate, was Sapir's first published work. Student records at
Columbia University (Murray and Dynes 1986) show that Sapir enrolled in
Boas's introductory course on "American Languages" in 1903-04 while he was
an undergraduate in Columbia College, and that he continued with a second
course in 1904-05 during the time he was enrolled as a Master s candidate in
Germanic Philology. According to the Columbia catalogue for those years, the
"translation and grammatical interpretation of Indian myths" formed part of
the work of both courses, and it is entirely possible that Sapir edited "The Rival
Chiefs" for this purpose. The text was one of those transcribed by George Hunt.
Boas's native Kwakiutl collaborator, and the original manuscript is preserved
among Hunt's Kwakiutl materials in the Columbia University Library.
352 VI American Indian Languages 2
George Hunt, a native speaker of Kwakiutl, gathered and transcribed large
quantities of ethnographic and Hnguistic material for Boas during a 40-year
association. Boas first met Hunt during his initial field trip to the Kwakiutl in
1886. In 1893 Boas arranged for Hunt to visit the World Columbian Exposition
in Chicago as part of a delegation of Kwakiutls, and their close collaboration
dated from that time. To facilitate their work Boas taught Hunt to write
Kwakiutl phonetically, and Hunt began collecting texts for transmittal to Boas
for editing and publication, with two of the resulting monographs explicitly co-
authored (Boas and Hunt 1902-05, 1906). Boas described Hunt s work in some
detail, including his orthographic practices, in his introduction to The Religion
of the Kwakiutl Indians (1930: ix-xviii). See also Helen Codere, "George Hunt
and Boas" (in Boas 1966: xxviii-xxxi).
Sapir, impressed by the productiveness of the Boas-Hunt collaboration,
established similar working relationships with several native speakers, includ-
ing Pete McGuff (Wishram), Tony Tillohash (Southern Paiute), Albert San-
doval (Navajo), and Alex Thomas (Nootka). The orthographic practices of
these men later formed an important part of Sapir s famous discussion of the
psychological reality of phonemes (1933c).
Some Aspects of Nootka Language and Culture
[excerpt]
The Linguistic Relationship of Kwakiuti and Nootka
The Wakashan linguistic stock is divided into two main branches, the
Kwakiuti and the Nootka or Aht; the former embraces Kwakiuti proper,
Xaisla, and He'ttsa^q", the latter Northern Nootka (from about Cape Beale north
to Cape Cook on the west coast of Vancouver Island) and Southern Nootka or
Nitinat (south of Cape Beale to Cape Flattery). By careful comparison of the two
Wakashan branches one can in part reconstruct a Wakashan "Ursprache," but the
actual differences between Kwakiuti and Nootka are in fact very great; they differ
perhaps as much as Slavic and Latin.
[16] As regards phonetics, Kwakiuti and Nootka, while both showing charac-
teristic Northwest Coast features, differ rather considerably. The sonant or inter-
mediate stop series of Kwakiuti is absent in Nootka, Kwakiuti p and b for instance
being replaced by Nootka p. Besides the ^-series, which Kwakiuti and Nootka
possess in common, Nootka has a c-series, which is doubtless derived from the
Kwakiuti and Wakashan /:--series, which in turn Nootka lacks; thus Kwakiuti g-
and k- are cognate with Nootka tc, k-f with tcf, and x- with c. There is no / in
Nootka, n corresponding to both Kwakiuti / and n. The velars q! and x, while
somewhat infrequently found in Nootka, are not the regular Nootka represent-
atives of Kwakiuti q! and x; q! has developed into a peculiarly harsh and choky
glottal stop, which I write ^, x into a strangulated-sounding h which I write //,
these two consonants respectively resembling Arabic 'ain and ha; ordinary ^
and h are also frequently found in Nootka. As regards phonetic processes,
Kwakiuti and Nootka agree in allowing no initial consonant clusters in words;
initial Kwakiuti and Nootka ^m, ^'n, ^ w, ^'y, and Kwakiuti ^/ are undoubtedly
related to ordinary Kwakiuti and Nootka m, n, w, y, and Kwakiuti / as are
Kwakiuti and Nootkap.^ t!, kf, iJ, tsf, qf, Kwakiuti k-!, and Nootka tcHo non-
fortis Kwakiuti and Nootka/?, t, k, l, ts, q, Kwakiuti k-, and Nootka tc. In both
Kwakiuti and Nootka certain derivative suffixes "harden" the final consonant of
the stem; thusp, q, and i, become Kwakiuti /7.^ qf, and ^7, Nootka /;.^ ^\ and ^'y.
The "softening" of Kwakiuti seems to be represented in Nootka by but a few
stray phonetic processes. Syllabically final glottal stops and glottally affected
consonants — such as -^/ and -p! — which are common in Kwakiuti, are entirely
absent in Nootka. Medial and final consonant clusters are not as freely allowed
in Nootka as in Kwakiuti, / often serving in Nootka to lighten them (cf. Nootka
-qEmit, 'round thing', with Kwakiuti -^Einl 'mask'). All final vowels and
354 VI American Indian Languages 2
Stopped consonants in Nootka are aspirated. Peculiar to Kwakiutl is the change
of A:-stops to spirants (x, x", x-) before consonants, whereas in Nootka they
remain; in this point Nootka seems more archaic than Kwakiutl.
In general morphology Kwakiutl and Nootka are quite similar, [17] despite
numerous differences of detail. In both the stem is, as far as its meaning allows,
indifferently verbal or nominal and one or more suffixes are required to give
rise to definitely verbal or nominal complexes; in Nootka a suffixed-'/' is often
used to substantivize a verb form. Both Kwakiutl and Nootka are absolutely
devoid of prefixes, most of the elaborate grammatical mechanism being carried
on by means of suffixes, to a lesser extent by means of initial reduplication, and,
in Nootka, consonantal changes. The suffixes of Nootka and Kwakiutl express
similar ideas and are used in more or less parallel fashion, though the number
of suffixes that are etymologically related form but a small percentage of those
found in either; so far about ninety Nootka suffixes have been discovered that
are entirely or in part cognate to Kwakiutl suffixes. Examples of local suffixes
shared by Kwakiutl and Nootka are: Kwakiutl -o-^yo 'in the middle', Nootka
-^'win'; Kwakiutl-A7e<7"'in the corner', Nootka -nikw-; Kwakiutl -atCls 'down
river', Nootka -atis; Kwakiutl -tsfo 'in', Nootka -ts!o'; Kwakiutl -k-E 'top of a
box', Nootka -tci' 'full'; Kwakiutl -!a^ 'on the rocks', Nootka -!a^a'\ Kwakiutl
-es'on the beach', Nootka -is; Kwakiutl -/7'in the house', Nootka-//; Kwakiutl
-xs 'in a canoe', Nootka -qs, -lans. A few examples of body-part suffixes are:
Kwakiutl -Ids 'cheek', Nootka -as; Kwakiutl -xo 'neck', Nootka -as-Haul
'chest'; Kwakiutl -dp! 'neck', Nootka -dpfaf 'back'. Important temporal ele-
ments held in common are: Kwakiutl -l 'future', Nootka -^dq-L, -^il; Kwakiutl
- x-^ld 'inceptive', Nootka -ci-L. There are some striking agreements in verbify-
ing derivative suffixes, as: Kwakiutl -lexst 'to desire', Nootka -Hh" 'to try to
get', -st!iH^ 'to have as goal'; Kwakiutl -!a 'to go in order to', Nootka -fas;
Kwakiutl -k-!dla 'to make a noise', Nootka -^en' (= Wakashan *-q!Ela);
Kwakiutl -g-a^i 'beginning of a noise', Nootka -^aL (= Wakashan *-q!a^l)\
Kwakiutl -qlES 'to eat', Nootka -lis; Kwakiutl -nuh' 'to have', Nootka -nak\
Examples of nominal suffixes are: Kwakiutl -aano 'rope', Nootka -d^nul 'long';
Kwakiutl -gas 'woman', Nootka -^aqs; Kwakiutl -asde 'meat', Nootka -act'
'dried meat'; Kwakiutl -mis 'useless', Nootka -mis 'mass'; Kwakiutl -p!e-q [18]
'stick, tree', Nootka -p.'U 'long board-like object', -q- 'tree'; Kwakiutl -(x)^'Enx
'year, season', Nootka -q'^itcH" 'year', -fitcH" 'season'. On the whole it seems
that Nootka has a rather larger number of derivative suffixes than Kwakiutl,
many quite special ideas being expressed by means of suffixes where there seem
to be no Kwakiutl equivalents. A few examples are -a/ 'blanket'; -'m//' 'son';-^^
'daughter'; -Htut 'to dream of; -!d^il 'to ask for as a gift in a girl's puberty
ceremony'; - tfota'' 'to give a potlatch for'; -yaqn'^ 'to sing a song'; -Hl 'to begin
to sing a song'; -lini 'to give a feast of; -Hd' 'to buy'.
Both Kwakiutl and Nootka make use of two kinds of reduplication, one in
which the first consonant, first vowel, and second consonant of the stem are
repeated, and one in which only the first consonant and vowel are repeated; the
1. ! denotes a "strengthening" of the preceding consonant.
Eight: Wakashan and Sallshan Languages 355
former type is employed in forming iteratives, the second in forming plurals or
distributives and with certain suffixes (such as Kwakiutl -fa, Nootka -fas 'to
endeavor, to go in order to'; Kwakiutl -^ydla 'to go to look for'; Nootka -////" 'to
try to get'; Nootka -kfok'" 'to look like'). In Nootka the repeated vowel is in all
cases the same as that of the stem, in Kwakiutl the second type of reduplication
has a definite vocalism {e in some cases, a in others) in the reduplicating sylla-
ble. In Kwakiutl verb stems ending in vowels inserts- after the first, k- after the
second syllable of the iterative, while Nootka iteratives of like form insert l and
y; Nootka sd- 'to crawl' forms iterative sd' Lsdtc, -tc being probably identical
with Kwakiutl -k-. One other striking resemblance of detail between Kwakiutl
and Nootka may be noted: both Kwakiutl diminutives in -Etn and Nootka nouns
in -kwin' 'toy' require reduplication of the stem.
In regard to pronominal development there is considerable difference
between Kwakiutl and Nootka. While there is, practically speaking, but one
series of personal pronominal suffixes in Kwakiutl, there are three in Nootka
(represented, for second person singular, by -e^its, -k\ and -sok'), of which the
second and third are etymologically related: the first Nootka series is used in
indicative forms of verbs, the second in subordinate clauses, interrogatives,
and possessive forms of nouns, while the third seems to be confined to [19]
certain modal forms. Kwakiutl has distinct forms for first person plural
inclusive and exclusive, while Nootka has only one form for both. Pronominal
objects are, to at least a considerable extent, incorporated in Kwakiutl; in
Nootka, however, only in the case of the first person (second series) of the
imperative. A great degree of complexity in pronominal forms is brought about
in Kwakiutl by the combination of the pronominal affixes with syntactic (sub-
jective, objective, and instrumental) and demonstrative elements. Nootka has
none of this syntactic and demonstrative complexity of the pronoun, but a
series of forms is found built up of the second pronominal series and an element
-tc implying that the statement is not made on the authority of the speaker.
Almost all Nootka and Kwakiutl words are noun or verb forms, there being
almost no particles properly speaking. Such apparent Nootka conjunctive and
case particles as ^onoi. 'because', ^oyi' 'when, if, and ^okwil 'to' are mor-
phologically verb forms built up of a stem^o- 'a certain one, thing' and deriva-
tive verbifying suffixes. There is, however, in Nootka a syntactically important
conjunctive element ^ani' 'that' to which may be appended pronominal affixes
of the second series and which may perhaps be considered a particle in the
proper sense of the word. The "empty stem," Nootka ^6-, is cognate with
Kwakiutl 6- 'something,' which, however, is used primarily in noun forms.
Other Wakashan "empty stems" are: Nootka ^ap-, ^am-, Kwakiutl dps-, used
chiefly in forming nouns of body-parts that occur in pairs, and Nootka hit-, hi-
'to be at', Kwakiutl he- 'that'; peculiar to Nootka is hin, hit- (before "harden-
ing" suffixes) 'to be or do (as indicated by derivative suffix)'.
In regard to vocabulary Kwakiutl and Nootka differ greatly. Considering the
very striking morphological agreements between them it is somewhat disap-
pointing to find comparatively few stems held in common. It is highly impor-
tant, however, to note that many of these are rather colorless in content and
356 Vf American Indian Languages 2
thus hardly to be suspected of having been borrowed in post-Wakashan times.
Such are Kwakiutl ^nd-, Nootka ^nds 'daylight'; Kwakiutl g-dl-, Nootka tcdn-
'to be first'; Kwakiutl ax- 'to do, be', Nootka 'o//- 'to be'; [20] Kwakiutl we-,
Nootka wi-, wik" 'not'; Kwakiutl g<?, Nootka qe' 'a long time'; Kwakiutl ^riEm-,
Nootka ^'nup- 'one'; Kwakiutl gwe- 'thus', Nootka qwi- 'to be or do thus';
Kwakiutl .s^o-, Nootka 50- 'you'; Kwakiutl ek-!-, Nootka '77c/- 'above'. Thus Dr.
Boas' first announcement in 1890 of the close relationship between Kwakiutl
and Nootka has been confirmed in every way by new evidence.
Editorial Note
Excerpt from "Some Aspects of Nootka Language and Culture", originally
published in the American Anthropologist 13, 15-28 (1911); the article appears
in full in Volume IV. Reprinted by permission of the American Anthropologi-
cal Association. Sapir's ms. corrections have been incorporated.
For a much fuller presentation of the data on which this paper is based, see
the "Wakashan Comparative Vocabulary," based on Morris Swadesh's rework-
ing of unpublished materials of Sapir (now in the Boas Collection, American
Philosophical Society Library), published for the first time in Volume XIL
Sapir also returned to comparative Wakashan in a late paper on "Glottalized
Continuants in Navaho, Nootka, and Kwakiutl, with a Note on Indo-Euro-
pean" (1938b, Volume II), where he showed that the glottalized sonorant con-
sonants (v, vv, /, and the nasals) of Wakashan have arisen from the coalescence
of ^ and h with following or preceding plain sonorants. Recent work on com-
parative Wakashan, which has largely been confined to studies within the
Kwakiutlan and Nootkan branches, is surveyed in Jacobsen (1979).
Abnormal Types of Speech
in Nootka
An interesting linguistic and cultural problem is the use
in speech of various devices implying something in regard to
the status, sex, age, or other characteristics of the speaker,
person addressed, or person spoken of, without any direct
statement as to such characteristics. When we say "big dog
make bow-wow" instead of "the dog barks," it is a fair inference
that we are talking to a baby, not to a serious-minded man of
experience. Further, when we hear one use "thee" where
most would say "you," we suspect that we are listening to an
orthodox Quaker. In neither of these cases is there an explicit
reference to a baby as person addressed or to a Quaker as person
speaking. Such implications are common in all languages and
are most often effected by means of the use of special words or
specific locutions. Thus, in Nootka there are special words
used in speaking of obscene matters to or in the presence of
women; a number of "baby- words" also exist. Generally it
is the speaker or person addressed that is thus signalized, but
it is quite possible, though less frequent, to thus imply some-
thing also in regard to the third person. A more specialized
type of these person-implications is comprised by all cases in
which the reference is brought about not by the use of special
words or locutions, that is, by lexical, stylistic, or syntactic
means, but by the employment of special grammatical elements,
consonant or vocalic changes, or addition of meaningless
sounds, that is, by morphologic or phonetic means.
358 Vf American Indian Languages 2
To enumerate all the possible types of person-implication
expressed in language, from the point of view of resulting
classifications of human beings, would lead one far afield.
Two types, however, seem to stand out most prominently —
those referring to sex-discrimination and to rank-discrimination.
Several languages make a distinction between words or forms
used by males and such as are restricted to females. Such a
distinction, for instance, is made by certain Eskimo dialects,
in which, at least in earlier times, according to Boas,^ final p,
t, k, and q^ were pronounced by the women as the correspond-
ing nasals m, n, V, and rj. In Yana, an isolated linguistic stock
of northern California, the forms used by the women, whether
in speaking to one another or to males, differ from the fuller
forms used by the latter in the unvoicing of final vowels; final
-na {-hi in Southern Yana), a common noun ending, is replaced
by aspiration in the speech of the women, who further lengthen
final vowels to express the interrogative, while the males suffix
an element -n. Most languages that make such sex distinctions
differentiate the sexes as speakers. In Yana, however, a further
discriminating factor is the sex of the person spoken to, in so
far as the men in speaking to the women use the forms charac-
teristic of the latter.
More widespread in language seems to be a discrimination of
forms according to the rank or social status of the person speak-
ing, addressed, or spoken of. Here belong the etiquette forms
characteristic of several East Asiatic and Indonesian languages,
by which the social grading of the speakers as inferiors or
superiors in reference to one another is clearly reflected in their
speech. An analogous American instance is the use in Nahuatl
of reverential forms to imply respect to the person addressed or
spoken of. These are morphologically nothing but indirectives
or causatives in -lia, -Ha, or -Ilia with reflexive pronominal
prefixes; "he sleeps" is thus more politely expressed as "he
causes himself to sleep." Here belongs also the use in so many
European languages (French, German, Russian, and others)
of second or third person plurals, instead of the more logical
second person singulars, in speaking to people with whom one
* Handbook of American Indian Languages, Bulletin 40 of Bureau of American Ethnology,
1911. p. 79.
' See Phonetic Key at end of this paper.
Eight: Wakashan and Salishan Languages 359
is not on the most intimate terms. This usage has its parallel
in Yana, where brothers and sisters address each other in the
pluraP; other Californian examples of a similar nature have
been given by Goddard^ and Kroeber.^
These preliminary remarks are intended merely to indicate
the general class of linguistic phenomena to which belong the
more specialized Nootka examples to be given presently. At
the same time they will serve to render these latter less glaringly
bizarre by providing them with parallels of a more general
character. The data here presented were chiefly obtained in
November, 1910, in the course of ethnologic and linguistic re-
search for the Geological Survey of Canada among the Nootka
Indians of Alberni canal, Vancouver island; the informant was
Dan Watts, the young chief of the Hopdtdas'atH" tribe. Further
data on this subject were obtained in the winter of 1913-14 from
Alex Thomas, a young Indian of the TsUcd'atH" tribe of the
same region.
It is possible and often customary in Nootka to imply in speech
some physical characteristic of the person addressed or spoken
of, partly by means of suflSxed elements, partly by means of
"consonantal play." Consonantal play consists either in alter-
ing certain consonants of a word, in this case sibilants, to other
consonants that are phonetically related to them, or in inserting
meaningless consonants or consonant clusters in the body of
the word. The physical classes indicated by these methods
are children, unusually fat or heavy people, unusually short
adults, those suffering from some defect of the eye, hunchbacks,
those that are lame, left-handed persons, and circumcised males.
In speaking to or about a child it is customary to add the
regular diminutive suflfix -'is to verb or other forms, even
though the word so affected connotes nothing intrinsically
diminutive; affection may also be denoted by it. The -'is
comes before temporal, modal, and pronominal suffixes. Thus,
the normal qwlstci" "do so!" {qwis- "to do thus;" -tci' second
person singular imperative, "go and . . , !") is changed to
qwis'istci" "do so, little one!" when speaking to a child.
' Sapir, Yana Texts, University of California Publications in Arncrican Archaeology and
Ethnology, vol. 9, 1910, p. 95, footnote 139; p. 101, footnote 150.
» Goddard, Kalo Texts, ibid., 1909, vol. 5, p. 143, footnote 185.
' Kroeber, The Languages of the Coast of California north of San Francisco, ibid., 1911,
vol. 9, p. 321 (Porno).
360 ^f American Indian Languages 2
Similarly, qwisma^ "he does so" {-ma'^ third person present
indicative) is changed to qurls'ismd' when one is speaking about
a child. In speaking about oneself or others when addressing a
child, it does not seem to bo customary to use the diminutive
suffix except to show affection at the same time. Thus, the
word waidLaH "I am going home" {wal- "to return home;"
-ciL- inceptive; -an "I") may be changed to walciL'tsan "I
am going home, little one" when addressed to a child for whom
one wants to show love, but this form would not be used in
speaking to a child that is a stranger. As might be expected,
diminutive verbal and other forms occur in lullabies, in some
of which the child is represented as speaking about itself.
Thus, in a lullaby supposed to be sung by a whale mother to
its child, occur the words 'oH'^'eso¥ '.emiti' ("my) little name
is" ('o//"- "to be;" -'is- diminutive; -ok^ "of, belonging to;"
lemiW "name"). Some people were said by Dan to have the
habit of using the diminutive suffix in order to belittle others, as
though the persons addressed or referred to were of no more
importance than children as compared to themselves. If a
chief does this to too great an extent, he is set down as haughty.
In talking to or about fat people or people of unusual size,
the suffixed element -aq' is used in a manner analogous to the
diminutive -'is. Thus, the normal hinV cihwe' in^ "he comes,
it is said" {hin- "empty" verb stem "to be, do;" -V-, shortened
form of -tw* "to come;" -a"L- inceptive; -we'in^ quotative)
becomes hinVciLaq'we'in^; 'otsdtciLma' "he goes to it" {'o-
"empty" noun stem meaning "something;" -tsa- "to start for,
go to;" -tciL- inceptive, used after vowels; -ma' third person
present indicative) becomes 'otsatciLdq'ma\ Other examples
are: ha'gkwaq'ma' "he, clumsy one, eats;" (ha'w- "to eat;"
-okw- intransitive verbal suffix); and ha'okwdqiVnak'^ "did you
eat, fatty?" {-iV tense suffix denoting past time; -na- inter-
rogative; -/c' second person singular).
People who are abnormally small are spoken of in forms
with the diminutive suffix; moreover, in such cases, all sibilant
consonants (s, is, is!; c, tc, id) become palatalized c- sounds
(s, th, t^!; compare, for s, Polish s and Sanskrit f ; for ts, com-
pare Polish c), which sound acoustically midway between s-
and c- sounds; the diminutive -'is itself becomes -'is. Thus,
Eight: Wakashan and Salishan Languages 361
hinVciLweHn* "he comes, they say" is changed to hinVkiL-
'Uwe'in* "he, Httle man, comes, they say." These §- forms are
also used to refer to small birds, such as sparrows and wrens.
Sometimes a meaningless ^ is added to the word, as in wiMh"^
t6Hauk' from wikdn" toHauk^ "I am not afraid" (wik- verb stem
"to be not;" -d//" first person singular present indicative; ton-
verb stem "to be afraid;" -uk\ diphthongized to -auk' because
of preceding a- timbred h, intransitive suffix). We shall meet
this consonantal change again further on in another connexion.
Quite analogously to dwarfs, are addressed or spoken of those
suffering from some defect of the eye. Under this category
are included cross-eyed people, those who squint, and such as
have one eye run out, but not the blind. Here again the
diminutive suffix is used, with the added feature that all s-
sounds and c- sounds are converted into the corresponding
voiceless lateral stops or spirants (s and c become I; ts and tc
become l; ts ! and tc ! become lI); the diminutive -'is itself
becomes -'il. This style of speech is termed daddtckHn'- "to
talk in sore-eyed fashion" (cf. Lladdtck'sul "one-eyed per-
son"). Thus, qimsma' "he does so" is changed to qwll-
'ilma'. Similarly, tditciLma' "he cuts" {tdi- "to cut;" -tciL-
inceptive; -ma' third person present indicative) becomes
LliLJL'ilma'. A full-grown Indian named Sammy (or Se'mi
as pronounced in Nootka), who is cross-eyed, is referred to
as ie'mi'il "little cross-eyed Sammy." Another Indian of the
same tribe, To'mic, who has only one good eye, is, in parallel
fashion, referred to as To'miVil "little one-eyed To'mic." It
should be remarked that such people, particularly when adult,
are apt to become offended if addressed in this fashion, and that
one would not use such forms in their presence unless with the
express purpose of showing contempt or of teasing. As will be
seen again later on, L!aL!dtck!in^ forms are used also in referring
to the deer^ and mink. Thus, the mythological Mink, tc/dsti-
mits'miC "Mink-son," is generally referred to as Lldltimifinit'.
Hunchbacks (k.'wdpi') are also addressed or spoken of in
forms provided with the diminutive suffix, a further peculiarity
in these being the change of ordinary s- sounds and c- sounds
* Deer is associated with sore eyes also in other Indian mythologies. An Ojibwa einmplo
may be found in P. Radin, Some Mylks and Tales of the Ojibwa of Southeaatern Ontario, Gooloji-
cal Survey of Canada, Memoir 48 (No. 2, Anthropological Series], p. 3 (episode d).
362 VI American Indian Languages 2
to peculiar thickish c- sounds, pronounced with the lower jaw
held in front of the upper; the diminutive -is appears as -He.
We may represent these c- sounds by c. In this hunchback
talk qivisma'^ becomes qwicHcma\ Other examples are: ydtcuk'-
'i^ma' "he is walking" (yats- "to walk;" -uk'- intransitive verb
suffix); tc.'dtck'^miniH'^'icma'^ "all of them are" {tddick"- "to
be all;" -miniH'^- plural); and tcldxciLHcma' "he spears"
{ts.'ax- "to spear"; -oil- inceptive). Here again these distinc-
tive forms are generally avoided when in the presence of hump-
backed people, for fear of giving offence. However, a hump-
backed child who is well known to the speaker would hardly
take offence and would be addressed as described. Or, if an
old humpbacked woman is good-natured, c- forms may well be
used when she is about, as though to show that she is happy
and not easily ruffled. Here the notions of contempt and
affection commingle.
In speaking of lame people the diminutive suffix is again
used, this time in its normal form. Besides this, the meaning-
less element lc or lci is inserted in the body of the word some-
where before the diminutive suffix, its exact position apparently
depending on the whim of the speaker. Thus, hinini'aLma"
"he comes now" {hin- "empty" verb stem; -ini- "to come;"
-ai- determinative suffix marking point of time, "now"; -ma'
third person present indicative) becomes hininiLcVitsfaLmd'
(diminutive -'is and -'ol regularly combine to form -itslab)
or hiLcnini'its.'aLma" "the lame chap is coming." Similarly,
the verb tdltd'aLma' "he cuts now" (inceptive -tciL and -'gl
combine into -tci'aL) is changed to tditciLditsIaLma' when a lame
person is spoken of. The word tfa'ne'is'i" "the child" {t.'a'na-
"child, son, daughter;" -is diminutive suffix, i causing preceding
a to become umlauted to e; -i" nominalizing element, about
equivalent to our definite article) becomes tlaLcne'is'i' "the
young lame fellow," which may be used in speaking to children.
In speaking of or to left-handed people the diminutive suffix
is used in its normal form, besides which the meaningless element
few" is inserted after the first syllable of the word. Thus, ydV-
aLma' "there now he is" (ydl- "to be there;" -'ai- and -ma' as
above) becomes ydltcn'^'itslaLma' (-'is and -aL combine to
form -iislaL) "there now he is, poor little left-handed chap!"
Eight: Wakashan and Salishan Languages 363
Similarly, from sukwi'ttLma' "now he takes it" {su- verb stem
"to take;" -kwiL inceptive suffix, changed to -kwi- before -ai)
is formed sutcH^kwiL^itslaLma'. The diminutive suffix may also
be omitted. Examples are: hitcH'^nin^ from hinln^ "to come";
and tHtcHHciLan from tlitcxLaH "I throw it down" {Hi- "to throw;"
-tcib inceptive suffix; -an first person singular indicative). Such
a form as the last might be appropriately used in speaking to a
left-handed person that one is well acquainted with and who will
not take offence at being thus twitted. It is customary, particu-
larly for jokers, to use these left-hand forms also in talking
about bears, who are supposed to be left-handed.^
In speaking of or to circumcised males, forms known as
H'icVklin'' "to make cC- sounds" are used. In these the mean-
ingless element cV is inserted after the first syllable of the word.
One of the TsUcd^atn'^ Indians, named T!6xmis "Slaying-while-
moving-from-beach-to-beach," is often humorously referred to as
Tfdctxmis because of his having been born circumcised. Other
examples of this class of forms are: hict'ninnna' from hininima'
"he comes;" and /idcf'pA;'" from hd'ok'"" "to eat."
Similar phonetic changes are made in forms used to refer to
one or two classes of individuals characterized by some mental
quality. Thus, greedy people are addressed or referred to in
forms having a meaningless tcx inserted after the first syllable of
the word. Thus, from 'ou'^sdmaH "I hunger for it" ('o-
"empty" stem which may be rendered by "something" or "so
and so;" -u^sd- verbifying suffix "to desire to eat;" -maH first
person singular present indicative, used after vowels) is formed
'utcxHsdmaH. Similarly, hinini'aLma" "now he comes" becomes
hitcxnini'aLma' "now he comes, greedy fellow that he is." These
tcx- forms are also used to refer to ravens, regularly to the mytho-
logical Raven, a character noted for his gluttony.
Cowards may be satirized by "making one's voice small" in
referring to or addressing them, in other words by speaking in a
thin piping voice that suggests timidity.
It is interesting to notice that in several of the above usages,
the notions of mere smallness, of contempt, and of affection are
found side by side, and doubtless the precise nuance of feeling
expressed depends much on the relations subsisting between
' According to Dr. Paul Radio, the Winnebago also consider the bear to be left-handed. !■
the bear clan feast of these Indians the guests eat with a spoon in their left hand.
364 ^J American Indian Languages 2
8
the speaker and the person addressed or spoken of. What is
meant in the spirit of pitying affection for a poor lame or hump-
backed child or for a good-natured squinting old grandpa,
might be intended to convey contempt when addressed to a
young man and would be promptly resented as an insult. It
is significant that the various types of abnormal forms of speech
that we have reviewed are used with little or no reserve when
speaking of the persons referred to or when addressing children,
but are, on the whole, avoided when within ear-shot of adults
so referred to. It seems further significant that the traits
satirized are chiefly such as are inherent in a person, not merely
acquired in the accidental course of events, whereby he is set
apart by nature as falling short in some respect of the normal
type of individual and is to that extent stamped as inferior.
This may explain why blindness, which is more often acquired
rather late in life than congenital, is not made the subject of
speech-mockery. Added to this may be the feeling that blind-
ness is too grave an affliction to be treated light-heartedly, an
explanation which gains weight when the well-known sensitive-
ness of the Indian is considered.
Outside of the normal use of the diminutive in addressing or
referring to children, the peculiar forms of speech that we have
seen to obtain in Nootka are not easily paralleled in America.
For diminutive verbal forms of the Nootka type Uto-Aztekan
affords a close parallel. In Southern Paiute the regular diminu-
tive suffix -tsi-, which is employed to form diminutive nouns
and adverbs of all sorts, is also used as a verb suffix when speak-
ing to or of a child. Cognate with this element is the diminutive
suffix -tzin{tli) of Nahuatl. Derived from this is the verb
suffix -tzinoa, "which," according to Remi Sim6on,^ "serves to
denote respect or love;" it is generally, like reverentials of the
type already referred to, employed with reflexive prefixes.
Examples given by Remi Simeon are: otechmo-chiuilitzino in
Totecuyo "our Lord created us" (o preterit prefix; tech- first
person plural objective prefix; wo- third person reflexive
prefix; chiui-, from chiua, because of following -li-, verb stem
"to make;" -li dative suffix, mo- . . . .-li "for himself;"
-tzino reverential, final -a being dropped because of preterit
' Dictionnaire de la Langue Nahuatl ou Meiicaine, s.v. tzinoa.
Eight: Wakashan and Salishan Languages 365
tense; in definite article, "the"; to- first person plural posses-
sive prefix; tecuyo noun stem "lord"); and timo-qauhtzinoa
(quoted from Olmos) "you fast" {ti- second person singular
gubject; mo- reflexive;^ gauh-, from gaua verb stem "to fast;"
-tzinoa reverential). These forms may be rendered in some
such fashion as: "our Lord has created us for himself, revered
one," and "you fast, honoured sir."
Strikingly similar psychologically to the cases of consonantal
play in Nootka just considered are the peculiar consonant
changes characteristic of Chinookan, employed to convey
diminutive and augmentative notions respectively in all parts
of speech.^ The change here of c- consonants to s- consonants
to express the idea of diminution further illustrates the tendency
of sibilants in America to be subject to consonantal play. In
Yana the phenomenon of diminutive consonantism is illustrated
in the change of I to n. This process takes place regularly in
forming diminutive nouns in -p.'o; thus, ntnimaupla "little
nose," from ltlimau{na) "nose." The l-n type of consonantal
play is another one of some currency in America, and seems to
obtain also in Sahaptin. This matter of consonantal play to
express modalities of attitude is doubtless a fruitful field for
investigation in American linguistics and should receive more
attention than has hitherto been accorded it. It may be
expected to turn up particularly in connexion with notions of
smallness, largeness, contempt, affection, respect, and sex-
differences.
Such consonant changes and increments as have been con-
sidered are evidently of a rhetorical or stylistic as much as of a
purely grammatical sort. This is borne out by the fact that
quite analogous processes are found employed as literary
devices in American myths and songs. I have already drawn
attention to the fact,^ that in American mythology certain
beings are apt to be definitely characterized by speech peculiari-
ties. The employment of consonantal play or of similar devices
in such cases seems always to have a decidedly humorous effect.
• This verb is intrinsically reflexive.
* See Sapir, Preliminary Report on the Language and Mythology of the Upper Chinook,
American Anthropologist, N.S., 9, 1907, pp. 537, 538; and, in greater detail, Sapir, section
on "Diminutive and Augmentative Consonantism in Wishram," in Boas, Handbook of American
Indian Languages, pp. 638-645.
• Sapir, Song Recitative in Paiute Mythology, Journal of American Folk-Lore, XXIII,
1910, pp. 455-472. Takelma, Ute, Chinookan, and Nootka examples are there given, p. 471.
366 VI American Indian Languages 2
10
The culture-hero KwdtiydV of Nootka mythology is in the habit
of inserting a meaningless x after the first vowel of a word;
thus, the normal form htnuse'i" ''come up out of the water!"
{hln- empty stem "to do, be;" -use-, umlauted from -usa-
because of following i, "to move up out of the water;" -'z'
imperative singular) becomes, at the same time, inasmuch as it
occurs in a song, with song-vocalism, hlxnusa'e. In the speech
of the Deer and Mink all sibilants, whether of the s or c series,
are transformed into the corresponding laterals (s and c to I, ts
and tc to l, tsf and tc! to l/). Thus, the Deer says Limil for tdmis
"black bear;" U.dpaL for tddpats "canoe." The Nootka Deer
and Mink style of talking is of particular interest for two rea-
sons. In the first place, it will have been noticed that the
consonantal changes are identical with those employed in speech
about or addressed to those that have some defect of the eye,
the latter type of forms, of course, being further characterized
by the use of the diminutive suffix -'il (from -'is). Here we
see at once the intimate connexion between the two types of
consonant play. In the second place, the speech of the
Nootka Deer and Mink offers an interesting parallel,
or rather contrast, to that of the Kwakiutl Mink. This
latter character regularly transforms all laterals to cor-
responding s- sounds {I, L, I, and lI become respectively s, is,
dz, and tsl), the exact reverse of the Nootka process. From the
• point of view of the psychology of phonetics, it is significant to
observe that both Nootka and Kwakiutl have a feeling for the
interchangeability of the sibilant and lateral series of consonants.
But the Mink of the Kwakiutl is not content with this. He
also regularly transforms all anterior palatals to corresponding
sibilants (x'. A;', g , and Ml become respectively s, is, dz, and ts!).
There are still other phonetic changes to be found in Boas'
Mink texts, but they seem less regular in character than these
two; the changes at times of I and 'I to y and 'y may be instanced
as one of these (thus ss'yt for h'lt "dead").^ Now it is per-
haps significant that the change in Kwakiutl of anterior palatals
to sibilants is curiously like the change of original Wakashan
(Kwakiutl-Nootka) anterior palatals, as preserved in Kwakiutl,
• For data on Mink's peculiarities of speech, see F. Boas and G. Hunt, Kwakiutl Texts —
Second Series, Publications of the Jesup North Pacific Expedition, vol. X, 1906, footnotes o
pages 82 to 154: and Boas, Kwakiutl Tales, Columbia University Contributions to Anthr
pology, volume II, 1910, footnotes on pp. 126-154.
Eight: Wakashan and Salishan Languages 367
11
to c- consonants in Nootka.^ Thus, a Mink form nsdzt in
Kwakiutl for normal UEg't "mountain" is strikingly similar to
the regular Nootka cognate nutci\ Suggestive also, a propos of
the use by Mink of sonant palatal spirants {y and 'y)
for normal sonant laterals {I and 'Z), is the fact that in
Nootka so-called "hardening" suffixes change immediately
preceding I to 'y, corresponding in such cases to Kwakiutl 'Z.^
The bearing of these facts on mythological consonant play in
Kwakiutl is not easy to determine; a possibility will be
suggested farther on.
Consonant play as a device in mythology is not confined to
America. In reading some recently published Bushman litera-
ture the writer came across striking parallels. The Bushman
Mantis, who, like the Kwakiutl Mink, is a trickster, consistently
changes all the cerebral clicks of normal speech into lateral
clicks.^ Similarly, the Baboon transforms all the clicks of
ordinary speech into a compound click, consisting of cerebral
followed by dental click.^ Evidently a comic effect is aimed at
in both these cases.
The phenomenon of consonant and vocalic play is also well
illustrated in Indian songs. Song diction is an extremely
important, though rather neglected, field of primitive lore, and
only one phase of it can be touched on here. Song texts often
represent a "mutilated" form of the language, but study of the
peculiarities of song forms generally shows that the normal forms
of speech are modified according to definite stylistic conventions,
which may vary for different types of songs. Sometimes sounds
are found in songs which do not otherwise occur in the language.
Where the texts of a type of songs are in the language of another
tribe, as happens so often in America, such an abnormal sound
may be simply borrowed from the foreign language, as is the
case with the mourning songs of the Southern Paiute, which,
sung to supposedly Mohave texts, contain many examples of I,
a sound otherwise unknown in Paiute. On the other hand,
new sounds may be developed spontaneously or in imitation of
foreign sounds. The former is probably the case in the frequent
1 See Sapir, Some Aspects of Nootka Language and Culture, American Anthropologist,
N.S., 13, 1911, p. 16.
^ See Boas, Handbook of American Indian Languages, pp. 430, 435; Sapir, loc. cit.
' Bleek and Lloyd, Specimens of Bushman Folklore, 1911, footnotes on pp. 6 and 8.
* Ibid. Footnotes on pp. 18 and 22. At least this is indicjited by Block's orthography, though
possibly the compound sign is meant to indicate a special click not otherwise found.
368 VI American Indian Languages 2
12
Nootka use of 77, a sound quite foreign to normal Nootka speech,
in certain classes of songs; the latter explanation is more
plausible in the case of the regular Nootka change of n to Z in
many songs. This n-l interchange, again, is significant in so far
as Kwakiutl, doubtless agreeing in this respect with primitive
Wakashan, has both n and I, while Nootka, when cognate words
are compared, is seen to have only n to correspond to both.
Of particular interest in this connexion is the fact that such
special song-sounds (Paiute I; Nootka I and rj) are, at least so
it would seem, pronounced with difficulty by Indians under
ordinary circumstances, as in the handling of English words
that contain them. The obvious inference is that one may
react quite differently to the same speech-sound entering into
dissimilar associations. This fact, has, of course, a much wider
psychological significance.^ Conventional consonant changes
in songs are no more restricted to America than, as we have
seen, are parallel changes in mythology. An example that
happens to have come to the writer's attention lately is
the change of voiceless stops to corresponding nasals plus
voiced stops in the songs of the Karesau-Papua of German New
Guinea. Thus, the normal apil becomes ambil in songs.'
In seeking some comparatively simple basic phenomenon,
from which, as a starting point, the various types of consonant
play we have illustrated from Nootka could have originated,
• one easily thinks of the vocalic changes or consonant substitu-
tions that take place in the speech of those who have some
specific speech defect. The most familiar case of this sort in
English is lisping, which simply means that the ordinary
alveolar sibilants (sometimes also stops) are changed to the
corresponding dental sibilants or even interdental fricatives
(and sometimes correspondingly for stops). Information was
obtained of five types of speech defects found among the Nootka.
The first of these is called ninikHn^ {nini- reduplicated stem;
-k!in} "to make a sound of") and consists of the involuntary
' Sounds falling outside the regular phonetic system of the language may be spontane-
ously developed also by the operation of other systems of consonantal (or vocalic) play than
are found in song diction. Thus, in Wishram (Upper Chinookan), the analogy of certain
consonant changes of augmentative value (as of p to 6, ( to d, k to g) brought about the
creation of dj, a sound otherwise unknown in Chinookan, as the augmentative correlate of
tc or Is sounds. See Handbook of AmeTxran Irnlian Languages, pp. 638, 639, 640.
' See Father W. Schmidt, abstract of Ober Musik und Gesange der Karesau-Papuas ,
Deulsch Neu-Guinea, Bericht iiber den III. Kongress der Internationalen MusikgesalLschaft,
1909, p 297.
Eight: Wakashan and Salishan Languages 369
13
nasalizing of all vowels and continuants. Thus, the normal
hayd'akaH "I do not know" (-an first person singular present
indicative) is pronounced by people who have this defect
hqijq 'qkqii. The father-in law of Dan Watts, who is a Ucluelet
Indian that came to visit his son-in-law, was observed by the
writer and definitely stated by Dan to have this "nasal twang,"
which is due to an inability, muscular or nervous, to raise the
velum so as to shut off the passage of the outgoing breath
through the nose. In speaking of the elk, ninik/in* forms are
used.
A second type of defective articulation is termed hahdCklin^
or hahdt/in^ {hahaC- reduplicated stem; -k!in^ "to make a
sound of"), and is supposed to be due to a hole in the palate.
I have no clear idea as to just what the organic basis of the
faulty articulation is, but, judging from the examples given of
it, it seems evident that those subject to it have difficulty in
articulating against the hard palate. Perhaps the speech defect
is due to cleft palate. All ^s and tc affricatives (presumably also
lateral aflfricatives) become simple t- sounds (dental), while s, c,
and I become interdental fricatives (6). The acoustic effect is
that of an exaggerated lisp. Thus, tclotck" "all" becomes t!6Vk';
'otsVyukwaH "I go to it" ('o- empty noun stem "some-
thing;" -tsi'yukw- "to go to;" -an "I") becomes ' otV yukwan ; and
tc.'op'tdop'cinil "stretch around the neck; sweater" (tc.'op'-
tdop'c- reduplicated stem; -inil "at the neck") becomes
t/op^t/op'dinid. This latter rests on the authority of Dan Watts;
Alex Thomas, starting from a form tcfop'tcfop'cimil for
''sweater," gave tlop'tlop'timil as its hahdtUn^ correspondent.
Those who are hahdt'kfin^ thus confound three distinct series of
consonants in a single dental or interdental series. Such per-
sons are imitated when addressed. The outward resemblance
with the phenomena of consonant play is quite striking here.
This resemblance becomes even stronger in the case of the
third Nootka speech defect of which information was obtained,
that known as tstska' (tsisk- verb stem; -a' verb suffix of con-
tinuative significance) or istskaq'sul (tsisk- verb stem; -aq'sul,
perhaps misheard for -ak'sul "at the lips"). Such as are
subject to it are supposed always to keep their teeth open and
to be saying ts-]-. As a matter of fact, those who are tstska
370 VI American Indian Languages 2
14
change all s and c- sounds to palatalized sibilants (s). Thus,
'otsl'yukwaH "1 go to it" becomes 'ot^Vyukwan; si'ydsaH "it is
mine" (si'yds- "to be mine," from independent pronoun si'ya'
"I;" -aH first person singular present indicative) becomes
si'y&saH. It will be remembered that these consonant changes
are characteristic of the forms used in addressing or speaking
about abnormally small adults, except that such discourse is
further characterized by the use of the diminutive suffix -i^
(from -'is). Here there is a tangible connexion between the
involuntary consonant changes brought about by a speech
defect and the consonant play used to symbolize a body defect,
though it is far from obvious in this particular case what asso-
ciation there can be between a kind of lisp and a dwarfed condi-
tion of the body. A further point of interest is that those who
are tsiskd' are generally imitated when spoken of. The signi-
ficance of this in the argument is obvious.
Somewhat similar to the hahdtfin* speech defect, yet not to be
confused with it, is that known as kakaVwin"^ "to talk as one
with missing teeth" (cf. kdlxwak" sul ''to have teeth missing in
one's mouth"). Such persons speak with a decided lisp, sub-
stituting B for s and c, id for ts, id! for ts! and td, but, it would
seem, t for ic. Examples are: 'e'pinid from 'e'pinis "apples;"
'd'yintad from '6'yintcas "oranges;" timid from tdmis "bear;"
Id/otk' from tclotck' "all;" tdfdpatB for tcfdpats "canoe"
(contrast the corresponding hahdtfin^ form: tfdpaV). Here
again, one who is afflicted with this speech defect is imitated
when addressed; thus, Alex Thomas, before he had caps put on
his vestiges of teeth, used to be mocked A;aA;dr'tt;m'-fashion.
A fifth, not uncommon, speech defect among the Nootka is
stuttering. Stutterers, like all other persons who have some-
thing abnormal about their speech, are derided by being imitated.
The West Greenland speech defect known as kutdt'oq^ is
particularly instructive in that an individual speech-peculi-
arity, which, however, seems to be a common one in the Eskimo
settlements along the coast, has become one of the dialectic
peculiarities of the northern settlements of the Upernavik
district. The kuidt'oq habit consists in substituting ordinary
gutturals (fc- sounds) for velars {q- sounds), and is evidently due
' See W. Thalbitier, A Phonetical Study of the Eikimo Language, Meddelelser om Gron-
Und. XXXI, 1904, pp. 178-180.
Eight: Wakashan and Salishan Languages 371
15
to the greater difl&culty of bringing about a contact between the
root of the tongue and the velum than farther front in the mouth.
This defect, it should be noted, brings with it the confusion of
two etymologically distinct series of consonants with resulting
grammatical or lexical ambiguities, at least theoretically. In
this respect kutdt'oq forms are parallel to the forms resulting
in Nootka from speech defects or the use of consonantal play.
Children are particularly apt to be kutdt'oq, but generally lose
the habit as they grow older. However, certain adults, particu-
larly women, always remain kutdt'oq, whether because of the
mere force of habit or because of a physiological or anatomical
impediment. As for the Upernavik peculiarity, it seems clear
that the kutdt'oq habit can hardly be due to the individual
disability or carelessness of all the members of the district,
but that what was originally a speech defect has become social-
ized into a dialectic peculiarity. The analogy with the forms
employed in Nootka in speaking of or addressing certain classes
of people that are ill-favoured by nature is striking.
The explanation and genesis of the various types of speech
mutilation in Nootka can hardly be more than guessed at, yet
certain probabilities, in part already suggested, seem to stand
out. In the first place, the use of definite morphological elements
to indicate some characteristic of the person spoken to or of
(Nootka -is and -aq*; Paiute -tsi-; Nahuatl -tzinoa) needs no
particular comment, at least from the purely linguistic point of
view. Further, definite points of contact have been estab-
lished between speech defects and "mocking-forms," with
consonantal play, on the one hand, and between the latter and
myth-character forms with consonantal play, on the other.
I am inclined to believe that the observation of consonant
substitutions such as take place, with involuntarily humorous
effect, in the speech of those that articulate incorrectly, has set
the pace for the consciously humorous use of the same or similar
substitutions in both mocking and, directly or indirectly, myth-
character forms. The Nootka mocking-forms, with their use
of the diminutive affix and of consonant play, represent a
combination, both linguistically and psychologically, of the
pity and affection symbolized by the use of the diminutive
element and of the contempt or jesting attitude implied by the
372 VI American Indian Languages 2
16
imitation of a speech defect. A myth character whom it is
desired to treat humorously may, among other possibilities,
be relegated either to the class of poor talkers or to that of nat-
ure's step-children. Hence the consonant play of such characters
is in part traceable either to speech defects or to mocking-forms.
In passing it may be observed that the "enfant terrible" motive
is fairly clear in the treatment of many humorous characters of
American mythology, and that consonant play may in some
cases be taken to symbolize this attitude. The socializing of
the kutdt'oq habit among certain of the Eskimo forcibly suggests
the influence of the speech of children as a contributing factor
in the creation of myth-character forms. The Kwakiutl Mink
is a very likely example of the "enfant terrible," both in action
and speech. The possibility should not be lost sight of, of the
use of myth-character forms to apply to a class of people or to
an individual in ordinary life. This would be an extension of
the well-known American Indian habit of comparing one that is
marked by some peculiarity of temper or habit with a favourite
mythological character.^
There is, however, another factor which has undoubtedly
exercised a great influence both on the forms of speech used
by myth-characters and on the forms peculiar to songs. This
is the comic or novel effect produced by the imitation of the
speech of foreigners, particularly of such as speak a dialect
divergent enough from the home-dialect to be funny or impres-
sive, yet not so different as to be unintelligible and, therefore,
lacking in interest. Hence we often find mythological characters
in America making use of a neighbouring dialect of the language,
as in the case of the Nass River TxdmsEm and other characters,
who talk in the dialect of the Tsimshian proper of Skeena
river.^ Examples of songs whose texts are in a divergent
dialect, not to speak of the common use of a totally distinct
language, are frequently met with in and out of America. A
well-known instance is the use by Melanesian tribes, according
to Codrington, of the dialect of some neighbouring tribe for their
own song diction; thus, the Melanesians of Mota (Norfolk
' A few interesting examplea are given by A. Skinner, Notes on the Eastern Cree and
Northern SauUeaux, Anthropological Papers of the American Museum of Natural History,
vol. IX, 1912. p. 82.
' See Boas, Tsimshian Texts, Bulletin 27 of Bureau of American Ethnology, 1902, pp. 8,
18. 20. 30. 35, 46. 61-64, 78, 171.
Eight: Wakashan and Salishan Languages yiZ
17
island of Banks islands) use for their songs the dialect of Saddle
island. Also in the clownish episodes of rituals, which are so
characteristic of America, the impersonation and imitation of
the speech peculiarities of foreigners are often resorted to and
never fail to arouse a hearty laugh. In all these cases, it is
rather important to observe, real accuracy of imitation is not
generallj' attained or even aimed at, so that the foreign style
often tends to reduce itself to a number of conventional vocalic
and consonantal displacements. In dealing above with the
change of anterior palatal k- counds to ts- sounds in the
language of the Kwakiutl Mink, I pointed out that a similar
change was involved in the passage of original Wakashan
anterior palatal k- sounds to Nootka tc- sounds. It is
just possible that the Mink ts- sounds are in such cases due
to an imitation of the speech of the northern Nootka tribes.
The difficulty with this interpretation is that Nootka and
Kwakiutl are altogether too divergent to afford more than a
quite inconsiderable number of illustrative cases of the k- tc
change, and of these but few would strike the naive mind. It
seems more plausible, on the whole, to assume that both the Mink
and Nootka consonant changes rest on a common Kwakiutl-
Nootka tendency, perhaps a tendency on the part of children to
pronounce anterior palatals as sibilants. Data on the speech
peculiarities of Kwakiutl children would be valuable here.
The Nootka Indians of one tribe frequently imitate the real
or supposed speech peculiarities of those belonging to other
Nootka tribes, the stress being primarily laid not so much on
peculiarities of vocabulary and grammatical form as on general
traits of intonation or sound articulation (cf. our New England
"nasal twang" and Southern "drawl"). For the purposes of
this paper the Nootka now spoken by the TsUcd'atH" and
HopdtcIas'atH" of Barkley sound and the head of Alberni canal
may be taken as the normal form of Nootka speech; this is, of
course, purely arbitrary, but so would any other point of de-
parture be. It is instructive to note that one or two of these
tribal speech peculiarities coincide with individual speech defects.
According to the TsUcd'atH" Indians, the Houtcuq' Lis'atH" tribe
of Uchucklesit harbour, a western inlet of Alberni canal, speak or
spoke (for there are few of them left now) in a rumbling fashion
374 VI American Indian Languages 2
18
(JoL!o:en^) ; they are said to use their throat more than the other
tribes. The peculiarity referred to seems to be a more than
ordinary use of velar resonance, due to a tightening of the passage
between the root of the tongue and the velum or perhaps the
throat.
The Hoidi'atH'^ Indians of Sarita river and the southern
shore of Barkley sound are said to speak LfdLfatdin'^, a splutter-
ing effect being apparently referred to. As far as can be made
out, their speech peculiarity consists in a more liberal use of tc
sounds than ordinarily. Thus, according to Alex Thomas, the
Ho'.di'atH" say 'ndtcciL instead of 'ndcciL "to look at" (as a
matter of fact, this usage is probably etymologically justified, as
'nac- and, in other forms, 'natc- are both used as verb stems in
Ts.'ud'atH" itself); instead of pronouncing tdayi'is "give me
water" (tc.'a- noun stem "water;" -yi- verbifying suffix "to
give;" -is second person singular imperative with first person
singular object) they say something like tdatcyVis, though Alex
maintained that it was not a full clear-cut tc that was inserted.
At any rate, the TsUcd'atH'^ have seized upon the tc- insert as a
convenient means of poking fun at their Hdidi'atn'^ kinsmen,
using it in ways that are certainly not, nor meant to be, accurate
renderings of the tribal peculiarity. Thus, the tribe itself is
humorously referred to as HotcidVatH'^; Numdqemiyis, the main
inlet of their country, is similarly termed Nutcmdqemiyis.
Evidently, we have here an example of a mocking usage, based
on a tribal peculiarity, that is in form perfectly analogous to
certain myth character and cripple-mocking usages (cf. inserted
X for Kwatiyat and inserted tcu" for left-handed people.)
The northern Nootka tribes, beginning with the La'okwi'atH'^
of Clayoquot sound and proceeding north, are said to speak
tdHtdua', which refers to a drawling or long drawn out manner of
talking. Apparently the peculiarity, which is often imitated in
jest, consists not so much in lengthening out vowels as in a some-
what exaggerated rise in pitch towards the end of a sentence,
which gives the flow of speech a sliding cadence. The most
northern Nootka tribe, the Tc.'vq^Lis'atn'^, are said to be all
stutterers and are accordingly imitated in jest.
In imitating the Nitinats {Nlttna' atn") , a group of Nootka
tribes to the south of Barkley sound that speak a very divergent
Eight: Wakashan and Salishan Languages 375
19
dialect, the meaningless syllable -aq' is always added to the
word, as this syllable is supposed to be a very common one in
Nitinat. This device is strikingly similar to the use of sufiixed
-ag' for large persons.
The real old Hopdtdas'atH'* Indians, whose earliest homes were
in the interior of the island along Somass river and about Sproat
and Great Central lakes, were said to talk tstska', that is, to
confound s and c sounds. As we have seen, this is also a well-
recognized individual speech defect among the Nootka. In the
case of the Hopdtdas'atH'', the tstska' habit was simply due to the
fact that they carried over into Nootka speech a linguistic
peculiarity found in the Salish dialect which they originally spoke
(a dialect apparently identical with or closely related to Boas'
PEnLatc; recognized as Pindd'atc by Tyee Bob, the leading man
among the Hopdtdas'atH'* to-day and whose father is still re-
membered to have spoken tsiska^).
As for the TsUcd^atH" themselves, they are said by the other
tribes to talk very fast. If one anywhere among the Nootka
Indians talks too fast, the proverbial saying is that he is a
TsUcd'atH".
It will, as we have seen, have to be admitted, that mocking
forms for various classes of people are connected not only with
speech defects and mythological devices, but, to a large extent,
also with tribal speech peculiarities.
Finally, the possibility of a direct psychological relation
between the consonant change and the type of individual or
attitude it symbolizes should not be summarily ruled out of
court. That such an association once established by historical
causes will be felt as a direct and simple psychological associa-
tion is quite obvious, also that it may become productive, by
analogy, of further associations of a related sort. I would,
however, even be inclined to suppose, though proof may be
difficult or impossible, that certain associations of sound and
character or form arose more or less spontaneously, or, to put it
more correctly, by virtue of the inherent associative value of
the otherwise unconnected phenomena in the mind of a par-
ticular individual or group of individuals. Such an individual
association, if given outward expression, can become socialized
in the same way in which any individual idea becomes socialized.
376 VI American Indian Languages 2
20
The type of association here thought of is quite parallel to the
sound-colour associations familiar enough in psychology. It
may be not uninteresting as a psychological datum to note that
the writer himself feels, or thinks he feels, the intrinsically
diminutive or augmentative value of certain consonant changes
in Wishram. Moreover, the association of c- consonants with
humpbackedness in Nootka seems not so far-fetched after all.
The thickish quality of these consonants, together with the pro-
trusion of the lower jaw in pronouncing them, suggests to me
the same squat clumsiness as the image of a hunchback. All
this may, of course, be merely auto-suggestion ad hoc.
To summarize, evidence has been presented of the historical
connexion between various linguistic and stylistic processes
involving the symbolic use of sounds. These are diminutive
and augmentative forms of speech, mocking-forms, myth-
character and animal forms, and song forms. Moreover, further
evidence has been presented to show the historical connexion of
these quite specialized tricks of language with the far simpler
phenomena of speech defects, children's language, and imitation
of the phonetic peculiarities of foreigners. The direct association
of some of the former with the types they symbolize, after the
manner of primary association between data of distinct sense,
has also been suggested as a possibility.
The data brought forward in this paper as to the associations
obtaining in Nootka between various classes of persons, mytho-
logical beings and animals, linguistic devices designed to satirize
or characterize, speech defects, and tribal speech peculiarities,
may be most conveniently grouped in tabular form. The arrange-
ment in the table is intended to emphasize the purely linguistic
similarities.
Phonetic Key.
a, short as in German Mann; e, short and open as in English
met; i, short and open as in English it; o, short and open as in
German voll; u, short and open as in English put; e, short and
close as in French ete) i, short and close as in French jini; o,
short and close as in French chaud.
a, long as in German Bahn; e, long and close as in German
See; i, long and close as in German Sie; o, long and close as in
Eight: Wakashan and Salishan Languages 371
21
German ro/i; e, long and open as in French /e/e; d, long and open
as in English saw, yet with back of tongue not so low.
E (Kwakiutl), short obscure vowel like e of German Rose;
I (Nootka), short open i-vowel of rather unclear quality; *
(Nootka), occurring as syllabic final after n and m, barely
articulated or murmured (yet not voiceless or whispered) /;
" (Nootka), denotes a-timbre of preceding h (see below).
c, like sh in English ship: tc. corresponding voiceless affrica-
tive, ch of English church (in Nahuatl ch is used for tc); dj,
corresponding voiced affricative, j of English joy; s and is, as
in English sit and hats (in Nahuatl z and tz are respectively
used instead); s and ts, palatal voiceless sibilant and affricative,
acoustically midway between s-c and ts-tc respectively; c and
tc, c and tc pronounced with lower teeth in front of upper; 6,
interdental voiceless spirant, like th in English thin.
q, voiceless velar stop like Semitic qof; qw, labialized form of
same; x, voiceless spirant of g-position; x, voiceless spirant of
fc-position, not pronounced as far back as German ch of Bach;
k • and g • (Kwakiutl), anterior palatal stops (palatalized k-stops),
approximately ky and gy; x- (Kwakiutl), voiceless spirant of
fc- -position, ch of German ich; 17, voiced nasal of k- position, ng
of English sing; rj (Eskimo), voiced nasal of q- position.
I, voiceless lateral spirant; l, corresponding voiceless lateral
affricative (written tl in Nahuatl); i (Kwakiutl), corresponding
voiced affricative.
' , glottal stop; ; (Nootka), strangulated-sounding laryngeal
stop, similar in resonance to Arabic 'ain; h (Nootka), strangu-
lated-sounding laryngeal spirant, Arabic ha; \ aspiration or
breath-release of preceding vowel or consonant (p", V, k\ and
g' are aspirated voiceless stops); .' denotes glottalized stops
and aflfricatives {p!, t!, kf, q!, l!, ts!, tc!, ts!, tc!, k-!), that is, such
as are pronounced with simultaneous closure of glottis, but with
oral release prior to that of glottal release. All other con-
sonants as in English.
' , stress accent; *, denotes preceding long consonant (except
in Kwakiutl k'- sounds); ,, denotes nasalization of vowel under
which it is placed; +, denotes excessive length of preceding
vowel or consonant.
378
VI American Indian Languages 2
"
g
-0 «>
a
'S
C b
■J
rt«
.9
s •■
£
1
«,7
"3
1
"S
GJ
0]
•*j 1
"O
a
fl
•g
t£0
5
b
(U
§
"S'm
d
3
d
■0 1
a
CO
0-S
So
"
a
3
■3
fl
•c
3
■3
ft
~
3
1
V.S *^
e
•«
i
.c
c
rt u S
Si
s
5
Q
':; i
Js^a
^
'^.
^
s
5
c
3
"a
■:3
■^
"a
":4
^
^
e
S
1
>
1
>
•2 §"*
■'1
•1 §
fl
fl
■5
S
fl
OS
J)
73 •»'
•
?"^
g
:"
^2
>•
fl
_fl
§^'
TJf;
•2^
"1
>
i
1 ■-T3
•Beg
3 3
.9 "
3 "„, *•
:§l.8
.tJT3
3 fl
■5t)
3 d _,
>
"1
>
"0
a
■ft
*ft
08
2
•^-0
'3 *
3s.
'•2
■■V
a
B
■ — aj 3
a "
•oSg
3 v>
3
fl
a
5
1
>
g"
'.a
■♦^
_fl
2
"3
>
J3
T3
■5
^ fl
3
i
>
S 3'S
9> .. "
2 S'S'
•0
■0
■0
T3
?-g5
?-g2
•a 4)
t-H
1
■3
I
•3
•3-0
C9
gg§
S fl {
<
-»!
J<_
<
-«!
<
««!
_■<
M
^
H_
W
,H_
H
H
Pui
E
i°
.2 fl
■
g
a
T3
^
5
1
jS
£a
SS
'■3
'•z
K.
1
a:
5
01 .=
1-1
_ "J
Jo
a
Xi
fc
OS
0.
1
'•5
i
g^-3
3-
t5
1
.2£ fc> c
U) c
^ 'c
i
V
r
c
— V
e*
w
OS
a
3
^<I>
u
sg>
$ *"
u
a-^
a
E
V
D.
"0
a
a
>.
H
ft*)
al
3 —
S.-H
it c «
0-5 a
ill
a
0.
a
"5
i
15
1=
<
g-3 2
OS
13
3
K
fl
1h
0.
V
E
d
g
2
«)
a
■g
T)
fl
ot
XI
I
0.
I
a
3
u
fl
£
4)
a
■a
V
OS
1
Eight: Wakashan and Salishan Languages 379
Editorial Note
Originally published as Memoir 62, Anthropological Series 5, Geological
Survey, Department of Mines, Canada. Ottawa (1915).
Although the formal patterns of consonantal symbolism, particularly dimin-
utive/augmentative devices, have been much discussed in Americanist work
(for an excellent survey see Nichols 1971), the rhetorical and literary uses of
these have been noted only sporadically. Some attention has been given
recently to the phonetic stereotyping of the Coyote and similar figures in myth
(Aoki 1975: 190; Toelken 1969). Hymes's study of the expressive resources of
Takelma traditional literature (1979; revised in Hymes 1981: 65-76) is closest to
the spirit of Sapir's work.
Noun Reduplication in Comox,
a Salish Language of
Vancouver Island
INTRODUCTION.
One of the most characteristic grammatical processes of a
group of Northwest Pacific Coast languages, embracing the
Tsimshian, Kwakiutl-Nootka, Salish, and Chemakum linguistic
stocks, is initial reduplication, employed in both noun and verb
forms to indicate a variety of grammatical concepts, chiefly
those of plurality, distribution, and iteration. The Salish
languages in particular are known to make exuberant use of
reduplication for grammatical purposes, but the subject, which
seems to bristle with irregularities and intricacies of detail, has
never been adequately treated for any of the numerous dialects
of the stock. Indeed, a thorough grammatical study, at the same
time phonetically adequate, of a Salish language, is still one of the
desiderata of American linguistics.
During the autumn of 1910, while prosecuting ethnologic and
linguistic research for the Geological Survey of Canada
among the Nootka Indians now living in two reserves near
Alberni, B.C., opportunity was incidentally found to gather
some linguistic data on Comox, a Salish language spoken on
the east coast of Vancouver island near the present town of
Comox. The dialect represented in these notes seems to be
382 ^f American Indian Languages 2
Comox proper (Q!6mgx"s), with which Uohos, spoken on the
mainland of British Columbia, was stated to be identical.
SdlolV'' was stated to be a northern dialect of the same language.
This term is evidently identical with Boas' Qatloltq, which he
uses to apply to the most northern group of Coast Salish tribes,
excluding Bella Coola, inhabiting "Discovery Passage, Valdes
Island, Bute and Mala-^pina Inlets."^ Boas adds, "The
Catl61tq are called K'omoks by the Lekwiltok'" (southernmost
Kwakiutl tribe.)
The informant was Tommy Bill, an Indian of mixed blood,
whose father belongs to the Ts'.ictiatH" tribe of Nootka Indians,
while his mother was a Comox, he himself living with and being
to all intents and purposes a member of the Hopdtdas'atH"
tribe of Nootkas. His knowledge of Comox was obtained in
his earher years, when living among his mother's people, whom
he visits from time to time; it is only fair to add that he speaks
mainly Nootka and English nowadays and does not claim to
have a perfect command of Comox. However, the rather
elementary character of the data obtained, together with
convincing internal evidence derived from their study, leaves
no room for doubt as to the essential accuracy of the material
here presented. Most of the time spent on Comox was taken
up with securing material pertinent to the problem of reduplica-
tion in nouns. For most of the nouns obtained, plural, diminu-
tive, and diminutive plural forms were secured, all of which
involve various types of reduplication. Our linguistic material
thus naturally divides itself into three heads, not to speak of a
small number of nouns that are always used in reduplicated
form. A few introductory remarks on Comox phonetics and
some supplementary data are also added.
I. PHONETICS.
Vowels. The short vowels found in the Comox material
secured are: a (as in German Mann); a (as in English bat);
e (short and open as in English 7net); e (short and close as in
French ete); i (short and open as in English hit); i (short and
close as in French fini); o (short and open as in German dort);
' See F. Boas, First General Report on the Indians of British Columbia, Report B. A. A. S.,
1889, 5th Report on North-Weatern Tribes of Canada, p. 10.
Eight: Wakashan and Salishan Languages 383
(short and close as in French beau); and u (short and open as
in English 'put). Of these vowels, e, i, and i are etymologically
one sound, which is modified by phonetic surroundings; simi-
larly, o and u. Velar consonants tend to lower preceding or
following i to e (possibly sometimes e), while certain consonants
(particularly s and I) tend to palatalize i to i. e and o, which
latter does not occur often, are doubtless etymologically related
to e and o respectively, but seem in every case to be clearly
kept distinct from these, a is not common.
Corresponding to each of the short vowels is a long vowel
(long d, however, has not been found). These are indicated as:
d (as in German Bahn); e (long and open as in French mere, or
as in English hear, but without "r-vanish"); e (long and close
as in German See); i (long and open as in English beer, but
without "r-vanish"); i (long and close as in English see); 6
(long and close as in English roll, or as in German Sohn) ; 6
(long and open as in English born, but without "r-vanish");
u (long and close as in English rule) ; and u (long and open as in
English poor, but without "r-vanish"). Similarly to the
corresponding short vowels, and under parallel phonetic circum-
stances, e, %, and I are variants of one sound, etymologically
speaking, though i is often to be interpreted as lengthened form
of inorganic vowels, in which case it does not seem to vary with
e and l; 6, u, and u are likewise representatives of what is
etymologically a single sound. 6 does not often occur; it is
probably etymologically related to 6. e occurs often and cannot
be considered a mere variant of e.
As not infrequently happens in American Indian languages,
the long vowels are not always held out with even stress, but
end with short rearticulations which give the whole vowel in
each case a quasi-diphthongal effect. Such vowels have been
noted by the writer in Takelma, Southern Paiute, and, at least
to a moderate extent, in Nootka; Boas has noted them in
Tsimshian. While they occur to a considerable extent in
Comox, they cannot as in Takelma be considered the normal
forms of the long vowels; sometimes the short rearticulations
seem to serve as glides to following consonants, particularly
velars. The quasi-diphthongal long vowels are here indicated
by long vowels followed by superior short vowels, the vocalic
384 VI American Indian Languages 2
quality of the latter being indicated as in normal short vowels.
There are found: a"; e%' e^; e' (occurs before anterior palatal
consonants) ; 1'; i'; i^ (occurs before velar consonants) ;
0° and o"; and w". A number of cases also occur of short
vowels followed by weak rearticulating vowels; such are e",
q^, and i' (here the * is a glide to the following velar consonant).
Some of these may well represent secondarily shortened long
vowels. Differing from such long or short vowels with quasi-
diphthongal character are vowels that are secondarily diph-
thongized by a vocalic glide whose timbre depends wholly on
the following consonant; such is i" in /cwpw"mi"a;" "hill," in which
the second " is a glide due to the u- timbre of the final consonant.
Short vowels of somewhat obscure quality are also found,
either representing dulled forms of normal short vowels or being
of inorganic origin and meant to lighten consonant clusters or
serve as glides. Such vowels are: a (as in English but, yet
sometimes less clearly marked in quality), which is sometimes
inorganic, sometimes dulled from a; e (obscure vowel with e-
quality) ; and / (very short rather unclear i).
At times short vowels are so weakly articulated as to be
barely audible; these are rather "murmured" short vowels
of etymological significance than merely glides, timbre-echos
of preceding consonants, or voiceless vowels. Examples are:
? in Idl^bo'm' "small clam" (-^?6- reduced from ld'"b- in I6"°bnm'
"clam"; yet in this case ? can just as well be morphologically
dispensed with and phonetically explained as a timbre-echo of
-ol-); ^ in qt'w^x "steel-head salmon" (that ^ is organic,
despite its dull quality and extreme brevity, and reduced from a,
is indicated by Nootka qe'wan "steel-head salmon," with which
Comox qt^w^x is evidently identical; borrowing has doubtless
taken place); "^ and " in hew^qen" "swan" and its diminutive
hew'^qAdol.
Another class of "murmured" vowels (German 'Murmel-
vokale") is formed by weakly articulated, yet not voiceless,
vowels occurring in syllabically final position after glottal
stops ('). Such vowels are only in part "murmured echoes,"
i.e., reduced repetitions of immediately preceding fully voiced
vowels (such are a'", e'*, V\ ai'\ o'?, d'"; vowel breakings of this
type occur often in American languages) ; in some cases we have
Eight: Wakashan and Salishan Languages 385
also murmured vowels after glottal stops that are of different
quality and etymologically distinct from immediately preceding
vowels (such are a'' and a'O-
Some consonants, notably glottalized ("fortis") consonants,
are apt to be followed by timbre-echoes dependent in quality
on the preceding vowel. This simply means that the oral
resonance chamber characteristic of a vowel may, failing to be
materially disturbed by the following consonant position,
linger on and thus become acoustically noticeable as a voiceless
(sometimes aspirated) vocalic echo; if the consonant is a
spirant, the vocalic timbre may be audible during its production.
Examples of such unaspirated timbre-echoes after glottalized
consonants are: " in pld'alats!'^ "skunk" and ° in k!6°ddt!°
"porpoise." In Id^g^et!" "herring" the t! was heard with definite
a-timbre despite preceding e. After u (o)-vowels syllabically
final A;-sounds are regularly followed by echoes (aspirations
when consonant is not glottalized) with w-timbre. Hence A;'",
A;.'", X", q'", q!", and x" (see below for orthography of fc-sounds).
These sounds, however, are also very frequent after unrounded
vowels, as in Id'^dak'"' "skin;" in such cases they represent
original labialized A;-sounds (see below). Aspiration with
definite w-timbre is also found after t, as in sdW" "woman."
Excluding such inorganic diphthongs as are formed by vowels
and following glides (e.g., i"), there have been found as true
short diphthongs ai, au (also au), di, ei, and long diphthongs
di, du. Vowels normally forming diphthongs that do not so
unite, each preserving its full value, are separated by . (thus,
a.i as distinct from true diphthong ai). Stress accent is indi-
cated by ' over vowels.
Consonants. The consonant system of Comox is fairly
full, including, as it does, eleven distinct series that differ
according to place of articulation. As regards manner of articu-
lation, six distinct series are to be recognized (voiceless stops,
glottalized or "fortis" stops, voiced stops, voiced nasals, voiceless
spirants, and voiced spirants), though by no means all of these
are represented for all places of articulation. The voiceless
stop and glottalized stop series are complete, the voiceless
spirants nearly so, while the others are quite defective. All
these consonants may be represented in the form of a table : —
386
VI American Indian Languages 2
6
Laryngeal glottal)
Velar
Imbialiekd velar
Guttural
Labialized guttural
Pbe-quttural (anterior pala
tal)
Dorsal lateral
Palatal sibilant
Alteolar aiBILANr
Alveolar
Labial
Voice-
less
8TOPS
Q
qw
k
kw
kv
Glot-
TALIZED
STOPS
q.'w
k!
k.'w
tr!
is!
a
Voiced
STOPS
Nasals
I
(voiced
contin-
uant)
dj
id)
(b)
Voice-
less
spirants
Voiced
spirants
c is pronounced like sh of English ship; x^ like ch of German
ich. tc, td, dj (like j of PJnglish jam), ts, and ts! are affricatives
(stop plus corresponding spirant; no simple stops correspond
to <c-series). l and l! are also affricatives, but with lateral
(voiceless spirant /) release.
h and d are phonetic variants of m and n; h and d were often,
though not consistently, heard between vowels, m and n rather
consistently as initials, while m and n were more often heard as
syllabic finals than h and d. These h-m and d-n sounds have
been at various times analysed by Boas as "semi-nasalized"
consonants. "The nasal opening," he v/rites, "may differ in
width, and the stricture of the upper nares may produce semi-
nasalized consonants. "2) Again, in speaking more definitely
of Coast Salish, "... the b sound ... is produced with
half-closed nose by the Indians of the Strait of Fuca, in the State
of Washington. . . . The characteristic trait of the sound is a
scmiclosure of the nose, similiar to the effect produced by a cold
ill the head."' These remarks doubtless apply to Comox as
Eight: Wakashan and Salishan Languages 387
well as to more southern Coast Salish languages, yet it seems
likely to the writer that under certain phonetic conditions these
semi-nasals become true nasals. No attempt will here be made
to normalize orthography on this point, a faithful record of
what was heard, or thought to be heard, being presented.
Eliminating h and d as of secondary origin {g^ and dj, it should
be carefully noted, are true sonant stops, not "intermediates"),
all the other consonants listed in the table are etymologically
distinct, that is, none of them are mere variants, {k, k!, and
X, however, may prove to be merely secondary forms of kw,
k!w, and xw.) This gives us no less than thirty-six (or thirty-
three) organically distinct consonants to operate with. A
secondary series of aspirated surds (voiceless stops followed by
aspiration) arises when voiceless stops occur as syllabic finals
(written p\ t\ k\ fc^', q\ tc^); kw and qw become fc'" and 5'",
that is, their aspiration-release has w-timbre; similarly, k!w
and q!w in this position become fc.'" and g/". q, it may be noted,
is often released into a weak spirant glide x (written ^) before
the following vowel is attacked (thus, q^'a for qa). Final vowels
and m and n are also often followed by aspiration (-a' and
similarly for other vowels, m' or less often h\ n' or less often d"),
though this was not consistently heard. Final m and n are
etymologically distinct from final glottally affected m and n,
which are written wi' and n' (sometimes breath release is heard
after glottal release, when they are written m" and n"). Long
consonants (indicated by ' after consonant) were noted, but
seem to be of no etymological significance (examples are q', d').
Sound Changes. Lengthening and reduction of vowels are
important phonological processes in Comox, also, though to
less extent, changes of vowel quality. As these, however, are
generally of grammatical significance, they are best taken up in
their proper place under types of reduplication. As more
strictly phonetic pure and simple in character is to be considered
the palatalizing of a to i in the neighbourhood of g", also the change
of A to w and i in appropriate phonetic circumstances. These
changes also, however, are most clearly brought out in connex-
ion with morphological processes.
Many cases of g", perhaps all, are undoubtedly due to original
w. It seems that w, when it came to stand between vowels
388 VI American Indian Languages 2
8
(not, it would seem, including cases of preceding vowel plus
glottal stop), also initially in many cases, regularly passed into
g^. Thus, as diminutive of xAucin* "bone" is found xt^xig^icin'
<*xtxAwidirC {-Ag^- becomes -ig^-, as noted above). Similarly,
from qt'w^x "steel-head salmon" is formed qeqeg^e^x "little steel-
head salmon" and qeqAuqd'''g"e^x "little steel-head salmon
(plur.)." This phonetic law explains a class of plurals, formed
by reduplicating with o- vowel, derived from stems in internal
-g"-. Thus, from t!eg"em {<*t!ewem) "sun, moon" is formed
plur. UoHleg^em {<*t!AWt!ewem); other examples will be
given in their proper place. So also is explained suffix -dg^il
"canoe" in such forms as tcddd^g^il "three canoes," seyatsd^g^il
"five canoes", as compared with -dul in mdsdui "four canoes;"
-dgHl is evidently from *-dwil (cf. KwantlEn, of Cowichan
group of Coast Salish, -agitl "canoe" in numerals,'^ i.e., -axwil;
perhaps cf. Comox n6xwU "canoe"). An interesting test case
is qPgyas "deer," doubtless a loanword from Kwakiutl (cf.
Kwakiutl gtwas "deer"^). Another such test case is afforded
by Comox tigH^x" "nine" < *tAWAX'' or *tAwux'' (cf. KwantlEn
tuq "nine,"^ i.e., tux or tux, contracted from *tuwux). Compare
also Comox ht'g^os "chief" with PEntlatc and Siciatl htwus*.
On the other hand a number of words have been found with w
between vowels. Such are ts.'atsldwicin^ "hail," xwd'awVC
"fire," and 'dwdk'" "tobacco." It is not clear how this -w- is
. related to -w- > -g''-.
Just as g^ and w are related, so there is reason to believe that
dj and y are related, though there is perhaps not quite as con-
vincing internal evidence at hand. See Type VIII of plural
formations for such evidence. Moreover, with Comox djidis
"tooth" compare KwantlEn yenis "tooth;"^ with Comox djicin*
"foot" compare Siciatl yicin.^
* See C. Hill-Tout, Ethnological Studies of the Mainland IlalkomilBm, a division of the
Salish of British Columbia, Report of British Association for the Advancement of Science,
1902, Ethnological Survey of Canada, p. 65.
' See F. Boas, Kwakiutl, Handbook of American Indian Languages, Bulletin 40, Bureau
of American Ethnology, 1911, p. 447.
' C. Hill-Tout, ibid., p. 64.
* F. Boas, Comparative Vocabulary of Eighteen Languages spoken in British Columbia,
Report of British Association for the Advancement of Science, 1890, 6th Report on the North-
western Tribes of Canada, p. 148.
' C. Hill-Tout, Ethnological Studies of the Mainland HalkomilBm, a division of the Salish
of British Columbia, Report of British Association for the Advancement of Science, 1902,
Ethnological Survey of Canada, p. 86.
' F. Boas, Comparative Vocabulary of Eighteen Languages spoken in British Columbia,
Report of British Association for the Advancement of Science, 1890, 6th Report on the North-
western Tribes of Canada, p. 14'
Eight: Wakashan and Salishan Languages 389
9
II. NOUNS NORMALLY REDUPLICATED.
A considerable number of Comox nouns always appear in
reduplicated form, reduplication in these cases being of no
grammatical significance, but belonging to the noun as such.
Many of them are animal names, and of these some are quite
evidently onomatopoetic. Ten fairly distinct types of redupli-
cation seem to be illustrated in the rather limited material
available. Very likely others exist.
Type I. Completely Reduplicating.
hd^mho^m blue grouse xop'xop' humming-bird
k'^dck'^ac bluejay tsU'xHsHx" fish-hawk
pok'^'pok"'' liver gH'gH'- panther
qe'n'qen"^ duck qwVqwi' sea-gull
"Duck" and "sea-gull" have both syllables with vowels alike
in quality but with short vowel in the second.
Type II. Completely Reduplicating with e.
te'ltoV small butter-ball duck hd'ihei' arrow
"Arrow" belongs perhaps rather with Type I. Both of these
nouns lose a glottal stop in the reduplicating syllable.
Type III. Reduplicating Syllable: cvci.'
titctitcV c owl kiva'kwd'^djo' grey squirrel
t.'Aq't.'Aqdi dog-wood
Type IV. Reduplicating Syllable: ce.
mf'mau cat k^It'k^Idk^! crow
td'itca.iq' salt-water hunter
In "salt-water hunter" reduplicating icl- is broken into tcl'i-.
' In these formulae c represents first consonant of stem, v first vowel, ci accond consonant
of stem, vi second vowel, and so on. V represents any long vowel, <■ any shortened vowel.
390 VI American Indian Languages 2
10
Type V. Reduplicating Syllable: ci.
Only one or two certain examples have been found of this type.
They differ from the preceding in that the vowel of the redupli-
cating syllable is short.
qwi'^qwdH! .dd'^^k' butterfly we'wdlos young man
(form probably diminutive in).
Possibly also: —
e'ddjam' young woman
Type VI. Reduplicating Syllable: cd or ca.
LdLdpx pocket-knife qwdqumi^s marten
xdxe'* nit mdmstco^m mink
Type VII. Reduplicating Syllable: cv.
ts!ats!dwicin' hail tdatddH'.dn" mouse
xwdxwadjo'm fly (word probably diminutive in form).
q.iq'td'amas game with wooden ball^
qoqgwVm^ down (of bird)
Type VIII. Reduplicating Syllable: cv'.
qd''^qa'^ rush mat djd'''dja' tree
Type IX. Reduplicating Syllable: eo.
Only one example has been found of this type: —
tofxHal necklace
Type X. Reduplicating Syllable: cec.
Of this very peculiar type (doubly reduplicating consonant,
otherwise like Type IV) also only one example has been found: —
qliqlq.'d'adje'uk''^ butter-ball duck
' Formed from q'td'ahas "wooden ball covered with spruce-roots." There were two
sides in the game, with the same number on each. Each side had a goal consisting of a little
pit, which was guarded by one man. All but the two guards gathered in the centre. One
man threw up the wooden ball and everyone tried to catch it, run with it to the goal of the
opponents, and put it into the pit. Those of the other side tried to take the ball away from
the one that had it. The side that 6rst made ten goals won the game. After four goals had
been made, the game was suspended for a while and a general free-for-all fight took place.
Eight: Wakashan and Salishan Languages
391
11
Here may also be given: —
qldq.'tux'* big fire (form is augmentative?) :
scattered around.
cf. q!6tix'' fires
III. REDUPLICATED PLURALS OF NOUNS.
By far the larger number of Comox nouns form their plural by
reduplication, in a few cases different stems are used for singular
and plural, while still other nouns seem to form no plural. The
most persistent type of plural reduplication is that in which
both first and second consonants of stem are repeated, though
'ess numerously represented types also occur.
Type I. Reduplicating Syllable: cvci
lAkoni"^ beaver
kuvidqin^ sea-lion
qwAdPs humpbacked whale
qwASAm woolly grouse
xop'xdp' humming-bird
ts!oxd"° codfish
LlAxwd'"' dog salmon
sd^arC cohoe salmon
qlwAt'lHcin" humpback salmon
xd'd big clam
Lp'Am' cockle
xApd'^ red cedar
qo'^'a''^ hemlock
q!dp!xwai oak
p'.tHxdi alder
t.'t'ibdi wild cherry bush
'dwdk''' tobacco
qfwA'ix wood
xd'a.idatc stump
plural tlAk^HlAkom"^
kumkumdqirC
qwxd'qw Adit's
qwAsqwASAm
xop'xop'xop^
ts!6xts!oxd'°
lIaxlIaxwo''-
sd^'^sa'an"
qlwAVq'.wAVlHcirC
xd'^xa'd
Lp'Lp'Am'^ (type viii?)
xAp'xApd^"^
qo'^qd'^'a''
q!ap!q!dp!exwai (with
lengthening of first
stem-vowel; -e- is in-
organic)
p.'e'p.'t'ixdi (type viii?)
HeH.'t'ibdi (type viii?)
'au'dwdk'^ many bun-
ches of tobacco
qlwAiqfwA'ix
xd'''xa'a.idaic
392
VI American Indian Languages 2
12
mAqsm nose
djiciti^ foot
djidis tooth
dikuiriKS heart
xAucirC bone
k^it! little finger
tslAmdla" index finger
qlwdt'Am river
pdxai' creek
Lfdqe'nac spring
kupu^'mi^x'' hill
Lldxai' old man
qAl'q! warrior
L.'Ams house
ajiSAm box
kwd'am coiled storage basket
L.'pdtjl basket bag
q!dk'" board
k^Iik^dyu oar
SAq'Ak'" war-club
lAq!" bow
tcHCqdmin knife
sipfAvitn" shinny stick
lAq'.AS mountain-goat blanket
L.'pi'ts.'d''^ yellow-cedar
q.'As'Addi buckskin shirt
Lldq.'acin" moccasin
pdq'dos white-eyed
tdxdos red-eyed
plural niAqimAqsin'
djicdjicirC
djiddjidis
L.'^k'L.'ikuinAS
xAuxAUcirC
kHV.kHt!
ts! Amis! AmaW
q!wdVq!waVAm
pdxpaxai'
ildq'L.'aqe'^nac
kup'kupumt"x" (with
shortening of second
stem-vowel)
LfdxLlaxai'
qAl'qAlq!
LfAmL.'Ams
xAsxASAm
kwd'^kwa'am
l! Ap' l! Apdtil
qlak'^'qlak""
k^Hk^'k^Hk^dyu
sAq'sAqAk'""
lAq.'HAq!"
tcHVtdit' qdmin"
siplsiplAmi^n'
lAqUAq'.AS
L.'Ap^LfApitsfd'"
q! Asq! As' addi
LlAq/LlAqfadn'^
pdq'paq'dos
tdxtcixdos
Type II. Reduplicating Syllable: cac.
This type differs from the preceding in that, while both first
and second stem-consonants are reduplicated, the stem vowel
between these consonants is not, but is replaced by an inorganic
A-vowel. If the vowel is followed or broken by a glottal stop,
or if there are two successive vowels, the second consonant is
Eight: Wakashan and Salishan Languages
393
13
repeated just the same, the glottal stop being neglected in the
reduplicating syllable. Thus, tc'.e'dd- and L.'d'al- reduplicate as
tdin- and l!aI- respectively. Several nouns with stem-A and
reduplicating-A, listed under Type I, should perhaps belong
here. Three sub-types are to be recognized, according to
whether a remains as such (sub-type a), is palatalized by s, tc,
td, k^, I, or y to i (/) (sub-type 6), or is labialized by xw to u
(sub-type c).
Sub-type II a.
mi'xdl bear
L/d^ard'm' wolf
qld^L.' land otter
qld^sa" sea otter
xd'^'wa fur seal
dsx" hair seal
k!6°ddt!° porpoise
pIdqiAddtc goose
qen^qerC duck
hew^qen' swan
qt'w'^x steel-head salmon
td^qfwa' devil-fish
mdt'.di horse clam
sd^^ba" mussel
mdHdirC louse
dsd'i huckleberry bush
xwdsAbdi soapberry bush
t'.e'^^de'qwai salmon-berry bush
t/d'abuxwdi gooseberry bush
qex" ring finger
bld^qlwdi fish-gill
sopAdatc tail
tsldmuql cloud
plural mAxmi'xdl
LlAiLld'aVd'nC
qlALlqld^L!
qlAsqld^sa'
xAuxd'wa
^As^dsx''
klwAd'k.'wd^dot!'? (with
shortening of second
vowel of stem)
plAqlp'.dq'.Addtc
qAd' qeri' qerC
hAuhew^qerC
qAuqt'w^x
tAqftd^qlwa'
mAt.'mdHldi (with length-
ening of first vowel
of stem)
SAmsd^^ba^
mAtdimdHdirC
^ As'dsdH
xwAsxwdSAbdi
t!Ant!e'''de'qwai
t! Amt! Amuxwdi (with re-
duction of d'a of stem
to a)
qAx^qex^
L!Aq!''L!d''q!wdi
sAp'so'^pAdatc
tslAmits'.dmuql
394
VI American Indian Languages 2
14
tld'^qfaV mountain
st'qeV dug hole, well
td'mic man
xd''p! baby basket
t.'o'mV paddle
waxdHsU pipe
toVxHal necklace
q'td'ahas wooden ball used in
game
mitdli beaver-tooth die
plural t!Ag!t!d''q!aC
sAq'se'qeV
tAmto'mic
XAplxd^p!
V.AmV.Q^mV
wAxwaxdHsli
tAt'oVxHal
qAfq^td^ahas
mAfmiHdli (with length-
ening of first stem
vowel)
qAq'qd^^qa"
lAqfHdq.'wdinop'
qd'^qa rush mat
Idq.'wdinop cedar-bark mat
L.'dxe oldest dAXhldxe
Lldlsdmi strong LfAlLldlsdmi
An irregular example of this sub-type is: —
sdts'.Am tyee salmon SAmsdHs! Am
Here the first and third, instead of first and second, consonants
are reduplicated.
Sub-type II 6.
tde'ddg dog
k^dck^dc bluejay
l6"''honC small clam
tsfdtc.'ilbai spruce
sdsin'^ mouth
sdpdxos horn
ko^sAd' star
ydxai^^ pack-basket
tcHntde'ddo
k^ick^dck^dc
lnnld'^°hoirC
taHtc'ts.'dtcHlbai {tsUtc"-
instead of tsUtc!-)
sissosirC (with shorten-
ing of second stem-
vowel)
sipsdpdxos
kwiskosAd'
yixiydxai^^
Irregular examples of this sub-type are:
idatddH.'dn" mouse
Itx^sal tongue
tclUHddH.'dn" (for tdiC-
instead of tdit.'-see
"spruce" above)
Hsiix"sal
Eight: Wakashan and Salishan Languages
395
15
In the first of these the plural is built not on the already redupli-
cated simplex (as e.g., in "bluejay" above), but on a simpler
unreduplicated stem abstracted from it. In the second example
the first and third, unstead of the first and second consonants,
are reduplicated (cf. "tyee salmon" above).
Sub-type II c. Only one example is available: —
xwdtoqo'm "falls" plural xuCxwdtoqo^m
Type III. Reduplicating Syllable: co or co.
Nearly all of these nouns have g" as their second consonant,
representing, as we have already seen, original w. These nouns
could be considered a sub-type of Type II, were it not that they
form their reduplicating syllable not in -au, as might perhaps
be expected (cf. xAuxAUciii'^ under Type I), but in -o- (-w- after
dj- and g^-) or -o- (probably due to contraction of original -yiw-).
Two sub-types can be recognized, according to whether the
reduplicating vowel is short (sub-type a) or long (sub-type 6).
Sub-type III a.
td^ag^ax"^ fern
td'agHn salmon spear
Sub-type III b.
Id^g^et!" herring
pleg^di halibut
gT/figvp panther
t.'cg^em sun, moon
Mg^os chief
djigHn' song
Id^dak'" skin
totd'ag^ax''
tqtd'agHn
Idla^gyet!"
p!d'^p!eg^di
*g^ug''Vg"l^ (not obtained
as such, but implied
in diminutive plural
gyVg^ugH'gH' "pan-
ther cubs")
tldH.'eg^em sun and moon
hd'^he'gvqs
djudjig^in'
loHd^dak'"
It is not clear why "skin" should reduplicate with o-vowel.
396 ^I American Indian Languages 2
16
Type IV. Reduplicating Syllable: cv; Syncope of First Stem
Vowel.
Only one example has been found of this type. As it begins
with g"-, the stem -g^ of the plural, coming immediately before
another consonant, reverts to it;, uniting with preceding a to
form au.
g^dq'dhas married woman plural g"duq'dhas
That *wdq'dhas is to be presupposed is corroborated by com-
parison with KwantlEn s-wd-wskus "married woman. "^
Type V. Reduplicating Syllable: cvc.
Nouns belonging to this group have long stem-vowels and
differ from Type I in that the reduplicated vowel is shortened,
though it keeps its quality.
xdug"as grizzly bear xduxdug^as
qd'um' eye qduqd'um'
qd^'mai" snow on ground qiirnqo"' mai'
}6kd°min bailer luk'loko'^min
Type VI. Reduplicating Syllable: caci.
ti'hd''ddn' chief's wife tdhtihd''ddn'
heq'sd^min pole for poling canoe hdq'heq'sd°min'
6lqai'' snake 'dVolqai'' (with shorten-
ing of first stem-
vowel)
aL leggings 'dt^aL
"Leggings" may, of course, just as well belong to Type I.
Type VII. Reduplicating Syllable: cv.
q!6a'dda ear q!oq!oa'dda
tt (L.'.-ims) big (house) titl (LfAvis) big (houses)
dx" snow-flake a'dx" falling snow
' C. Hill-Tout, Ethnological Studies of the Mainland HalkomilBm, a division of the Salish
of British Columbia, Report of British Association for the Advancement of Science, 1902,
Ethnological Survey of Canada, p. 89.
Eight: Wakashan and Salishan Languages
39ri
17
Type VIII. Reduplicating Syllable: ce.
According to varying phonetic circumstances we have either
I or e, the latter occurring after q, g! and x. The examples of
this type obtained are: —
qldik''' eagle
kwudjak''^ trout
ft'x" yellow cedar
djd^'dja' tree
sd'idJA^ leaf
tcdyac hand
sayd^ada neck
qd^ya' water
s&'yal lake
xd'adjaic stone
tcu"i child
k.'oyokobPn (or -mVd) fisherman
sidjdqo'p' basket hat
IdidatdAn woman's cedar-bark
skirt
plural q!t'q!dik'"
kivPkwudjdk''^
tltPxwai' (may belong
also to type vii; note
-ai' in plural)
*djidjd'''dja" (not ob-
tained as such, but
implied by diminu-
tive plural djedjidjd-
'"dja)
stsa'idJA'
tcitcdyac
stsayd'ada
qtqd'ya^
stsd'yal
xtxd'adjaic
tcttcu'i
kfwtk! oy okomi 'n
stsidjdqo'p'
ItldidatctAn
Eliminating "yellow cedar," which, as was pointed out, may
just as well be reckoned as belonging to Type VII (there is
reason, however, to believe that il'x" goes back to *tiyix"; see
diminutive type I b and diminutive plural type II f), all these
plurals may be plausibly explained as cases of Type II, redupli-
cating -I- or -e- being the contracted result of -Ay-. It will be
observed that the stems of these nouns contain either i- diph-
thongs, including broken groups (-di-, -a'l-, -u'i-), vowel plus y
{-ay-, -ay-, -a'y-, -oy-), or vowel plus dj {-udj-, -d'^dj-, -d^adj-,
-idj-) ; dj, as we saw above, is probably a resultant of original y.
398 VI American Indian Languages 2
18
Type IX. Reduplicating Syllable: cd (or ca).
Sub-type IX a (with d).
tdd rain plural tcldtclel
qd'"qwai speaker qiuaq6'"qwai
ylp'i'x" hole ydyipl'x'*
Sub-type IX b {with a).
tct'itca.iq^ salt-water hunter tcatcVitca.iq*
Type X. First Stem-vowel Changed to e.
These nouns are reduplicated to begin with, and substitute for
plural reduplication a change of the first stem-vowel to e (long
and open).
The few examples are: —
we'wdlos young man we"wdlgs
e'ddja7n' young woman e"ddjam'
ky.'t'ky.'dky! crow k«!e''k«!dk«!
Type XI. Reduplicating Syllables: cdCAC.
Only two examples have been found of this doubly reduplicat-
ing type of plural formation. In the first, the a, coming after
g", is palatalized to i; in the second, the reduplicating -Ay-
becomes -I- (see Type VIII).
g^d^di'm slave gyagndg^ddPrn
tdyac killer-whale idtltdyac
Irregular Plurals.
Several plurals listed above are somewhat irregular, but there
has been no difficulty in assigning them to definite types. The
two that follow are quite irregular. The second shows not only
reduplication but breaking of -a- to d'a-.
djd'°dja' tree djadj'id'm
inAVq" fawn mamd' aliq' "
Eight: Wakashan and Salishan Languages 399
19
A few nouns change the stem entirely in passing from singular
to plural. Such are: —
sQ.lV'^ woman plural nig'-^dp'tai
sa'as?r" girl (diminutive of ssiir") ninig^ap'lai (diminutive
of nig^dp'tai)
Involving this same change of stem is: —
sdHux'^ married man nig^dpHahai'
Rather different, presumably, is: —
tatfndtcap^ leg tcuklu'ndtcap'
which keeps the same suffix in the plural, while changing the
stem.
Nouns without Plurals.
Quite a number of nouns were secured which form no plural.
Some of these are reduplicated to begin with, and there is clearly
a feeling, though one by no means consistently applied, against
re-reduplication in forming plurals. Others, however, are such
as might easily be reduplicated, were it usage to do so. It is
possible that reduplicated plurals might have been given for
some of these by other informants. Reduplicated nouns that
form no plural are: —
gypgyp panther qivdqumi^s marten
titditcVc owl mdmstco'm mink
is'.i' xHs'.ix'' fish-hawk qwVqwV sea-gull
qwi^qwdH! Ald^'k" butterfly xivdxwadjo'77i' fly
(probably diminutive: dim. plur.
is found)
pok^^pok'"^ liver xdxe'i nit
LaLdpx"^ knife hdiihei' arrow
Non-reduplicated nouns for which my informant would give
no plurals are: —
mdyos raccoon pVk! ground-hog
qlt'etc elk p.'d'alats!" skunk
v!dxd'° \ ^dmaxHdjo'" ant
^ ■■ Vraven ,. • •"
p. 'ah ) qeix salmon-egg
tdeq" robin mo'os head
For "robin," tdAq'tcfeq', which might well enough l)e expected
as plural, was explicitly denied. If necessary to express plurality
in these nouns, qax or q.ix "many" can be juxtaposed before
any of thorn.
400 VI American Indian Languages 2
20
IV. REDUPLICATED DIMINUTIVES OF NOUNS.
Diminutives in Comox, as in other Salish languages, are formed
by means of reduplication. Reduplicated diminutive forms,
however, differ from reduplicated plurals in that the reduplicat-
ing syllable rei:>eats the first consonant of the stem, never also
the second. Moreover, the vowel of the reduplicating syllable
is formed according to different rules from that of the redupli-
cating syllable of plural forms. Further complications result
from the internal changes to which the stem is often subjected,
so that altogether a large number of more or less distinct types
of diminutive formations may be recognized. It will be advan-
tageous to list in a purely analytical way the various features
that are found in diminutives, so that ready reference may be
made to them when discussing the types as^such.
Diminutivizing characteristics are: —
{I. ) Reduplication of initial consonant of stem, followed by
a. Short e (i or i). Two types of c- reduplication may be
recognized, according to whether e is or is not accented.
Thus, mimosas from yno^os "head"; qeqn^ya' from qd^ya^
"water.'
b. Long e (i or i), always accented. Thus LllLlAXWd'*
from lIaxwo''- "dog-salmon."
c. e, always accented. Thus qfe'^q'.e^L! from q.'d'^L!" land-
otter."
d. V, which may or may not be accented. Thus, lolko°min
from loko^min "bailer."
e. V, which is regularly accented. Thus, k!6k!gddf!° from
k!6"ddt!'^ "porpoise."
f. Short a, accented or not. Thus, LdLV'tm'^ from lV' attC
"cockle."
g. Long a. Thus, djddjd''gHn' from djig^in' "song."
h. Long d'a. Thus, sd^aslV^ from sd/f" "woman."
i. Short 0. Thus, LloLld'ami's from dAms "house."
(2.) Glottal stop inserted in stem. This may occur as
a. Breaking of (non-final) yowel or diphthong. Thus,
tcitcd^'^yac from tcdyac "hand."
b. Glottalizing of final consonant (generally m or n); this
should probably include breaking of vowel when final.
Thus. Idtbo'm' from 16'^^hom' "small clam."
Eight: Wakashan and Salishan Languages 401
21
(3.) Quantitative vocalic changes (increments). These include
a. Lengthening of (last) stem voweL Thus, tdtig'^dx" from
td'agyax"" "fern."
b. Change to wd or wa of u of stem. Thus, diminutive
plural kwikumkwd'''mdqin' from plural kumkumdqin"
"sea-lions."
c. Lengthening of inorganic a (or i. e) to i. Thus, xexsim'
from XASAm "box." Less often, full a is changed to i
(cf. 4b), as in kwekwi'im' from kwd'am "coiled storage
basket."
d. Insertion of i. This is probably but another form of 3c,
inorganic a and absence of vowel being perhaps con-
sidered as phonologically equivalent. Thus, qeqAli'q!
from qAl'q! "warrior."
e. Insertion of short vowel (a, i) before syllable with length-
ened vowel. Thus, xt^xigHctn^ (note second i) from
xAucin^ "bone."
f. Lengthening of a or a (non-final) to d. Thns, qlwdq.'wd''^-
djix from q'.wAHx "wood."
(4.) Qualitative vocalic changes. These include
a. Umlaut of a to short e (i). Thus, xexd'adjeHc from
xd'adjaic "stone."
b. Umlaut of a {or d), rarely o, to long e {i, i). Thus,
qle^qle^k'^ from qfak^"- "board."
c. Change of stem vowel to a'a. Thus, tot&'amic from td'mic
"man."
(5.) Vocalic reduction. Under this head may be grouped
a. Shortening of stem vowel before syllable with lengthened
vowel (or inserted i). This shortening before lengthening
is doubtless due to quantitative rhythm. Thus, qicrl''-
qwi'qwV* (note second i) from gmV/iw' "sea-gull. Such
shortened syllables regularly lose their glottal stop, if
there is one present, as in qeqawem" from qd'um' "eye."
b. Syncope of stem vowel after reduplicating syllable with
accented vowel. Long vowels may thus fall out quite as
readily as short ones. Thus, se''sp'xos from sdpdxos
"horn."
These twenty-two diminutivizing features occur in various
combinations, so that a large number of possible types of
402 VI American Indian Languages 2
22
diminutive formation maj' result. A considerable number of
such types can be constructed from the available material, but
this need not exemplify all that actually occur. As to which of
the features listed are fundamental to Salish and which merely
secondary in Comox or several Coast Salish languages, it is
useless to speculate. Adequate comparative data are necessary.
A few points of a comparative nature will be brought forward at
the end of the paper. The various diminutive types will now
be taken up in order, the main stress being laid on the form of
the reduplicating syllable.
Type I. Reduplicating Syllable: ce.
Various sub-types occur, according to whether the stem
vowels remain unmodified or are subjected to comparatively
slight changes.
Sub-type I a. Diminutive feature la only: —
mo' OS head diminutive mimo'os
qtaabas wooden ball qeq'td'abas
q.'ga'dda ear q!weq!oa'dda
'dwdk^'' tobacco 'e'awdk'"
qwAdi^s whale qweqwAdi^s
qo'^'a'^ hemlock qioiqd'^'a''
sidjdqo'p' basket cap sisidjdqo'p'
qeix salmon-egg qeqeyix {-eyi- prob-
ably merely variant
of -ei-)
Sub-type I b. Diminutive features la, 3c (or d) : —
mAqsirC nose mimAqsi^n'
djidis tooth djldjidVs
^I'x" yellow cedar {<i*tiyix") titiylx'^
qAl'q! warrior qeqAWq!
Sub-type I c. Diminutive features la, 3a, 5a: —
/■ey^os chief heheg'-'d"s
Eight: Wakashan and Salishan Languages 403
23
Type II. Reduplicating Syllable: ce; stem: feature 2a.
In these diminutives the first vowel of the stem is broken, the
broken vowel taking the form v'v. If the final vowel is long, it
seems to be shortened (-'d becomes -'").
pi'k! ground-hog diminutive pipi'ik!
qfs^etc elk gleq'/e'e'etc
xd'd big clam xexA'a'"
Though the last diminutive seems to correspond exactly in
form and rhythm to the second, the final -a'" may perhaps here
be better explained as breaking of the last vowel (-d) of the stem
(feature 2b).
Type III. Reduplicating Syllable: ce: stem: features 3a
or d, 5a, 2b.
qd.'unC eye qeqaweni" {-e- doubt-
less merely variant
of -1-)
qwPqui" sea-gull qv/i^qwi'qwi"^
Type IV. Reduplicating Syllable: ce; stem: features 4a, 3c, 26.
kwd'am coiled storage basket kwekwi'im'
Type V . Reduplicating Syllable: ce; stem: feature 5b.
hHkuinAS heart LlidkuinAS
xApd^^ red cedar xexpd'''
Type VI . Reduplicating Syllable: ce; stem: features 5b,
46.
qwAsAm wooll}^ grouse qweq'"se''m-ol
pdxai' creek pfp'xe'*
Type VII. Reduplicating Syllable: ce; stem: features 5b, dc, 2a.
XASAVi box xexsim'
404 VI American Indian Languages 2
24
Type VIII. Reduplicating Syllable: ce.
In this type the redupHcating e is unaccented. According to
whether or not the stem is modified in regard to vocalic length or
quality, various sub-types may be recognized.
Sub-type VIII a. Diminutive feature la only: —
tsldtdilbai spruce diminutive tsUts.'dtcHlbai
qd'ya' water qeqd'ya'
qd'^qwai speaker qwiq6'''qwai
Here probably also belongs qwi'qwdH! Ala^'k" "butterfly."
Sub-type VIII b. Diminutive features la, 5a (accent on third
syllable of diminutive) : —
sayd'ada neck dsiyd'ada (sa-short-
ened to sa-, which,
coming before y,
has to be palatal-
ized to si-)
xa'aidaic stump xexA'd.idatc (-a.i-
probably equiva-
lent to -di-)
Sub-type VIII c. Diminutive features la, 3a (or c) : —
sd'yal lake sisd'ydl
mdHdirC louse mimdHdin'
Sub-type VIII d. Diminutive features la, 4b: —
pdh'^'pok'"' liver pipuk'^pVk'*'
Type IX. Reduplicating Syllable: ce; stem: feature 2a.
Here again the reduplicating vowel is an unaccented e. The
stem, however, is characterized by the breaking of one of its
vowels. According to whether or not umlaut also takes place,
two sub-types are to be recognized.
Sub-type IX a. Diminutive features la, 2a: —
tcdyac hand tcitcd'''yac
Eight: Wakashan and Salishan Languages 405
25
Sub-type IX b. Diminutive features la, 2a, 4a: —
xd'adjaic stone diminutive xexd'adje'ic
As irregular representative of this type may perhaps be con-
sidered : —
qe'n'qen' duck qeqA'dd-ol (built on
unreduplicated sim-
plex)
Type X. Reduplicating Syllable: ct.
Various sub-types are to be recognized, according to whether
or not the stem vowels are quantitatively modified.
Sub-type X a. Diminutive feature lb only: —
tclel rain tcHtdel
pleg^di halibut plVpHg^di
tsfoxd"" codfish ts!i'ts!ox6'°
LlAxwd'^ dog-salmon LltiJAXwd'^
x&^p! baby-basket xt'xd^p!
p!oxo^° raven pH'p.'oxd''
yip'Vx" hole ytyipVx'^
L.'pi'ts.'d'"- yellow-cedar bark l!i' dpttsfa'^
blanket
titctitcl'c little owl tititctitcl^c
knacky dc bluejay kH'^k^dc
qoqgwVm' small breast feathers qivi'^quwihn'
In the last two examples the diminutive is formed, not from
the already reduplicated simplex, but from the unreduplicated
form abstracted from it.
Sub-type X b. Diminutive features lb, 3c: —
Llpdtil bag hlt'Llpdlil
tcleq'' robin tdiHcH^q'
kumdqin' sea-lion kwi'kumdqip'^
djicin' foot dji'djichC
Uq!'' bow li'Hi'q!'*
Sub-type X c. Diminutive features lb, 5a, 3c (or 3a) : —
qlwAt'lHcin" humpback salmon qlwtq'.utVtcin"
t.'d'abuxwdi gooseberry bush tUt! Amuxwdi
pld'alats!" skunk pltp'.A^ldts! (mis-
heard for -p!aI-?)
406 VI American Indian Languages 2
26
Sub-type X d. Diminutive features lb, 3e, 3c:—
xAucin' howe diminutive arg'xrg^/dn' {-igH-<
*-AWA-)
It should be noted that this type of diminutive formation,
while externally similar to Type VIII of plural formation (of., e.g.,
xt'Td^p! "little basket" with qlt'q.'dik'" "eagles"), is in reality
quite distinct in origin, the latter, as we have seen, tracing its
reduplicating -e- to -Ay- and being limited to nouns with z'-diph-
thongs.
Type XI. Reduplicating Syllable: ct; steyn: feature 26.
mdV.ai horse clam me'mAtfa''- (me'- per-
haps misheard for
me-)
hdihei {hdihei'7) arrow htheihei'^
qlAs^Addi buckskin shirt q!t''q!as'add'i
Type XII. Reduplicating Syllable: ct; stem: 46.
g.'dfc'" board qlt'qle'k"''
aL leggings t'eL
Type XIII. Reduplicating Syllable: ct; stem: feature 5b.
There are two sub-types, according to whether or not the stem
vowel is modified.
Sub-type XIII a. Diminutive features 1 b, 5 b: —
q'.dpfxwai oak qliq.'plxivai
pleixdi alder p'.i'plxdi
L!d"q!wdi fish-gill Lli'^Uqlwdi
kup-il''nn"x'' hill kivik'"p-VV
Idqlwdinop' cedar-bark mat H'lq.'wdinqp'^
V.eHbdi wild-cherry bush t'.i'V.bdi
i'.Aq't'.Aqdi dog-wood t'.et'.qdi
In the last example the diminutive is built up on the unre-
duplicatod stem abstracted from the already reduplicated
simplex. The broken stem vowels -ei- of "alder" and "wild-
Eight: Wakashan and Salishan Languages 407
27
cherry bush" disappear in the diminutive apparently without
trace of ',. but this may in part be due to following q! and p! ,
which imply '. With these contrast: —
sd'an' cohoe salmon diminutive sts'ad-ol
Here the -a'a- is treated, not as a broken vowel, but as two
vowels with intervening consonant.
Sub-type XIII b. Diminutive features lb, 5b, 3c: —
lAq'.AS mountain-goat blanket li'l'qUs
LlAqlacin"^ moccasins Llt^Uqlaccn'' (mis-
heard for -en"?)
Type XIV. Reduplicating Syllable: eg; stem: features 5a,
Sc, 2b.
tUg^evi sun, moon tH'tHgHm^
-i- is for -A-, because of following g^.
Type XV . Reduplicating Syllable: ce; stem: features 5b, 4a.
SAg'iA;'" war-club sVsqek'^
Type XVI. Reduplicating Syllable: ce'.
qd'^qa' rush mat qe'^qa''^
toVxHal necklace te'H'x"lal
The diminutive of "necklace," as often happens with nouns
reduplicated to begin with, is built uf) on the implied unredupli-
cated stem. The same applies to the diminutive of "rush mat,"
except that here it is the reduplicating syllable of the simplex,
which doubtless more nearly represents the simple stem, that is
taken as the base of the diminutive form.
Type XVII. Reduplicating Syllable: ce'; stem: feature 4a orb.
Two sub-types are found, according to whether or not there
areat the same time quantitative changes in the stem.
408 VI American Indian Languages 2
28
Sub-type XVII a. Diminutive features Ic, 4b: —
qfa^t! land-otter diminutive qle'^qle'L.'
qld^sa" sea-otter q!e''q!e^s (note loss of
-a')
Sub-type XVII b. Diminutive features, Ic, 5a, 4a (or b) :—
qt'W^x steel-head salmon qe'qeg^e'x
-gy- is from original -w-. It is not clear whether -qeg^e^x repre-
sents *-qewe'x or *-qewex.
Type XVIII. Reduplicating Syllable: ce; stem: features Sc, 2b .
q.'wdt'Am river q!we'q!wat'im' {qlwe'-
not equivalent to
q!we^-; see diminu-
tive plural type iv)
Type XIX. Reduplicating Syllable: ce'; stem: feature 6b.
There are two sub-types, the latter with modified stem vowel.
Sub-type XIX a. Diminutive features Ic, 5b: —
sdpaxos horn se''sp'xos
heg'sd^min" pole for poling canoe he'^hq^sd^min"
Sub-type XIX b. Diminutive features Ic, 5b, 3c: —
tlAkom" beaver (-/;p- doubt- Ue'Hikwim'
less for -kwA-)
Type XX. Reduplicating Syllable: cv.
Here again there are two sub-types, the latter with vocalic
reduction.
Sub-type XX a. Diminutive feature Id: —
xdug^as grizzly bear xdxdug^as
Here probably belongs also xwdxadjo'm' "fly."
Sub-type XX b. Diminutive features Id, 5a, 3c or d: —
Id^gyet!" herring Idlig^et!" {-i- < -a-)
IdidatctAu woman's cedar-bark laltdatctin {-I- <-Ai-)
skirt
t!6'mV paddle tlgUAbiH'
Eight: Wakashan and Salishan Languages 409
29
Type XXI. Reduplicating Syllable: cv; stem: feature 5b.
There are three sub-types, based on diflferences in the further
treatment of the stem.
Sub-type XXI a. Diminutive features Id, 5b: —
yaxai'* pack-basket diminutive yd.ixai'*
Sub-type XXI b. Diminutive features Id, 5b, 3a: —
waxdHsU pipe wauxdHsli*
Sub-type XXI c. Diminutive features Id, 5a, 5b: —
I6kd°mtn bailer l6lko"m%n
Type XXII. Reduplicating Syllable c'^; stem: features 3a or c,
and 26.
There are two sub-types, depending on whether or not the
first vowel of the stem is reduced.
Sub-type XXII a. Diminutive features Id, 3c, 2b: —
sdtslAin tyee salmon sas'dHs'.VnC
Sub-type XXII b. Diminutive features Id, 5a, 3a, 2b: —
sd'idJA^ leaf sasidjd'" (-1- reduced
from -a'i-)
Type XXIII. Reduplicating Syllable: cv; stem: features 5a,
Sa or c.
kfd'^ddt!'? porpoise k!6k!od6th
mdyos raccoon mdmiyo°s (-?'- palat-
alized from -A-, re-
duced from -a-)
td'ag''ax'' fern tdtig'-'dx"" {-i- palatal-
ized from -A-, re-
duced from -d'a-)
td'ag^in salmon-spear tdtigHn (dit.)
y^d^di^m slave g^dg^idVm (-i- pala-
talized from -.1-, re-
duced from -d"-)
410 VI American Indian Languages 2
30
tn^dak'" skin diminutive idlidd'^k''' (dit.)
(Vf.r" hair-seal 'a'asix"
ts.'amuql cloud ts!dts!imAgwil (-/-
palatalized from-x-,
reduced from -a-;
-wiA- merely vari-
ant of -mu-)
olqai'"^ snake ^d^olqai'^
'dmax^idjd'" ant 'd^imax^idjo'"
In the last two examples the final vowel is considered quanti-
tatively long and hence cannot be further lengthened. Quite
irregular is: —
tdyac killer-whale tdtlyac
The long -i- and the short -d- of the stem are the exact reverse
of what would be expected (*tdtiydc, cf. tdtig^dx'' above).
Type XXIV. Reduplicating Syllable: cv; stem: features
5a, 46.
tdatddH.'dn" mouse tddtcHt.'in" (-i- pal-
atalized form of -A-,
reduced from -d"-)
The diminutive, as often, is based on the unreduplicated
stem abstracted from the already reduplicated simplex.
Type XXV. Reduplicating Syllable: cv; stem: features 5a, 2b
Two sub-types are to be recognized, depending on the treat"
ment of the last vowel of the stem.
Sub-type XXV a. Diminutive features la, 5a, 2b: —
l6"°bom' small clam lol'^bo'm"
kloyokobVn fisherman kfokloyokobVn"
Sub-type XXV b. Diminutive features le, 5a, 3a, 2b: —
xd^'wa fur seal xdxAwd'a
sd'^^ba' mussel sdsAbd'"
tc.'e'ddo dog tde'dtdidd'" {-i- pala-
talized from -.4-, re-
duced from -e'd-)
In the last example -e'd is treated as a reduplicating long vowel.
Eight: Wakashan and Salishan Languages 411
31
Type XXVI. Reduplicating Syllable: cv; stem: feature 5b.
Three sub-types are to be recognized, according to whether
the stem undergoes no further change or is further modified.
Sub-type XXVI a. Diminutive features le, 5b: —
sOstn' mouth diminutive sossin"
pldqlAddtc goose p.'dp.'qiAddtc
ti'hd'^ddn' chief's wife tiVhdddn'
sopAdatc tail so^'spAdatc
xwdsAbdi soapberry bush xwdx"sabdi
tix'^sal tongue tlHx'^sal
osd'i huckleberry bush '6'ASd'i {-'os- cannot
be further reduced
than -'as-)
mi'xdl bear mt'mExdl {-e- is mere-
ly glide)
siplAmirC shinny stick sPsp! Ami^n"
mitdli beaver-tooth die mi'm{i)tdli i-i- is
merely glide)
ky'.ikydyu oar k^!t'k«!k^dyu
si'^qeV dug hole, well si'^sqeV
"Bear," "shinny stick," "beaver-tooth die," and "oar,"
which have short stem-vowels, are perhaps better listed with
type X.
Sub-type XXVI b. Diminutive features le, 5b, 3c: —
kd'^SAd' star kok'sid'
Sub-type XXVI c. Diminutive features le, 5b, 5a, 3a: —
V.e'^'de^qwai salmon-berry bush V.eV.dAqwdi
Type XXVII. Reduplicating Syllable: cv; stem: features
5 b, 4 b.
tld^qlaV mountain t!dt!q!eH"
Type XXVIII. Reduplicating Syllable: cv; stem: features 5b
(or a), 3a, 2a.
td'^qlwa' devil-fish tdCqlwd^"
djd'^dja' tree djddjidjd^" {-i- pal-
atalized from -.4-
reduced from -a"-)
412 ^f American Indian Languages 2
32
Type XXIX. Reduplicating Syllable: ca.
Two sub-types have been found illustrated, each represented
by but one example in the material obtained.
Sub-type XXIX a. Diminutive features If, 3c, 2b: —
Lp'Am' cockle diminutive LdLV'im''
Sub-type XXIX b. Diminutive features If, 3b, 2a: —
kwudjdk'" trout kwakwd'^djak'''
Type XXX. Reduplicating Syllable: ca; stem: feature Sf.
Two sub-types may be recognized, the second with further
modification of the stem.
Sub-type XXX a. Diminutive features Ig, 3f : —
qlwA'ix wood qlwdqlwcC^djix {-dj-
<*-y-, glide be-
tween -a'"- and -i-).
Sub-type XXX b. Diminutive features Ig, 3f, 3c: —
djig^in" song {<*dJAWAn') djddjd'gHn"
Type XXXI. Reduplicating Syllable: cv'v; stem: feature 5b.
sdlV"* woman sd'aslV'' girl
Lfd'ard'm" wolf Ud'aLfrd'm'
Type XXXII. Reduplicating Syllable: co; stem: feature J^c.
Two sub-types, each represented by one example, are found,
the second involving a further change of stem.
Sub-type XXXII a. Diminutive features li (perhaps rather 1
d), 4c:-
td'mic man totd'amic boy
Sub-type XXXII b. Diminutive features li, 4c, 3d: —
L'.Ams house Lfoild'amPs
Diminutive in -ol, -ol.
Besides forming diminutives by means of reduplication and
internal stem change, Comox can also make diminutives of
animal nouns by means of a suffix -dl{t'") or -oi(r"). Some of
the diminutives in -o?(r") or -ol{V") are nouns whose simplex is
Eight: Wakashan and Salishan Languages
413
33
already reduplicated (cf. reduplicated nouns which form no
redupHcated plural), yet not all. Of those formed from unre-
duplicated nouns, somo have diminutive reduplication at the
same time, others not. By an interesting phonetic law of
rhythmic balance -o?(r") is suffixed to stems whose last vowel
is short, -pZ(r") to those whose last vowel is long. The examples
obtained of the suffix are: —
1. -dl{V-)
hew^qen' swan diminutive Jiew^qAdol
mi'^mau cat
q.'dik'" eagle
qe'n'qen' duck
tsfitsqU'^nas chicken hawk
tdeq' robin
sd'arC cohoe salmon
xdp'xop'^ humming bird
gvfigvii panther
qldikol
^q!eq!Auq!dikdl
little eagles
qeqA'ddol
ts!itfiq!e'^nasdlV "
tditdeqHdeqolV" little
robins
jsis^adol
[sisosg^dddl plur.
xop'xdpoIC'^
The last two seem irregular as regards rhythmic balance;
perhaps they were respectively misheard for *x6p'xopdlt''' and
*gH^gHyul. -ol has also been found in mim'ini'dl A;'" mdmstco'm
"little mink."
2. -p^ (r")
hdhnhd'm blue grouse hd^mho^mol
qwdqumi^s marten qwdqunii^solV"
qwAsAm woolly grouse qweq'"se^mol
tsU'x^tsHx^ fish-hawk tsU'xHsUxwol
kwa'kwd'^djo' grey-squirrel kwa'kwd'^djgl
414 Vf American Indian Languages 2
34
V. DOUBLY REDUPLICATED DIMINUTIVE PLURALS
OF NOUNS.
The plurals of diminutives are, as a rule, doubly reduplicated,
the first reduplicating syllable expressing the diminutive idea,
the second that of plurality; the first reduplicating syllable is
almost invariably of diminutive type, the second of plural type.
Hence diminutive plurals are morphologically, and psycholo
gically, diminutivized plurals, not pluralized diminutives.
While they may be said, on the whole, to be formed from the
plural of the simplex, the diminutive singular has often influence
on the form of the diminutive plural, both as regards the inner
stem changes and the vowel of the reduplicating syllable. Thus
diminutive plurals may be said to combine, roughly speaking,
the characteristics of both the plural and diminutive of the
simplex. In order better to understand the formation of the
diminutive plural and to assist in cross-referencing, the types to
which the non-diminutive plural and the diminutive singular
belong will be indicated in the following lists.
Type I. Reduplicating Syllable: ce; followed by plural of simplex.
The reduplicating syllable is analogous to that of diminutive
types X, XI, XII, XIII, XIV, and XV. According to whether
or not the remaining part of the word is somewhat modified
from the plural of the simplex, sub-types may be recognized.
Sub-type I a. Plural of simplex unchanged: —
LllkuiriAS heart
plur.
I.
dim. V.
dim. plur. lU'lIeJc"-
LlikuiriAs
tsloxo"" codfish
dit.
x a
tslUsloxtsloxd'"
lIAxwo''' dog-salmon
dit.
dit.
L.'tL.'aXL.'AXWd'''
Lfpt'tsld'" yellow
cedar
bark basket
dit.
dit.
LHL.'Ap'L.'ApitS.'d'"
tc.'eq'' robin
no plur.
xb.
(type
I implied
tditdeqHdeqdlt'''
in dim. plur.)
aL leggings
]
[. (or VI.)
XII.
e'aL'aL
kupu''mi"x" hill
dit.
XIII
a. kwlkup'kup-lH'
(with -IH' as in
diminutive singu-
lar)
Eight: Wakashan and Salishan Languages
415
k"!ik''dyu oar
35
plur. dit. dim.
sip.'Amtn' shinny-stick dit.
qwdqumi's marten no plur.
(type I, based
on stem-form of
unreduplicated
simplex, im-
plied in dim .
plur.)
xd^p! baby basket ii a.
t.'d'abuxwdi goose-
berry bush dit.
Idqfwdinop' cedar-
bark mat dit.
dd'qlwdi fish-gill dit.
k.'o^dot!'? porpoise dit.
tsldmuql cloud dit.
mitdli beaver-tooth
XXVI a. dim. plur. k^Hk^li-
kv'k^Hkmyu
dit. sisip!sip!amVn'
-olV'* qwtqumqwdqumVs
X a. xexAptxd'^p!
X c. t!it! Amt! Abuxwdi
XIII a. lilAqfHdqfwdinop'^
dit. lUl! Aq!" ild^qlwdi
XXIII, k!wik!wAd'k!6°doi!^
dit. tsUts.'Amtsldmuql
die
dit.
XXVI a.
mVniAt'mi'Hdli
mi^xdl bear
II a.
XXVI a.
mi'mAxEmixdl
{-E- is glide)
st^eV well
dit.
dit.
sisAq'se^qet'
Dsd'i huckleberry
bush
dit.
dit.
'e'As'osd'i
sdpAdatc tail
dit.
dit.
slSAp'so'^pAdalc
td'mic man
dit.
XXXII a.
titAmtd'omic
g^d'di'm slave
XI. (i
iside
XXIII.
gHgndg^d^dl'm
from
g"d- belongs
(based on -gHdg^d
to II
b.)
dVm of plur.)
ydxai'^ pack-basket
lib.
XXI a.
ylyixiydxai'*
tdatc.'dHIdn" mouse
dit.
XXIV.
tcHlcHVtddH'.dn'"
kd^'SAd' star
dit.
XXVI b.
kwi'kwiskd"SAd'
pleg^di halibut
Ill b.
X a.
p!ip!6"p!(:g"di
gypgvii panther
no pi
ur.
-ul
gyi^gyugypgH'
(type
Ill b
im-
plied
in dim
plur.
)
olqai'* snake
VI.
XXIII.
'e'Al'olqai''
416
VI American Indian Languages 2
36
sd'idJA' leaf plur. viii. dim. xxii b. dim. plur. smsd'trfj^'
djd'^dja' tree irregular xxviii. djedjidjd^''dja''
(built on plur. of
type viii)
yipi^x'* hole ix. x a. yeyip'yip'l'x''
(built on plur. of
type i)
Sub-type I b.
feature 3a, c, or d
djidis tooth
L.'pdtil bag
djicin' foot
lAq.'AS mountain-
goat blanket
L.'Aqlacin^' mocca-
sin
tfAkom"^ beaver
tfo'mV paddle
waxdHs.'i pipe
td^qlwa' devil-fish
djig'^in' song
Sub-type I c.
feature 5a: —
I6k6°min bailer
ti'hd''ddn' chief's
wife
Plural of simplex modified by diminutive
I.
dit.
dit.
dit.
dit.
I.
II a.
dit.
dit.
Ill b.
Plural of simplex modified by diminutive
lb.
djldjiddjidVs
xb.
lUl! Ap^ l! Apdtil
dit.
djidjisdjidtn'
XIII b.
lilAqUAqHs
dit.
L.'eLlAq.'LlAq.'actn'^
(-tn' misheard for
-tn" ?)
XIX b.
tat! Ak'H! Akwim'
XX b.
t!lt!Amt!6''biH'
XXI b.
wiwAxwaxdHsH'
XXVIII.
titA^q'td^qlwd"
(-'q' misheard for
-q! ?)
XXX b.
djidjudjigHrC
V.
VI.
XXI c.
XXVI a.
liluk'ldko^mtn
titahtihd'^ddn'
Sub-type I d. Plural of simplex modified by diminutive
feature 2b: —
W'^bonC small clam ii b.
kloyokobVn fisher-
man VIII.
XXV a. lilimld"°bo^7rC
dit. k!wtk!wlk!oyoko-
bi'n"
Eight: Wakashan and Salishan Languages
All
37
Sub-type I e. Plural of simplex modified by diminutive
feature 4b: —
tld'^qlaV mountain plur. ii a. dim. xxviii.
dim. plur. t!et!Aq!t!d''q!eH'
Sub-type I f. Plural of simplex modified by diminutive
features 3b, 2a, 3c: —
kumaqirC sea-lion i. x b. kwikurnkwd"' mdqirC
A couple of aberrant diminutive plurals with ce- are given
under type ii f.
Type II. Reduplicating Syllable: ce; followed by plural of
simplex.
The reduplicating syllable is analogous to that of diminutive
types I, II, III, IV, V, VI, VII, VIII, IX. Sub-types are
to be recognized here as in type i.
Sub-type II a.
Plural of simplex
unchanged : —
qwAdi^s hump-
backed whale
I.
I a.
qweqwAd ' qwAdVs
qo^'a'^ hemlock
dit.
dit.
qwiqoqo'^'a'^
'dwdk''' tobacco
dit.
dit.
'e'Au'dwdk'"'
xApd'^ red cedar
dit.
V.
xexAp'xApd'^
q!dp!xwai oak
dit.
XIII a.
q!eq!Ap!q!dp!xwai
q'td'abas wooden ball
used in game
II a.
I a.
qeqAVq'td'abas
qe'n'qerC duck
dit.
IX b.
qeqAd'qen" (based
on unreduplicated
simplex)
qd'^qa" rush mat
dit.
XVI.
qeqAq^qd''^qa'
toVx^lal necklace
dit.
XVI.
tftAxHdVxHal (re-
duplicating sylla-
ble for plurality
based on unredu-
plicated form of
simplex)
418
VI American Indian Languages 2
38
qfa^L! land-otter plur.
II a. dim. xvii a. dim
. plur. qleqlALfqld^d
qld'sa" sea-otter
dit.
dit.
qfeqlAsqld^s (with
loss of -a\ as in
dim. sing.)
xd^'wa fur seal
dit.
XXV b.
xexAuxd'wa
sd^'ba' mussel
dit.
dit.
sisAmsd^'ba'
xwdsAbai soapberry
bush
dit.
XXVI a.
xwexwASXwdsAbdi
p.'dqfAddtc goose
dit.
dit.
pfeplAqlpfd^qlA-
ddtc
L'.d'aVo'm" wolf
dit.
XXXI.
lIcl! AIl! d' aV 6' rrC
ts'.dtcHlhai spruce
lib.
VIII a.
ts!its!itc'ts!dtc!ilbai
k^dck^dc blue jay
dit.
X a.
k^dky'ickydc (based
on unreduplicated
-form of simplex)
sdpdxos horn
dit.
XIX a.
sisipsdpdxos
t&'ag^ax'' fern
Ill a.
XXIII.
titgtd'ag^ax"
htg^os chief
Ill b.
I c.
heho^he^g^os
xdug^as grizzly bear
V.
XX a.
xexauxdug"as
heq'sd''min' pole for
poling canoe
VI.
XIX a.
hehdq' heq' sd^min'^
qtod'dda ear
VII.
I a.
q!weq!oq!oa'dda
sidjdqo'p' basket
hat
VIII.
I a.
sistsidjdqd'p'^
tdyac killer whale
XI
XXIII
tititd'^yac
(plur. of type
VIII
implied in
dim.
plur.)
Sub-type II b. Plural of simplex modified by diminutive
feature 3c or d: —
qAl'q! warrior
I.
lb.
qeqAlqAli'^q!
Ug.'" bow
dit.
xb.
lUAq'Hi'q!''
q!wAt ■ iHcin' hump-
back salmon
dit.
X c.
q!weq!wAVq!wA-
VVtdirC
xAucin' bone
dit.
xd.
xexAUXAudirC
mdHdin' louse
II a.
VIII c.
mimAtdrndHdirC
<isx" hair seal
dit.
XXIII.
'e'As^dsix"
td'agHn salmon spear
Ill a.
dit.
titotd'agHn
Eight: Wakashan and Salishan Languages
419
39
Sub-type II c. Plural of simplex modified by diminutive
features 3a or c, and 2b: —
xAsAm box
plur. I.
dim
VII.
dim. plur.
xexASXAsim'
qwAsAm woolly
dit.
VI.
qwequsqusim"
grouse
{-qus- probably
merely variant
of -qwAS-)
qfAs'Addi buckskin
shirt
dit.
XI.
qleq.'AsqfAs'add'i
LV'Am' cockle
dit. (or VIII.)
XXIX a.
LiLVLdi'im" (with
irregular lengthen-
ing of -V- = -Al- to
-di-)
htw^qerC swan
II a.
-ol
hehAuhew''qen"
mdV.di horse clam
dit.
XI.
memAt!mdH!d'*
qd'um' eye
V.
III.
qeqoqd'om"^ (-qo-
heard for -qau-, or
perhaps for -qAu-
reduced hom-qau-
— see type in)
Sub-type II d. Plural of simplex modified by diminutive
feature 4a or b : —
pdxai' creek i.
q!dk'" board dit.
SAg'ifc'" war-club dit.
VI.
XII.
XV.
pipAxpaxe'*
qfeqlak'^'qltk'*
Sub-type II e. Reduplicating syllable of plural of simplex
changed to cau-: —
xd'a big clam
II. xexAuxd'A (note
change of xa'd- to
-xd'a, perhaps due
to rhythmic anal-
ogy of dim. sing.
xexA^d'A)
qfdik'" eagle
VIII.
-ol q!eqfAuq!dik-dl
420
VI American Indian Languages 2
40
These strange diminutive plurals can hardly be explained
otherwise than as formed by analogy of such diminutive plurals
as xexAUxd'wa "little fur seals," xexauxdug^as "little bears,"
and xexAUXAUcin' "little bones," where -xAU-{-xau-) is etymo-
logically justified. The parallelism of xd'd "big clam" and
xd^'wa' "fur seal" seems particularly plausible.
Sub-type II /. Plural of simplex modified by diminutive
feature 4c (for convenience of comparison one form with ce is
included) : —
V.t'ihdi wild cherry plur. i (or viii), dim. xiii a. dim. plur.
bush
qe'w^x steel-head
salmon
ft*x" yellow cedar
II.
VIII.
XVII b.
lb.
t!et!Amt!d'abdi
(really belongs to
type i; based on
reduplicated plu-
ral of type ii)
I.
qeqAuqd'''g^e*x
{-g^e^x as in dim.
sing.)
titold'ayix'' {tVx*
> Hiyix'', U- being
modified to ta'a- ;
-to-, cf . type II e,
is peculiar and is
probably due to
analogy of titotd'-
ag^ax'' "little
ferns")
Another diminutive plural with erratic -o- vowel (in both
reduplicating syllable for plurality and stem) belonging to
type I, is: —
safari' cohoe sal-
mon
XIII a. sisoso'dd-ol
The material at hand does not permit to see what analogies
have operated here.
Eight: Wakashan and Salishan Languages 421
41
Type III. Reduplicating Syllable: ee; reduplicating vowel of
plural of simplex shortened.
A new feature is here introduced, the shortening of the long
reduplicating vowel characteristic of the plural. Sub-types
are here also to be recognized.
Sub-type III a. Plural of simplex not otherwise modified : —
xdp'xop' hum- plur. i. dim. -olV" dim. plur.
ming bird xwexop'xo^p'^ (bas-
ed on unredupli-
cated simplex)
^"^"6^'" herring iii b. xxb. lilola^g^et!"
Id^dak'"* skin iii b. xxiii. lei old" dak""
qd'ya* water viii. viii a. qeqeqd'ya'
xd'a.idatc stump i. viii b. xexexd' aidatc
(type VIII im-
plied in dim.
plur.)
sd'yal lake viii . viii c. sjsjsd'yal
IdidatctAn woman's dit. xx b . lilildidatctAn
cedar-bark skirt
Sub-type III b. Plural of simplex modified by diminutive
feature 2a: —
tcdyac hand viii. ix a. tcitcUcd'yac
Sub-type III c. Plural of simplex modified by diminutive
features 2a, and 3b or f : —
sayd^ada neck viii. viiib. sisfsd'ya'ada
kwudjak'"^ trout dit. xxix b. kwikwikwd'^djak'''
Sub-type III d. Plural of simplex modified by diminutive
features 4a and 2a: —
xd'adjatc stone viii. ixb. xexexd' adje'ic
422
VI American Indian Languages 2
42
Type IV. Reduplicating Syllable: ce; followed by simplex.
It seems that a reduplicating syllable with e tends to be
considered the morphological equivalent of double reduplication
(see plural type X, diminutive type XVI), in this case of combined
diminutive and plural reduplication. Various sub-types are to
be recognized, according to whether the reduplicating syllable
is followed by the unmodified (or modified) simplex, the modified
form characteristic of the diminutive, or by a form still further
modified.
Sub-type IV a.
Simplex unc
changed : —
pWixdi alder
plur. I.
dim. XIII a. dim. plur.
(or viii).
plepft'^^xdi
L.'Ams house
dit.
XXXII b. hle'^LlAms
mo' OS hand
no plur.
I a. me' mo' OS (may
also be considered
as belonging to
type IV b)
Sub-type IV b. Simplex modified by diminutive feature 5a:-
ic.'e'ddo dog ii b . xxvb . tcle'tdin'am" (ir-
regular in that -o
of stem is dropp-
ed; with -dm' cf.
perhaps -d'm of
djddjid'm 'trees')
Sub-type IV c. Reduplicating vowel of diminutive changed
to e: —
xwdxwadjo'm' fly
(dim. in form)
qwi^qwdH.'Ald'k* butterfly
(dim. in form)
mAqsirC nose i.
kwd'am coiled
storage basket dit.
q!wdt' Am river dit.
ttx'^sal tongue ii b .
xwe''xwAdjd'm'
qwe^qwdHfAla'k'
I b. me'mAqsi^n
IV. kw^'^kwi'im'
xviii . qlwe'^q'.wat ' tm'
XXVI a. te'Hx'^sal
Eight: Wakashan and Salishan Languages
423
43
Sub-type IV d. Reduplicating vowel of diminutive changed
to I; stem further modified by diminutive features 5a and 3c: —
q'.wAHx wood plur. i. dim. xxxa. dim. plur.
q!we''q!wadjix
Sub-type IV e. Reduplicating vowel of diminutive changed
to e; stem further modified by diminutive feature 5b: —
sdtslAm tyee sal-
mon
II a.
XXII a.
se'^stsli^m'^
Type V. Reduplicating Syllable: ce; followed by plural of
simplex modified by diminutive features 5a and 3a: —
tlc'^'de'qwai salmon- plur. ii a. dim. xxvi c. dim. plur.
berry bush t'.e't'.AnV.An'qwai
{-€'- is lost, cf.
diminutive feature
5 b)
sdstn^ mouth lib. xxvi a. se'sossin'
VI. MISCELLANEOUS LINGUISTIC MATERIAL.
Numerals.
1. pd'a
2. sd'a
3. tcdlas
4. mos
5. siydtcis
6. tfdxam (oT-ab)
7. ts'.o'HcVs
td'atcPs
tigH^'x''
8.
9.
10.
IL op'an haik'" pd'"
20. simcyd'a 200.
30. tcanaux'^cyd^a 300.
40. mosalcyd'a 400.
50. seyatsfalcyd'a 500.
60. V.dxamalcyd'a 600.
70. ts.'otci'alcyd'a 700.
80. td'atdsalcyd'a 800.
90. tigHxwalcyd'a 900.
100. Vsd'ntc 1000.
6p' an
2000 is sdba Vsd'agHtc or sd'a Vsd'agHtc.
sd'mitc
tcd'adagHtc
mosd'ag'^itc
seatsd'agHtc
tlaxamd'ag^itc
tslotcisd'ag^ltc
td'atclsd'ag^ltc
tig^ixwd' ag^itc
fsd'agHtc
424
VI American Indian Languages 2
44
Numerals with classifying suffixes, referring to class of objects
counted, are:
People
Canoes
Fathoms
Houses
Dollars
1. pipd'a
natdd^gni
no<cM"?"fal
natddfwdutr"
pdp'p*
2. sisd'a
sabagvil
sAtntdl
ad'abautf
sdQ'vS
3. tcdldy
tcddd'gvil
teadd-fldl
tcd'dautf
tcdlaa(fs
4. mOsdvi
m&sdul
mosdltdl
mdsautf'
mbsQS
5. s^yatsdyi
si'vatsd'oxit
siyatsaltdl
siyatsautf
a^yalsos
6. tfdzamdyi
t!dxam.d<'gvU
7. ts!6tcisdvi
8. ta'dtcisdyi
9. llgt/'iTWdyi
10. dpdndyi
The series for "dollars" refers, properly speaking, to round
objects, including such objects as heads and turnips.
Body-part suffixes. Examples of body-part "substanti-
vals," as they have been termed by Boas, which occur only in
composition (better perhaps derivation), are: —
head: ■pdq-e'^q'warC white-headed
idx • e'q'warC red-headed
(or -ad')
hand: pdq'd'''dja' white-handed
tcixo'"dja' red-handed
eye: pdq-dos white-eyed
pdq'paq'dos white-eyed (plur.; refers to several persons
or to two eyes of one person)
tcixdos red-eyed
tdxtcixdos red-eyed (plur.)
nose: ts!dts!e'miq'"^ red-nosed
paqe'^q'"" white-nosed
t!dt'ts!d'''miq'" nose bleeds
foot: pdq'dn' white-footed
pdq'paq'cin' white-footed (plur.)
With these contrast independent use of "ear" in pdq^paq*
qloa'dda "white ears."
Eight: Wakashan and Salishan Languages 425
45
Possessive and subjective pronouns. Only very frag-
mentary data were secured on Comox pronouns. I do not
consider them as particularly reliable.
tAtsi md'os my head tAmsi md'gs our heads
tAn mo^os your head Ia mo^osap' your
(plur.) heads (vis-
ible)
tA mo'gss his head (visible) ku mo'osap' your
(plur.) heads (in-
visible)
ku mo'oss his head (invisible)
tA and ku are articles implying visibility and invisibility respect-
ively. Possessive pronouns modifying verb subjects are: —
'd tsi mo' OS my head is sore ('a' to be sore)
'd' tAn mo' OS your head is sore
'd' tA mo'oss his head is sore
'd tA mo'oss tA 6-d?r" the woman has headache (literally, sore
the her-head the woman)
Possessive pronouns modifying verb objects are: —
tc'kfudA wad tsi mo'os I see my head
tc'kludAxwad das mo'os I see your head
tc'kludAxwad dA mo'oss I see his head
tc'kludAxwad das tciHcdyac I see your hands
tc^k/udAxwad dA tdHcdyacs I see his hands
Subjective pronominal suffixes are: —
titc to'mic I am a big man (tl big)
ttHc'"" to'mic you are a big man
tV" to'mic he is a big man
tt'^djan sdlV" I am a big woman
tV^'djaux" sdW'' you are a big woman
426
VI American Indian Languages 2
46
VII. COMPARATIVE NOTES ON SALISH NOUN
REDUPLICATION.
This is not the place to enter into anything like a systematic
comparative treatment of Salish reduplication, the more so as the
phonetics of most of the material available for comparison are
not such as to allow one to make definitive classifications of plural
and diminutive types (this remark applies particularly to vocalic
quantity and glottal stops, both of which, as we have seen, are
important for our present purpose). Certain facts of a compara-
tive nature, however, come out quite clearly and may be briefly
noted here.
Plural reduplication. It is evident that all Salish lang-
uages make use, like Comox, of different types of plural reduplica-
tion. Both types I and II are plentifully illustrated and are
without doubt the fundamental Salish processes. Examples of
type I are: —
Bella Coola
s-t7i tree
plur. s-tntn^ {s-, as
often in Salish, is
prefix)
Tcil'Q^uk
s-kwomdi dog
s-kwomkwomdi^
(Cowichan group)
Shuswap
sk'dqa dog
s-k'aqk'dqa^
noqonuq woman
noqn6qonu<f
Okanagan
s-k' EltEmtQ man
s-k' Elk' EltEmtQ*
Thompson River
s-k'um
s-k'umk'urw'
mountain
s-nukoa friend
s-nukEnukoa^
s-k6um crumpled
s-koumkoum^
Examples of type
II are: —
Nanaimo
s-pdl raven
s-pslpd'l^
s-tdlo river
s-tEltd'lo^
' F. Boas, The Salish Languayes of British Columbia, Report of British Association for
the Advancement of Science, 1890, 6th Report on the Northwestern Tribes of Canada, p.
127.
' C. Hill-Tout, Report of British Association for the Advancement of Science, 1902,
Report on the Ethnological Survey of Canada, p. 20.
• F. Boas, ibid., p. 131. k' is here and in other forms equivalent to our q; q to our i;
Q to our f, tl to our I (and l); IV to our l!.
« Ibxd., p. 135.
' F. Boas, Report of British Association for the Advancement of Science, 1898, 12th
and Final Report on tlie Northwestern Tribes of Canada, p. 28.
• F. Boss. Report B.A.A.S., 6th Report on N.W. Tribes, p. 129.
Eight: Wakashan and Salishan Languages
427
Shuswap
Okanagan
Thompson River
47
s-kdpk'Eii head plur. s-k' Epkdpqsn^ (prob-
ably misprint for
-kspkapk' En)
k'est bad ky'Eskest^ (probably
misprint for -k^est)
s-k'slk'tlo^
s-k'elg Indian
cdsuQ stone
s-pam camp fire
s-nikidp coj'ote
CEncaEnQ"^
s-pEmpdm^
s-niknikidp^ (-t- is
very open and short,
-I- is close and equi-
valent to our -t-;
hence type iib)
s-Quasit to walk s-QUsquasit^ {type II c)
An interesting Thompson River example of type II is: —
cirdp tree cipcirdp^
An example of type III (reduphcating -aw- contracted to -p-
or -0-), but with retained -w- (Comox -g^-) is: —
Okanagan iEiuwU boy totuit* (based on unre-
duplicated form of
simplex ; final vowel
of stem apparently
shortened)
This example follows type III b.
usage in the treatment of the same
languages, compare with this: —
Lower Lillooet tu'u''wut' boy
As illustrating diversity of
stem in different Salish
tutu'u'^wuV^ (-U- is short
and close)
This follows type III a, besides which the stem itself seems to
differ markedly in regard to vocalic quantity and rhythm from
the cognate Okanagan stem. Shuswap agrees better with
Okanagan: —
tuwtui boy tutuwtut^
i Ibid., p. 131.
* Ibid., p. 135.
' F. Boas, Report B.A.A.S.. 12th Report on N.W. Tribes, p. 28.
* F. Boas, Report B.A.A.S., 6th Report on N.W. Tribes, p. 135.
' Some Lower Lillooet linguistic material was obtained iu January, 1912, from I nace
Jacob (Indian name Yisp).
•F. Boas. ibid., p. 131.
428
VI American Indian Languages 2
48
It would seem that type VII, which is only sporadically repre
sented in Comox, is more typically developed in Interior Salish
Examples are: —
tsitQ house
gitia old woman
tcltQ house
s-tsuk' picture
s-k'dk'qa dog
s-pEzuzo bird
Shuswap
Thompson River
s-kikElUQoa musk-
plur. tsUsttQ^
gigitia^
IciidtQ^
s-tsutsuk'^
s-k'ak'dk'qa^
s-pEpEzuzo^ (this form,
however, may really
be diminutive plural,
s-pEzuzo being dimin-
utive, with final re-
duplication, of s-pEz6
"animal," whose plu-
ral is normally form-
ed : s-pEzpEzo,^ type I)
rat
s-kikikEldQoa^
Lower Lillooet
tctC'^x house
tcitctV''x^
gd'? water
qdqo'^^
Note also : —
Nanaimo
k'unes whale
k'okuinis* (probably mis-
(i.e. qunes)
print for -k'uinis)
It is interesting to contrast with this plural (qogwinis in our
orthography) Comox qwAd'qwAdi^s humpbacked whales ( <qwAn-
qwAnis) of type I. Here again we see the tendency for different
Salish languages to form the plural of the same stem according
to different types.
Type IX also is illustrated outside of Comox. Examples are : —
Nanaimo
Tcil'Qeuk
lalEm house
wuqas frog
mela son
laldlEni'^
hdiiweqas* (-u- presum-
ably glide; haw- dis-
similated from *wdw-?)
mdmela^
' ibid., p. 131.
« F. Boas, Report B.A.A.S.. 12 Report on N.W. Tribes, p. 28.
' Obtained from Ignace Jacob.
< F. Boas, Report B.A.A.S., 6th Report on N.W. Tribes, p. 129.
»C. Hill-Tout. Report B.A.A.S.. 1902, Ethnological Survey of Canada, p. 20.
Eight: Wakashan and Salishan Languages 429
49
Type X is illustrated in : —
Nanaimo k'dk'En post plur. k'&lak'En} (d is
apparentl}' our e)
The last example, with its inserted -la-, shows also another
method of plural formation, one not found, at least as far as
can be judged from available material, in Comox. Other
examples of this inserted -lia)- are: —
Nanaimo hd'pet deer hala'pst^ (type IX)
tcitdek'an mink tciletciek'an^ (type VII)
spdk'Em flower spdlak'Em}
Tcil'Qeuk k''dmi maid k''dlami^
stEktyu horse stElEktyu^
ydsuk hat ydlsuk^
There seem to be still other types of plural formation in
Salish that are not represented in the Comox material given in
this paper. One of these is to prefix -a- (Boas and Hill-Tout
write -E-), which may be palatalized to -j-, to the stem. Examples
of this type are : —
Nanaimo s-mtysg deer s-Emtyeg^
Tcil'Qeuk s-wtEka man s-iwtska^ {-a- palata-
lized to -i-, -T- by S-?)
This type is perhaps a reduced form of another one that occurs
with some frequency, reduplication with ca-. Examples are: —
Tcil'Qeuk IdlEm house lEldlEm^
s-mdlt stone s-mEvidW
Shuswap la good IeW
Nanaimo laldlEm "houses," as compared with Tcil'Qeuk
IsldlEm, suggests, in turn, that CA-reduplication is reduced from
ca-reduplication (type IX). Tcil'Qeuk yESidm "chiefs"^ from
sldm may be dissimilated from *SEsld7n (or does y- reduplicate
-I- of stem?). Vocalic changes {e to o and d) are illustrated in: —
Tcil'Qeuk s-weEkdtl boy woEkdil^
' F. Boas, Report B.A.A.S., 6th Report on N.W. Tribes, p. 129.
* ibid., p 128.
» C. Hill-Tout, Report B.A.A.S., 1902, Ethnolocical Survey of Canada, p. 20.
< F Boas, Report B.A.A.S , 6th Report on N.W. Tribe-, p. 128.
* Ibid., p. 131.
430 VI American Indian Languages 2
50
s-wtwilus youth s-wAvrilus^ (this may
be considered, how-
ever, as formed from
unreduplicated sim-
plex according to
Type IX)
With the latter example compare Coraox we"wdlos "young
men" from we^wdlos.
To sum up, it is clear that there are a number of wide-spread
Salish methods of forming the plural, which may, however, at
last analysis turn out to be capable of reduction to Type I (of
which Type II may be a reduced form). It is conceivable that
sub-types, which have developed in particular cases from this
by secondary phonetic processes (cf., e.g., Comox Types III and
VIII), set the pace for new purely analogical, not etymologically
justifiable, forms, so that now any one Salish language exhibits
great irregularity. Certain of these secondary types seem to be
favoured in one language, others in another, so that, as we have
seen, the same stem is sometimes differently treated in different
languages. To unravel the history of reduplicated (and other)
plurals in Salish, however, requires a far more abundant body of
material, for purposes of comparison, than has as yet been made
accessible.
Diminutive reduplication. The last remark applies even
more forcibly to the study of Salish diminutive formations, for
here there is a still greater variety of types represented. Avail-
able comparative data are quite scanty, so that only a few
points can here be referred to. The most consistently carried
out difference between plural and diminutive reduplication in
Salish is that in the former the first two consonants of the stem
(though not infrequently only the first) are reduplicated, while
in the latter only the first is reduplicated, never also the second.
At the same time there is a marked tendency, as in so many
Comox examples, for vocalic reduction of the stem. Redupli-
cation with e- vowel seems also characteristic of many forms;
also breaking of stem vowel and umlaut of a to e or e seem
to be found.
Some of the types represented, outside of Comox, are: —
' C. Hill-Tout, Report B.A.A.S., 1902, Ethnological Survey of Canada, p. 20
Eight: Wakashan and Salishan Languages
431
Type X.
Nanaimo
Okanagan
51
IdlEm house
diminutive lUEm^
(based on unredupli-
cated simplex)
HtnotEm little girl*
(H = our x")
Tyye XII .
Nanaimo
wuqas frog
wtweqas^
Type XIX a.
Nanaimo
k'dk'En post
k'dk'k'ETi^
Type XXI a.
Shuswap
pasitlkua lake
papsitlkua*
Thompson River
s-nukoa friend
nunkoa'
Type XXIII.
Tcil'Qeuk
s-tdlo river
s-tdtElo^
Nanaimo
s-td'lo river
S-tdtElo^
Type XXVI a.
Nanaimo s-pak'sm flower s-pdpk'sm^
Comparable perhaps to Comox Type XXX a is: —
Thompson River s-pee'tc s-pdpaats' {-aa-^-d'a-
black bear (e = our e) ?)
Other diminutive types than those listed for Comox un-
doubtedly exist in Salish. Among these is reduplication with
CA- (cf. plural types above), as examples of which may be given: —
Thompson River
Tcil'qeuk
c-mtits deer
IdlEm house
c-mEmeiis^
Isldm? (based on unre-
duplicated form of
simplex; change of
-E- to -a- is perhaps
parallel to that of
Comox -A- to -%-)
' F. Boas, Report B.A.A.S.. 6th Report on N.W. Tribes, p. 129.
' C. Hill-Tout, Report on the Ethnology of the Okandkin of British Columbia, Journal of
th« Royal Anthropological Institute of Great Britain and Ireland, vol. xli, 1911, p. 143.
' F. Boas, ibid.
* Boas, ibid., p. 131.
• C. Hill-Tout, Report B.A.A.S., 1902 Ethnological Survey of Canada, p. 20.
• Boas, ibid., p. 129.
' Boas, Report B.A.A.S., 12th Report on N.W. Tribes, p. 29.
' Boas, ibid.
* Hill-Tout, ibid.
432 VI American Indian Languages 2
52
Similar apparently to Comox type VII (but without diminutive
feature 2a), except for its incomplete reduplication (loss of
reduplicating consonant after s-, cf. plural types above), is: —
Tcil'Qeuk s-mdlt stone diminutive s-eniElet^
Reduplicating with cv-, and with breaking of stem-vowel, is: —
Thompson River kss bad kEkEES-f^ (? = -kA'As)
This tj^pe may well exist in Comox, but not happen to be
represented in the material collected. Such diminutive forms as
Thompson River qEzuzum^ , with interior reduplication, from
qzum "large," and Thompson River spEyiizu^, with change
of -z- to -y-, from spEzuzu "bird," are evidently representatives
of very specialized tj^pes. Neither of these, so far as known,
has a Comox counterpart.
Judging from the analogy of Comox and from a few Interior
Salish forms obtained by the writer, it seems very likely that
glottal stops are frequently employed in Salish as diminutivizing
elements, though this is not apparent from most of the material
that has been published. Examples are: —
Upper Lillooet' s-mulcitc woman sE-m'Em'letc girl
Thompson River* c-muldtc woman c-mu'ni'ldtc
(type XXVIa)
Comparative data on diminutive plurals are too scanty to
enable us to gather much of interest. Some Interior Salish
forms obtained by the writer seem to indicate quite clearly
that in those languages the diminutive plural is not, as in Comox,
a diminutivized plural, but a pluralized diminutive; in other
words, of the tv/o reduplicating sj'Uables, the first contains the
first two consonants of the stem (plural type), the second syllable
the first consonant only (diminutive type). Examples are: —
Upper Lillooet plur. s-muhnu'fdtc dim. sE-tnErn'letc girl
women
dim. plural
SE-mEVmEm'letc
Thompson River c-muhnuldtc dim. c-mu'm'ldtc
women dim. pi. c-mElmu'm'ldtc
' Hill-Tout, ibid.
' Boas, ibid.
' L'ppcr Lillooet forms were obtained in January, 1912, from Chief Jim (Indian name
Aide^i'q.'t). F has here been used to indicate very short obscure vowel of undeiined quality.
* .Some Thompson River forms were obtained in January, 1912, from Chief Tetlenitsa.
Eight: Wakashan and Salishan Languages 433
53
This difference of treatment again indicates that in many
respects each dialectic division of SaHsh has gone its own way
in the use of morphologic features common to Salish generally.
Editorial Note
Originally published as Memoir 63, Anthropological Series 6, Geological
Survey, Department of Mines, Canada. Ottawa (1915).
Pluralizing and diminutive reduplication is nearly universal in Salishan
(Kuipers 1978: 612). Haeberlin's survey of Salishan reduplication (1918), com-
piled shortly after the appearance of Sapir's paper and undoubtedly stimulated
by it, summarizes the Comox data (169-170) and cites similar patterns from a
number of other languages, but does not go much beyond Sapir's own com-
parative remarks. Paul Kroeber (1988) has recently published a study of incep-
tive reduplication in Comox. The reduplicative morphology of several other
Salishan languages has been described in detail; see in particular Hess's study of
"chameleons" in the Snohomish dialect of Lushootseed (1966).
THE RIVAL WHALERS, A NITINAT STORY
(Nootka Text witli Translation and Grammatical Analysis).
The following text was dictated to me in
November, 191 5, by Tom (Sa'ya't'capis
" Stands up high on the beach "), one of the
oldest and best informed men of the T'sica'-
'ath tribe of Nootka. The grammatical analysis
should give a serviceable idea of Nootka
structure, pending the appearance of a full
grammar of the language. The phonetic system
used in this paper is explained in " Phonetic
Transcription of Indian Languages, Report of
Committee of American Anthropological Asso-
ciation " (Smithsonian Miscellaneous Collec-
tions, vol. 66, no. 6, 19 16, particularly pp. 7-
15) ; my u, however, is always open, as in
English full, and varies freely with close 0.
The tale is Nitinat (Nootka dialects south of
Cape Beale, including Makah of Cape Flattery,
Washington), but its linguistic form is Nootka
proper (T'sica'^ath dialect), except for the
names of the rivals, which are unmodified
Nitinat.
TEXT AND INTERLINEAR
TRANSLATION
'o''si.m't'cat} ' ma""ak' -
Now trained secretly for sue- California whale
cess in so and so
''.•'h-'to"p' ' k'waTs'.ts. •♦ wz"'.tcu"ati >
humpbacked K'walisits. Now went to sleep,
whale
hawt"atl '' 'o''stm'tc. ' t'si't'k'pi'atl *^
now train secretly Now lay down in
finished for success in the house on
so and so. (l^is) back
'uvk"t'.9 w-.'k'at}uk''° t'si't'k'p-.tl"
the fire. Now of (him) lie down in the house
was not on (her) back
lo''ts"ma''^ t'l'qwd'atL'' 'na''ts'a"ti''*
woman, now sat in the Now was look-
house ing at
t'aci'''ak"i'5 'a'the'".'^ ya-'i"
the door of (them) be night. There
ka'tlh^'cttt'* t'aci-"i'"^ t'^'h't'sai'.'"
appeared the door head .
qwa^' qw:""tt'q'''^ k'a'yu"min'.^'
Was in quality as is in panther,
quality
yai hi'Lat'^4 si't'a'^^ tb'hn's'.tat"-.''*^
There of (him) tail the head of (him),
was at
'ah^'a-.^«
this.
wik'^9
Was not
qwa* ^'
was in quality
t'hi"ya-x.'" ya"l h-.stsa'q't'so'^'
move quickly. There was provided at
each end with
tb'h't'sit'.'. su'kwitL'' tca'kopokw"t.''
head Took hold of the husband of
(her)
lo-'ts-nu''.'.>^ tci''tC'.ti'> tlup'k'sa"'p'at}."'*'
the woman. Pulled, now caused to
wake up.
tiu'p'k'citL'7 k'waTsas. yai
Woke up K'walisits. There was
t'o'tbh'tsaq't'so''" su'kw<.tl tS'/kuinn'-
Head-at-each-end. Took hold of the iron
436
VI American Indian Languages 2
'ak''-.'''' k'wuTsits t'ci'tcah" 'a'h''^'
of(Him) K'walisits, cut this,
hi's'a'p'^^ 'ah*ko-^> 'a"ap'tsa'atcat"'.'.+'*
caused to bleed this the thigh of (him).
tuxtspa''-<5 k'waTst.ts. t'c'.'tc.tiiia'*^
Jumped over K'waHsits. Cut also
kw'.sa"'s'at''». ■»^ tli'cth.n'.*^ hz's'c-.ti. *'^
the other side of (him) foot. Bled.
kwr'spanoici.'atl^" lu's*i,m''yawi'atl> '
Now began successively now became blood-
(to jump) from side to side, covered
t'o't'-">h*tsaq't'so"'<.'.5' qah^c/'Eti^' t'o't':>h-"-
the Head-at-each-end. Now died the Head-
tsaq't'so"'-,'. ?'u'y'.tci"lc'.'ztl.>^ qw-.'sh^'atl^'
at-each-end. Now began to Now was acting
mal>e medicine. thus that
'a'yMii'k'c'.'atl^'' ma-"ak'.
now began to obtain many California whale,
in hunting
'napxta'"atlqo"W£''.n'5' t'saxci";tiqo' .>**
Now would die imme- whenever now speared,
diately, it is said,
'ah'''a-"atKv£'uVi'' tta'o-"atltta'''"
Now was thus, it is said, now another was also
t'so"ach-'citl*" W£"it'cah's'atl'^'
winter take place. Now was sleeping in (his)
canoe
k'wa'l'S'.ts h'.'l 'anah*'i.s^' t'ca'pats*"*
K'walisits, was at little canoe
'athe''.'*' 'oyo"aratl^> kwi'stsatcith'ca'^^
be night. Now perceived the, as they say, " go
so and so off elsewhither "
'yh-^'atF'' tca"ats--.-b'^~* 'ukta'^?
now the one was Cha'atssib' was named so
and so
.'£'m'ti.na"k'"" tla"o" 'o"o"tah\''
have as name, was another hunt such and such
sea-mammals.
t'i'.'"wi.n''ap'at'''^ kwaTsas 'ani">
Was caused to be laughable K'walisits that
qa"'ya*panatcqa'^ Wz"itcqa' wi'k'arattqa"*
being drifting being sleeping, being now not
aimlessly aware of
kw'.'stsatcittqa''7 'na^'s."** hac'i"'tC'.ti'''
being " going off else- day. Heard about
whither "
k'wal'S'.ts 'an-.'"' t"h.""win''ap'at'qa'.''^"
K'walisits that being caused to be laughable.
ya'k'ci"zt''*' h/'maqst'.'*'^ wi'k'
Of (him) became sore heart, was not
citlsti^'s**' hi."'sasa^^ tlo""ow's.*^
move inland, stayed right on Tlo'owis.
the beach
citistr's'atl'"*^' ni-ti-'na'ath-*^' 'utsatc' 'ii\^^
Now moved Nitinat now wentoff to so
inland people, and so
t'sa"akokw'i'.''' w-.-'napatl''" k'wa'l's-.ts
the stream of Now remained K'walisits
(them).
tto'"ow's hd'atl. W'.k'r't'c.tl^'
Tlo'owis now was at. Became non-existent
qo-"as^^ 'o-Sf.m'tc-i"atl9'
person. Now began to train secretly for
success in so and so
kNva'l's'.ts ma'"ak'.
K'walisits California whale.
hr'ms'o"- j
Beaan to
'uk'c.ti'-*. wa-'lak'''i
move up and down (like a Was bound for
blowing whale).
say£""i''^^ mo''tci'l''' lv.''niso"'uk' ''*'
the far- was for four move up and down
distant ; days whale-fashion
mo- 9'* 'na-'s mo" 'a'thai' ""'. haia's-af'°'
four daylight four be night. Arrived nt
kaxi-'k'.s '"^ 'ukt£-"e '"' nis-'ma'"°^
Kahikis the so-and-so-namcd land.
Eight: Wakashan and Salishan Languages
437
78
t'-.'q'satt '°5 k'wa'l'stts. yai
Now sat down on the beach K'walisits. There
hi'tahnis '°^ t'ca-'pok'" '"7 t'sa"ak'i'. '°**
came down-stream canoeman the river.
tbwi-tci"£t' '°' h'.-'s'it'q' "°
Now was approached where (he) was on the
beach
k'wa'l'3its na\va*'yis"'
K'waUsits be seated on the beach looking
around.
hmii'ha "^ h'.'iyin''-.' '" t'ca"'pokw'r.'.
Went out of the the one at the canoeman.
canoe the bow
si'ma'tsy.n'"^ mi'h'yi.' "5.
(They) had sticking up in the bow spear.
t't'q'stC'.sa'at' "^ k'waTsits 'o'h^'at' ""
Was sat alongside of K'walisits was soed by
on beach
t'ca^'pokw''.'. qatcct"zt' "*'. 'o"'quni'him'a "^
the canoeman. Now was " It is fine
nudged with weather,
(his) elbow.
takha'^'^ 'ah' 'na-'sT '^' wa-"at' '".
is it not ? this the day," was said to (him).
'moqwt'"yu*tl 'V k'wa'l'sas h£'s*iH*tcr,'a'h* '^+
Became speechless K'walisits, became unable
in (his) throat in any way
tS','q'c'.tl'^> qalrkwatC'."at' '^^ qa"yap't'at"t'. '^"
speak, of (him) the limbs of
completely died (him).
tb'k-icitl "** qo'"as''.' ya-'tscai "^
Stood up the person, walked off
ho'a'tsatcai '5" t'ca'patsukw't' '•'. haa'qsal "^
go off back the canoe of (him). Went into
(his) canoe
t'ca'patsukw'-.'. th'h'c.t} ''' ys-f-'"
the canoe of (him). Paddled off yonder,
ta'kh'tatcat ''^ '.^'h^'atl ^^
became far out at sea. Now was the one
t'c'.-n'.-"ath't'ca' ''^ 'oya-'tl ti-'tcatc.'sii '^^
the, as they say. Now was at now come
Ch'ini-person. such and such a to life
time
th'h'C'.'str-.t'q' ■5'^ t*ca-'pokw''.' t'c',*n'.-"ath^'e' '■'°.
when now the canoeman the Ch'ini-
paddled off person.
su'kwal'^ ''.•'nax'mak''r.'.'4' tr.mi"'s'ap' ^'>^.
Took hold of the regalia of rubbed about
(him), on the beach.
hr's '■♦5 'ah-*'.-!" '■»■> mo"'tci"y'.s '■*>.
Was there on the that be for four
beach days on the beach.
h'.'tats--">hnatltla ''^ qwa'yst's'.'k' ■•♦"
Now came out of the woods also wolf.
tiaw'.-tC'.";t'tla '*** k'wa'l's-.ts. tsusk'c'."£tuk'"'t^
Was approached K'walisits. Of (him) was
again urinated on
mu'tsmDhaq' '5° k'wa'l'sas. qwa""ak' '5'
bearskin K'walisits. Of(him)wasin
quality
qwz""tt'q' '>^ kats"o"'m'.n' ''5. 'ah'''a""atl
as is in quality hail. Now was thus
haw'.""ati ho"atsatC'.tl k'waTs-.ts
was now finished, went offback K'walisits
walc'."£tl''4 maht'f"ak"'.' ■>> ha'wi-t}'>^
Now returned the house of Was finished
home (him).
t'so"'.tcly\ t'cukwe"ati ■'" ma""ak'.
winter." Now began to run California whale.
t'sa'xcal tca"ats"',"b'. qa'h-'ba'p' '>^ ma""ak"i
Speared Cha'atssib'. Caused to die the Califor-
nia whale
ya'.a"'n'.t "'.t'q' "''' t'i'.'"wtn''ap'at' k'waTs'.ts.
the one by whom caused to be K'walisits.
had been — ed laughable
t'sa'.xc'.tl 'yo"qwa" ""' k'wa'l's-.ts. t'sa'xcaitia '^'
Speared likewise K'walisits. Speared again
438
VI American Indian Languages 2
79
tca"ats-'.-b' h-.tr-.-'p'tb' '^'. t's.i'xca^ 'yo-qwa'
Cha'atssib', again obtained. Speared likewise
k'wal's'.ts h'.n-'.-'p' yo-'qvva-' mo-'y-.p' '^'
K'walisits, obtained likewise. Obtained four
tca"ats''.'b' ma""ak' 'an'a'^+
Cha'atssib' California whale, was only
kwa'l'S'.ts c-.iVV.-r^'ifii
K'walisits
sutC'.'p '"' ma'"ak'.
obtain five California whale.
t'tu'p'i'tch*C'.tt"^^ 'o'o'i''h^C'.ti'^' ''.-'h'to'p'.
Became summer, began to be in whale,
pursuit of so and so
t'sa'xc'.thla tca"ats"'.'b' ''.''h^to'p'.
Speared again Cha'atssib' whale.
'o"\v'.'a'i'ap' ''"'^ t'sa'xc.tl tv:a"ats''.'b'.
Repeatedly caused so and spear Cha'atssib'.
so to be first
t'saxc'."£thla 'yo'qwa' k'waTsas ''.''h-^to'p "•.'.
Now speared likewise K'walisits the whale.
again
nio""yf.p'athla '^^ '-.•'h'to'p' tca"ats''."b',
Nowobtained four again whale Cha'atssib',
W'.ki'm'.'tC',"atl ■''' h'.n''."'p' tca"ats''."b'.
now became unable obtain Cha'atssib'.
sut'c'.'"y'.p'athla' '7' k'waTs'.ts ''."'h^to'p'-
Now obtained five again K'walisits whale.
su't'ci"'y'.p'tk nia""ak'. hayo'"y'.p'c'.t} ''^
Obtained five also California Had obtained ten
whale.
k'wa'l'sits 'a'tiakwah.'p' '■' tca"ats"'.-b'
K'walisits, obtained eight Cha'atssib'
'oyo"ai'.t "f. ■■•» 'o'h* "'> kwi'stsatcitl.
the one who 'had be the one " go oft else-
perceived so and so wither".
'a'ttakwal'.-p' 'o'h* 'ana k'waTs-.ts
Obtained eight was the one, was only K'walisits
hayu""y'.p' ''^. ha"okw'.'ctt '7' k'wa'l's-.ts
obtain ten. Now took revenge K'walisits
ya!a-'na"'.t'q' '''^ t'}'.""w'.n''ap'at'.
the one by whom had been caused to be laugh-
— ed able,
sa'tckok't'cac ''* .'^'u'yi' '" tca"ats''.-b'.
"Of (him) is sharp medicine Cha'atssib'.
evidently
'a-'q£n'qha'^° 'a'tlakwal-.-p' tca"ats-'.-b'
For what reason is obtain eight Cha'atssib'.
he that
'a'naqa '^' 'p'u'yiwatl ■'- hayo-"y'.p'ath'. '*'
being alone have medicine fall Now I obtained
to (his) lot? ten
si"ya' '^* yaqe''s '^> w-.'k' .'^'u'yiwatl
I I who am not have medicine fall
to (my) lot, "
wa-"att "*^ k'wa'l's'.ts ha"okw'.'£th
now said K'walisits now take revenge.
TRANSLATION
K'walisits was training in secret for success
in hunting California whales and humpbacked
whales. And he went to sleep, having complet-
ed his ritual training. He lay down on his back
near the fire, but his wife did not lie down on
her back but sat up. She was looking at their
door at night. There at the door appeared a
head, looking just like a panther's. There was
his head with a tail attached to it, that was
what it was like. He did not move briskly.
There he was with a head at each end.
The woman took quick hold of her husband,
and pulled at him, endeavouring to wake him.
K'walisits woke up. There was the superna-
tural being known as Head-at-each-end.
K'walisits seized his iron knife and cut here,
making bleed this thigh of his own. K'walisits
jumped over him. He made a cut in his other
leg. It bled. And so he continued, jumping
from one side to the other, until the Head-at-
each-end became all covered with blood and
Eight: Wakashan and Salishan Languages
439
80
died '. K'walisits proceeded to make medicine
.of him. And this is how he began to capture
many Cahfornia whales when he went out to
sea. Whenever lie speared, the}^ say, they
would die at once.
And then there came another winter. K'walis-
its was sleeping in his canoe, was in a little
canoe through the night. Now there was one
bearing the name of Cha'ats^ib', another
whaler, who saw the thing they call " going
off to another place " ^
K'walisits was laughed at because he was
drifting about aimlessly, asleep, and because he
was not aware that it was the season of
" going off to another place ". K'walisits heard
about how he was being laughed at. His heart
grew sore and he did not move inland with his
people for the drying of salmon but stayed
right at Tlo'owis. The Nitinats moved inland,
they went off to their river, but K'walisits
remained behind there at Tlo'owis. All the
people had gone.
And then K'walisits began to train secretly
for success in hunting California whales. He
began to imitate the movements, up and down,
of a blowing whale, while on his way to a far-
distant place ; four days he made motions as
of a whale, four spans of daylight and four of
1. Tlic onlvjW.u' toliill this bciiig is to sprinkle luiman
blood over iiim.
2. The Nootka Indians believe that twice during tlie
year, at unknown dates, a big liikev^'arni tide conies in at
night and shifts everything about in the village, houses
and all. After a short time everything is restored to its
proper place. If one happens to be awake at such a time,
he can train (Jo-S'-iii'lc) for anvthiug he pleases, such as
wealth or success in whaling, and be sure of attaining
his desire. Should he step into the water, however, he
becomes paralyzed for life. He must step into a canoe or
move back to higher ground. Signs of the approach of
the mysterious " shift " are the birds starting in to sing
and the mice running through the house. The people
become very drowsy just before the " shift ", so that
few are fortunate enough to be awake during the spell
and make " medicine " of it. K'walisits too was caught
napping, great wlialer though he was. His rival was
more fortunate.
night. He came to the land which is called
Kahikis.
K'walisits sat down upon the beach. Yonder
on the river was a canoe-party coming down-
stream. They came near to where K'walisits
w^as, seated on the beach and looking around.
The one of the canoe-party that was at the
bow came out of the canoe. They had a spear
sticking up in the bow. Someone sat down
alongside of K'walisits — it was the canoeman,
who nudged him with his elbow. " It is fine
weather today, what do you think ? " he said
to him, but K'walisits' voice stuck in his throat.
He became unable to speak and his limbs
became lifeless. The person stood up and walk-
ed back to his canoe. He went into his canoe,
paddled off way yonder, until he was far out
at sea. Now this one was he whom they call
Ch'ini-person.
It was when the canoeman, the Ch'ini-person,
set off that K'walisits came to life. He took his
bearskin robe and rubbed it about on the beach ' .
For four days he stayed at that place. And then
also a wolf came out of the woods. He came
near to K'walisits and urinated upon his
bearskin, and his urine was like hail. And
then K'walisits was done and started back ; he
returned to his home.
When winter was over, the California
whales began to run ^ Cha'atssib' threw his
harpoon and killed the California whale, he
who had laughed at K'walisits. K'walisits too
speared a whale. Again Cha'atssib' threw his
harpoon and again he got his quarry. K'walis-
1. Wherever the (Ch'ini-person had been in direct con-
tact with tlie ground was medicine. In most Nootka
tales, legendary or modern, of the acquirement of power,
the seeker carries away with him an object granted bv
the supernatural being, some part or effluvium of his
body, or some tangible evidence of direct contact with
his bod\ .
2. They migrate nortii with the coming of warm
weather and touch ar various points on the west coast ot
Vancouver Island at fairly regularly recurring dates.
440
VI American Indian Languages 2
8i
its too speared a whale, he too got his quarry.
Cha'atssib' secured four California whales, but
it was K'walisits alone who got five of them.
The summer came and they started to hunt
humpbacked whales. Once more Cha'atssib'
threw his harpoon at a humpbacked whale.
K'walisits too speared a whale. Each time he
allowed Cha'atssib' to be the first to throw his
harpoon. And then, once more, K'walisits too
speared his whale. This time too Cha'atssib
secured four whales, but then he proved unable
to get another. And K'walisits, once more,
obtained five humpbacked whales. And he had
obtained five California whales.
K'walisits had ten ; Cha'atssib', the one who
had seen the " going off to another place ",
had eight. He was the one who got eight
whales, but it was K'walisits alone who obtain-
ed ten of them. Now K'walisits had his revenge
on him by whom he had been laughed at.
" I have found out that his medicine is sharp.
Why has Cha'atssib' obtained but eight whales,
seeing that he alone has had real medicine fall
to his lot? And I have obtained ten, I who
have had no medicine come to me ", said
K'walisits. He had his revenge.
GRAMMATICAL ANALYSIS
I. 'o'-sun't'c-ati consists of radical element
'o'-, derivative stem-suffix -son'tc-, and word-
suffix -ati, 'o'-, lengthened from 'o- because of
following Simple-, is exceedingly common in
Nootkain both noun and verb forms; it appears
as 'o-, 'o*-, and as reduplicated 'o'o-, 'o"'o-, 'o'o'-,
and 'o''o'-, each of these forms being determined
by the following element. It cannot be used as
an independent element but needs always to
be followed by a stem-suffix to specify its
meaning. Its function is relational; it indicates
the person, object, or activity required to
limit the following element and frequently, as
here, anticipates a word of specific content
(here nta''ak' and 'clfto'p'). The nature of the
relation between 'o- and the suffixed element is
implicit in the latter ; thus, it may be con-
strued objectively, as here, subjectively, geni-
tively, causally, and in other ways. A form in
'o- always implies that the psychological inte-
rest centers in the person or object or idea
with which the logically significant concept is
connected, not in this concept itself. It may be
translated a certain (pi-rson, thing), such
AND SUCH A (person, thing), SO AND SO. Verbs
in 'o- should be conceived of as answering
questions of the type '' IFho is it that. . . ? ''
or " What did he. . . ? " in contrast to ques-
tions of the type " What did he do ? " Thus,
'o'swi'tc ma''a¥ means not so much to train
SECRETLY FOR SUCCESS IN HUNTING CALIEORNIA
WHALES as WHAT ONE TRAINS SECRETLY FOR IS
CALIFORNIA WHALES, RATHER THAN SOMETHING
ELSE. California whales, not secret training,
is the psychological predicate and is anticipated
by a place-filling 'o- ; one can also say, more
synthetically, ma'^ah'san'tc and \'h"lo-p's<.m'tc.
There are practically as many verbs and nouns
in 'o- as there are derivative suffixes and com-
binations of derivative suffixes to append to it,
in other words, many hundreds. If the signifi-
cant content of an 'o- word, that is, the idea
expressed by the derivative suffix, is the true
center of interest, 'o- is replaced oy hin(a)-,
h'.l(ay, or an entirely different word is used.
A few examples of 'o- words, with parallel and
contrastive forms, will make these remarks
clearer : 'o-i's to eat so and so, like t^sis'.-"is
to E.^T meat, contrast ha'w- to eat (as such) j
'o'o'-/«/ (reduplicated) TO dream of so and so,
WHAT one drea.ms OF IS..., like fuiul'c-itul to
DREAM OF A WOMAN, Contrast po'huds- TO
DREAM (as such) ; 'j-ha-' to buy a certain
THING, like futc-ha'' to buy a woman, to
MARRY, contrast mahv- to buy ; 'o-so'tf the
one who dies is..., like ivik'-so'ti nobody dies,
******
Eight: Wakashan and Salishan Languages
441
82
contrast ^Z^/;- to die ; 'o-'ycha' thk cause of
one's dying is..., like t'sax-'yrha' to die from
BEING SPEARED ; 'o-yi' ' TO GIVE A CERTAIN THING,
Hkc nio'-yc' TO give four things, contrast
hi.n-K' TO give (as such); 'o-iiwt' so and so's
SON, as verb the one of whom (he) is son...,
contrast /'fl';/fl' child.
-sun'tc, derivative verbifying suffix following
stem form, to undergo secret ritual train-
ing IN ORDER to gain success IN...; it lengthens
stem vowel if short. (Derivative suffixes leave
the stem vowel unaffected, lengthen it if short,
shorten it if long, or reduplicate the stem
according to varying quantitative patterns).
Cf. further 'titc-s\m'lc to train for long
life {trie TO BE alive) ; ha'ivii-sun'tc to train
FOR WEALTH (ha lud CHIEF, ha\ud-mis wealth);
tO'k'-SWl'tC TO TRAIN FOR SUCCESS IN HUNTING
SEA-LIONS (tok-o'k'" sea-lion) ; hi'tc-sr.m'fc to
TRAIN FOR SUCZESS IN FISHING BY TORCHLIGHT
(hitc-ina' torch). There are probably several
hundred such verbifying suffixed elements in
Nootka, many of them very specific in content,
which differ from primary verb-stems not
only in their position but in that they are
always construed, according to an implicit
syntactic relation, with a preceding denomina-
tive term (which may be a " noun " or a
" verb "stem). Composition of primary stems
is as good as unknown, 'o'-suii'tc is durative in
aspect (all verbs have durative and momenta-
neous, or inceptive, aspects, most have also at
least one iterative aspect, and m.uiy have still
other aspects).
-all, word-sutlix (i. c., attached to complete
word, not stem) of colorless content. It may
be translated now, then, and then, so ; it
seems to indicate state or activitv at a given
moment of time and implies that the verb
form is finite. Its use is not obligatory, however.
The 'of -'<!// combines with precedingsiop (/>, /,
k, hu, q, (}w,ts, tc, if) mto glottalized stop (p",
', k\ k'lr, ', '., fs, t'cy I'i respectively ; origi-
nal q' and q'lv have become .', a peculiar glot-
tal stop of strangulated articulation and velar
resonance) ; other consonants remain unaffect-
ed.
'o-sivi'i'cati is absolute (of undetermined
tense-mode) in form. Absolutes, withor with-
out -'all, are freely used in narrative ; without
-all they are frequently used as complemen-
tary infinitives (see note 7). Absolutes with
3d personal subject have no personal suffix
(contrast 3d person indicative forms : present
'o-sun'fcati-nia' ; preterital 'o'sun't'catf-U-a' ;
future 'o'sitn't'c-a'q'li-ma'). The personal end-
ings of absolute (or narrative) forms differ from
those of indicative forms :
Sing. I. 'o'smi'Uatf-S'.
Absolute Present Indicative
'o'swi'fcati-ah
2. 'o'sun'l'cati-.uik' 'o'sun'l'calf-tits
3. 'o'suii'fcali 'o'S'.nVfcati-ma'
Plur. I. \rswi'fcat{-n\' 'o's<jn'l'cat{-.n'
2. \r Sim' t' calf -so-' 'o-suii't'catt-i'Uso''
3. 'o'sun'l'catf(''at) 'o's<jit'l'catf-)iia{-'tif)
A third set of personal endings, used in
various modal and subordinate forms and in
possessives of nouns, is clearly related to the
absolute series :
Interrogative Possessive
Sing. i.\ys'.m't'catf-ha-s t'a'tia'k'-qa-s
MY child
2. \rs\ui'Ccalf-}M-k Caiurk'-d'-qa-k'
3. 'o-S'.niifcalf-lm' Cana'lc^-'.')
Plur. \.\)SKin'lcatf-hi-n' t'ana1;'-q\.-n'
2. \rs<.)n'l\atf-f_m-
so'' Ca'na'k' -^'.I'-q'-io' '
3. 'o'SKin't'i-alf-lui
(-'rt/) t\ina'k\-\-af).
2. Noun of uncertain analysis, -ak' is pro-
bably durative intransitive {d. qah-ak' to be
dead; 'nio-'ak to burn [intr.J) or, what at last
442
VI American Indian Languages 2
83
analysis amounts to the same thing, absolutive
noun suffix (d. I'sa-'ak' to flow, hence
STREAM ; t'ca-ak' water ; tca-'ak' island), but
is now petrified. iiur~, which does not occur
alone, may be an old verb-stem (cf. Kwakiutl
7na' TO CRAWL, TO SWIM ?) that has become
obsolete ; tna-'ak' originally to swim about,
SLOWLY SWIMMING ANIMAL ?
3. ''.-If-to-p', literally big thing, big variety
(OE animal). \-h", durative intransitive big,
TO BE BIG ; inceptive \'iv-a-tciti TO get big,
GROW UP (Nootka h \ w <. Wakashan xw : w ;
" of h" indicates voiceless rt-timbre of h, which
colors and lowers following high vowels, e. g.
his- < h"\s-, }ps- < }fus-). -to'p' (after conso-
nants), -cto-p' (after vowels), shortened to
-{c)liip under appropriate rhythmic circum-
stances, noun forming suffix added to verb
stems, THING, KIND, CLASS (c. g. Jlh'-tO'p
diving kind, SEA-MAMMAL, kwiS'tO'p' DIFFERENT
KIND, SUPERNATURAL, ONE WHOSE NATURE IS
ABNORMAL, Sd'CtO'p' CRAWL-KIND, ANIMAL WALK-
ING ON ALL FOURS, 'o-CtO'p SUCH AND SUCH A
KIND, kiu\ktu>.s-tup'-sa-p' [red.] to cause various
THINGS to be of A DIFFERENT SORT, TRANSFOR-
MER, -lo'p' < -to'in-, cf. kui'.sio'm-ah i am a
DIFFERENT KIND OF BEING, AM SURERNATURAL.
ma-'ak' 'vh"krp' are object of 'o-S'.ni'Uati,
which they follow, as regularly ; or, perhaps
more accurately, they may be looked upon as
merely appositional to 'o'-. Like all nouns not
specifically distributive or plural in form, they
are indeterminate in number. " And " is gene-
rally omitted in Nootka ; \c, a conjunctive
particle, may be placed between the two
nouns.
4. A Nitinat name. / does not occur in
Nootka except in songs for n. K'lual'S'.ls is the
subject oi'o'sun'tcati ; verb, object, subject —
this is the most common Nootka order.
5. ¥rom vji':tc-iiti, momentaneous form cor-
responding to durative wiitc to sleep, -f -'ati.
Momentaneous -nil is uncommon for primary
verbs (cf. also durative -ap</ standing, up in
THE AIR : momentaneous -apuli ; dur. -tcict'
on the surface of the water : mom. -Icic-
tiili ; dur. -'aq'li inside : mom. -aq'stuli^.
Nearly all momentaneous or inceptive forms
end in //. This -// disappears before -'all, pas-
sive (or possessive) -'rt/', imperative -'i', finalis
-a'-, irrealis -V///' (see note 124), future -<.k' ;
e. g. -sa-ali (mom.) on the beach <C -sati -j-
-'ali, -o'-at' (mom. passive) on the face
<C -o'tl (cf. dur. -0'/) + - (it\ -'t'l' (mom.
imperative) on the ground < -\ti (cf. dur.
-'as) + -'i', -C'/a'-h" in order that i may...
■< -Cili- (see note 18), -c.-ik-ah i shall...
■<-C',//-. -'rt// (see note i). luiUcu'atl is narrative
absolute, 3d personal subject, as in note i ; this
is true of all verb forms commented on in
this text, unless otherwise explained.
6. From haioi'-tf, momentaneous form, to
cease. -//, momentaneous suffix, drops before
-ali (see note 5) ; comparatively few primary
verbs add simple -// in their momentaneous
form (cf. also 'a'k^o'-ti to borrow ; na'o'-ti to
have for e.\ting). Certain verbs, like hazvc-ti,
are basically momentaneous because of their
radical significance ; they can form only a
secondary quasi-durative by making an incep-
tive, more properly graduative, out of the
momentaneous by lengthening its stem vowel,
e. g. ha'ivi-tf TO begin to cease, to be finish-
ing (ct. momentaneous wai-citi to be gone
home, to return home : graduative lua'i-citi
to be going home, to be on the point of
RETURN home; moui. tiawoi'ti to approach:
graduative tia-uvi'lf to be approaching, cf.
static durative llaiua'' to be near), -ati, see
note I.
7. See note 1. Depends as complementary
infinitive upon preceding haw\'atl. This use
of an absolute verb form to complete the
meaning of a preceding finite form is exceed-
ingly common in Nootka. Note that o"- does
not need to be specified by a following noun.
I
Eight: Wakashan and Salishan Languages
443
84
8. From t'sit'k'-pdf, momentaneous form,
-\--'ati ; -ti lost as in note 5. t'sit'k-, verb-stem
TO LIE (in bed) on one's BACK (aspirated stops
are of purely secondary origin in Nootka,
being developed from un aspirated surds when
they stand at the end of a syllable or immedia-
tely before another consonant at the end of a
syllable ; to the two Kwakiutl series d [inter-
mediate] : /{aspirated surd] corresponds a single
Nootka series/, /'). -/)•.//, momentaneous form of
-//, locative suffix in the house. Examples of
dur. -// : mom. -pdi are wt\tc-ii to sleep in
BED : lUi:tc-pUl TO GO TO BED ; t\qw-\.l TO BE
SEATED ON THE FLOOR : t''.q'-p\tf TO SIT DOWN
ON THE FLOOR ; 'liatC-H TO BE LOOKING (WHEN
INSIDE THE house) : 'natC-p'.tl TO LOOK DOWN
(when inside). Every local and body-part
suffix has a durative and a momentaneous form
in -//, further an iterative with lengthened stem-
vowel, lengthened suffix vowel, and change of
momentaneous -// to -/, e. g. t\'q'-pi'i to sit
DOWN OM the floor SEVERAL TIMES, -'flf//, Sec
note I.
9. \n'k' , noun or durative intransitive, fire,
TO BE BURNING (saidof fire, not burning object).
From original *'an-aV ; -a¥ , durative intran-
sitive suffix, see note 2. The common Nootka
groups in' and iui\ in which the ' represents a
murmured i-vowel, go back to fuller forms of
type a (or i, //) + n or ;// + a (or ■., u), in
which the second vowel is unaccented (e.g.
-all momentaneous sound : -\in' continuous
sound < -.'"r.«' < *-.ana, in Wakashan terms
*-qa-ti : *-qa'-la ; 'ana'h'-is small : reduplicat-
ed W'n*h'^-\s SEVERALLY SMALL < *'a'a'nah-).
-\ (-' is merely a breath-release after all final
vovv'els ; it is not heard if the word is pronoun-
ced in close contact with the following word),
suffixed definite article, often used as nomina-
lizing element. Properly speaking, all " nouns"
are indeterminately such, being formally iden-
tical with durative intransitives (e. g. qo''as a
PERSON, TO BE A person) Until noniinalized by
-C or an equivalent element. Syntactically,
\u'k'''.' is objectively related to the preceding
verb ; the difference between a direct object
and an indirect object or local phrase (at the
fire) does not exist for Nootka, because the
indirect or local relation is generally expressed
by a suffixed element in the verb or is other-
wise absorbed in the verb ; the fire here ampli-
fies the more general local idea of in the house
conveyed by -pi-(ti), no specific rendering of
our AT or near being therefore necessary.
10. u'lk-, durative intransitive, to be not ;
wik' not, no ! is really verbal in form, w.k'
consists of archaic stem lui- and durative intran-
sitive -k', d. -ak', -ak' (notes 2, 9) ; most
Nootka derivatives of to be not are based on
Will- (e. g. luiFnit' to be not-stocked, to have
NO FISH RUNNING UP IT ; IVik'-faq'yu' TO HAVE
HAD NO VISITATION, TO BE UNINITIATED), but
there are also a number of more archaic forma-
tions based on wi- (e. g. wi-ma-h'ti to be
UNABLE ; lui-.aq'ti a man is angry), -all,
see note. i. -uk' (after consonants), -ak'
(after vowels ; often contracts with prece-
ding vowel to -a'k', -ak'), possessive word-
suffix indicating that the subject of the verb
is the owner of the following alienable noun
(here woman) : his (wife) did not, he had (a
wife) WHO DIDNOT, W/d''fl//M^'ti;df/» nowmine isnot,
NOW I HAVE WHAT IS NOT ; cf. nOtCS I 5, 24, 26.
11. See note 8. Complementary infinitive
dependent on preceding verb. Negatived state-
ments are always expressed by treating the
negative as the main verb and having the verb
proper follow as an infinitive, e.g.ivikila himn'
HE-WAS-NOT COME, HE DID NOT COME ; U'ik\
hiq'citl DO NOT (imperative) to-^speak! do not
speak!
12. Absolute form of noun, assimilated from
io'lc-sma. io-tc-, lengthened from stem iutc-
WOMAN ; examples of derivatives are iut'c-i'h" to
want, BE after a WOMAN, iiitc'-na'k' -citl to
BEGIN TO HAVE A WOMAN, MARRY A WOMAN, lltlc-
444
VI American Indian Languages 2
85
tci' TO LIVE AT oxe's wife's HOME, fo'tc-azu'/qc
TO CALL FOR A WOMAN, fufulc-atah (red.) TO GO
AND SLEEP WITH A WOMAN IN ANOTHER HOUSE-
-sma\ -s'ma\ absolutive stem-suffix lengthening
stem vowel ; not freely used as derivative ele-
ment, probably compounded of absolutive ele-
ments -S-, -s'- (c{. absolutive suffixes -s-yi',
■s-'yup' , -s-yin', -s'-fnnni') and -ma' (cf. li'tc'-
ma' REDHEADED WOODPECKER), presumably
reduced from older *-sa- as indicated by irregu-
lar plural io'ts-sa-mch" women. Object of
tuik'atfuk\ though logically subject of iv.k'atf
i's'.t'k'pdi.
13. t'<.qw-, verb stem to sit ; cannot be used
without following local suffix, e. g. t'e'.-as (<i
*t'\q\i'-as) TO SIT ON the ground, Ciqw-a's to
BE seated on, i'e\-ih"ta to sit at the end,
t\q'-s'ato--'asTo siton thegroundatthe door.
-t/, durative local suffix on the floor, in the
house ; for corresponding momentaneous form
see note i.
14. Assimilated from 'na'lc-sa'ti{c{. note 12).
'invic-, lengthened from 'natc-, verb-stem to
look ; other derivatives are 'natc-fso' to look
into (a barrel), 'natc-i'itf to look into the
house, 'natc-a'yii to be looking up in the air,
'natc-mai-apC to look all around, 'nat'c-
aq'ff-a'a' to be looking into the fire,
'na'tc-uk' to look for, 'natc-n'af to see, 'tm-
'na't'c-a'i {red.) to watch, 'na'ts-sa' to look
at, watch, see is durative ; as momentaneous
is used 'natc-ii'af (-n'at, -yiiai to get sight
of, perceive), -sa' , durative suffix lengthening
stem-vowel, not freely used, cf. durative -a' ;
perhaps identical with stem-lengthening -sa'
very, just, -most, too. -sa-il is contracted from
-sa-'alf (^-'atf contracts with certain preceding
vowels to -a'tf, -ati, according to rhytiimic cir-
cumstances ; e. g. -ape standing + -'ati >
-apati, -'alo' into the water -f -ati > -'alaii,
durative -a' -f -aii > -a'ti, -ati). -'ati, see
note I .
13. t'aci-', noun trail, doorway ; absolutive
in -/-'jStem t'ac- (e. g. t'ac-kuin' with a trail
IN the center), -'ak' (after vowels ; often con-
tracts with preceding vowel to -a'k\ -ak'), -uk'
(after consonants), possessive word-suffix for
alienable nouns ; for inalienable possession see
note 26; for possessive paradigm, see note i.
Their is ordinarily not distinguished from his,
her. its. -'i', nominalizing element, see note 9 ;
faci''\ik' alone would mean to be one's door.
t\ici'"ak'\ is object oVna'tsiVtt.
16. 'alh"-, verb-stem to be night, 'ntl/'-citi
night comes, 'ath-e-' is durative in aspect ; verb
stems endingin /'take -e'', -e' as durative suffix
instead of normal -a'', -n' . Like other absolute
durative forms, \ilhc'' can be used adverbially ;
properly speaking, it is a complementary infini-
tive following 'na'ts'a'li.
17. Demonstrative pronoun and advert, that,
there ; properly a verb to be there. It is bas-
ed on simpler demonstrative 3'^', ya that; for
-/, probably an old local suffix no longer freely
employed, cf. /;// (note 24).
18. katih"-, verb-stem to appear, come into
view, -ff.//, momentaneous (or inceptive) suffix.
Most verbs form their momentaneous aspec-
by adding -c<.-li to the stem if it ends in a cont
sonant, -ta-ti if it ends in a vowel, often -kw\.-
ti (see note 32) if it ends in w, o' . Complemen-
tary infinitive depending on ya'i : there-was
to-appear ; ya'isi katilfcdi there i appeared,
not ya-i katiJfatisC.
19. Seenote 1 5. Object ofyfl"/^a////Vi//,/W''t'
amplifying the local idea in ya'i ; cf. note 9.
20. Absolutive. Noun-stem t'oh"- head (e.g.
t'j}f-n<.' cod-head that drifts to shore, t'o''w-
rs to eat a fish-head). -fsitC, absolutive suf-
fix, not otherwise found. Subject of ya'i
katih"ati.
21. Properly qwa'', verb stem and durative
absolute to be in quality, to be like.
22. Umlauted from *qzva'-\t'q' ; a and a'
immediately followed by i or r are umlauted
to open z and z' (these vowels are felt as dis-
Eight: Wakashan and Salishan Languages
445
86
tinct from secondary e, i and e' , '' that are
merely lowered from i, c because of preceding
or following velar consonant), qwa'-, see note
21. -\t'q' third person relative or subordinate,
indicating various subordinating relations, such
as comparison, time, place, relative clause (cf.
notes no, 139, 159) ; the precise nature of
the subordination depends on the verb. -'1/ q'
may be considered as a nominalizedform, paral-
rel to -'i' (see note 9), of the subordinate -qa-
series (see notes i, 7^1).
23. Absolutive form of noun. Stem k'ayntn-,
k'ayiip'- (intervocalic -m-, -n- become stopped
to -/)', -t' at the end of a syllable), e. g.
k'ayup'-q-inakTOimT.\TE A panther in a dance.
-in', absolutive noun suffix ; other examples are
bttC->.n' SMALL CLAM, ho'p-m' SALMON TROUT,
mxl-m' SALMON-EGG, Miv-in' GRISTLE, tca'skw-
i«' BACKBONE, hd-f sn' t-'.n' SEA. Subjcct of subor-
dinate clause qivi'\t'q' .
24. Jhf, verb-stem and durative absolute to
BE HERE, to BE THERE, TO BE AT ; probably
consists of old demonstrative stem /;',- {d. bin-,
h'.t-, notes 1,101 \h.s- to be at such and such
A place) -f -/-, petrified local element (see
note 17). -'at' (often contracts with preceding
vowel to-rt*/', -al' ; affects preceding consonants
like -'fl//, d. note i), possessive word-suffix
referring to possession of inalienable noun,
nearly always body-part (cf. note 10 for cor-
responding alienable possessive suffix), hfat'
is complementary infinitive depending on ya'i :
there (he) was with his (tail) at, there he
(was) having his at.
25. Absolutive noun, tail (or mammal).
Stem probably 5i/'- ; -a' absolutive noun suffix,
identical with durative intransitive -a' (other
nouns in -a' are faha' ghost, qama' trap,
pa-liprftfa' substance for face paint, nat'ca'
tail [of fish], kap't'a' pointed stick). Object
of Ihfat', th :)ugh logically subject of /;-./ ; d-
note 12.
26. Contracted from t'oh^l'sdr. head (see
note 20) -\- -at'-V. -'at', identical with verb-
suffix -'a/' (see note 24), inalienable possessive
suffix appended to nouns, chiefly bodv-part
nouns ; for corresponding verbal and nominal
alienable possessive suffix, see notes 10, 15.
With pronominal elements -'at' combines
exactly as does -uk' , -ak' (see note i) :
Sing. i.t'ok't's\tat' my head Plur. i. t'ojft'sdat'
-qa-5 -qi-n'
2. t'jh"t's'.tat' 2. I'oh^i'sdat'
-d'-qa-k' -'>.t'-q'-so''
3. t'oh^t'sdat' 3. t'oJfl'sKtai'
t'jJft's'Jat"r.' is local object of /?>./ (syntax as in
notes 9, 19), while s\t'a' is direct object of -Vr/'
in hifat' to have. . . being at. . .
27. See note 21.
28. Demonstrative pronoun (this), adverb
(thus), or verb (to be thus), used as general
demonstrative ('al/'ko' ' is more specifically this,
ya-' that); syntactically a complementary infi-
nitive defining the preceding verb, which gives
the nature of the relation, such as time, place,
or manner, as here. 'ah"'a' is compounded of
demonstrative stem 'aJf this, that, which may
occur alone, and -a' , probably petrified demons-
trative or local element.
29. See note 10.
30. Durative form of verb. No etymological
analysis suggests itself.
31. Assimilated from hc-tsaq't'so' ; cf. notes
12, 14. h\c-, verb-stem to bf. all, to be both ;
other derivatives are hic-im'f to be assembled,
h'X-sa'tso' TO BE everywhere, -tsaq'fso' , local
suffix AT THE END (see also note 38). Many
verbs with local or body-part suffixes are to be
interpreted as "bahuvrihi" compounds, i. e.
the radical element expresses a concept which
is possessed by the subject ; e. g. fwk'^-t'so' to
BE SOUND-INSIDED, TO HAVE MUSIC INSIDE, PHO-
NOGRAPH, 'ayaqs to have much (game) in the
CANOE. The object of the verb is head.
446
V/ American Indian Languages 2
87
32. SU-, verb-stem to hold, get hold of ;
its aspects are Jurative so' ' to hold (other mono-
syllabic duratives with lengthened stem-vowel
are /<:/•' TO pull, drag, t'fo'' to remember, qiun''
to be in Q.UAL1TY), momentaneous su-kwjf to
GET HOLD OF, graduative scr-huUJ to begin to
GET HOLD OF, TO BE GbTTING HOLD OF, dumtive-
iterative so'tiso'ya' to hold time and again,
SUtisok' TO GET HOLD OF TIME AND AGAIN.
-kwitf, momentaneous sufhx (cf. note 18),
etymologically identical with postvocalic -tcuf
(cf. -tci' AT : 'o-hvi to be at such and such a
place) ; Nootka tc: hw (after u) goes back to
Wakashan ^':/«c; (^d. Kwakiutl change of Oi;;- to
ogw-).
33. tcakop' MALE, HUSBAND, irregular absolu-
tive to which corresponds as stem Irapxw- (e.g.
tcapx-nak'-f'.tf to marry a man), -o^w- (final
form-o^'', -ok'"; ^sounds are labialized aftero),
alienable possessive suffix after consonants (cf.
notes 10, 15) ; -'i, see notes 9, 15. Object of
suhviti. Note that -okvj-'i refers to possession
by the subject ; if HER husband had referred to
another woman ,than the subject, siikivitftcp'
tcakop'^i took-hold-of-another's the-husband
would have had to be used.
34. Umlauted from *lo'ts'ma-\\ cf. note 22.
See notes 12, 9. Subject^of sukwiti.
35. tci'-, verb-stem to pull ; durative tci'',
momentaneous tci'-tc.ti. -Icdi, post-vocalic
form of -citi, momentaneous suffix, see
note 18.
36. tlup'k- or ti'.m'k-, verb-stem to be awake;
see note 37. -sa'p' (alternates for rhythmical
reasons with -sap'), causative of momentaneous
-all, while causative -'yap' corresponds to -tcUi
(other examples of 'Sa'p\ -sap' are qah"--sa'p'
TO KILL : qah"~Cili to die ; 'utcq' -sa'p' to cause
TO BE A FOG ; hh''-sa'p' TO MAKE A HOLE ;
hamat'-sap' to cause to be known, to find
out) ; alternates with -sa'm-, -sam- (e. g. tiup'k'-
sa'in-ah i cause to wake. up). Every intran-
sitive aspect has its corresponding causative in
-p' , -in- ; momentaneous -tl always drops in
causative forms (e. g. -pdi in the house,
mom. : caus. -pitap' ; -as\ti on, mom. : caus.
-asip'). -ati, see note i. him is understood as
object ; third personal subjective and objective
pronominal ideas are not specifically expressed
in Nootka.
37. Momentaneous intransitive, see notes 36,
18 Aspects recorded of tlup'k- : dur. tfnp'k-a''
or tiim'k-a'' to be awake, mom. tlup'k' -ati to
WAKE UP, graduative llo'p'k'-all or thm'k'-c<ji
to be waking UP, durative-iterative //t>-/)7v''-n'-/
TO be waking up time and again, momenta-
neous-iterative tlup'k' tliip'k-c or thm'k'lhm'k-c
to keep waking up by fits and starts.
38. Cot'olf-, reduplicated from f^h"- head,
.see note 20 ; reduplication expresses distribu-
tion, HEAD HERE AND THERE. -tSUq't'so', SeC
note 31. This word is a "bahuvrihi" : having
A HKAD AT EACH END, d. note 3 I.
39. tsikwun', noun iron ; borrowed from
Chinook Jargon, -'ak -'<.', see note 15.
40. /V/-, verb-stem 10 cut (e. g. t'ci-maht'
TO CUT A BODY, tC\~h"tiVk TO BE CUT APART, fci'-
mcf.' mussel-shell knife) ; dur. t'c'rya to be
cutting, mom. Cci-tcitl, cf. note 35.
41. General demonstrative, object of /V/Zr-.// ;
see note 28.
42. = his-sa'p . his-, verb-stem to bleed and
noun-stem blood (e.g. his-m\s blood, hihcs-sitl
[red.] TO be bloody-eyed, hey-i's to drink
one's blood), -sa'p', see note 36 ; his-c.tl to
BLEED, mom.
43. Demonstrative pronoun, object of /i^^v/'/)'.
Compounded of \ih" (see notes 41 , 28) and -ko',
not otherwise found.
-14. Contracted from 'a'ap'-ts\.fata-al'-\', d.
note 26. 'aap'-, reduplicated from 'ap'-, 'am-,
noun-stem of general locality part, body-part
(e. 'g. 'ap'-qe' summit, ^ap-p'.q' tl-il on a mat
NEAR THE FIRE TOWARDS THE DOOR, 'ailL-ashaul
chest, \ip"-'win' WAIST, 'am-ak'th.' butt end);
suffixes indicating body-parts occurring in pairs
I
Eight: Wakashan and Salishan Languages
447
88
generally reduplicate preceding stem even if only
one of two is actually referred to (c. g. \{ap'-
p\qa' KNEE, 'a'''ap-s"'iuin'i arm-pit, 'a am- as
CHEEK, 'aa'ni-anid shin ; from other stems,
e. g. yaya'k'-'nuk'" to be sore-handed, yaya'k'tv-
wi'i TO be sore-eared, lotop'k'-atsoh'^ to be
BLACK ON THE SOLES, 'nilts'}lO'q'"-Sui TO HAVE A
BOIL ON THE eye). -ts-fatcC, body-part suffix
THIGH ; compounded of -tS'.t- side, apparently
not found uncompounded (other examples are
'fPq'"-tSlf<.m' TO HAVE A BOIL ON THE SIDE [OF
THE head], hopai-tsu'ak'lii to be moon-si de-
REARED, TO HAVE A CRESCENT PAINTED ON THE
UPPER THIGH AND RUMP), and -a'tci, -atCi PRI-
VATE PARTS, VULVA (c. g. t'h'.a'tc' TO SHOOT AT
THE PRIVATE PARTS ; compounded, e. g., in
'tiJq' "-C sa'.alcC to have a boil below the navel,
yaya'k'-si'o'w-atcr to be sore on the inner
PART OF THE thigh), lieuce properly on the
SIDE OF THE PRIVATE PARTS. -at'-C see notc 26.
45. tuxw-, verb-stem to jump (cf., further,
tUXW-rtCltl to JUMP ON one OUTSIDE THE HOUSE,
UlXW-ii FALLS, tlllU-ik' TO BE FOND OF JUMPING) ;
mom. tux"-citi, iterative lo'x"to-xw-a' to jump
UP AND DOWN, -tspa' , -tspa'\ local suffix over,
PAST, durative aspect (other examples are fsax-
tSpn'' A SPEAR GOES OVER, h\ta-tspa' TO GO OVER,
TO PASS, kannt'q'-tspa' to run past) ; corres-
ponding momentaneous aspect, -tspm'ii.
/\G. See note 40. -tia, word-suffix (or enclitic
particle) also, too, again.
47. klUlS-, stem the other, DIFFtRENT (c. g.
kw\s-i-'y-as to be atthe far end of the village,
kwi.s-aq't'so' [a house] stands opposite, kwis-t-
i'ya' to be at another time, Inw.s-to'p' to be
of a different class, abnormal), -a's, -as, -s
(after vowels), local suffix of durative aspect
on, at (e. g. hin-a's to be on, fiqw-a's to be
sitting on [a box], k'wa'i-as branches are on
[the logs]), mom. -a'siti, -{a)siti ; kims-a's to
be different, the other On, kw.sa's'al'i' iiiclh.n'
the leg of him which is otherwise, elsewhere
on, atfached to [him], i. e. his other leg.
-'at-W see notes 26, 44; -'i' relates kiuisa' s' al '
to tiictii.n'.
48. Absolutive form of noun, probably an
irregular reduplication. Object oiCcitc.tl.
49. Momentaneous intransitive. See note 42.
50. kwcs-, see note 47; lengthened form ot
stem because of iterative aspect, -pano'i, uera-
tive form of -pa'' side end {kwis-pa'- to be
on the other side, at the other end, often
with -s- ON. : kwis-pa'-s-'); to durative -pa'"
corresponds momentaneous -/>'.«'// from *-paimif,
whence iterative -pano'i by change of-// to -/
and lengthening of u to o' (momentaneous
forms in -in'ti correspond to iteratives in
-ano'i or -m'i ; another example is dur. -misa'
MOVING UP, mom. -tmsmHi, iter, -nnsano'f).
-ci-'atf, from -ati -\ — 'ati, see notes 18, i ; for
loss of-// in inceptive suffixes, see note 5.
kivi'spano'icitf is iterative-inceptive in aspect :
TO BEGIN TO (move) FROM SIDE TO SIDE; Other
examples of this aspect are isu'tstsu'tsati to
START SCRATCHING, SUtlsO'Vall TO BEGIN TO
TAKE TIME AND AGAIN (sCC HOte 32).
51. his-, see notes 42, 49. -an^'yaiui-, longer
form of -\.tn''yo'-tf, -uni'yu-li, momentaneous
form of dur. ->,;//'/ (after' vowels, -q-wi'f), used
partly as classifying suffix (round object ;
moon), partly as local suffix all over, cover-
ing A ROUNDED OR BULKY SURFACE (e. g. f hs-
iw'/to be WHITE ON THE OUTSIDE; ya'k-iini'f TO
BE SORE-HEADED \tn*' m\.-qonH TO HAVE SNAILS
ALL OVER IT ; b£-Wl'i TO BE ASSEMBLED, mom.
/;'.f-iw''j'()'//TO ASSEMBLE, COME together). When
momentaneous -// drops, as before -'«// (see
note 5), modifications often appear in the pre-
ceding derivative suffix (e. g. -tmsin'li up,
mom., but -tmsano'alf ; -o'ti on the face,
mom., but -aw.- ati, causative -azvtip' ;. -so'li
so AND so DIES, butsa'wCati, -sawi-'ati, causa-
tive -sawup' ; such cascsof -(r-(-()-) : -au'i-,-au'u-
suggcst that Nootka o' is sometimes contracted
from older rfzi^, an (cf. (/o/ slave : qaqo'i slaves,
reduplicated, from *qaqaiii). -ati see note 1.
448
VI American Indian Languages 2
52. See notes 38, 9. -t'so-- with long vowel,
-t'so' with short ; prope.ly long vowels are
frequently heard shortened in final posi-
tion.
53. qaJf-, verb-stem to be dead, to die ;
dur. qah-ak' to be dead, mom. qalf-cdt to
DIE. -a-\ti, see note 50 ; t phonetic variant of
a, X, because of preceding i.
54. \:>uyi' (.f'ji' ; o' and i' are often brok-
en to Jii, ail, and ti, ai after h and .), noun
medicine; probably derivative in -y\ (cf. mih-
SPEAR, absolutive mds-'yi ; further, derivatives
in -s'yC , e. g. 'mukw- stones lie : 'nuik-s'yi'
STONE, ''.«'/r' fire, to burn : '^n'k-s'y.' wood,
stick). -Ici'i (after a and -.), -kwi'i (after 0-
vowels), -r/ (after consonants), derivative suf-
fix attached to noun stems, to make (other
examples are'o-fef/"/, to make so and so, iihai-
i'f TO MAKE A CEDAR-BARK MAT). -C\.-\tl^ see
note 53 ; -ati is here inceptive.
55. qwis, durative verb to do thus, to act
as described ; perhaps related to qwa'' (see
note 21) and to qwK-, stem used in relative
verb forms to make indirect questions (e. g.
qu'iyifftaqak-i'tc WHAT, as they say, it is made
OY , qwe-sa'h'^i-vjos-C why he would be..., qwe-
y.-n'-yi's at what time i come) ; -s-, possibly
identical with local -s on, see note 47. -//' (after
all consonants but h), -qff (after vowels and h),
word-suffixattachedtoabsoluteverbform(comes
before all other word-suffixes) and indicating
that the activity or state predicated by its verb is
accompanied by or in some way conditions the
activity or state predicated by the following
verb or an understood verb (e. g. h.t-ii-h-ah
ho'ya'l i-am-in-the-house-while dancing, uut-
qh-atl-ni now we say it while [thus occu-
pied!) ; here it implies that the following verb
(to obtain many in hunting) results from
the activity (to do thus, i.e. to make medi-
cine) of its own verb, -ati, see note i.
56. 'a'y.m'k'y from *\rya-mik' , see note 9.
'a'ya-, lengthened from 'a\a-, verb-stem to be
89
MUCH, MANY (durative absolute '(iya) (e. g.
'aya-qs to have much [game] in one's canoe,
'a-y-ip TO secure much, 'ayir-p'd' to be many
times), -niik', -mi'k\ derivative verb suffix
lengthening preceding stem-vowel, to succeed
IN hunting... (e. g. 'o'-ini'k' to get so and
so IN hunting, to be a successful hunter).
-a-<xtl, see note 50.
57. 'napxta'', durative intransitive to die
IMMEDIATELY (aFTER BEING STRUCK). -'olf, SeC
note I. -qo'-, conditional suffix, see note 58,
frequently used in main clause as past usita-
tive (e. g. mala'- ati -qo'-k' whenever, if you
FLY ABOUT or YOU would, USED TO FLY ABOUT) ;
-wt\n' shows that-(/o'- cannot be here under-
stood as subordinating, as -tc- is quotative in
subordinate clauses (e. g. 'napxta'-ati-qo'-lc
IF, AS IS said, he dies IMMEDIATELY). -lVt\.n\
quotative word-suffix it is said in main clau-
ses ; replaces third personal indicative -ma\ e.
g. qaipak' -wiai' he is dead, they say : qaipak'-
ma' HE IS dead (ist pers. sing. -wi\-S'.' , 2nd
per. sing, -u'z'oi-tsuk') ; << *-wa-\n', probably
petrified nominal derivative (omwa'- {wa-) to
SAY.
58. t'saxw-, t'sax- (labializations regularly
disappear in syllabically final position), verb-
stem to spear (e. g. fsaxw-i''nak' to imitate
A SPEARER IN A DANCE, fsax-tSpa'' SPEAR GOES
OVER, t'sax-yak' spearing-ikstrument, speak);
mom. t'sax-C'.ti, iter, t'sa'xt'sa'xzv-a' . -c.-eti,
see note 50, 53. -qo'', conditional suffix; its
paradigm is :
Sing. I. -qo'-s
2. -qo'-k'
3. -qo-'
Plur. I . -qu-ti'
2. -qo'-so''
3. qo(-a{)
With quotative -tc (cf. note 57) it forms :
Sing. I. -qo'-ts-i Plur. i. -qo'-t'a-n'
2. -qo'-tc-k' 2. -qo'-ts-so''
3. -qo'-tc }.-qo--tc(-'ai)
Eight: Wakashan and Salishan Languages
449
90
59. 'ah'^'a', see note 28. -'ati, see note i.
-wz\n\ quotative, see note 57.
60. ttao'' ANOTHER, as durative verb and
verb-stem tla'o'- to be another (e. g. tia'o'-fsiq '
ANOTHER LONG OBJECT, CANOE, tlaQ-'yiy-a'
TO BE ANOTHER TIME, NEXT TIME, tiaO'-tClti
TO BECOME another). -'«//, see note i. ~iia\
see note 46.
6 1 . f so- itch" to be winter ; mom. t'so\tch"citi.
t'so-\lch" is explained by theNootka as washed-
SEASON, i. e. season when everything is washed
clean by rain and snow, cf. t'so-, verb-stem
TO WASH, -itch", -'i'tclf, stem suffix season (e.
g- 'mil'i-i'lflf rainy season, la'y-i'lcJf rotting
SEASON, FALL; cf. ^\sO -(]'-\tch" YEAR, Q. g
nio'-q'\tch" four years), -cilf, momentaneous
suffix, see note 18.
62.iuz\tc, see note 5. -'ah"s, dur. local suffix
IN A RECEPTACLE, IN THE VULVA, IN A CANOE,
mom. -'al/'sitf, caus. -ah"-s\p' (cf. further W/-
ah"S COLD IN A RECEPTACLE, \. e. COLD WATER,
hayu.-ah"s lo have 200 (salmon) brought
HOME IN A canoe) ; -alfs is one of those suf-
fixes that " harden " preceding final consonants
of stems, i. e. preceding p, t, k, hv, q, qiu, ts,
ic, // become glottalized to p\ /', k\ k'u', ', !,
t's, fc, /'/ respectively (cf. note i), s, c, / to 'y
(-/-'- above is irregular), .v, xiu to \v, n to ';/,
in to 'w, h to }/ (sometimes 'zu) ; see -^.tch",
note 61, for another " hardening" sutfix. -ati,
see note i.
63. 'annh-, probably identical with \inah to
BE (so AND so) IN SIZE ; -\s, diminutive suffix.
'annh- ' is doubtless based on \via- (dur. abso-
lute 'ana') only, see note 164 ;cf. further V/z/^t-
(with interrogative -ha) how many ? \rna-
TO last, to be IN EXTENT, 'ani-ts- as long as,
'ana'-...-\s to be near. Diminutive -'a is
freely used as word -suffix, less frequently stem-
suffix, in both noun and verb forms, e. g.
la' nz~ '.s cuiLD, ha'hua'ti-\s little girl, tand-
q'tl-'nnkw-'is TO BE child-in-handed-little,
to hold a child in the hand. In 'anah-'is
little, -\s may be separated from 'anah-, e. g.
'rt"jm/>(r;)//i''-'u to BE LITTLE-HEADED. Here used
as adjective qualifying canoe.
64. Absolutive noun t'capats canoe (irregu-
lar plur. tcn'ya'pats), local object of hd. t'cap-
(sometimes t'catn-), noun-stem canoe (e, g.
t'ca'p-ok'i canoeman, see note 107, t'cam-e.'ali
TO return in a canoe after going in some
OTHER way) ; in most derivatives t'capais is
treated as stem (e. g. fcapats-nak' to have a
CANOE, t\apals-diua' canoe-place on the
rocks), -ats, derivative noun-torming suffix
receptacle (e. g. \oqw-ats urine-receptacle,
bladder) ; cf. also -sats receptacle (e. g.
k'o'f-sats vessel for eating a relish out
of).
65. V, see note i ; refers to following noun.
-yoaf (after vowels), -o'ai (after consojiants),
derivative suffix to get sight of, perceive
(e. g. wawa'-yn'af to hear what one says,
qo'ats-u\if to SEE A person), -atf, see note i.
66. kw.stsatcili, momentaneous verb form to
GO off elsewhither, nominalized by -t'ca\
kwis-, see notes 47, 50, -tsa-tcitf, mom. suffix
TO GO off to, cf. note 35 (e. g. 'u-tsa-tcilf to
GO TO so and so, ho a-tsa-tcili to turn back);
corresponding caus. is -tsa-'ap' , e. g. 'n-isa-ap'
TO take so AND SO UP To) ; as corresponding
dur. intr. is used -lsuuk\ -tsc'nk' (e. g.
lua-S'lsu'nk'-ha-k' where are you going ?).
-tea' , suffixed article or nominalizing particle
with quotntive color, the..., as they say (for
related forms in -ic- see note 58); syntacti-
cally equivalent to non-quotative -\' (see
note 9).
67. 'jh", durative verb to be the one, thh
one who... is so AND SO, indicates that the
noun following or understood is the one refer-
red to in a preceding verb ; here : the one who
(perceived the shift) was (cha'atssib'). This
verb probably consists of '()-, see note i, and
petrilied suffix //', possibly to be {d. perhaps
-/.'" in 'aJf, note 28, and 'aiia-h", note 63).
450
VI American Indian Languages 2
91
68. NitiiKU name ; h and final glottal stop
show at one that it cannot be Nootka proper.
69. 'u-, variant of '0-, see note i. -kia(^y
(after u- vowels), -/^/rtf()' (after other vowels),
-iaCy ('"^f^ter consonants), dur. verbifying suf-
fix TO BE NAMED... (c. g. 'ag'-tctu-ha' HOW IS HE
NAMHD ? ko'0(J-fl-\' THE OBSCENELY NAMED ONe).
70. lim'h.-, noun-stem name, absolutive
.im'tc' ; cf. verb-stem Um'/- to sing the signi-
ficant syllables OF A song. -«a"A'', -Mfl//, verb-
ifying suffix TO have... (e. g. 'u-na'k' to
HAVE so and so, t'atia-nak' to have a child,
imy-tci'i-uak' to have for four days). [=i)i'li-
na'k , complementary infinitive depending on
'iikla' ; literally, (cha'atssib') is what he was
called TO have a name (or in having a name).
The idea conveyed by a derivative suffix is fre-
quently supplemented by a following primary
element conveying the same notion, e. g. 'o'cs
ha ok'" to-consume-so-and-so to-eat, 'utshtn'tf
hifcna'k'ati to-marrYtSO-and-so to-begin-to-
have-as-wife.
71. 'o'o'-, reduplicated form of '0-, see note
I. -laff, derivative suffix reduplicating stem
and lengthening stem-vov/el to hunt (such
AND such) SE.\-MAMMALS, TO BE A WHALER.
J2.l'h-'u".n', laughing-stock, TO BE A CAUSE
OF LAUGHTER, derivative of verb-stem t'ii'xiu-
TO LAUGH (e. g. mom. fii'x-citi to laugh
ONCE, dur. t'H-xzv-a' to be laughing, I'ii'x-p'i-
tdf to laugh while ENGAGED IN SOMETHING
else) ; -\u- is " hardened " from -xw-. -in',
" hardening " derivative suffix, apparently
m.ikes passive or resultative nouns out of
verb-stems (e. g. t'coc- to suspect : t'coc-'<.n'
ONE who suspected, haca'- to hear about :
hady-oi' famous). -ap\ -am-, causative suffix
(ci. note 36); t'ic'ivi.n''ap' to cause to to be
laughable, TO treat as a laughing-stock, i.
c. to laugh at. -'at' , passive suffix, see note
I ; identical with inalienable possessive -'al',
see note 2^.
73. Relative particle introducing staiemen.t
of cause or other attendant circumstance in
following subordinate verb. Probably petrified
relative in -i, as shown by its pronominal
forms, which are those of relative forms in -i-,
-.-, -3'..-, -_)V- :
Sing. I. 'atn-s
2. 'am-V
3. 'am'
Plur. I. 'aui-n'
2. 'ajH-so'
3. 'anC
Cf. relative paradigm in note 185.
74. qa'ya'- to drift (in a canoe) ; may
contain -^--outat sea in a canoe (e. g. hin-a'-
tc'.tf mom. TO GO out to sea in a canoe, ivik-
a'-nak' to have none in a canoe), -panatc,
derivative suffix generally lengthening stem
vowel, to wander acout aimlessly (e. g.
^o'tcq' -panatc to get lost in a fog, ya'is-panatc
TO BE OUT FOR A WALK, hita'-q' t'l-as-panotc to
SPEND one's time IN THE bush). -qa. Subordi-
nating suffix indicating cause or other atten-
dant circumstance, often, but not always, in-
troduced by \ui<. ; it is etymologically identical
with -qa- of possessive paradigm (see notes
I, 26). Its pronominal forms are ;
Plur. I. -qi-n'
2. -\t'-q'-ifl''
3. -qa{-'af)
Sing. I. -qa-s
2. -it'-qa-k'
3. -qa'
'an<. qa'ya' panatcqa indicates cause of t'fi'-
'u".?i''ap'at' ; more explicit causal statements
are rendered by 'o-no'tf 'am... -qa' to-be-for-
such-and-such-a-reason, that...
75. See notes 5, 74. Here -qa' denotes at-
tendant circumstance, while asleep ; iviilcqa'
follows clo.sely on qa'ya' panatc and needs no
formal 'am to introduce it.
76, u".k'-'ai NOT TO BE AWARE OF, dur., con-
sists of luik' (see note 10) and derivative suffix
-ai (certain suffixed elements beginning with
glottal stop do not glottally afl^ect preceding
consonants) to be aware of, have news OF(e.
g. 'o-'af TO HAVE news OF SO AND so), inceptive
-'<:-tc<.ti TO become aware of (contracted from
Eight: Wakashan and Salishan Languages
451
92
older *-'a/-, « softened » form of -'d/). -'atf, see
note I. -qa, see note 74. u'ik"afalfqa follows
'aw. (note 73) and is parallel to qa'ya'panatcqa ;
AND is not necessary (ci. note 3).
77. See notes GG, 74. Here -qa marks a
subordinate clause that is objectively related to
wJi"ai'ati.
78. Absolutive and stem-form : daylight,
DAY, also PERIOD, SEASON, WEATHER ; as dur.
verb, TO be day, mom. 'na's-atl day comes.
Possibly 'na's is composed of simple 'na'- (cf.
Kwakiutl'wrt'-/^ day) and petrified -s (on ?),
cf. qwis, note 55 ; this seems to be confirmed
by '^oolkd.^ na' - p' <.naq- the proper time comes.
'na's is subject o{ kwistsatcitiqa : that day was
GOING OFF elsewhither, THAT IT WAS THE SEA-
SON OF GOING OFF ELSEWHITHER..
79. Umlauted from haca'-tcili. haca'-, dur.
stem TO HEAR ABOUT; dur. -a' (but not short
-fl) is umlauted to -r- before momentaneous
-tati {c{. 'tiiitii'-tcitf to begin to rain from
'ni'.tia'' TO BE RAINING ; but 'i'wa-tcdi to get
big), -tciti, see note 35. had' tati is momen-
taneous; most forms in -r-/<:i// are definitely
inceptives.
80. See notes 72, 74. Here, as frequently,
am. . . -qa' mark an objective clause of indirect
discourse.
81. ya'kiu-, ya'k'- to be sore, used either
with body-part suffixes (e. g. ya'k-o'i to be
sore-faced, ya'liiv-in'i to be sore-necked,
yaya'k'-niik'" to be sore-handed) or with ina-
lienable possessive -at' (see note 24) and fol-
lowing body-part noun (e. g. ya'k'-at-ah qasc'
mine-is-sore eye, I have a sore on the eye;
d. yaya'k-sui to be sore-eyed) ; dur. ya'k'-at',
inceptive ya'k'-ci-'at\ -ci-tt', see notes 18, 5,
24, 53 ; -at' is treated analogously to -'ati (see
note i), and they combine into -'atlat' .
82. Absolutive noun « heart, » seat of
intelligence and feeling ; heart in its ana-
tomical sense is tiUc-nia' (from ti'tc to be
ai.iVe). iCiiiaqslC consists of radical iini-, of
unknown meaning, and " hardening " nomi-
nalizing suffix -aqstC what is inside (one's
body) (d. dur. -'aqii inside, mom. -aqstiili).
83. citi-, verb-stem to move from one place
TO ANOTHER (e. g. ci'tf-uk'" TO BE MOVING,
ci'ya't'f-aqa' several move in a canoe, ci't'l-
ik' TO BE ALWAYS ON THE MOVe). StCS, local
suffix INTO THE INTERIOR, UP WHERE SALMON
RUN (cf. hii-StCS TO BE IN THE INTERIOR), after
a- and /-vowels -tsstvs, after //-vowels -ksti's
(e. g. 'o-kstcs so AND so IS UP country); -st<.'s
is correlative to -zcr'ti (e. g. hd-ivi'':s to be
ALONG THE SHORE, 'o-Wi'\s SO AND SO IS DOWN
AT THE S£a).
84. =hcs-sasa. hcs, dur. verb to be at the
beach, contracted from *hiy-is, " softened
from hd (see note 24) -|- -15, dur. local suffix at
THE BEACH, ON A LEVEL AND OPEN STRETCH (e.
g. 'yaq-'.s long level stetch of country,
to'h"--maq-\s spruce trees are scatiered on
the BEACH, ' yayak-is heads show of people
seated on the beach), mom. -saii ; h's like
hi'i TO be in the house, contracted from
*hy-ii (see note I3)C' softening, " as contrast-
ed with " hardening ", is no longer a live
process in Nootka, but survivals occur of /, s,
and c softening to y or u> [after 7/], ot h and xu<
to lu, and of .v to ts <C original d^; d. k'wa'y-i.^
TO HAVE BRANCHES ON THE SHORE, frOtU k'wat
bra-nch). -sasa', word suffix lengthening stem
vowel, emphasizes verbal idea, right along,
JUST so, IN VERY TRUTH (e. g. 'o'lfsasa' TO BE
the very ONE WHO..., TO BE THE ONE WHO...
RIGHT along) ; reduplicated (uniquely so for a
suffixed element) from -sa\ emphatic word suf-
fix lengthening stem vowel, very, too, just,
MOST (e. g. 'o'-ak'th-sa' to be the very last,
'a-n-uaf-sa' to see merely..., 'a-ya-kwai-
C'.ti-sa' TO become too much-missing, to
spend too much).
85. Place name, local object ot h'ssasa'.
Radical element ihr-, of unknown meaning.
-ow\s (after vowels), -m'\s (after consonants)
452
VI American Indian Languages 2
i
93
nominalizing sufhx place on the beach (d.
liukwat' q-(yw\s place on the beach for hold-
ing A WOLF RITUAL, \aS-OlUKS PLACE ON THE
BEACH WHERE CARPENTER WORK IS DONE) ;
" softened " (cf. note 84) from -ul place (cf.
})0-'a'-q-ul PLACE WHERE CORMORANTS CONGRE-
GATE, WtilC-ui PLACE TO sleep) + -l? ON THE
BEACH (see note 84) (parallel forms in -w- and
-'w- from -/ are -uw-d place in the house,
-uw-a' PLACE ON the ROCKS, -uw-as place on
the ground; -anow-<.s along on the beach <
-anid along + -ij).
86. See notes 84, 1.
87. Contracted from ni'trna'a + -lUh".
ni'lvna'a' , village name containing, in contract-
ed form, " hardening " -\ia' on the rocks;
exact form of stem and its meaning are unde-
termined, -ath'', nominalizing suffix people
OF... (e. g. fsica'-atfj" people of ts'isha,
pj'sh.n'-ath"- boston people, Americans), com-
monly used as ending in tribal names (whence
our Aht as general term for all Nootka tribes);
also used as verb suffix, " hardening, " to
dwell, to live together, to be peopled by
(e. g. hiy-atJf to dwell [probably from hic-
all, see note 31], tuk'w-atff sea-lions stay,
TO be inhabited by sea-lions, yaq'-tct-ath"-
'•J'g' with whom he lives, his neighbors).
88. '«-, V, see note i. -tsa-tci-zti, from
-tsa-!citi (see note 66) + -'ali (see notes 1,5}^.
89. t'sa-, verb-stem to flow (e. g. fsa-
ts'ivi' A current goes through, t'sa-htw'a'
A creek runs down a kocky bed), -'ak',
durative suffix and absolutive noun sutiix
(see note 2) ; Csaak' to bf flowing ns
verb, CREEK as noun, -okiu- (labialized because
ofo preceding k\ -ok', see note 33. -''.', see
notes 9, 15, 33.
90. Contracted from *iucnapi-all. wi'napi ,
durative to stay, remain ; w.'n-, with leng-
thened stem vowel because of following suffix,
verb-stem not apparently found without fol-
lowing -np:' ; -ap'.' (after consonants), -/?-.' (after
vowels), dur. local suffix, lengthens stem
vowel, erect, standing, placed up (e. g.
hii-api' TO BE IN THE SKY, RAISED UP, 'o'-p'.'
SO AND SO STANDS OUT, IS AHEAD, 'nO'p'-t\siq-
apis [= -apf. -\- -\s] one-long-object-standing-
ON-THE-BEACH, BEACH WITH A LONE TREe),
mom. -{a)pntl. -'atl, see notes i, 14.
91. Inceptive form in -c.tf (see notes 18, 54)
ofWi-'l"/' TO BE NON-EXISTENT, THERE IS NOT.
zv\.k-, see note 10. -''."/', " hardening "stem-
suffix used only with wik-, meaning probably
THERE IS IN E.\.iSTENCE ; -''."/' trom -*•.•«-, cf. pre-
terit lUlk'-l'n-lta' THERE WAS NONE.
92. Absolutive form ; distributive qoqiua's
PERSONS, but here qo''as person is taken collec-
tively, hence people became non existent,
THERE CEASED TO BE ANYONE THERE. Stem fomi
xsqoats- (e. g. qdats-ina' soul, personal double,
qo''at's-iti TOGO for a person, qoq-wa' ts-h"la'
to be PERSON- FOOTED, TO HAVE A PERSON ON
THE foot).
93. = ' 0' s\m' Ic-ci- 3.li , inceptive aspect. See
notes I, 50.
94. /?'."«-, lengthened forni of /;•.;/-, " emptv "
stem to be, do ( as described by suffixed ele-
ment), possibly demonstrative in origin (see
note i; d. hi'n-a'^uiti to be up river, hm-usa'
TO come up out of the water, h.n-ustas to
GO AWAY prom THE BEACH, hoi-U'k'wai TO GO
AWAY EXPECTING TO RETURN, hlfl-iu' TO COME,
bm-i''as TO go outside, Inn-a'sdi to go up on ;
also in body-part nouns, e. g. hm-o'f face,
b.n-aksui lips, hui-i'fiaksiii chin ; varies, for
phonetic reasons, with hit-). -<.s-, -i's, verbal
suffix TO TAKE ALONG, CARRY (e. g. hoi-CS TO
TAKE ALONG, b.n-i' S-v' itl TO TAKE INSIDE THE
HOUSE, hop' q-.s-aqo''' ol' to round a point of
LAND CARRYING A ROUND OBJECT, ROCK ; hcil-lS-
is lengthened from hin-<.'s-, -cs- shortening to
-IS- as rhythmic counterpart of stem lengthe-
ning) ; d. also related -/7^^ (after consonants),
-Iss (with lengthening of immediately preced-
ing vowel) TO CARRY (e. g. ':ixiu-i'tss to carry
Eight: Wakashan and Salishan Languages
453
94
A PADDLK, hayU'-tSS TO CARRY TEN OBJECTS IN
THE HAND, 'o'-tSS-'aq' H-.ti TO TAKE SO AND SO UP
TO THE woods), mom . -rsiili (e. g. .'^dlc-i'sut}
TO TAKE A DOG ALONG, '."-'.ftC-i tSS TO HAVE A
DOG along), -o-'uk' , dur. suffix, lengthening
stem, TO BE ON one's way (see note 66); hcn\.
so''uk' means literally to take along while
on one's way and refers possibly to the canoe
dragged along by the lunging whale, though
the verb is used simply tor the movements of
a blowing and diving whale and for ceremonial
imitations of these movements, -c.li, inceptive,
see note 50.
95. wa'iak'y lengthened form oduaia'k', dur.
verb TO go to, be bound for a place, probably
contracted {rom*tvaia-'ak' (cf. analogous forms
like t'sawa'k' TO be one <C *t'sawa-'ak' and note
14); waia- probably related to loaf-, verb-stem
TO return home (mom. wai-ati to be gone
home, graduativeti^d'/r-A// to be going home) ;
-ak', dur. intransitive, see notes 2, 10, 89.
96. \Jm\zu\.e(i {xom * saya' -\' ; see notes 22,
34. saya-', dur. verb to be far away, distant
(other aspects are inceptive sayzi'' to get
distant, graduative sa'yzi'' to begetting dis-
tant, durativeiterativcyfl^a'i'a'/jmomcntaneous
iterative sast'i'y, saya'- is also used as verb
stem (e. g. saya'-Cca' to be high up, sa'ya-
tch"saa' to be a long rocky shore). -\\ see
note 9 ; the far-distant (place) is local object
of iva'iak' .
97. mo'' , dur. verb to be four, also verb-
stem nio'-{e. g. nnr-p'ai TO have four objects
on one's back, mo'-tc'.n'k' four are on the
WAY, mo'-sivtso' TO be in four places), -ici'i
-tcii, numeral classifier for day(s) (e. g. ^luip'-
tci-l one day, 'ati-cq'-tcii two-twenty-da Ys,
forty days) ; -tci'i does not become -kiui'f
after u- vowels (ci. note 32).
98. Durative form ; see note 94.
99. Durative form of numeral; seee note 97.
mo' 'ua-s is specifically I'OUr periods of day-
light, while nwtci'i means four days, spans
OF DAY AND NIGHT. Syntactically, nw 'tz^-j (and
nw 'athai') is best considered an absolute -|-
complementary infinitive : it was four in being
DAYLIGHT (and IT WAS FOUR FN BEING NIGHT).
100. Phonetic variant of '<r//>e*', see note 16;
both are properly 'ath"'.-'.
10 1. hoi-, see note 94. -as<.t{, mom. local
suffix (arriving) at ; not to be confused with
-(a)5'.ti, -a'sitf, mom. form of -{a)s, -a's on
(e. g. t'<.(]iv-a'S'.t{ to SIT down on, h'.ii-a'S'.lf 10
get up on [a platform I).
102. Place name, probablv Niiinat, of un-
known etymology. -a is doubtlcs ON THE beach,
cf. note 84.
103. Umlauted {rom*'iikfa'-\, d. note 96.
'tikfa'-, see note 59. -'-., see note 9. 'iikii'^c is
in apposition to kaxi'h.s.
104. Absolutive noun; there is no evident
analysis. Syntactically, ms'^nin' amplifies '/<- of
preceding nominalized verb.
105. '/'.(/'-, syllabically final form of 'tuju'-,
see note 13. -sali, mom. form of -'.s on the
beach (e. g. hda-satijo land on the beach, /rt'-
Satt TO DRIFT ON TO THE BEACh).
106. /;-./-, h'.ta-, phonetic variant of /.»•.;;-, see
note 98, found particularly, but by no means
exclusivelv, before " hardening" suffixes (e. g.
h'Ja-t'.w' TO be in the bay, /;•./-;■/ to be inside,
h'.l-J/'ta' TO be at the tip, end h.t-aq'ti-nuk'
TO HOLD IN THE HANDS, h<.ta-t]SO} TO GO INTO A
CANOE, h.ta-ts^o}fia' to appear coming out of
THE woods; also in body-p.irt nouns, e. g. h\t-
(fk'lf'.' rump). -(rt)/'"/'..v, local sufiix downstream;
probablv compounded of (n)h"l- and -j on the
beach, on a level spot. Evidently -{(i)h"l'.s is
correlative to -hf-ir'a' downstream on the
rocks (c. g. fsa-hC-a-\i a creek runs down a
ROCKY bed) and -nht'-as out of the woods
(c. g. cans. h\ta-hCas-ap' to cause to come
OUT OF the woods) ; hence -{a)lfhj probably
means, more accurately, moving into the open
TO A BEACii-i.iKE PLACE, leaving -(<;)/'"/- to mean
out into 1 iik open, seaward. Not to be confus-
454
VI American Indian Languages 2
95
ed with -{a)h"hs is -at'.s downstream (e. g.
hd-ahs TO BE DOWN THE RIVER, t'ias-at<.s TO
SLIDE DOWNSTREAM, laq' maq-at<.s grass moves
downstream).
107. t'ca-p- lengthened komt'cap-, see note
64. ~ok'>', intransitive suffix, lengthening stem-
vowel, TO MOVE along (e. g. ya'ts-uk'" to walk
< yats- TO step , scv'-ok''^ to crawl on all
FOURS, 'na-lc-uk'" to look for, ci'ti-iik'" to
MOVE FROM ONE PLACE TO ANOTHER). CciVpok'''
thus literally means TO MOVE ALONG IN a canoe;
nominalized, canoe-party, canoeman.
108. See note 89. Local object of h.tah"tis.
109. Umlauted kon-\*liawa'-to/at' , cf. note
79. tfaiva'', dur. verb to be near (other aspects,
some of ^vhich are parallel to those oisaya'',
see note 96, are : mom. thiv^n'ti or llaivii''
TO GET NEAR, Inceptive tia-wz-tatf to begin to
be near, to approach, graduative tia'ivr.nHi or
tfa'tui''i'' TO be getting near, iterative //a*-
wano-f or tio'iuui'iy -to.-, for -tc.tl, inceptive
suffix after vowels, see notes 35, 50, 54; for
loss of -tl before -at' , see note 5. -tt\ palatali-
zed phonetic variant of -at' (cf. note 33),
passive suffix, see note 72.
no. h's, see note 84. -'it'q', marks subor-
dinate clause of locality, cf. note 22.
111. naiua'y-, probably " softened " from
some stem no longer ascertainable {*nazva'i-,
*naiua's-, oY*nawa-c-), to sit looking auound,
TO sit around without particular purpose
{d. nawa-'y-as to sit outside of the house
looking around, used particularly of old men
sitting on the platform in front of the house in
the morning and gossiping as they look around).
KS, local suffix ON the beach, cf. note 84.
112. Iku-, see note 94. -nita' (after conso-
nants), -'iv'.ita' (after vowels), local sulfix out
OF THE CANOE.
113. /;i/-, see note 2^. -yin', local suffix at
THE bow OF A CANOE, also -a'tsyiu' , -alsyin' (e.
g. hop~a'tsy<.n' TO HA\E a round object in the
BOW, i/dA'/V-^/vy..^' tostandin the bow, see note
1 14). -'i' , see notes 9, 103 .
1 14. s<jn- , verb stem a pole-like object has
POSITION (cf. SKp'-to'p' stick), -tt'lsym*, local
suffix in the bow, see note 113. sona'tsym* is
to be understood as a " bahuvrihi " : to be
pole-bowed, to have a pole-like object in the
Bcnv.
115. Absolutive form of noun, object of pre-
ceding '• baiiuvrihi " verb. hkJs-, noun stem .
spear (e. g. inds'-na'k' to have a spear, tni'is- I
\ti to go for a spear, mrls-i't to make a
spear). -'yC, absolutive noun suffix, see note
54- ^
116. I'^q'-, see note 105. -stc<.sa-'af' from
-stcisati -\- -at\ cf. note 109. -sto.sali, mom.
form (cf. note 105) of dur. -j/«".^,next to on
the beach, contracted from -stci- (which does
not occur uncompounded) and -is on the
BEACH (after u- vowel -k'tci's, e. g. ' n-k'tci's
so and so is next to one on THE BEACH), cf.
-tci-'ath", -A'7n-'fl///' TO BE NEXT-PEOPLED, neigh-
boring (e. g. 'o-k'tci"'ath"'-st'at to be neighbors
to one another); this -stci's, -k'tci's is not to
be confused with stem-lengthening -stci's fur-
ther up on the beach (e. g. 'o'-stci's to be the
ONE that is further UPON THE BEACH; cf.
'o'-iica'S to BE THE FURTHER ONE UP OUTSIDE OF
THE house), -at', passive suffix, see note 72.
117. 'jh", see note 67. -at', passive suffix,
see note 72. 'jh'"at' may be interpreted as it
was (the canoeman) by whom (k'walisits)
was (neighbor)ed. 'jf/^at' + noun corresponds
to our English agentive phrase (by...) alter
passive verbs. It should be understood that
every Nootka verb, transitive or intransitive,
can be made passive in form.
118. Momentaneous form, qatc-, verb stem
TO NUDGE WITH THE ELBOW. -C.-zt', see UOtCS
18, 109.
119. 'oqto-,s\.em good weather (e. g. 'oqw-
at\S FINE WEATHER COMES DOWN ON THE BEACH,
'oqzu-i''nak' to imitate fair weather in a
Eight: Wakashan and Salishan Languages
455
96
dance), -um'h (perhaps properly -an'h- ; labiali-
zed ^-sound + -iw', -',«', often develops to k--
sound H — urn', -un'), stem-lengthening verbal
suffix of unknown meaning (1 have found it
only with 'oqw-). -z-, for -i- (^''i > ht), durative
suffix after h (see note 16 ; -r', sometimes
shortened to -e' , is absolute durative, but before
indicative -m- this element is always shortened).
-ma, third person present indicative (see para-
digm in note i ; -m^ also occurs in other per-
sons if preceded by a vowel, e. g. haokw-ah i
EAT but so'-m-ah i a.m holding).
120. Interrogative adverb of modality, is it
NOT ? DO YOU THINK ? tak-, possibly identical
with stem of /a^^x*' nevertheless, still, yet.
-ha, interrogative word-suffix, third personal
subject (see note i for paradigm of interroga-
tive forms).
121. See note 78 -\', see note 9; ihisthe-
DAY is equivalent to today, 'ah" 'na's'i is
probably better taken as subject of 'o'gtim'htvra
(this day is fine-weathered) than as indepen-
dent adverb (it is fine weather today).
122. lua'', verb to say (also used as verb-
stem iva'-y, ww' does not seem to be used as
durative (like certain other monosyllabic long-
voweled duratives, see note 32), but as
momentnneous : to say (a single thing); its
corresponding durative is waiuw' to be saying
(not necessarily several times, despite its
iterative form), inceptive ivawi'-lcui to begin
TO BH SAYING, -'at', passive suffix, see note 72 ;
passives of zua'- refer to the person addressed,
not the thing spoken of (e. g. wa'-at-ah i am
told) .
123. 'tnogw-, verb stem whose precise mean-
ing is not yet clear; presumably to bh dumb or,
less specifically, to be barred, hemmed, rigid
or something of that sort, -c'yu'lf, -c'yiili, mom.
form of dur. -c^yuf, -Cyui, local suffix in the
throat, distributive form -i"V//', -Cy:Jh" (/ of
suffixes frequently changes to /'" in distributive
forms, ci. further -o'i faced: distr. -o'h") (cl.
dur. '.aq'maq-eyid to have grass in the
throat : distr. fht'hh-atyolf red here and
there at the throat, marten ; mom. 'atf-taq-
e'yuti TO BE DOUBLE IN THE THROAT, BIRD SINGS
DOUBLE notes) .
124. hi- (from /'"i"), stein to be unable in
ANY WAY followed by irrealis -a'lf (e. g. hi-
/jfl-'fl/)"-fl-/7-n' I COULD NOT take ITTO ANY place).
-s'lii-, i. e. -ssdi'-, umlauted from -ss>Ja'- (see
notes 79, 109) and lengthened, as generally in
short-vowelcd durative-inceptive forms, from
-ss'Ja- ; -ssda' (after vowels), -sda' (after con-
sonants), dur. verbal suffix to act in such and
SUCH a way (e. g. p'<.c-s<Ja' to do something
bad); -ssiia' assimilated from *-cs\la' {c\. per-
haps Kwakiutl -ysi'Ja to take care of ; this
original -c- ■< --/- explains 'ani'-sda' to only
DO so <C *'anaisiia' <C *'ana-ysda' , cf. qawi'' :
qawac- salmonberry <C *qaii'a'i). -ta-, from
-tC'.ii, inceptive suffix, see notes 35, 109, 5.
-a'h", modal word-suffix indicatinguncertaiiuy
or unreality, particularly in negative clauses,
act phonetically like -'atf, see note i (e.g.
wik\ni' fw'wiFa'h do-not steal-at-any-time !
hayim*hath\ hisi'k'a'hrs now-i-do-not-know
which-way-i-should-go).
125. ts<.q-, verb stem to speak (e. g. dur.-
iterative tscq' tscq-a' to be talking, ts>.'.-"ik'
to be fond of talking, is'.'q'-f'.lch" TO talk
while doing something else). -cUf, momcn-
taneous suffix, see note 18.
126. qalf-, see note 53. -kca-li^-, for -ktm-
tcitl, sec note 5 ; -kwa-lali, momentancous
suffix -Ic'.li preceded by element -kca- denoting
completeness, totality (e. g. qai/'-lru'a-tc<.ti
PEOPLE DIE 01 f, xwak'-kwa-tC'.tf to become all
swollen up), causative -hva-'ap' (e. g. -Vi/.v-
kwa-ap to crush to pieces, fsax-hva-'ap'
TO SPEAR AT SEVERAL PEOPLE). - xt' , for -at' ,
passive suffix, see notes 72, 109, 118.
127. Contracted from *qa'yap't'a-at'-\\ see
notes 14, 26. qa'yap' t\i' , absolutive noun leg ;
no analysis suggests itself. -'al'-\' , see note 26.
456
VI American Indian Languages 2
128. = tiakic-c\tl, momentaneous aspect, see
note 18, from verb stem ilakic- to stand.
There is no simple durative corresponding to
tlakic-C'.li, as liakic- ("softened " to //a/.'/'-) as
durative is regularly followed by some local
suffix (e. g. tiaki'-s to stand on, used as
nearest equivalent for to stand ; tiaki'-qs to
stand in a canoe, wagon, box ; lia'kic-s'alo'
TO stand at the door), liakic- is based on
simpler tia- stick-like object stands (e.g.
thi-yw' stake for a weir, tia-a'^a a stick
stands up on the rocks).
129. Graduative in -cdi, with lengthened
stem-vowel, from verb-stem yats- to step, to
WALK {yats-all to take a step), see note 6 ;
practically, ya'ts-ati may be looked upon as
inceptive to va/j-H^'" TO WALK, see note 107.
Other derivatives o{yats- a.re yats-haid to walk
ON THE side (of A WILL), yaf s-a'q' ll\ti TO walk
INTO the bush, yats-i'tis to track, yats-tspa''
TO GET PAST, ya't's-J/' to try to reach by
following in one's footsteps.
i}0. ho' a-, stem back, returning (e.g. ho'i-
n' TO come back, ho'^a-l'safo' to repeat a wolf
RITUAL A YEAR L.\TER IN ABBREVIATED FORm).
-tsa-tciti, see note 66 ; as momentaneous-ite-
rative of mom. hc'atsatcdi, is found hotlha'lsalc
TO TURN BACK SEVERAL TIMES.
131. See notes 64, 33. Local object of pre-
ceding verb, completing -tsa-.
132. h\ta-, see note 106. -qsiti, mom. form
of -qs IN A CANOE (e. g. dur. tco-qs to kneel in
A CANOE WITH BENT BACK, mO'-qS TO BE FOUR-IN-
CANOED, TO HAVE FOUR [saLMOn] IN ONe's
CANOE ; iterative hi' la'qsi'i to go into a canoe
SEVERAL times) ; -^/i('.//) is uscd only after vowels,
-alfs^yti) after consonants, see note 62.
133. Momentaneous form of //i/'"-, verb
stem to paddle, travel in a canoe (dur.
tkh-ak' TO paddle, mom. tiJf-cUf to set off
In a canoe, graduative ti'/l/'-o.ti to be paddling
off). -C'.ti, see note 18.
97
134. Rhetorical lengthening of demonstra-
tive stem^w THAT YONDER (cf. note 17).
135. tdk-, from verb stem takzc- to do
NOTHING but, EXCLUSIVELY, dur. Ink-ok' TO DO
so at ONCE. -h"la-to.ii, inceptive local suffix
TO BECOME OUT AT SEA ; -h"ta- probably identi-
cal with -/.'"/- of note 106, in which case-ffta-
is best analyzed as -h"t- + dur. ■a-,tak-}ft-a-
to.ti being properly to begin to be altogether
OUT on the sea .
136. fc'.'iiK'-''ath" name of a gnome-like being
of the woods ; t'cvtw-, stem of unknown
meaning; -ath" person, people, .see note 87.
-Cca' , quotative article, see note 66.
137. \)-, see note i ; anticipates following
temporal clause ttih"o/ iti' itq' t'ca'pokw\'
t'ci'm'atlf'e'. -ya'ti, contracted from -yi-'atf,
cf. notes 14, 26 ; -y.' derivative verbal suffix at
the time of (e. g. 'o-yi\ at such and such a
TIME, when, if, qiue-yi-n'-m-Kt-i'-s. at what
time I CAME, WHEN I came), also -yiya' , -yi'ya'
after vowels (e. g. 'ah"ko'-yi'ya' to be at this
time, tiao'-yi'ya ' to be next time, 'ah"'a'-yi'yati
<C -yi'ya -\- -'ati now at that time, passive
qoxwa-yiyat' -< *-yiya-at' to be so done to
WHEN freezing), -iya' , -i'ya after consonants
(e. g. 'alh"-iya-sati to come to the beach when
IT IS NIGHT, ku'ist-i'ya' to be at a different
time:).
138. f/'/.-^-Zo.//, inceptive aspect (to become
alive, to get well) of dur. ti'tc to be alive
{li'tc-a- is used as durative base for other
aspects, but not as independent durative; cf.,
further, iterative ti' Ica'-tci'l to become well
SEVERAL times, distributive mcepuvetdti' Ica-tati
SEVERAL BECOME well), also used as verb stem
ti'lc- (e.g. ti'tc-ma' heart, tdit'c-oi' to pray
FOR health). -tCK-\ti, for -tci-'ati, see notes
5> I-
139. See notes 133, 5, i, 22. -d'q' is here
temporal not because of any inherent value
but because time subordination is demanded
by preceding 'oya'ti.
Eight: Wakashan and Salishan Languages
457
98
140. See note 136. -e', phonetic variant of
-'i', see note 9. Note that quotative nominaH-
zing -^'m' (note 136) changes to non-quotative
-'i' because, once introduced, the t'ci'nc'ath"-
is conceived as known.
141. Contracted from *'i'ny..\iiia-'ak'-\\ see
notes 1^,26. 'r;zxv-;;/a',absolutivenoun cloth-
ing, REGALIA, derived from verb stem 'i'fiax-
TO BE PREPARED, READY (dur. 'Vfly.xa'' TO BE
READY, DRESSED UP, graduative 'in'xa'' to be
GETTING READY, gi'aduative CAUsa.t[ve'in'xa'-ap'
TO BE GETTING SOMETHING READY, momenta-
neous ^i'liyixi'-tall to get ready, momenta-
neous causative 'i'nxxi'-yoLp' to get something
ready) ; -f)ia\ absolutive noun suffix, cf. note
12. -ak'-\', see note 15.
1^2. /-.-, t'r-, verb-stem to rub (oneself) (e.
g. tr-tati TO rub once, tr-tcw' to rub one-
self down, iterative tititry-a to rub oneself
all over,- t\ti-n^kiim' hand-rubbing-object,
towel), -tnvs-ap, causative in ~ap' (see note
72) of -m\s, -mis, local suffix moving on the
BEACH (e. g. tla-mi's a stick-like object is put
UP several times on the beach, wik'-maif-mi's
TO go about the beach and be UNMANLY,
'o-'na'h-m\s to look for so and so on the
BEACH, 'ak'ivai-'nah-mis to look about the
beach to borrow) ; -ni'/s is contracted from
*-m\yij, *-mayis, " softened " from -mai-
MOviNG ABOUT and -IS ON THE BEACH, c(. note
84 (analogous forms are -mi'i, -mil moving
ABOUT IN THE HOUSE, -ma'S, -UldS MOVING ABOUT
ON THE GROUND, FROM HOUSE TO HOUSE, -Uli'^a' ,
-niCa' MOVING ABOUT ON .THE ROCKS ; with pre-
served -nial-, which does not seem to occur
alone, e. g. -mai-ap\ wovmc about standing,
-mai-it MOVING about on the body, -mai-m'
MOVING about in THE water), and is not a
true iterative (the true iterative of tit-\s a
STICK-LIKK OBJECT STANDS ON THE BEACH, e.g.,
is tia'-si's), though often practically used as
such.
143. See note 84.
144. See note 28. Specifies the local idea
contained in the preceding verb : he was
ON THAT BEACH.
145. " Softened " from nw-tci'i to be for
FOUR DAYS, see note 97, and -i^ on the beach.
146. Contracted from *lKtatssoh'Ha- ati-tia , stt
notes 14, 137. b.ta-, see note 106. -Issjfft-a'
(after vowels), -sohVa' (after consonants), local
suffix OUT OF the woods, probably containing
element -/?"/- discussed in note 106 (e. g. yats-
soh'^ta TO WALK OUT OF THE BUSH^ mat'-sohHa'
"TO FLY OUT OF THE BUSH, Sa'-tSSOh'^ta' [a WOLf]
COMES OUT OF THE WOODS, iterative sa'-tssJh'^tm'i
[wolves] KEEP COMING OUT OF THE WOODS).
-''ati, see note i. -tia, see note 46.
147. Phonetic variant of gwayafsi'k', abso-
lutive noun WOLF from stem qwayats- (e. g.
qivayiis-inak' to imitate a wolf in a dance,
(]zvayzts-ox"sim' wolf head-.mask). -'i'k' abso-
lutive noun suffix " hardening " preceding
consonants (cf. hivCi-vk lightning serpent,
literally ALWAYS coiling down to the ground?);
perhaps ultimately identical with " harde-
ning '' -ik' to be fond of — ing, to be
ALWAYS — ING (e. g. tuii'-lk' TO BE FOND OF
JUMPING, tS'.\-"ik' TO BE FOND OF TALKING,
na\-"ik' to be a drunkard), -i'k' , redupli-
cates with long vowel in reduplicating syllable,
TO BE ALWAYS ING (e. g. t'cO'fcilC-i'k' TO
BE ALWAYS SUSPICIOUS, \'\'qhaiik'zv-i'k' to bk
FOND of TELLING THINGs).
148. See notes 109, 46.
149. fsiisk-, verb-Stem to urinate on (used
only of animals), -ci-'et-, see note 118. -uk",
see note 10.
130. luiiiswJjp-, irregularly reduplicated stem
with -Is- after reduplicating vowel ; other exam-
ples of this inserted -ts- in reduplicated forms
are matsmay.xwvn' s^ikytswhosvekk passers-by,
iterative nilsfii'ya' to shoot several times,
and forms \n-sui in the eye, -s\m' what is at
the EYE (e. g. ^nutino'q'-sni to have a boil
ON THE EYE, yolsya' k' -sui to be sore inside the
458
VI American Indian Languages 2
99
EYE, halshim'k-son' eyelashes, k'utsk'ow'-s'.tn'
SCALES IN THE eye). -acj' , derivative noun-
suffix HIDE, SKIN (e. g. Itikw-aq' skin); more
common is stem-shortening -aaq' (e. g. [attic-
a'aq' deer-hide < la'luc deer, ' at" axiu-a' aq'
beaver-hide).
151. qwa'-, see notes 21, 27. -ak', postvo-
caiic form of word suffix denoting alienable
possession, see notes 10, 149, also notes 15,
39, 141. What is possessed is here urine (un-
derstood).
152. See note 22.
i53./cfl/5-, verb-stem to uml (mom. kats-atl
TO hail, dur. kats-a'' to be hailing, haily,
mom. -iterative katska'ts-c to hail at ^REauENT
intervals), -o'tnin', derivative noun suffix of
unknown meaning (mass of globules ? cf.
t'iaq-omvi' disease like chicken-pox) ; con-
tains absolutive noun suffix -i«', cf. note 23.
154. luai-, verb-stem to return home, d.
note 95. -C'.-tti, see note 53.
155. maht'i'' noun house, ma-, stem group
of people, to dwell as a community (e. g.
ma-'as tribe, village, ma'-ifsa's a house
stands against the hill, ma-mai-n:' dwelling
while moving about in the water, white
M\}i, ma-lci'{ TO be inside the house, mx-stc.m'
dwelling next to, common man). -/'/''."', deri-
vative noun suffix of unknown meaning,
perhaps related to -ht-, see notes 106, 135. ina-
ht'-c' dwelling place out in the open,
AWAY from the WOODS ?
156. See note 6.
157. feu-, verb-stem fish, sea-mammals
RUN ; dur. t'cu-ak'. -kive-oLii, phonetic variant
of -kw'.-'atf, for -kwUf -\- -atf, see note 5 ;
-kw'.tf, momentaneous suffix, see note 32 ; -'ati,
see note i.
i58.</a/?"-, see note 53. -5flp', momentaneous
causative, see note ^6.
i^^. yaia'mt- =^ yaq{i.u)- -\- " hardening"
-a'n-'f, see note 62. yaqw- (before stem-
suffixes), yaq- (before word suffixes, ci. note
185), general denominating stem for relative
clauses, as is '0- (see note i) for main clauses
and qw.- {sQt note 55) for indirect questions;
it may be translated who, which, what, and
is always completed by a relative suffix (j\t'q\
as here, some form in -qa-, see note 74, or
some form in -t-, -i-, -y., -yc-, see note 73);
examples of stem-forin yaqw- are yaq' -tci- ath"--
'-./'</' next to whom one lives, one's neigh-
bors, yaqw-ats-Ct'q' to whom it belongs, yaq-
tsh'i-it'q' to whom he is married. -a'n-\t-,
form taken by passive -'a/', see note 72, when
combined with -t/-, perfective suffix (for active
perfectives cf. ho'ya'i-it-ah i have been dan-
cing, ha\ud-it'-Wiai' there was a chief, it
is said, ho'ya'i-U-a' he has been, had been
dancing ; for passive perfectives cf. ha'ok'w-a'n-
it-a' it was eaten up); without perfective -'./-
this form would be yalat"it'q' . -\t'q' , see
notes 22, no, 139. ya\a'n<.l'-, as passive, is
analogous to 'jh"'at', note 117; syntactically,
ya'.a'U'.t" it'q' tl'/'w.n^'ap'at' k\vahs<.ts is subject
of qah'sa'p' .
160. Modal adverb ; properly, durative
absolute used as complementary infinitive to
preceding verb, 'yv-qiua' is evidently com-
pounded of 'yo'-, probably petrified demons-
trative stem found only with following qica',
and durative qiva'' to be in quality, see notes
21, 27. It is one of the exceedingly rare cases
in which a primary stem comes in second
position.
161. t'sax-cit{, see note 58. -//^;, see notes 46,
146.
162. bi.n-, see note loi. -'.'p' (after conso-
nants), -'y.p' (after vowels, see note 163),
verbal suffix to get, to obtain (e. g. p'w-i'p'
TO get some bad [sickness], see also notes 165,
I73)j ~'-'P' seems to be momentaneous in
aspect, while -i/>', lengthening stem-vowel, is
durative (e. g. 'ay-cp' to obtain many things
AT A certain time, but 'a'y-\p' to be getting
much right along). -i\a , see note 46.
Eight: Wakashan and Salishan Languages
459
100
163. })io'-, see note 97. -yip' posivocalic
form o{ -I'p', see note 162 (see also notes 171,
172, 176).
164. Durative absolute in -a' (cf. inceptive
'ani'-tciti to get to be the only one, to
ALONE become) ; 'ana- is freely used as verb
stem TO be only, formally analogous to 'aya-
to be much (e. g. 'ana-tsstno\s to be the only
one along the beach, 'a'na-lcii to do so and
so only to, 'a'n-i'S'Cai'Sa' to eat nothing
BUT • . . ) •
165. 5M/V-, verb-stem to be five (form of
stem used before vowels, e. g. sut'c-cq' five-
twenties, ONE hundred, snt'c-i'tss to carry
FIVE objects in THE HAND, Sufc-Q-Irwaf FIVE
ARE MISSING, SUSUt'c-m'k' [r^^-] TO BE FIVE
FINGER-WIDTHS LONG; beforcconsonantal suffixes
it appears as siifca-, e. g. stit'ca-qmi'i five
ROUND OBJECTS, Sllfca-t'siq' FIVE LONG CYLIN-
DRICAL OBJECTS, Sufca-p'ii FIVE LONG FLAT
OBJECTS, siifca-h"tak' five bags full, sufca-to.i
FIVE days, sut'ca-qs to have five [salmonJ in
THE canoe, sulca'-p'ai to carry five objects
ON^ the back), as durative absolute snt'ca' . -cp' ,
see note 162 (but also note 171).
166. tlnp'ctdf TO be summer consists of
verb-stem tlnp- and " hardening " -'etch" ;
t'iup-, verb-stem to be hot (e. g. dur. t'iop-
a' weather is hot, t'hp-a'i object is hot,
iHop'-yvha to die of heat, i. e. to sweat,
t'iop'-qi TO be hot on the top end; less fre-
quently tlom-, e. g. t'io'm-ahs to be hot in a
receptacle, i. e. hot water), -cidf, see note
61. -atl, see note 6r.
167. 'do-y reduplicated form of V, see note
I. -'(.-//', properly -\'h", " hardening " verb
suffix which causes reduplication of stem, to
be in pursuit of, to try to get, tohunt (such
animals or objects as may be obtained in num-
ber) (e. g. tiutiu'y-ch" to go herring-fishing
<C ^/"^-Wl/' HbRRING, tutuck'-i'h" to GO COD-
fishing < tuck-Q-h" cod, t'ioli'ontip"."\.'h'^ to
HUNT elk < I'fo-nun' elk, bihe.'\i'-\'h" to
go for sea-eggs < he-x, stem heww- sea-egg) ;
-\'h" also occurs after both noun and verb
stems, unreduplicated, to mean to try to get,
TO BE AFTER, generally with lengthened siem-
vowel(e.g. 'mz--</fp" to try to get the one
HELD IN the mouth [bY THE WOLf], i. e. TO
attempt to CAPTURE A NOVICE IN THE WOLF
RITUAL, ya't's-i'lf TO TRACK, TO TRY TO GET BY
FOLLOWING IN one's STEPS, he-'y-'.'h" TO BE AFTER
BLOOD < hes-nus blood), also with unlengthe-
ned Stem-vowel, in which case it means to
want (e. g. k'nt's-</h" to want mussels, iiit'c-
ch" TO WANT A woman) ; finally, -\h" forms
derivatives from certain verb-stems with both
reduplication and lengthening of stem-vowel
(e. g. 'na'na't'c-i'h"' to watch for <; 'natc-
To look), -citf, inceptive suffix, see note 18 {c(.
also h.n'-'clf-ati to go after).
168. 'o'-, lengtheijed from V, see note i,
because of iterative aspect (cf. note 50). -wCa'i
iterative form of -zi'i'' to be the first (e. g.
'o-u'c' so and so is first, 'o'-wi-tsaqa' to cook
so and so first, causative 'o-wi'-'ap' to let
so and so be the first), -ap', causative
suffix, see note 72 (all iteratives in final -/ form
their causative in -'ap\ e. g. -o'i-'ap' to cause
repeatedly to be on the rocks, -ap'uU.n'l-
'ap' tocause repeatedly to be under, -a'l-ap'
TO CAUSE repeatedly TO BE OFF, -i'tf-'ap' TO
cause repeatedly into the house).
169. nio''y\p'ati, see notes 16^, i. -//</, see
note 46.
170. zvik-, see note 10. -fnv.'-tC'.-'xli, from
umlauted *-/';/w-/ri// -|- -'all, see notes 79, 5.
-njur', derivative suffix denoting ability
(e.g. u'd'-fnia'' 10 be unable), -tc.tl, in-
ceptive suffix, see notes J09, i2.).-V///, see
note I.
171. snl\\'^yip , parallel to and apparently
entirely synonymous with sut\-<.'p' (see note
165) ; based on sul'ca-, which is lengthened
and umlauted to sut'cc- before -'\<.p' (as before
mom. ox inceptive -tc.ti and its causative
460
Vf American Indian Languages 2
I
lOI
-'yap', cf. notes 79, 109). -'y.p', see note '63.
-ali, see note i. -tin', see note 46.
172. hayiY-, with lengthening of final vowel
before -'_}'-./)' (cf. note 171) from Ipayn-, verb-
stem TO BE TEN, durative absolute hayu' (exam-
ples of stem hayu- are hayu'-if ten-tvv>:nties,
TWO HUN'DRED, hayu-cjiuon'i T?M round objects,
hayu-fs<.q' ten long cylindrical objectsj
hayi{'-p\{ tem long flat objkcts, hciyif-p'ai to
HAVE TEN OBJECTS ON THE BACK, hayu'-tcd TEN
DAYS, hayif-tss to carry ten objects in the
hand), -cdly see note 18 ; form in -cili apparent-
ly synonymous with simpler form without it
(cf. note 176), unless -citi is here perfective :
HAD obtained TEN.
173. 'atiakivaf, numeral eight, literally two
absent (from ten), 'ati-, verb stem to be two
(e. g. ^aff-qmi'f two round objects, 'ati-i'f
to be two in the house, 'ati-h"tak' two bags
full, 'ati-tc'ri two days, 'a'a'ti-alptn' to be
TWO ON the end, 'ati-sa'tso' to be in two
places), before certain suffixes 'atia- (e. g.
distributive 'a'atta-kwai there are two missing
out of each [pile], 'athi-qs-d to be two in a
receptacle in the house), durative absolute
'atia' . -kwai, -kwa'i, verb suffix to be absent,
missing (e. g. qe'-kwa'i to stay away for a
long time, ino-hva'f four are gone, yaq'-
hua'l-\t'q' WHO is absent, Ihta-hcai-c.tt to get
TO be away, to go away, 'a'ya-kwai-atf-sa'
TO get to be too much-missing, to spend too
much) ; analogous to 'atia-kivat eight is t'sawa'-
/cZffl/ONE absent (from TEn), NINE. -'.•/)', seC
note 162.
174. '()-, see note i. -yoaf, see note 65. -■./',
perfective suffix, see note 159. -'-., nominalizing
suffix, see note 9. In apposition with Ica'ats'iF.
175. See note 67. Formally, 'oh" is comple-
mentary infinitive to 'oyo' a ht"r. ; it serves to
emphasize tca'ats'^'F in contrast to k'lval'S'Js :
he for his part.
176. See note 172. /vrv/r- merely phonetic
variant of hayo'-.
177. ha'o-, verb-stem TO do in turn, to pass
ON to the next, metaphorically to take
revenge (e. g. iterative ha'tlho'-qsa-tc genera- ,
tion passes after generation). -kw<-'ill, for I
-kwi-ali,see notes 157, 32, 5.
178. satck-, verb-stem to be sharp (e. g.
red. sasalck"-wi to be sharp-clawed, salck-
aq'sni to be sharp-toothed, sa'.ck'-ahs comb,
literally being sharp inside), here metaphoric-
ally for strong, -ok', see note 10. -fcac,
modal verb suffixdenoting INFERENCE, it seems,
here ironically ; consists of -fca- and -c, of
which -fca- is inferential (^d. nominalizing
-fca', note GG^ and -c emphasizing (e g.
satckok' -si-c mine is indeed sharp). The pro-
nominal paradigm for -fca-c (ci. interrogative
paradigm in note i) is :
Sing. I. -fca-si-c
2. -tca-k-c
3. -fca-c
Plur. I. -fci-n'-c
2. -fca-so'-c
3. -fca-c{-ai)
179. See note 54.
iSo.'a'qvi'qha, (or \r qui' -qh-ha/ a' q- length-
ened from 'aq- (before vowels), 'aq\-, 'aq\'-
(before consonants), interrogative verb-stem
what ? always followed by interrogative -ba-
it, g. dur. absolute 'aq-nk'-ha' what is it ?
'aqi-yJftaqak'-ha' whAT is it made of ? 'au'qi-
yuk'-ha-k' why do you cry ? ' a' q\.-f sus-ha-k"
WHAT ARE YOU LAUGHING ABOUT ? 'uqi-sk'-ha'
WHY DID HE GO AWAY ? 'aq-i' S-kl-k' WHAT ARE
YOU EATING ? ' o" aq-0'' i' s-ha-k' what are you
SAYING ? 'aqi-yuai-fpa-k' what do you see ?
'aqe'-yun-'J-ha-k' what did you find ? 'aqi-qs-
ha-k' what [game] have you in your canoe ?
'aqe'-s-ha' what is on it ? 'aqe'-tciclnti-ha' what
CAME TO BE ON THE SURFACE OF THE WATER ?).
-in'-, verb suffix lengthening stem-vowel,
probably TO be because of, cf. causative 'a'qm'-
'ap-ha-k' what are you doing ? (this stem
lengthening -m'-, which does not seem to occur
freely, may be related to postvocalic -fio'ti,
postconsonantal -w'ti, with movable momen-
Eight: Wakashan and Salishan Languages
461
102
taneous-//, TO BE BECAUSE or, e. g. 'o'->iLr-'<.s-
'a'ia' DO so to me customarily for the rea-
son THAT. . \cjiL'a'-no'li-il-ah for that reason
I WAS, yaqiu-\yi' ll-\t' q' for which reason
HE. . .). -qh (after vowels and h), -h- (after
consonants), see note 55. -ha, interrogative
suffix, third person, see note 120. 'a'qui^qh{fp)a
would seem to mean literally for what reason
IS he while (loing so) ? i. e. why does he,
why did he. . . ? -qh- dnncipaung'atlakwaf'/ p' ,
which follows complementary infinitive.
181. See notes 164, 74.
182. ]ju\i-, broken from "."o'^yi-, see notes
179, ^^.-woiti, i. e. -zy^// (after vowels), -awalf
(after consonants), verbal suffix which lengthens
stem-vowel (so and so) falls to one's lot,
TO have something come to one (e. g.
'o'-wati it is so and so that falls to
one's lot, lui'k-aivali to have nothing come
to one); los<is -// before -'ati and analogous
elements (see note 5), e. g. ivi-k-awa-atl.
183. See notes 172, 176. -'atly see note i.
-s:, first person singular subject of absolute
paradigm, see note i.
184. Independent first person singular
pronoun, here used because of emphasis. si\a-
is also used as base in certain verbal forms, e.
g. si'ya'-q- to be i, si'va'-s- to be mine (gene-
rally with repeated first person singular prono-
minal suffi.x, e. g. sfya'-q-ah it is i, literally']
AM I ; sCya'-s-ah it is mine, literally i am i-
HAViN'G ; causative si'ya\'~ap'-is let it heme!
literally cause me [-5J to be 1). si-'ya' is evi-
dently based on absolute -sk, sometimes reduced
to -s (see notes 183, i); si- is freely used as
stem in derivatives, of which si-'\a' is really
one (d. si'-tcd to do to me, sv-altip' to do
FOR me, si'-putf-qa-s that iget to be i who am
erect, that I HAVE TAKEN FIRST RANK; -pUtf is
momentaneous of -/)•.' stem-lengthening suffix,
see note 90). Parallel to sfya' , si- are so-'wa'
You{so'^iva'-q-e\ts it is Yon, so'^wa'-s-tUs it is
yours), sot- (e. g. so't-'J to do to you, siit'-
nak' TO HAVE you [as chief]) ; ni-'zva' we
(niica'-q-in' n is we, iii'wa'-s-m' it is ouks),
ni^h)- (e. g. rtch-zi to do to us, nOp-a-'atup'
to -make for us) ; si'^wa' ye {si^iua--q-z\lso''
IT is ye, si'u'a--s-t'<.tSLr' it is yours), s:h- (e. g.
si/h zf to do to you, S'.'h-a''atup' to do for
you).
185. yaq-, see note 159; note that absolute
yaq-, not stem-form yaqw-, is used because it
is followed by a word-suffix (relative -<'-),
not a stem-suffix, -e'-s, relative suffix with
first person singular subjective element, d.
note 73. Paradigm :
Sing. I. yaq-e'-s
2. yaq-e'-k'
3. yaq-e-'
Plur. I. yaq-i-ti'
2. yaq-e'-so''
3. yaq-f{-'af)
Quotative -tc- {d. notes 66, 178, 58) fre-
quently unites with subordinating suffixes (e. g.
interrogative -ha-tc he, as they say. . . ? \jn'.-
t'ci-ii' THAT WE, AS IS SAID,. . " ; -qa-ts-s that
I, AS THEY SAY ; -qO'-tC-k' IF YOU, AS IS SAID,...).
For relative forms o( yaq-we have, as quotative
paradigm :
Sing. I. yaq-e'-ls-s
2. yaq-c-tc-k'
3. yaq-c'-lc
Plur. 1. \tiq-f-t'r:-n'
2. yaq-e'-ts-sc'
3. yaq-f-!c(''al)
Other examples of first person singular re-
latives in -r.f, -yis, -v/'.fare \^\ nial'-iifl-i's rmi-
TIME THAl-I-FLE\V-OFF, WHEN I 1 1 K\V OFF ;
hamnCamah qiui-ycha-yis i-know of-what-i-
DiE ; 'oy\. iiinla'-yi's when i was flying.
1 86. u'rt"-, see note 122. -a//, see note i.
462
VI American Indian Languages 2
Editorial Note
Originally published in International Journal of American Linguistics 3,
76-102 (1924).
Sapir s ms. corrigenda on his copy are as follows:
Original
p. 77, col
78, col.
78, col.
79, col.
79, col.
81, col.
82, col.
83, col.
83, col.
83, col.
84, col.
84, col.
84, col.
85, col.
1, I. 14
1, 1. 3
1, I. 28
2,1.4
2,1.8
2,1. 2
1 , last
1, I. 27
1,1.39
2,1. 15
1, I. 34
2, 1. 17
2, I. 23
1,1. 9
p. 85, col.
p. 85, col.
p. 85, col.
p. 86, col.
p. 87, col.
p. 88, col.
p. 88, col.
p. 88, col.
p. 88, col.
p. 89, col.
p. 89, col.
p. 89, col.
p. 89, col.
p. 90, col.
p. 90, col.
p. 90, col.
p. 90, col.
p. 90, col.
p. 91, col.
p. 91, col.
p. 91, col.
p. 91, col.
1,
1. 16
1,
1. 27
2,
1. 26
2,
2nd last
1,
1. 14
1,
1. 15
1,
1. 17
2
1. 8
2
1. 39
1
1. 9
1
1. 9
1
1. 10
1
1. 25
1
, 1. 4
1
, 1. 5
1
1. 22
1
1. 34
1
1. 42
1
, 1. 2
1
1. 36
2
, 1. 5
2
, 1. 7
For:
Du'yLtci-tcL'ett
t'sa'ak't'
qa'yap't'at'i'
sa'tckok't'cac
Cha'atssib'.
'i-h^to-p'
', k',
surernatural
'o-sim'tcatt
(see note 1).
'anah^-'is
wik'ntt'
note.
ma' redheaded
woodpecker
*tLq'w-as
'nat'c-
cont
'ayaqa
to
t'lt'.a-tCL'
'noq'"-t'sa\atci
side end
sa-WL^att
'Du'yt'
.o-'yt'
and .
qwi'yi...
tla'o-
tta'o--yiy-a'
hayu.-ah''s
'anah-'
q'tl-'nukw-'ts
at one
who suspected
-L-
2. 'ant-so-
Read:
'6u'yLtci-lcL'ett
t'sa'ak'\'
qa'yap't'af'i'
sa'tckok't'cac
Cha'atssib',
'fh^o-p'
t', k',
supernatural
'o-sim't'catl
see note 1,
'anah'*-'LS
wik'-ntt'
note
ma' heart, t'leh'^ma' red-
headed woodpecker
*tiq-'w-as
'nat'c-
con-
'aya-qs
to
t'k'.-a-tct'
'nDq'"-t'sa'.- atci
side, end
sawt'-atl
lou'yt'
lo'yi'
and !
qwL-'yi...
tla'o--
tla'o-yiy-a'
hayu!-ah^s
'anah-
q'tl-'nukw-'is
at once
who is suspected
-L--
2. 'ant-so-'
Eight: Wakashan and Salishan Languages 463
Original
For:
Read:
p. 92, col.
2, 1. 47
most
-most
p. 94, col.
2, 1. 21
'ttq-
't\q-
p. 94, col.
2, 1. 21
'ttq'w-
't\q'w-
p. 95, col.
1,1. 8
sa-'-ok'"
sa-'-'ok'"
p. 95, col.
1, 1. 36
LS
-LS
p. 96, col.
1, 1. 1
-um'h
-um'h-
p. 96, col.
2, 1. 23
act
acts
p. 97, col.
2, 1. 16
tkh'«CL^etrttq'
tteh^CL^etra'q'
p. 98, col.
1, 1. 19
tittti-y-a'
tittti-y-a'
p. 99, col.
1, 1. 5
'at'axw-a'aq'
!at'axw-a'aq'
p. 99, col.
2,1. 10
yaqw-ats-i't'q'
yaqw-ats-'tt'q'
p. 100, CO
. 1, 2nd last
t'lott'onup'.'^L-h^
t'lotl'onup-'.'^L
p. 100, CO
. 2, 1. 28
-'a-l-ap'
-'al-'ap'
p. 102, CO
.2,1. 10
sih el
sih-el
Nootka Baby Words
The Nootka Indians have a number of words which are used only by or in
speaking to infants and Httle children. Some of these are based on the regular
vocabulary, others are quite unrelated to any other words in the language, so
far as known. The following examples of these baby words were obtained from
an Indian of the ts'ica-'ath tribe of Nootkas, originally belonging to Barkley
Sound.
'o.'ec 'dirty! don't do it!' (a warning to a child that handles dirt); normal word
for 'dirty' is tci-c.
hox, exclamation to scare off a child that is in the way.
h-h- 'white man' (often used to scare a child, like our "boogie man"; this
word is said to imitate the white man's talk, and note that / is not a normal
Nootka sound, though often used in songs for n)\ normal word for 'white man'
is mamal'ni 'dwelling while in motion on the water'.
ho-c 'to sleep'; normal word we'itc.
ka-x- 'sore, hurt' (used by children as a conjugatable stem, e.g. ka-xatah T
am hurt'); normal word ya-'kat\ stem ya-kw-.
ma-h 'to drink'; normal stem naq-.
ta-'ta' 'father' (said to be the first word pronounced by a child); normal word
'nu'wL''q'so\ vocative '/7D-"wt'.
ma-'ma' 'mother'; normal word 'um'\-'q'so\ vocative '3"mt'.
ma-'mac 'to begin to suck', based on ma-'ma\
pa-' pa' 'toeat',/7fl-'/?ac'to begin to eat'; normal ha'ok'" 'to cat' (stem lui'w-),
ha'ok'"-citt 'to begin to eat'.
nan-Cc 'grandpa, grandma'; based on normal luuiL-'q'so' 'grandparent',
vocative /ie-/i', irregular possessive nan-V 'your grandparent'.
'ma-"mic 'older brother, sister'; based on normal ^nia-'mL-'q\s()\ irregular
possessive 'ma-"nn' 'your older brother, sister'.
'yukwic 'younger brother, sister'; based on normal \'uk\\'LUfs()\ irregular
possessive 'yukwC 'your younger brother, sister'.
A few linguistic remarks may be of some interest. The Icmhi '.^.'er "dirty!'
reconstructs to Wakashan *'oijex, while the normal /c/c 'dirty, dirt, excrement"
goes back to *ki-x. perhaps palatalized and lengthened from * kcx. Now it seems
very likely that the Kwakiutl 'Aa- (in Boas' orthography k!ix-), palatalized
from 'A:9x (k-fEx--), 'to defecate (dog salmon or halibut)', is related to Notuka
tci-c. The consonantal alternation of k : 'k is parallel to that of r/ . q in Nootka
^/o/;-; Kwakiutl qo-p- (qfop-) 'to copulate'. The old ' -r/rv of Nootka '.">.'ec may
be a sort of emphasizing velarization of Wakashan ' kcx (Kwak. 'Aa-)- If this is
so, Nootka 'o.'ec and tci-c would be related words.
466 VI American Indian Languages 2
It is even more likely that ho-c 'to sleep' and its normal correspondent we\tc
are related. The latter goes back to *waik, while the baby word may reconstruct
to *haux, reduced from *hawax.
The -c of ma-'ma-c 'to begin to suck' and oi pa'pa-c 'to begin to eat' is
reduced from the -ci- of the normal postvocalic inceptive and momentaneous
suffix -ci-tf (({ drops before certain word-suffixes, e.g. -ci-\iti). In normal
Nootka this -c is used in reduplicated verb [ 1 19] forms to express the "momen-
taneous iterative" aspect, -a' being used in corresponding "durative iterative"
forms (e.g. 'mitrmitl-c 'to rain repeatedly by brief spells', 'mi-ti'mitl-a' 'to rain
continuously time after time').
The -c oinan-i-c, 'ma-'^mt-c, and 'yti/cwi-cis apparently a form of diminutive.
The normal diminutive suffix is -'ls (-'its-). Perhaps -c is a baby-form of the
normal -s.
Editorial Note
Originally published in International Journal of American Linguistics 5,
118-119 (1929). Several diacritic marks have been added, following Sapir's ms.
notes in his copy.
There is a substantial recent literature on baby talk, particularly concerning
the relationship that baby talk and other "motherese" registers have to lan-
guage acquisition. For surveys see Ferguson (1964) and Snow and Ferguson
(1977). Kess and Kess (1986) have recently supplemented Sapir's Nootka study
with observations on baby talk in the Ahousaht dialect (only minimally distinct
from the Tsishaht described by Sapir). They were able to collect 19 fully sup-
pletive baby talk words and several reduplicative derivations. It is noteworthy
that several of the baby talk words cited by Sapir as being common in Tsishaht
in 1910-14 are not in use in modern Ahousaht.
Salish-Wakashan Comparisons
Edited by Morris Swadesh
These twenty-six Salish-Wakashan comparisons were found in the late
Edward Sapir's materials on small cards, preceded by a guide-card, simply
labeled "Salish and Wakashan." Eighteen of the items involve only one Salish
language, namely, Comox, along with KwakiutI or Nootka of the Wakashan
group; the other eight items involve two or more Salish languages. It is likely
that the Comox comparisons were a by-product of Sapir s personal study of
Noun Reduplication in that language, published in 1915 as Memoir 63 of the
Geological Survey of Canada. These comparisons include a few which could
easily be due to recent borrowing, and others which hardly could have arisen in
that manner. The words cited in Sapir's Comox paper include at least a few
additional obvious similarities with Nootka, indicating that Sapir did not make
a systematic search. It is important to note this fact in order to [172] realize that
the present comparisons are a casually collected sample rather than the full
measure of the Mosan theory. I have simplified the spelling along phonemic
lines, following recent practice. Salish forms are listed first and Wakashan
forms after the Salish. The spelling of languages is modernized, but Sapir's
abbreviations are retained. "T.R." is evidently Thompson River; "Hail." is the
Hailtsk dialect of KwakiutI. B. evidently indicates Boas, as a source.
— Morris Swadesh.
Salish -s 'his'. Kw. -s 3d. per. vis. pron. suffix, postnominal 3d. per. vis. pos.
suffix, poss. and instr. suf. ; N. sij'd-s-an 'it is mine'.
Bella C. ia 'good' (B.) (= ya'?), Satl. ai (B.), Pent, aim (B.), Sis. ai (B.),
Snan. a/(B.), Lkun. aii(B.), TR. ia (B.) (= ya?, cf. Shusw. la); Kw. 'aA,- 'good'
(B.), Hail. aikH (B.).
Bella Coola Hmltl 'we'; Satl. -Pent. -Sis. ni mi'itl; Snan. ti thiimitl; Skk. //
nimatl; Lkun. tlningitl. N. nlwa 'we'; Kw. nu T (B.), -^nux^^ 'excl.' (B.).
Comox jigin 'song', < *ydwdn; Bella Coola nu yam 'to sing' (B.); Pcntlatch
lulum 'to sing' (B.). N. nu- 'to sing', nunuk 'to sing', nuk 'song', nuydwa 'to be
singing'; Hail, ninuya 'to sing' (B.).
Shuswap -kin 'head' (B.) (= -qin)\ Okan. -dyaksn (B.) (= -qsn): Comox
md-qin 'hair'; Pent, sqikin 'hair' (B.) ( = sxiqin); Snan. smdqin (B.); Lil. mdqin
(B.). N. -(^ri" 'head', causative -qinup; Hail, -kia (B.) ( = -qia).
Comox susln 'mouth'; Pent. -Sis. susin (B.); Snan. -Lkun. sosin (B.); Skk.
cuc^n (B.); Lil. cucin (B.). Kw.-Hail. ssms 'mouth' (B.).
Comox higus 'chief; Pent. -Sis. hiwus (B.). Hail, himas 'chief" (B.).
Comox waxdti 'pipe'; Pent, waq'ac^n (B.). Cf. Kw. wax'aci 'pipe, smoke
receptacle' (B.).
468 Vf American Indian Languages 2
Comox iaq"^ 'bow\ Kw. hk""- 'to pull', hk'^isi 'bow'.
Comox tipatil 'bag'. Kw. ftabat 'basket'; N. t'iapdt 'big cedarbark basket for
storing dried tish'.
Comox mafdy 'horse clam', Kw. ma/- 'large clam'.
Comox mus 'four'. N. mu, Kw. mu.
Comox pigdy 'halibut', < *piwdy. Kw. po'i 'halibut'; N. pui.
Comox q'lgas 'deer'. Kw. cAwas 'deer'.
Comox qdsa 'sea-otter'. Kw. qds- 'sea-otter'. [173]
Comoxsdcam 'tyee salmon', ^a 'a/7 'cohoe salmon' (dim. si-sad-ul, dim. plur.
si-su-sii'ad-id). N. sacup 'tyee salmon', 5flcm 'small salmon'; Kw. sac^m 'spring
salmon' (stem sas-).
Comox tdq^a 'devil-fish'. Kw. t^q''' 'octopus'.
Comox tictids 'owl'. Kw. d^xd^xdlit 'owl' (perhaps borrowed from Comox
with folk-etymological -Id).
Comox tu-tx^lal 'necklace' (dim. plur. titax'''tutx'''lat). Kw. t^x^- 'to fasten';
N. tux'^'i 'fastening strings in cradle-board'.
Comox fakum' 'beaver'. N. 'afax^'- 'beaver'.
Comox xdp 'baby basket'. Kw. x^'dp 'cradle'.
Comox xasam 'box'. Kw. xac^m 'box'; N. napim 'bailer', < *xas-'im 'little
box' (?).
Comox /c wJMr"porpoise'. Kw. /:'w/wr"porpoise'.
Comox A:/r"little finger'. Kw. k'dd- 'third finger' (Kw. s^lt'- 'little finger').
Comox k'ikdyu 'oar' (Kw. loan-word with instrumental -ayu). Kw.
k'ik'^w^lx'^'^n 'bar'.
Comox laqas 'mt. goat blanket'. Kw. hq- 'to put down soft things'.
Editorial Note
Excerpt from Swadesh, Morris, "The Linguistic Approach to Salish Pre-
history," in Smith, Marian W. (ed.), Indians of the Urban Northwest, 171-173.
New York: Columbia University Press (1949). Reprinted by permission of
Columbia University Press.
Section Nine:
Other American Languages
A TUTELO VOCABULARY^
WHILE on Six Nations Reserve, Ontario, in August, 191 1,
I was told of a Cayuga Indian named Andrew Sprague
who had had opportunity during his childhood to hear
Tutelo spoken fluently and who was supposed to remember con-
siderable of it. As Tutelo is an extinct language, I thought it
imperative to rescue from oblivion what was still to be obtained and
thus add, if only a mite, to what had already been put on record.
As a matter of fact, it turned out that Andrew remembered only
very little indeed of Tutelo, and what small amount of material
could be obtained from him was extorted with some difficulty.
No attempt will here be made to discuss the data. They are given
for what they are worth in the hope that they may at some future
time prove of use to the student of comparative Siouan linguistics.
If in nothing else, perhaps the words listed are of value because they
have been recorded with greater phonetic accuracy than is generally
attained in mere vocabularies.
Persons
mihqii) stjk'' man niska' child, children-
wdrewd^ woman kukdk^ grandfather
marjgidd^ my cousin
urjgidd'^ his cousin
Animals
map^ayerjk^ cow, ox, cattle ts^u-qgide^ horse (literally, "big
ts^urjk^ dog dog")
mqsgolp"^ pig ts^mjgnve'^ wolf (literally, "like a
hemp frog dog")
dalusgik'' cat
babosgp'° deer buck
Objects
maksdpd'"' bread mdO^'" knife
* Published by permission of the Geological Survey of Canada.
295
472
VI American Indian Languages 2
296
matsiggyg salt
bebahe'". pepper
wddeztn sugar
wohe'" soup
wdsakse'^ shorts in grinding corn
gdk^saginagg'' silver armlet
Adverbs, Adjectives
ddbada''^ down river
udaphdi''' up river
btwd good
nyAgutsgA.htli' dizzy woman
nihetsgah^li' dizzy man
Pronoun
hentgu I
Verbs
u'dluC maksapd'"^ eat bread
wdgetfi'^ to dance
hadit^ gileda he's gone home
weCgil'lda let's go home
Numerals
1, bigs
2, ngs
3, nd
4, til
5, ^^
yqhe'"- lacrosse stick
wahgk^mf^ false face
meni'lgdCe'q''^ kettle (evidently con-
tains meni- "water")
meni'lgodg jug, jar, glass, bottle
(dit.)
kdembdi'^ pipe
kcembdi ngni' tobacco
mg.^ gang,^ ga'°' white
mihg mg^'ganq'ga''^ white man
mg'gang'kasit'' negro
givA^giltda I must go home
wdk^niendabewd I'm going to bed
bllahuk^ thank you!
h^h§ giddya k^ek^iik^arewaya surely,
everything is all right
6, agds
7, sakii
8, pelqk'
9, s^k'
10, bitskd^
Phonetic Note
Short vowels
a, as in German Mann
a, as in English hat
e, as in English met
e, as in French ete
Long Vowels
o, as in German Kahn
e, as in German See
i, as in German Sie
i, as in English bit
i, as in French fini
q, as in French beau
u, as in English put
6, as in German Sohn
u, as in German gut
Nine: Other American Languages 473
297
Nasalized vowels
g., as in French quand
4, as in French vin
§, nasalized open e (not as open as in French vin)
g, nasalized close (0), not open as in French bon
ji, nasalized open u
q, long g,
g, long g
Diphthong
ai, as in German mein
Stopped Consonants
b, d, g, sonant stops as in English
p, t, k, tj, "intermediate" stops
p, -, k, ts, unaspirated surds
p\ t\ k\ ts\ aspirated surds
ts\ aspirated ts (midway between ts and tc of English church)
Continuants
w, as in English
y, as in English
h, as in English
s, as in English
c, as in English she
6, interdental spirant, as in English thick
I, (approximately) as in English
r, trilled r (alveolar?). In k^ek^^ik^arewaya r sounded much like /
m, as in English
n, as in English
t], as in English sing
Miscellaneous
', glottal stop
-a'" (and similarly for other vowels), broken vowel, second part of
which is murmured ("echoed")
*, final breath release
*, occurs once as glide vowel from c to e
', main stress
^, secondary stress
Geological Survey of Canada
Ottawa, Ontario
474 VI American Indian Languages 2
Editorial Note
Originally published in American Anthropologist 15, 295-297 (1913).
Reprinted by permission of the American Anthropological Association.
Tutelo, Ofo, and Biloxi have been grouped as a distinct branch of Siouan,
called Ohio Valley or Southeastern Siouan (Voegelin 1941; Rood 1979:
243-248). None of these languages is well attested, and every scrap of data on
them is valuable for comparative work. Sapir s vocabulary, although brief, is
especially valuable for the accuracy of its phonetic transcription. Although
Sapir believed Tutelo to be extinct in 1911, Frachtenberg (1913a) published
some Tutelo material that he obtained from a speaker in 1907, and Mithun
discovered a speaker at the Six Nations Reserve as late as 1982 (Mithun 1983).
i
Review of
Benigno Bibolotti, Moseteno Vocabulary
and Treatises
Moseteno Vocabulary and Treatises. Benigno Bibolotti, Priest of the Fran-
ciscan Mission of Immaculada Concepcion de Covendo in Bolivia. From an
unpublished manuscript in possession of Northwestern University Library.
With an Introduction by Rudolph Schuller. Northwestern University: Evans-
ton and Chicago, 1917. pp. cxiii, 141, facsimile, map of Bolivia.
The external facts leading up to the publication of this sumptuously printed
volume are given by Dr. Schuller in his preface: "Northwestern University
Library possesses a fairly large collection of unpublished Spanish manuscripts
which are probably unique in the United States... Professor Lichtenstein,
Librarian of Northwestern University, acquired this material, consisting of
books, pamphlets, early periodicals and the like, from Sehor Donato Lanza y
Lanza during a sojourn in Bolivia. In September, 1916, Professor Lichtenstein
asked me to arrange and collate the manuscripts and prepare them for the
binder. While examining the different packages in order to make a preliminary
selection of the papers according to the subjects treated in them, I found
Bibolotti's manuscript dealing with the Moseteno language. The unexpected
discovery is all the more important since it concerns extensive materials
gathered together by a yet unknown author of a relatively little studied Bolivian
aboriginal idiom spoken by Indians who have almost vanished. If there are still
a few of them remaining without foreign admixture, they are destined to be
absorbed completely in the near future by the process of amalgama-
tion... Within a few years the name of the Moseteno will be added to the alarm-
ingly long list of extinct South American Indian tribes." The manuscript is the
work of an Italian Franciscan, concerning whom very little is known: it was
written some time between 1857 and 1868.
The Moseteno, also known as Chumanos or Chomanes, are or were one of
the Andean tribes of western Bolivia; their territory was embraced within the
present province of Yungas. More exactly, to quote from Dr. Schuller. "the
habitat of the Moseteno-Chumano embraced the mountainous regions to the
east of the Beni, more or less between 15° and 16° south latitude, and 69° to 71°
longitude west of Paris. Their eastern neighbors were the Yurucare; in the
north they reached as far as the territories occupied by Mobima and Moxo, or
Mojo, tribes, and in the northwest they touched lacana and Leco speaking
peoples. The natural border to the south and the west is the range of the higher
Andes."
476 VI American Indian Languages 2
Dr. Schuller s editorial work has been most painstaking, and the volume is a
highly welcome addition to our knowledge of the exceedingly tangled and
obscure problems of Bolivian linguistics. In his lengthy introduction Dr.
Schuller discusses first the manuscript; the author; the Moseteno Indians and
the Franciscan Missions ("in spite of uninterrupted intercourse for many cen-
turies with the more highly developed culture of Peru and Bolivia, the primitive
tribes, like the Moseteno, Tacana, Leco, Araona, etc., were not much influ-
enced''); and gives a critical analysis of previous writings on Moseteno. Pages
xxviii to xcv of the introduction give a digest in English of our present knowl-
edge of Moseteno, as based on Bibolotti and other writers (Wendell, Heath,
Armentia). This section includes notes on phonetics; vocabularies; gram-
matical processes (nouns: number, gender, formation of nouns, grammatical
cases; adjectives; pronouns: personal, possessive, relative and demonstrative,
indefinite, interrogative; numerals; adverbs; prepositions; conjunctions; [184]
verbs: verbal stems, classification of verbs, tenses, the imperative mood, the
participial mood, other moods); and general observations on suffixes,
reduplication, and affiliated languages and peoples. Of the three appendices,
one is devoted to a full bibliography of manuscript and printed sources. The
body of the work is a transcript of Bibolotti s Spanish text (Spanish-Moseteno
vocabulary and supplementary papers).
A few of the more interesting points may be noted here. Sex gender is indi-
cated in nouns and adjectives by distinctive suffixes (e.g., izanqui-t "baby boy':
izanqui-s 'baby girl'; moci-t 'new' m.: moci-s 'new' f.). There is a genitive suffix
in -s or -si, also a number of local case suffixes. The curiously widespread Amer-
ican second person singular in m- meets us here once more {mi 'thou'). Pro-
nouns are not welded with the verb stem, but occur independently (e.g., ye
queti 'I plant'). A considerable number of verbal suffixes have been isolated by
Dr. Schuller, but more intensive study of Moseteno, at first hand, if possible, is
needed to make clear their functions. Phonetically, Moseteno would seem to be
"far from agreeable to the ear"; it has many "clusters of totally heterogeneous
consonants." In this respect it differs from Tacana, Cavineno and other lan-
guages of the Bolivian highlands, approaching the "Chaco-Guaycuru linguistic
family, although it does not have the slightest affinity with the latter." Nev-
ertheless, Dr. Schuller finds that "the morphological and syntactical structure
convey the impression that the Moseteno is related to the Tacana group, and
particularly to the Cavineno."
Dr. Schuller leaves no doubt of the thoroughness of his task, and students of
American linguistics owe him a very real debt of gratitude. Perhaps one maybe
pardoned, however, for expressing the wish that penetrating first-hand
phonetic and morphological studies of a number of South American languages,
of a standard corresponding to some already accessible for certain North
American languages, be vouchsafed to us in the course of time. These inter-
minable vocabularies, grammatical notes, and classificatory speculations are,
let us hope, but the harbingers of more substantial meals.
Nine: Other American Languages All
Editorial Note
Originally published in International Journal of American Linguistics 1,
183-184(1918).
Moseteno is a linguistic isolate, although Greenberg (1987: 13-13) includes it
with Panoan, Tacanan, and a number of smaller groups in his "Macro-Panoan"
phylum. Sapir s passing remark on "the curiously widespread American second
person singular in m-" (p. 184) is his only published allusion to the possible
linguistic unity of all (or nearly all) American Indian languages; see his
unpublished notes on "proto American" in Golla (1984: 452), and a remark in a
letter to Frank Speck cited in Darnell and Sherzer (1971: 27). Greenberg has
cited m- 'second person', together with/?- 'first person', as evidence for a hemi-
sphere-wide genetic group, "Amerind" (1987: 49-57).
Appendix
Review of
Edward Sapir: Notes on Chasta Costa Piumology
and Morphology
A. G. Morice
Dr. E. Sapir achieved lately what the French could not help calling a veritable
tour de force. Enlarging upon linguistic material incidentally derived from an
Indian, mere bits of an aboriginal language which would not fill one common-
sized page, he managed to write in explanation of the same no fewer than sixty-
seven pages of first-class philological literature. His Notes on the Chasta Costa
Phonology and Morphology, are perfectly illuminating, and betray not only a
very keen ear but a quite creditable analytic acumen. After thirty-two-years'
study of the Dene group of languages, to which his "Chasta Costa" belongs, I
am tempted to pronounce Dr. Sapir's essay one of the most satisfactory mono-
graphs of its kind ever issued on any of the southern Dene languages.
The work does not pretend to be more than mere notes, and so far as com-
pleteness is concerned it could not compare with more elaborate productions
already published on the Hupa and Navajo dialects. Some might also object to
the graphic signs the author has resorted to in transcribing his texts and disap-
prove of the strange appearance which sometimes results therefrom. His own
language might furthermore have been simpler and less Hellenic or Latin in
complexion. But I maintain that, with few unimportant exceptions, he has
grasped and faithfully rendered not only the phonetics but the morphology of
an idiom whose intricacies must be above the average, unless it be not Dene.
I note with special satisfaction in his paper those particular sounds, such as
the aspirated r s and A: s, as well as the lingual and glottal explosions, which I had
always thought, and sometimes asserted, must exist in the southern Dene lan-
guages, in spite of the inability of former students to perceive thcni or oi their
carelessness in noting down their texts — a presumption for which 1 was e\en
taken to task.
These are as many essential points of the Dene phonetics, so very essential.
indeed, that I felt they could not possibly be missing in any dialect clainnng
relationship to the Dene languages of the North. In Dene the vowels are the
flesh of the body: they vary according to the dialect in the same way as the Hesh
is different in quantity or texture according to the individuals. The consonants
are the bones of its makeup, therefore much more important, ht)mogeneous,
and persistent, while the grammar may be compared to the arteries, without
which blood, that is life, could not circulate therein. Hut the "clicks" which
affect letters or groups of letters are the very nerves w liich alone enable the
Dene body to stand.
482 ^f American Indian Languages 2
I have so often insisted on this point that this simple remark must suffice. I
may nevertheless be pardoned for confessing a feeling of satisfaction at seeing
my contention of former years borne out by the researches of the latest investi-
gator in the southern field.
Only a very few hiatuses seem to have escaped the notice of Dr. Sapir, and I
am all the more free to call his attention to this point, as throughout his paper
he seems animated by that sense of diffidence which behooves a genuine
scholar who enters a new field with the knowledge that he is more of a phi-
lologist than of a linguist. Moreover, a mere passing acquaintance with a lan-
guage cannot, of course, shield one against possible oversights.
I shall therefore make bold to remark that in Dene the desinential radicals of
the verbs of vision (--i, --in, - -en) areimmediately preceded by a stop, or hiatus,
which prevents them from being merged into the consonantal element of the
pronoun. Thus I do not think I am mistaken when I observe that in neii{\), 'you
are looking at him' (nil-en in Carrier), the desinence -/"must be separated from
the preceding / by a hiatus, which should be shown on paper as prominently as it
is expressed by the native speaker.
The same remark applies to yiin (Carrier yi-en), at, 'wife' (Carrier and
Chilcotin -at), as well as to the verb aIaz, 'he sneezes,' which, barring the
medial / and final z, is the exact equivalent of the Carrier cei-ces (same
signification).
This hiatus plays a double role in the North. While at times it merely cuts
asunder, as it were, articulations which would otherwise coalesce into one
sound (nii-en, i.e., nil + en, not ni-len), or is prefixed to monosyllables often
expressive of distance, remoteness, or even repulsion {cen, Chasta Costa An)\
it also denotes the disappearance of a weak vowel through contact with a
stronger one. Thus the indicative present of the verb 'to work' is ces-en in Car-
rier, and, normally, its pronominal element ces should be developed into cezces
in the negative. But the e of the negative particle ie is stronger than the initial ce
oi cezces; hence the negative of<^5reAz is merely /e-z^sre/t, the hiatus (•) standing
here for the vanished ce.
The same happens even when a stronger desinential vowel of a word comes in
contact with a weaker initial vowel of another. Example: spa- hutqa huni, I am
lucky indeed' (for .vpa cehutqa huni, literally, me-for it-has-happened it-is-so).
The consonant q reminds me of Dr. Sapir's dj. If my own letter represents the
same sound which that gentleman has in mind when he uses his double conso-
nant, I must be allowed to object to the latter as misleading. Pronounce it as you
will, you are bound to have a double operation of the tongue and mouth when
you utter the sound dja, the dental one being always distinct from that caused
by the fricative/, whilst in pronouncing the sound I render by q but one opera-
tion is needed.
Were one bent on ultra-criticism, he might remark that in Dene all such parts
of the body as are naturally twofold are normally dual in meaning, the singular
being formed by suffixing a syllable which is [350] generally synonymous of our
word 'half (i.e., half of two = one). According to this rule Dr. Sapir's /zwa must
Appendix 483
not mean foot, but feet, in the same way as his la should be translated hands,
not hand.
Dr. Sapir s analysis of the short text which closes his essay is simply admira-
ble, and betrays an insight into the morphology of his material which one is at a
loss to know where, or how, it was acquired. Scarcely more than one criticism
have I to offer on this part of his paper. The last syllable of the compound do- at-
ti is not a "noun suffix," as he believes (30). It is a regular verb, or rather a
verbal stem, since the pronominal element of the same has disappeared
through the process of word formation. Do is the Carrier negation an, the
Babine so-, Chilcotin tla, Sekanais usse, Nahanais cptii. These particles or
words can, in the North, conveniently be omitted in many cases. They are the
equivalents of the French ne...pas. -At, as we have seen, means 'wife'; -// is the
root of the verb ceti 'he has'.
This is about the sum total of the criticisms I have to make on that author s
rendering and interpreting of the southern dialect he introduces to the phi-
lological world.
Editorial Note
Originally published in American Anthropologist 17, 347-350 (1913).
Reprinted by permission of the American Anthropological Association.
CHASTA COSTA AND THE DENE LANGUAGES OF THE
NORTHi
By a. G. MORICE, O.M.I.
ALL English scholars are familiar with the fact that the growth
of a language is evidenced not only by the alterations in the
material make-up, the morphology, of its component parts,
but by the remarkable evolution which those parts occasionally
undergo in their meaning while they remain unaltered in their
structure. For instance, the word "villain" had in Chaucer's
time a quite different value from that which modern usage attributes
to it. No Christian of our days would feel flattered by being called
the "villain of the Lord," an expression which originally meant the
"servant of the Lord."
Likewise, instances of such alterations in the meaning of words
are not wanting in American aboriginal philology, though said
alterations may not be the result of time, but rather due to other
circumstances such as, for instance, linguistic borrowing or changed
cultural environment.
An example will make my meaning clearer. There never was
any buffalo within British Columbia. When Cree-speaking half-
breeds reached the northern interior of that region, in the wake of
the Northwest Company traders, they told the natives of a wonder-
ful animal they called mustus, which roamed by the million o\'er the
immense plains east of the Rockies. Later on, domestic cattle
were introduced in the same country and dubbed mcestus by analogy
with the game the natives had heard so much about — the ii of mus
being altered to ce conformably to the requirements of the Dene
law of phonetic sequence.
So that with time mcestus came to be universally understood of
domestic cattle, instead of the buffalo to which alone the name is
1 A further discussion of Noles on Chasla Cosla Phonology and Morphology, by
Edward Sapir. See p. 347 above.
559
486 VI American Indian Languages 2
560
strictly speaking applicable, and when the Carrier had to refer to
the latter they called it tlokcst moestus, or prairie cattle (lit. grass-on
cattle). Meantime, buffalo remained mustiis to the Cree who, in
turn, gave the name of aivokhdni-mustus, slave ox, to domestic
cattle.
Here we have, therefore, significatory evolution resulting from
the importation of a loan word. In Dr Sapir's "Notes on Chasta
Costa Philology and Morphology," there are several terms or roots
the meaning of which seems to have undergone an analogous
transformation, through the action of time, contact with alien
populations or the shifting of environment, unless, of course, we
choose to ascribe it to imperfect information on the part of the
essayist.
A characteristic instance of this I find in the word tdac,^ which
Sapir gives us as the equivalent of the English term "bird." This
is evidently none other than the Carrier iscpz, which in that language
means not "bird" but "feather-down." Is it possible that in this
instance a part of one thing should have come to represent the
whole thing? This seems to be all the more likely as that one
thing is too well known to have ever been taken for another by
Dr Sapir's informant.
Color would also seem to be lent to this probability by the
fact that pcen, which means "roof" in Carrier, denotes a whole
house in Chasta Costa. However, it would require no great
flight of the imagination to see in that word something like a
reminiscence of the time when a Dene habitation consisted almost
entirely of a double shelter in the shape of a roof squatting on the
ground, as we still see some in the recesses of the North.
A still more curious terminological mutation, which affects
both the structure and the sense of a word and is perhaps the result
of accident or of growth along lines unknown in the latter regions,
is to be found in the Chasta Costa possessive form of the term for
"dog," li, whose variants are everywhere unimportant. In the
' The exclamation mark denotes the "click" in Dr Sapir's texts, and with him as
with me c = sh. Dr Sapir's a is my ce, almost the sound of u in " but," more exactly
that of e in the French je, te, le.
Appendix 487
561
South as in the North, the sibilant I is converted by the possessive
into a common /; but the analogy does not go any further. Ac-
cording to Sapir, this possessive entails in Chasta Costa the accre-
tion of a sort of suffix which he writes tc!e, the equivalent of my tse.
Now litse means in Carrier, not somebody's dog, but she-dog! Is
this again a mere accident, the result of word development or of
some other cause?
Another change of meaning in a vocable, coupled, this time,
with an anomalous formation, which is perfectly recognizable to a
Carrier I find on p. 311 of Dr Sapir's "Notes." There he gives us
t'enl'lat as the equivalent of the English "you drown," and to this
he adds, by way of comparison, the verb tcnniiUat, which he repre-
sents as the Kato synonym of "it floated there." This circum-
stance is virtually a voucher that our author is conscious of a
diversity of significations: to drown is evidently not to float.
But if we turn to the Carrier of the North we meet with a some-
what homonymous counterpart of the former term in theniltlat,
which means not "you drown," or "you float," but "you sink,"
or rather "thou sinkest."
This may not be the exact synonym of Dr Sapir's verb, though
it is nearest to it in meaning; but how are we to explain that the
act of drowning is rendered in Chasta Costa by a word the two
roots of which seem self-exclusive, while the principal one refers
to an act which is the opposite of drowning, namely that of floating?
The- and -lat are self-exclusive, I have said. For the former, at
least in the North, does not merely mean "in the water," as Sapir
would have it — tha- p. 302, which he gives as "referring to the
water, "1 has really that signification — but it hints at the "bottom
of the water" (from the Carrier theR, "water-bottom"), while -laty
as we have seen, is expressive of the act of staying on the surface
of the same.
Theniltlat, on the other hand, is easy of explanation and of
quite logical construction. The desinence -tlat refers to any pre-
' Carrier nacya, "I walk along;" nce-lha-dxcya, "I walk in the water, I wade."
Compare: tha-Rcel, "deep" (water); the-husKa, "shallow" (may be analyzed: the
bottom, the-, near the surface, -Ka).
488 yf American Indian Languages 2
.S62
cipitate action, and, closely analyzed, the whole verb amounts to
thou art precipitately brought to the bottom, that is, thou sinkest.
" I drown" is said in Carrier thu scezcelreh, "water kills me."
Apropos of water, Sapir gives the verbal stems -el and -al as
representing the acts of bathing and coming, respectively, while to
him the desinence -ya is synonymous of going or coming. ^ There
must be a slight inaccuracy here. The first of those desinential
radicals {-el) certainly wants the initial hiatus (') ; for his nadayllel,
"we are bathing," is none other than the Carrier ncetsoel'il, which
has the same value.
As to the verbal stem -al, which he believes to mean "to come,"
I more than suspect that it is but a corresponding form of -ya,
which he represents as expressing the idea of "going, coming,"
and should be -yal. In the first place, the difference between going
and coming flows in Dene from the apposition of locative adverbs,
or results from a diversity of prefixes, not of a dissimilarity of
desinential roots. The sufifix -ya denotes merely the action of
moving about on both feet as a man (there is another for the
walk of birds). Such is the meaning in Carrier of the word
noecya.
But another form of the same verb, which I call the actualizing
form, changes this term and its characteristic desinence into cecyal,
which means " I am actually walking" (with both legs). This form
is also commonly used in connection with locative adverbs, as in
the phrases: tiz tnyal, "come here"; hwaz inyal, "go there";
'cen tnyal, go on (which recalls Sapir's An yi'al, the equivalent, he
claims, of "come on").
Dr Sapir furthermore quotes the verb stem -t!o (-to) as denotive
of the act of swimming, while, according to him, that of paddling
is rendered liy the radical -xe {-Ke). ' Now, in most northern Dene
dialects, the former {-to) refers to paddling, while the latter indicates
the act of navigating, or moving about in a canoe. Are those roots,
with their changed value, new evidence of evolution in the meaning
of words?
• P. 323 of his essay.
Appendix 489
563
The verbal stem -lal, or rather -//a// to which our author
attributes the sense of "to sleep," has in Carrier the value of "to
dream of" (with a complement). Might not Dr Sapir's informant
have misunderstood his questioner and thus unwittingly misled him?
If not, such changes in the meaning of words or roots otherwise
so closely related are well worth a moment's reflection.
If Dr Sapir will allow me, I will also observe that the desinence
-tc'ac (-thcec), which he gives as a distinct verbal element,^ is nothing
else than the plural stem of the same verb nanisthi, whose deriva-
tive nthcesthih effectively means "to lie down, go to bed." That
plural stem is in Carrier -thes for the present, -thez for the past,
and -thoes for the proximate future.
Likewise, Sapir's verbal desinence td!i (tsi) is simply the plural
form of the verbal stem -ta, "to sit," which he gives elsewhere as
-da.
Our essayist very properly represents ^ the root -xwi (-Kwi) as
denoting the idea of vomiting, and compares it with the Carrier
substantive ku, "vomiting." He will be pleased to learn that the
former is -Kwe in the Lower Carrier dialect.
Elsewhere he speculates on the value of the verbal elements /
and I, and, p. 332, he goes to some trouble in order to explain the
passive form of the Chasta Costa verbs. Unless I am very much
mistaken, what he adduces as the equivalent of " I am seen, you are
seen, he is seen," etc., really means simply: "people see me"
(French: on me voit), "people see thee," etc.
It may be of interest to state that in Carrier the passive form
' As may be seen by Sapir's rendering: t'illal, p. 315. The double consonants //
and ts (both of which may be affected by a click) are of frequent occurrence in Dene
and form as many indivisible groups. The syllables of all Sapir's verbs in the first
person plural are wrongly cut up: the / which he attributes to the penultimate syllable
should commence the last one: -tlal, -tsoel, -tlal, -these, etc. Hence several of his verb
stems on pp. 323-26 are incomplete. For instance, -se, "to cry," should be -tse
(Carrier -tso, though the first person singular of the verb is in -so); -si, "to cause,"
should be -tsi (Carrier -Isi, same remark as to first person sing.); -lo, "to laugh,"
cannot be understood without its t, as is shown even by the examples the Doctor
adduces in explanation. Were he familiar with the Denes' syllabic way of writing
their own language, he would have been spared this little inaccuracy.
2 P. 326.
'P. 325-
490
VI American Indian Languages 2
564
very often results from a mere change from the second to the third
conjugation. Here is an example:
A dive
Passive
as'en, I do (somethin
g)
cez'en, I am done,
il'en, thou doest
il'en
(B{yoe)l'en, he does it
cel'en
cetscel'en, we do
cetscel'en
osl'en, you do
cel'en
mR(e{yce)l'en, they do
it
cercel'en
il'en, both of us do
il'en
When the verb under its active form belongs to the first conjuga-
tion, its passive is sometimes represented by a verb which has
nothing but the radical desinence in common with the active, and
which offers the strange characteristic of being at the same time
pluripersonal and unipersonal. The following will be clearer than
all possible explanations:
A dive
ukwessi', I love him
ukweintsi' , thou lovest him
ycekentsi', he loves him
ukwetsintsi' , we love him
ukwehtsi', you love him
Rikentsi', they love him
ukwetlsi', both of us love him
uKwa'dcesni, I call him, etc.
Passive
skeidintsi' , I am loved, etc.
nkeidintsi'
ukweidinlsi'
nekeidintsi'
nukweidintsi'
poekeidintsi'
nekeidintsi'
SKa'hwotni, I am called, etc.
In the first verb 11- is the third personal pronoun in the singular,
which is here the complement of the verb -kessi', which cannot
stand alone; ^ kwe is the postposition ke, "by attraction to,"
inflected into kwe by the stronger vowel u; the first of the two 5
represents the first person singular present (which would be a^s,
were it not for the contact with the stronger vowel e immediately
before) ; the same pronominal element of the first person (s) in
turn modifies into -si' the verbal stem -tsi', denoting badness,
avarice. This verb, therefore, stands for ukwe(£stsi' , and, considered
in its component parts, yields the literal meaning of: "by attrac-
* Unless it is immediately preceded by its complement, formed of a single word.
Appendix 491
565
tion to him I am bad," or avaricious, that is, I feel so much incHned
towards him that I cannot share him with others.
As to the passive form of this verb, it is in reality a regular phrase
commencing by the personal pronouns (5-, «-, u-, ne-, nuh-, pcB-),
to which is added a unipersonal verb. For instance, the first person
can be thus decomposed: s-, "me;" ke-, "by attraction to,"
idtntsi' , "one is avaricious."^
The same analysis applies to UKwa'dcssni (lit., him for-the-
want-of I-say something) and to SKa'hwotni (me-for one-says
something — with a very impersonal meaning), in both of which
verbs the reader will not fail to remark the elision of initial a'- of
cedcesni and cekwotni represented as usual by the hiatus.
, Such passives, however, are not very numerous in Carrier.
Their verbal part -idtntsi', -hwotni, varies alone according to the
tense: past, skeidantsi', sKa'hwodani; proximate future, skeidoethi-
tsi', SKa'hwodcethatnil; eventual, skeidutsi' , SKa'hwodoni. The nega-
tive furthermore affects not only those verbal forms, but even the
postposition -ke of the first. Skeidintsi' then becomes skoeleidttsi' ,
etc. But enough of this. Entering into the question of the intri-
cacies and niceties of the Carrier verbs would lead us too far.
Lastly, many other verbs obtain their passive merely by chang-
ing the second conjugation into the first. Ex.: naincelneh, he dis-
tinguishes it; nanisno, it is extinguished. It is only right to remark
that in such cases the equivalent of our passive is just as much of a
primary form as that of our active, both being independent of each
other.
If we are to believe Dr Sapir, the Chasta Costa radical -to
expresses the ideas of both swimming and sucking. The former
is rendered by -pe, -pi in Carrier and the latter by -tuk (which is an
instance of onomatopeia). According to the same authority, -na
represents as well the act of drinking as that of lying on. In Carrier
"I drink" is said cesnai when the verb is transiti\e, while the same
language boasts an infinity of roots corresponding to the English
"he or it lies on," which change according to the nature of the
' The d of dinlsi' refers to self-interest, and belongs to a characteristic form which
affects most of the Carrier verbs.
492 VI American Indian Languages 2
566
subject. We are also told that in Chasta Costa the desinence -lee
does duty for "to wager" and "to smoke." In the first case, the
Carrier equivalent is -le, if the verb is intransitive, and, as to the
second, it all depends on what is meant by smoking. If this refers
to the favorite pastime of the votaries of the pipe, the Carriers have
the root -ta:t (another case of onomatopeia) therefor. If Dr Sapir
means the act of treating with smoke, for instance, meat, those
Indians then say: tscel a;ssi, "I make, or cause to be, soot;" if the
complement is a skin, they change this into loet pe cestleh, literally
smoke with I make.
To the root -ya our essayist attributes an even larger number of
significations. It refers, he says, to locomotion, manducation, and
the feeling of shame. Normal human locomotion on both legs is,
in fact, expressed by -ya in Carrier; but the act of eating is rendered
by the desinence -'al among the northern half of the tribe, while
the southern part of the same replace it by the root -yt. On the
other hand, the Chilcotin word for eatable is tsiyan, and the general
root for manducation is -yan in that idiom.
As to the verb "to be ashamed," it is in Carrier one of the two
or three whose structure exactly tallies with the English synonym.
Yuya a'stli, "ashamed I am," is what they say, and in some com-
pounds the root for shame (substantive, not verb) is ya.
With regard to the plural of verbs Dr E. Sapir writes: "Among
deictic elements are further to be reckoned certain prefixes that
serve to indicate either plurality as such or more specifically third
personal plurality." ^ Whereupon he gives us a few examples of
verbs commencing in ya-. A word or two on this subject, such as
we find it in the Carrier and other Dene dialects of British Columbia
may be of interest to him, and possibly to others as well.
In the first place, the third personal plural of all the Carrier
verbs is rendered by the particle r-, ros- or rI- prefixed to, or
coalescing with, the pronominal element — the equivalents of the
Chilcotin qe- and qce. Added to this is the particular inflexion of
the desinential radical for each of the three plural persons when it
is a question of a verb of human locomotion on both legs {-ya = -til
' "Notes," p. 306.
Appendix 493
567
in the plural), of locomotion on all fours {-kret = -'as), of running
{-kraih = -ras), of station (-/a = -tse), of cubation {-thi = -this),
of natation {-pi = -'il, first conjugation for the singular, third for
the plural), of navigation {-ke = -'il, first conjugation for both
singular and plural), of physical feeling (to suffer, to swoon: -zit =
-teh),oi ejection with a human complement (to throw: -neh = -til,
first conjugation for the singular, second for the plural), of bursting
into laughter {-tsit = -Kcek). The verbs of flotation make their
plural as the verbs of navigation.
All of these plural desinences, and indeed the singular desinences
as well, furthermore undergo material changes according to the
tense.
Then there are the objective verbs, some of which have for all
their tenses a desinential plural of their own. Ex.: nences'aih,
"I put (in a certain place) a single object with no special char-
acteristics;" nenoestle, "I put many such (in the same way)."
Aditai, "it (a single unspecialized object) is put in a hole;" adtlya,
"several such objects are put in the same place. "^
We now come to the real verbs of plurality. They are char-
acterized by the element ncB- (ne- and sometimes no-, when in direct
contact with stronger vowels) introduced before the pronominal
syllable, the reduplicative prefix, or again the negative particle.
Here are a few instances:
Common Verbs Verbs of Plurality
tha-dassel, I cut up in various lengths with an axe thanoe-dcessel
thcennaskrcBS, I wash (linen, skinb) thxnnoena-skroez
pe'-dcestcoez, I bind it pane'-dcestccez
Itse-soe'a, it is straight UsetKB-soe'a
As may be seen in the case of thcennaskrces, the pluralizing form
sometimes affect the verbal desinence. It changes that of the pre-
sent, and the two futures into that of the past, which then remains
unchangeable.
1 It will be noticed that the plural-forming particle of the third person is wanting
in this verb. It is used only in connection with human subjects, and even then it dis-
appears whenever the subject is itself in the plural. Ex.: tedasKul tsiyauh h-wcnii,
"all the girls went away" (instead of hweRanlil).
494 VI American Indian Languages 2
568
Should the subject or complement be evidently plural or imply
such number, the verb of plurality is not resorted to, unless one
wants to draw attention to the fact that he refers to a multiple
object or several persons.
Another kind of pluralizing verbs, which are perhaps the ones
Dr Sapir had in mind when he penned the above quoted remark,
do indeed commence with the prefix ya-; but they mean much
more than the common verbs of plurality of which I have just
given a few instances. They refer not merely to several, but to all.
In other words, they imply totality rather than simple plurality.
The very substantive verb to be ((sstli, I am) can be thus affected.
Rinli, "they are;" yaRinli, "they are all, or at least in very large
numbers."^
Sometimes the prefix ya- denotes also a repeated action (espe-
cially if followed by the crement -das-) rather than a plural comple-
ment. Examples:
Common Verbs Verbs of Totality
cestah, I cut with a knife in a slashing; yastah, I cut to pieces with a knife
way
cesqul, I tear yasqul, I tear to pieces
thoesKat, I throw away with a shovel yaidasKat, I throw away on all sides
with a shovel, I scatter, etc.
doesnat, I split yaidasnat, I split in very many places
Oftentimes these verbs combine the crement n<r- indicative of
plurality with the prefix ya- which denotes a large number, or even
totality, and may furthermore take the pluralizing particle ro;-
proper to the third person plural, as may be seen in the following:
Itcan, she is with child yanoltcan (contraction oi yanwRceltcan),
they are all with child
ucyul, I blow repeatedly on (as a yancepucyul, I blow repeatedly on all
shaman)
The desinence of the first verb means " womb " in Carrier. Den6
phonetics always contract a^Roe- into 0-; hence the peculiar form
of totality noticeable in that verb. In ucyiil, the real signification
• Strictly speaking, they imply totality, but the Indians are so addicted to the
practice of exaggerating that such verbs practically refer to large numbers only.
Appendix 495
569
of which is "I exorcize" in English, we have still another category
of verbs, the frequentative. The original form of the same is
cecyul, "I blow." When one is told that the pretended exorcism
of the shamans consists in incessant blowing on the part of the body
which is affected by disease, he will realize the appropriateness of
the frequentative form to express the nature of their operations.
In yanoepucyul we have a further instance of a weak vowel
{ce of pee, "them") disappearing under the influence of a stronger
vowel (initial u of ucytd). That compound stands for yanoepceuc-
yuL^
Very often also the verbs have not only a pluralizing but a
totalizing form, as is the case in the following: thaRcenisnai, "they
are drunk;" thanonisnai, "many are drunk;" yathaRtnatnai, they
are all drunk; ceRcenistcet, "they are drunk with tobacco;" nonistcet,
"many are drunk with tobacco," yaRinatcet, "they are all drunk
with tobacco."
The dissection of these two series of verbs is quite interesting.
The desinence -nai of the first denotes the act of drinking, while
the prefix tha- indicates that something stronger than milk, namely
"fire-water," has been absorbed. Rce- is the usual particle of the
third person plural, and -ms- shows that the above mentioned
potations have been excessive, being attended with fatal results.^
The reader will not fail to remark how this accretion -nis- is con-
verted into -7ia- under the influence of initial ya- of totality {yatha-
Rinatnai).
In the second verb -tcet is onomatopeic. It denotes the act of
smoking, and is intended to reproduce the peculiar noise made by
the lips when pulling at a pipe. As to nonistcet the reader will
have guessed that it is a contraction of nceRoenista't, prompted by
the phonetic rule already alluded to.
Before dismissing the question of the verbs in ya- referred to
' Were not the frequentative form intended, the sequence of two vowels in m
would be against the principles of Carrier phonetics.
* These verbs belong to still another class, that of the verbs of error, which
denotes an action with unforeseen, or fatal, results. Almost all the Carrier verbs can
be clothed with this significant form, which, in common with the frequentative and
other forms, constitutes a new scries of verbs.
496 VI American Indian Languages 2
570
by Dr Sapir, it may be worth the while to observe that, inde-
pendently from the value of that prefix as a multiplicative element,
it is also the root of numerous verbs meaning "to land, to go (or
take) ashore." Here are some examples illustrating that new role:
yascesKeh, I land in a boat.
yasoecyaih, I land on foot (as on the ice).
yascezkraih, I land while running.
yaseeskuh, I land in a sleigh.
yasceslzut, I land while skating.
yascezkret, I land on all fours.
yascezKuh, I land on one leg.
yasoestlat, I land floating on the water.
yasaspih, I land swimming.
yascestah, I land in a balloon, flying.
yasoEstzas, I land hopping about like a bird.
ya'nas'oes, I land while limping about.
yadcezquh, I land while throwing out the throwing stick (game).
yaswzqeh, I land under the influence of anger.
yasoestcth, I land with head erect.^
yascezUas, I land on crutches.^
yadoezthis, I land with a walking stick.^
yascsstzit, I land while chasing large game,
etc., etc.
ya-isi'-does'aih, I, being a worthless fellow, land on foot.*
ya-Ke-ncez'ih, I land by stealth.^
ya-cosn-dijyaih, I land on fooi and while singing.^
ya-tso-dtjyaih, I land on foot while crying.''
ya-tse-soeztles, I land nodding right and left.*
ya-kw(B-disleh, I land on my knees.^
1 The desinence -tcih of the present becomes in the past -Icoen, which means
"stick, tree," and hints at the subject of the verb landing while "stuck up" as the trunk
of a tree.
2 From the root -tla, "posterior," because in such a case the person seems to an
Indian to be walking with his posterior instead of his legs.
■' From IhcBz, "walking stick," which is the desinence assumed by the verb for the
past tense.
^ F"rom the root -Isi' , which, as we have already seen, is the Carrier for "bad."
^ Literally: I land hiding (noez'ih) my feet (Ke).
* Ccen means "song."
' From the root Iso, "crying."
* From Ise-, equivalent in compounds of -Isi, "head," and -lies, stem of the verb
"to throw."
' From -kwcEl, "knee."
Appendix 497
571
ya-Kct-uzoezKeh, I land in a canoe with the intention of buying.^
ya-tse-noestaih, I land on foot under the influence of fear.'
ya-na-tse-ncestaih, I land again on foot under the influence of fear.'
ya-na-hwe-ncesqa. I commenced again to land on foot/
etc., etc., etc.
The compound, do-at-ti,^ leads me to remark on the evident inferi-
ority of the Chasta Costa compared with the northern Dene lan-
guages, especially those of British Columbia. "Not-wife-he has"
appears almost childishly simple in construction by the side of the
negative forms prevailing there. Indeed, this phrase recalls to mind
the similarly inelegant mode of expressing themselves adopted by the
Sekanais of Bear lake (Fort Connolly), British Columbia, when
they attempt to speak Carrier. Awontuh 'at oeti, " no a wife he has,"
they would then say instead of the classical an 'at Ittih, which con-
tains three distinct negative elements: first, an, "not;" second, It-,
made up of negative I and pronominal crement x inflected into i\
and, third, the verbal stem -ti altered into -tih by the same negative
prefix.^
Moreover, if, as is well known, the natural tendency of languages
is to disintegrate with time their constitutive elements, that is,
to pass from synthesis to analysis, the material presented to the
public by Dr Sapir, meagre as it may be, sufifices to prove that the
Chasta Costa dialect is much less primitive, because more analytic,
than the Dene idioms of the Canadian North.
To mention but one point of the grammar of the latter, they
form their futures by means of particular inflections, or even
' From Ket, which denotes the act of buying. The tizoes- of the verb furthermore
indicates that the person has "arrived" at the place.
^ This is one of the verbs of error, as can be seen by the particle Isc-, which inflects
the -cyaih of naecyaih into -slaih (noestaih).
' The reduplicative prefix na- is here introduced.
* To the reduplicative particle the initiative prefi.x hwe- is here added. The
verb is in the past tense {-qa instead of -taih, which is itself altered from -yaih by na-),
because such verbs are never used in the present tense.
* See footnote 82, p. 337.
* Strange to say, while, considered in its constitutive elements, this compound
should mean "he has no wife, he is unmarried" in Carrier as well as in Chasta Costa,
usage has given it the sense of "he did not get married" in the former dialect, since
"he got married" is said a-ti (with a contraction similar to that of the Chasta Costa).
498 VI American Indian Languages 2
572
excrescences, of the pronominal crements. Thus the proximate
future of the Carriers changes the present ces'^s, "I sneeze," into
cethts'ces.
Quite often also, the desinential radical of said verbs undergoes
itself a characteristic transformation indicative of the new meaning
assumed. For instance, dcesni, "I say," becomes dcethasnil in that
same future.
On the other hand, the Chasta Costa decomposes and scatters,
as it were, the elements of its verbs, much after the way the modern
Romance languages have treated the parent Latin. Thus to form
the above mentioned future it simply adds the suffix, or rather
separate particle, the to the present.^ No more any of the two
or three original inflections of the Dene negative, no more any
internal growth, but instead an independent monosyllable, for
the future: such are, indeed, unmistakable tokens of analytic dis-
integration which bespeaks unfavorable ground or unpropitious
environment for the life of the language.
Saint Boniface,
Manitoba, Canada
' The words Sapir gives as synonymous of " I shall sneeze, I shall look at him,"
etc., should be translated: "// I sneeze, if I look at him," etc., in Carrier, the and te
being in that language the conjunction if.
Editorial Note
Originally published in American Anthropologist 17, 559-572 (1915).
Reprinted by permission of the American Anthropological Association.
Misconceptions Concerning Dene Morphology
Remarks on Dr. Sapir's Would-be Corrigenda
The fourth number of the American Anthropologist for 1915 contained
a series of remarks by Dr. Sapir on my review of his Notes on the Chasta
Costa Phonology and Morphology under a caption ^ to which exception
could well be taken. It is not usual to see errata or corrigenda thus
glaringly paraded by another than the party responsible for the pro-
duction in which they are supposed to be found. In the present case
such a title is a misnomer, and, even if warranted by the circumstances,
it would come with a better grace from an old hand in the field of D6n6
linguistics. Most of the eleven statements which Dr. Sapir seems to
criticize were simply proflfered in a tentative way, as mere suggestions
or bids for explanations.^ One "corrects" assertions, not suggestions.
Practically all of the few that were unequivocal affirmations are the
expression of real facts, and therefore not any more corrigeyida than the
others. One exception refers to my taking Dr. Sapir's A for my ce, an
error which is perhaps imputable to the printer, and should not certainly
take eight pages of small text to "correct".' Moreover, with regard to
two of his would-be corrigenda (Nos. 6 and 11), my critic admits that I
was right.
I trust that, even in his kind reference to what he is pleased to call
my "admirable mastery of the Carrier language," one is not to see some-
thing akin to a hint that I should confine myself to its exposition, a
suggestion that has already appeared in print. To ward off the ac-
cusation of unbecoming presumption in treating of other related dialects,
I am forced to remark, once for all, that a man who used to preach without
an interpreter in Chilcotin before he knew one word of Carrier or Babine,
^Corrigenda to Father Morice's "Chasta Costa and the Dene Languages of the
North," vol. 17, p. 765-
' "I more than suspect;" "might not Dr. Sapir's informant have misunderstood
the questioner?" "unless I am very much mistaken," etc.
» Dr. Sapir is very obscure with regard to the definition of the value of his graphic
signs. Nowhere can I find an explicit statement concerning the phonetic identity of
his a; the nearest approach to it is an incidental remark (p. 278): "a (probably iden-
tical with our a) in Carrier," he says.
132
500 VI American Indian Languages 2
133
and had acquired these two last idioms before he understood Sekanais,
or who could catechize unaided in Nah'ane, five Dene dialects differing
widely in their lexicon and grammar (with the exception of Carrier and
Babine), that man, I say, should be able to detect errors or misappre-
hensions concerning phonetic and morphological points which run through
the whole D6n6 linguistic family, especially when he has been, since 1882,
studying those idioms, twenty-four years among those who speak them,
and then with the help of all the literature bearing on the subject. It
must be admitted that this help is not always infallible, because, oc-
casionally, of an imperfect ear on the part of the transcriber, distractions
or printer's errors, such a? those which Dr. Sapir himself points out at
the end of his remarks under Nos. 3 and 11.
This scholar claims that Chasta Costa tclAc "can have nothing to do
with this td'ad, but must go back to Ath.* ^k^Iac," and he props up Tiis
contention by remarking that, according to Dr. Goddard, kiyauw happens
to have in Hupa the same signification, "birds," as his own tc'AC. He
likewise takes exception to my practically assimilating the suffix -tc!e
of li-tc!e with that of my li-tse, declaring at the same time that this
really corresponds to Ath. k^!e, which assertion he similarly bases on the
-ke of Hupa lin-kc.
He might not be far astray if the guttural sound was accompanied
by the glottal explosion which we find in C.C.- tclAC, Carrier and Mon-
tagnais tscez (according to my orthography), Navajo tsos (do.), Loucheux
tsaw (do.), as well a? in the various equivalents of "woman" {tse or tc!e),
an all-important explosion, or click, which Dr. Sapir renders by an ex-
clamation point and I by a dot under the letter affected thereby. Dr.
Sapir is himself so well aware of this that he obligingly supplies that
"click" (p. 766) to the ki- of the first word and the -ke of the second! If
Dr. Goddard really meant k (Sapir's k!) when he wrote k, may I ask how,
in that case, he rendered the common k sound?^
» Athapascan.
* Chasta Costa.
• Years ago I criticized that scholar for having failed, as I thought, to render in
his Hupa Texts the particular aspirated t (my th, the Franciscan Fathers' tq) common
to all the Dene dialects. As a matter of fact, his t was the equivalent of my th, and
he rendered the ordinary t by d. But I still fail to see how he expressed the real d
sound, which does exist in Dene, though the natives themselves do not differentiate
it from t. A. G. M.
The aspirated k, sound Father Morice writes K, becomes everywhere in Hupa a
continuant written x. The character k was therefore used for the sound Father
Morice calls "click," and writes k. What Father Morice calls the common k is
Appendix 501
134
As to the Hupa root -tsots or -tsos, which my critic adduces in this
connection (Carrier, Chilcotin, Sekanais and Nah'ane -tsus, Babine -tsos,
Montagnais and Loucheux -tsun, Hare -tsu), it has no affinity with -tsoe^.
With regard to his strictures under No. 4, I could not, with the best
of will, change what I wrote concerning the point he refers to therein.
He may refuse as stoutly as he can to agree with me; further study and a
little speaking knowledge of a few Dene dialects will ultimately convince
him that I am right in this connection. But we must understand each
other. When I said that the- "hinted" at the bottom of the water, I
did not mean that it denoted exclusively the ground under the same
(though that prefix has frequently reference thereto) ; I had especially
in mind the water that is near the bottom of the body of water.
Now the very words which Dr. Sapir quotes to dispose of my assertion
redound against him to the extent of proving even more than I meant.
I even unnecessarily qualified my statement when I said that this was the
case "at least in the north," since that prefix has the same value among
some of the southern tribes, as is unwittingly shown by my critic himself.
He adduces Navajo tqe-h (my the-lin), "water-horse," which, forgetful,
or unaware, of Indian exquisite accuracy in rendering linguistically the
individual characteristics of natural elements, he imagines must mean
literally "in-the-water horse." But I claim that he is mistaken in this.
The native mind could not possibly form the concept of a horse without
thinking of an animal with four feet. It is too radically exact for that
when it is a question of concrete ideas. Even though the object de-
nominated may be a real fish, the name given to it predicates the notion
of legs and feet, which cannot be used as fins " in the water," but as means
of locomotion "over" some hard substance, namely the bottom of the
water.
This is so true that when, referring in Dene to the Eucharistic ele-
ments, I used the verbal desinence implying a personal, not material,
complement, to show that these were none other than Our Lord Jesus-
Christ himself, I could never prevail upon the Indians to follow me in
this. They believed as firmly as I did in transsubstantiation; but their
language was too strictly logical to use a verb connoting a personal com-
plement when the -ivord expressing that complement referred to a thing
— a Sacrament, or the Eucharistic bread — not to a person.*
really intermediate in sonancy in Hupa and was written q when velar, g when
prepalatal, and sometimes^ when postpalatal as in Luk kai, "wliite. These facts
have been called to the attention of Father Morice in print previously. Ed.
> The following passage from a letter lately received from Rev. Father Leopold,
502 VI American Indian Languages 2
135
Dr. Sapir gives as another proof(?) that Dene the- simply refers to
the water, and not to the bottom of it, the fact that the Kato verb ex-
pressing the idea of washing is in f'e'-. In the first place, this is not to the
point, since my statement did not embrace southern linguistics. Then
here again his pretended proof turns against him. The Chilcotins have
indeed thenasqoss and the Carriers thosnnaskroes for "I wash"; but what
kind of action do these terms represent to the native mind? How do
the primitive aborigines wash? By laying over the clayey "bottom"
of the shore of a sheet of water the soiled skin or blanket, on which they
tread with bare feet and legs as do the wine-makers of Europe! I have
witnessed this operation a number of times during my twenty-eight years'
stay in British Columbia, and the Kato word referred to plainly hints
that, though the natives who use it may have discarded this aboriginal
method of washing, they none the less practised it formerly.
Did my diagnosis of the case stand in need of confirmation, I would
have but to produce the fact that the D6n6 have an altogether different
word to render the idea of washing either their hands or their face.
Yet, as these particular operations were very little practised formerly,
the notions of washing and water bottom have remained so inseparably
correlated in the Dene estimation that the term which they use to desig-
nate soap is radically identical with "mud" or "clay." The latter is
called kwoetles in Chilcotin, hwotloss in Carrier, wotloss in Sekanais, and
practically the same among the Eastern Dene of the north. The root,
which is a primary one (hence its striking similarity), is -ties, -tlces. Now
the same dialects have respectively for "soap" la-tles, la-tloes and ta-tlces,
which, in the first two cases, mean "hands-mud," and in the last "eyes-
mud."
The mode of washing skins or stuffs above referred to seems to have
been originally unknown of the Nah'ane of the Far Northwest (the so-
called Thalthan), who use in this connection a word, cerathas'ots,^ which
simply implies that the action is done in the water, or with water, while
the chief contributor to the Navaho Dictionary, who had not been told of my contention
concerning the true etymological meaning of southern tqe-lt, is especially illuminating:
"There are," he writes (April 27), "such words as tqe-lt, water-horse, and Iqe-hoUsodi,
water-ox. Both are mythological animals, that live in, and walk in, or on the bottom
of the water. This last is expressed by the prefix tqe." The italics are mine.
' Henceforth I shall represent the hiatus by an apostrophe ('), instead of an
upper dot (") as before, and shall retain the dot exclusively for the exploded sounds
t, k, I, etc.
Appendix 503
136
Chilcotin thenasqces is analyzed: "I drag, rub (sqcKs) repeatedly (-na-)^
over the bottom of the water (the-)."
"Even in northern Athabaskan I do not find Father Morice's remark
[on the real meaning of the-] to apply without qualification, even if cor-
rect for Carrier." This is from Dr. Sapir's (p. 767). Let us then have
recourse to his usual authorities, to which I shall add what I personally
know of the Western Dene dialects. For "bottom of a body of water"
the Montagnais say (with my spelling) iheRe; the Hares, tM^; the Lou-
cheux, the; the Chilcotin, Carrier and Babine theR; the Sekanais, tceRe.
This last is consistent with the foregoing, because in Sekanais th is con-
vertible into tc. Ex.: "the Big Water" (the name of a British Columbia
lake) : Carrier, Thu-thi, Sekanais, Tcii-tci.
As to tha-, my eagerness to clearly differentiate it from the- induced
me inadvertently to go a little too far. It is simply the equivalent in
compound? — and in compounds only — of the word thu, Iho, tcu, etc.,
"water," and may contribute to the formation of substantives, adjec-
tiver, verbs and adverbs. Here are a few examples:
Substantives: Navajo thd-ba, "shore" (for tho-ba, "water-edge");
Chilcotin tha-zcel, "soup" {ior tho-nezcel, "water-warm"): Carrier //za-i5/,
"wave" (for thu-tsi, lit. "water-head").
Adjectives: Chilcotin tha-dinsat} "deep (water)," from tho, "water,"
and ncezat, "far off"; Babine tha-ltuk, "shallow," from tho, "water," and
ntuk, "short"; Carrier tha-scekcez, "spring water" (for thil-soskcez,
"water-cold").
Verbs: Chilcotin tha-stnan, "I drink (water)"; Carrier tha-s'aih, "I
put in the water," etc. — too many verbs of that kind in all dialects for
enumeration.
Adverbs: tha-niz, "at large on the water" (lit. "in the middle," -niz,
"of the water," th-)\ tha-tlat, "at the further end (of the lake)," etc.
This element appears "in compounds only," T have said. By this I
meant to controvert Dr. Sapir's statement to the effect that "in several
Mackenzie Valley dialects Ath. ^i'a even occurs as uncompounded noun
stem" (p. 768). For this he relies, of course, on Petitot's dictionary,
* Etymologically, na- (a contraction of nal, "twice") implies reduplication; but
it is also frequently used simply to show that the action is not done for the first time.
In those cases, it even occasionally alters somewhat the meaning of the word. Thus
in Carrier ihe-tcestliis the equivalent of "I beseech"; changed to thena-dcestli, it assumes
the signification of "I say my prayers, I pray." because praying is an action which
is normally done more than once.
2 In Chilcotin iH has the phonetic value of in in French singe, not of the same in
English sing.
504 VI American Indian Languages 2
137
sub voce "flot." But here we have one of the capes when mere book
knowledge falls short of its purpose. I never saw but one Mackenzie
Valley Indian, and never spoke to her, any more than I ever addressed,
or heard, a Navajo or a Hupa; yet I am positive, and if necessary will
stake my reputation as a Carrier scholar (the only one which Sapir seems
willing to concede) that the prefix tha- meaning " water "^ as well as
."flot",2 is never used uncompounded anywhere. This would be against
the morphological laws of the Dene dialects, such as my speaking
knowledge of five of them and book study of many years* duration have
revealed them to me. Were I not anxious to be as brief as possible, I
might give my reasons for this.
In other words, you may see this terminological element written
alone, as in Petitot's dictionary, or preceding independently without
hyphen some other words, as in the work of the Franciscan Fathers; I
insist that it cannot stand by itself, and needs the support of some other
element to exist. It is to thil, etc., "water," what tse- is to Kwon,
"fire"; tse- to tsi, "head",^ za- to -ze, "mouth," -tze to -tzi, "heart,"
and ne- or na- to nen, noen, yxn, "earth," "ground."
But I have tarried too long on these two particles. Dr. Sapir almost
ends his No. 6 criticism by adopting my suggestion that his desinence -al
should be -yal. "I now incline to think that -yal is correct," he admits.
He is not quite so yielding in his No. 7, since he objects under that head
that "there is plenty of evidence to show that Ath. -*;feV, -yfe'ew, -/fe'en,
-Wi frequently refers to, or implies, paddling," which I had declared is
expressed by the root -to, Sapir's equivalent of the idea of swimming.
The instances he gives absolutely fail to convince me. Nay, some of
them must have appeared of very little weight even to him.
Thus when Goddard renders Chipewayan -ki (t. e., kI) by "to
paddle a canoe, to travel by canoe," he is quite right; but this refers to
' Sapir's "in the water" is not quite exact, for several dialects have a desinential
letter (generally -/) which is locative in intent (thul, "in the water"; Iha-llat, "at the
upper end of the lake," etc.), in the same way as final -z denotes recess from {thiiz,
Iha-tlaz, etc.).
* Petitot gives to this French word the signification of "water," which it has some-
times in French poetry, as may be seen by the words he produces as synonymous of
"vague."
»A somewhat extreme case of compounding intp which this particle enters is
tha-na-tle-le dees' <ZR. " 1 do not drink again flat on the ground." which is thus analyzed:
"I am not {-le-) doing (-ce/j) with my bent body (as replaced by the hiatus 'dees-) a
renewed (-wo-) action connected with water (jlha-) wherein my head (-Cse-) takes a
part."
Appendix 505
138
locomotion, or the act of moving on the water through the action of a
paddle, instead of, for instance, by swimming, -pih, "floating on a raft,"
-Hat, "poling up," -thcez, etc.
So is it with the example quoted from Father LegofT, whose remark
is quite appropriate, and applies to all the D4n^ dialects that I know,
Naviguer en ramant never meant "to paddle," any more than j'at'anc^
en ramant, which has the same signification. All these terms refer to
locomotion which accidentally is effected by paddling, or the use of oars,
on the water.
With regard to Sapir's criticism under No. 8, I merely stated the
value of -lal in Carrier and asked whether his informant might not
have misunderstood him. He shows me by examples that this was
not the case. I am satisfied. This point consisted of a statement by
me which was perfectly correct, and of a question which is now answered.
I do not see here any room for a corrigendum.
I wish I could be as brief in connection with his No. 9, which he sub-
divided into four parts. As usual, Sapir believes that he "abundantly"
proves his case by references to other dialects. Were I sure that I do
not impose on the reader's patience, I feel I could, by the same process,
"super-abundantly" prove the contrary. The main point at issue here
is either the absence of t- in my critic's root stems, such as -se, "to cry,"
-si, "to cause," or his cutting up his words in such a way that this same
letter is made to belong to the preceding pronominal crement instead of
the radical desinence. Thus he wrote -t-lat, for -tlat, root for the action
of floating (which is in the same category as his -lal); -t-lo, "to laugh,"
instead of -tlo, and now, in his last paper, -t-na (^'a-7^7-nd), which should
be -tnd (t^a-yi-tna, "thou drinkest").
Speaking of the root for sleeping in some dialects, he claims that
"not -tlal, but -lal must be considered as the root stem," because, for-
sooth, the dental happens to be absent in the second persons singular
and plural {t'i-lal and t'o-lal) . He adds : " Were -/- part of the stem, there
would be absolutely no reason for its disappearance in these forms"
(p. 769). May I ask the learned doctor, firstly, whether this / is not
found in the first person singular, or at least dual and plural, of his verb,
and, secondly, why we find it in these persons of the following analogous
verbs, and not in any of the second ones?
Carrier Chilcotin Navajo
I float I am I give him a horse
r nwstlat nestli bdnislds'
Sing, j ninlat ninli banlos
I noellat henli yei'los
506
VI American Indian Languages 2
139
{tuztsaellat
noehlat
nitli
nelli
qinli
baddntdlos' '
bdddno'los'
yddei'los
bdnldlos'
L noeRcellat
Dual nttlat, we float (both of us)
To get a proper idea of the real root stem in such verbs, we must go
back to those dialects which have preserved it for occasional independent
use. You ask, for instance, a Carrier: "How did you come? By canoe?"
Tatqahoh hwosainya? Ke pe a?^ He may have to answer you: Awontuh;
tlal pe cesqa; "No, I did it by floating (on a raft)." Lat would here have
absolutely no meaning.
In his Hupa material Goddard also gives this monosyllable instead
of tlat; but it will be noticed that all the examples he produces as contain-
ing that root are in the third person, which, as we have seen, does not
exhibit it in full. -Tlat, not -lat, is so evidently the true verbal stem that,
under the transitive or causative form, the former appears in all the
persons of all the tenses, both affirmative and negative, as we may see
hereunder. The verb "I make him float" is conjugated as follows:
Affirmative
Present
ncBsUat
niltlat
noe{y(B)ltlat
naetsoeltlat
ncehlat
nceR(B{y(B)ltlal
D. niltlat
Present
nceloezaestlat
noeloeziltlat
nceloe{y(B)ltlat
Tueltsaeltlat
nwlcezoslllat
naelosRce {yos:)ltlat
D. ncelceztltlat
Past
nceswstlat
noesiltlat
n(B{yce)ltlat
ncetsoeltlat
ruBsoeltlat
nceRce{yoe)ltlat
noesilllat
Past
nwlcestlal
nceliltlal
noel<£{yi)ltlal
naeltsiltlal
ncelaltlal
nwlceRce{yi)ltlal
nceliltlal
Proximate Future
ncethistlcel
noethaltloel
ne{i)thiltlml
ncezthiltlwl
nosthiltloel
noERe{i)thiltlcel
nathaltlal
Negative
Proximate Future
nalthceztstlcEl
noelthcezaltloel
ncele(i)lhtltlcBl
nceUsosthihloel
ncelthcezceltlasl
7icel(BRe{i)thiltlwl
nmltluBzaltloel
Eventual
nostlat
noltlat
nce{yu)ltlat *
noetsultlat
nultlat
n(BRce{yu)ltlat
noltlat
Eventual
noeloezostlat
nceloezoltlat
n<Bloe{yu)ltlat
nceltsultlat
ncelcBzultlat
ncelwRoeiyu)ltlat
nceloezoltlat
* It must be borne in mind that the Dene ear perceives no difference between
d and (.
* Of course, Ke does not literally mean "canoe." but "navigation."
* The particles within parentheses represent the personal complement without
which a native will scarcely ever pronounce the word. This element may be omitted
when the complement immediately precedes the verb; but even then it is not con-
sidered a useless redundance in Carrier. Ex.: "Paul will take all my cattle (from one
place to another) by floating," Pol stnoestus tsiyauh naeyulUat.
Appendix 507
140
The desinential roots -tlcel of both proximate futures and -tlal of the
negative past tense are nothing else than the general stem -tlat, inflected
in conformity with well known, and invariably observed, laws of Carrier
verbal morphology.
In the same way, the desinence of Chilcotin nestli, " I am." which is in
Carrier cestli, and similarly loses its t- in all but the first persons singular,
dual and plural, always takes it back when used substantively or as a
compound element; in other words, whenever it has to play independently
its role of a root. Thus "I am dressed" is said: e-toene-oestli; "thou art
dressed," e-tcene-tnli, etc., and the compound word for clothing will be
pe-toene-oetli, literally " that-wherewith man one-is."
So is it with the verbal stem -tlo, representing the idea of laughing,
which Sapir would fain have us believe to be really -lo, since he writes it
-t-lo, disassociating therefrom the i which he unduly refers to the pro-
nominal element of the verb. This radical quite often fills the role of a
genuine independent noun, in fact it is a regular noun, in all the dialects:
Carrier, tlo; Navajo, dlo; Chilcotin, tldr; Sekanais, tloh; Nah'ane, tlok;
Montagnais, tlo and dlor; Hare, tlo, kid, Loucheux, tlog, dlog.
Here are a few instances of their use: tlo soezilre, " laughter killed me," '
' An idiotism, the like of which we find again in Navajo. In this connection,
I think it worth while to quote again from Fr. Leopold's letter (written on receipt of a
copy of my review of Dr. Sapir's paper), were it only to show how, for those who are
familiar with them, the dialects of the south have really preserved even minor points
of the Dene morphology, such as we find it in the north.
"What you say on pages 348-49 regarding the hiatus preceding the root of the
verbs of vision obtains exactly in Navaho, where this root is: 'i, '1, 'tl for the three
tenses: nesh'l, "I look at;" nel't, "I looked at;" dinesh'tl, "I will look at." Your
remark concerning the word "wife" also applies to the Navaho word 'ad, which is
generally used with the possessive pronoun: sha'ad, na'ad, ba'ad, "my, your, his
wife." The same hiatus occurs in Navaho before other verbal roots beginning with a
vowel: shana'a, one of the forms for "give me." . . .
"Your remark on the double parts of the body holds good partly in Na^•aho: a
one-eyed-man would be described as bina'kis {'kus being a now obsolete word meaning
'half.' However, shikhe or shila is used in the singular or plural for "foot" or "hand."
. . . Navaho has, as the Carrier, such personifying expressions as: tqoshinilqi, "■water
kills me," or "lam drowning." In hke manner they say: dichi shinilqi, "hunger
is killing me;" dabd' shinilqi, 'thirst is killing me,' to express a high degree of hunger
or thirst."
Had my correspondent been familiar with the Dine dialects of the north, he would
have saved himself the trouble of giving me translations of his Navajo material.
There is scarcely more than one or two words (,dichi for "hunger" and the future of
nesh'l, i. e., nec'in: Chilcotin nis'in, Nah'ane nes'ih. Carrier naes'en) which would not
be understood in the north. Thus "my, thy, his wife" is said see' at, na'at, pa' at, in
508 VI American Indian Languages 2
141
that is, "I am dying of laughing"; Ho kcenne'doesni, "I am acting the
buffoon," literally "laughter in-conformity-with I-say-many-things"
(ne'dcesni); tlo dizta, "I sit laughing"; tlo dtjyin, "I stand laughing";
tlo dizthi, " I laugh while lying down," etc. Lo could not be thought of in
this connection.^
With regard to the root for such an idea as weeping, tso, tscer in most
dialects, which Sapir again writes without its t, here is one tense of the
verb to which it belongs, together with that of a Carrier verb which is
morphologically identical:
Chilcotin Carrier
hcessd, I cry cessi', I am bad
hientsd tntsV
hcetsd ntsi'
hetsd tsintsi'
hatsd eehtsi'
qelsd Rintsi'
Dual: liii', both of us are bad
The Carrier for hcessd is cesscer,^ and its past, instead of making reiser
as in Chilcotin, becomes isse, yielding the very same root as that of
Sapir's Chasta Costa; but, of course, it does not fail to assume a / at all
its other persons. Now tso, tscer (not so, which means "sleet," not
"weeping") is employed in exactly the same way as tlo. So is it with tsi',
not si', which has no signification. It is given, under its independent
form, the sense of badness, malice, wretchedness, while it is also used
depreciatively in compounds, as in ya-tsi' -dees' aih (see p. 570 of my last
paper).
Chilcotin; Navajo shana'a is none other than Chilcotin sani'an (last syllable as French
en); bina'kis is Simply Lower Carrier boena'koez, or pcenakaez, etc.
I repeat that this information was spontaneous on the part of Fr. Leopold, who
had not been told of Dr. Sapir's criticism.
As to the former's remark concerning shikhe (or ci-Ke: Chilcotin sceKe) and shila
(Chilcotin sce-lla), it applies likewise to Northern Dene. Thus a Carrier will say:
sla ilket, "my hand is swollen," without using the singularizing suffix — a point I ha*d
noted long ago in the outlines of a grammar which, duly developed, may later on be
published. But whenever it is necessary to specify that one hand only is intended,
this particle (-kmz in Carrier, -kis in Navajo, -qces, in Chilcotin, etc.) must be aggluti-
nated to the noun. You must then say: sla-kcez, "the half of my hands." Were you
to use in this case the numeral "one," as is done in our European languages, nobody
would understand you.
' Dr. Goddard duly transcribes that t when he gives us the Hupa correlatives
tcivu and Icwe for tso.
* This desinential -r is scarcely audible at all, and many Indians, especially among
the younger generation, say tso instead of tsoer.
Appendix 509
142
The part of Dr. Sapir's criticism where he is the most positive is in
connection with syllabification. Here he declares that he "must em-
phatically disagree with" me (p. 771), and to show his utter disregard of
what I wrote on this subject, he gives us t'-d-yit-nd, for t'd-yi-tnd; yit-lo
instead of yi-tlo, etc., thereby absolutely disfiguring the words and grant-
ing to their pronominal inflective part that which in reality pertains to
their stem. I must therefore be allowed to make a confession:
Out of sheer presumption, and imagining that my 34 years' study of
the Dene language had fitted me for the task of writing thereon from
mere personal knowledge, the thought of consulting a book or published
paper on the subject did not as much as occur to me when I wrote my
review of Sapir's able essay on Chasta Costa. Now that he "emphatic-
ally" rejects my criticism on his peculiar way of dividing syllables, which,
I claim, entails a wrong idea of the personal crements no less than of the
nature of the verbal stems, I must have recourse to a book. On the
other hand, since it is the fashion in certain quarters to have nothing to
do with the morphological notions of the North, when treating of southern
dialects, I open the Navaho Dictionary of the Franciscan Fathers, and find,
page ID of its second part, the following caution noted in as prominent
a manner as possible:
The digraphs and trigraphs used in this alphabet are not two or three distinct
consonants in juxtaposition; but represent a single sound whenever they occur
initially or otherwise. We hyphenize and pronounce bi-tqo, not bit-qo; bi-t'd",
not bit'-o", etc., a distinction which is noticed by a Navaho ear.
This remark applies to pronouns and nouns, which are always coupled
together as inseparably as the various elements of a verb, since the pos-
sessive pronoun ibi-), for instance, cannot any more stand without the
support of its noun {tqo) than we could dream of t'd-yi- existing without
the immediate adjunction of its sense-giving stem -tnd. The above
mentioned caution of the Navajo scholars is, therefore, just as much to
the point when it is a question of verbs, as in connection with a pronomi-
nalo-substantival compound.
Hence let me tell my critic that he cannot possibly be half as em-
phatic concerning the accuracy of his syllabic divisions as I am with
regard to their inaccuracy. His plea that his informant "was particu-
larly careful in syllabifying" merely betrays an inexperience with D6n6
teachers which time will not fail to reveal. To prepare my monumental
Dictionary of the Carrier language, which was destroyed by fire, I must
have had, during the sixteen years that I worked on it, between 4,500
and 5,000 lessons or consultations, generally of several hours' duration.
510 VI American Indian Languages 2
143
I likewise took down vocabularies of the Chilcotin, Babine, Sekanais
and Nah'ane dialects. Now I never once found a native instructor who
could syllabify correctly, or in the same way as the preceding one! This
task must be done by thescholar who has mastered the language sufficiently
to be in a position to judge for himself, after a careful deductive process.
Perhaps the conjugation of one tense, the eventual, of the very verb
of which Sapir gives one C. C. person {Cd-yi-lnd) will help opening his
■eyes to the truth of my contention.
Chilcotin Carrier
tha-ros-tnan, I shall drink iha-us-nai'
tha-ron-tnan tha-Hn-tnai
tha-ro-tnan tha-u-tnai'
tha-ro-tnan tha-tsu-tnai'
tha-roh-tnan tha-uh-tnai'
tha-qu-tnan tha-Ru-tnaV
Dual: tha-6-tnai'
Here we have at a glance the full root -tnan, -tnai\ the equivalent of
C. C. -tnd, Sekanais and Hare -ton, Montagnais -dan, fully distinct from
the pronominal crements -ros-, -us-; -roil-, -un- etc., which latter are, in
turn, clearly differentiated from the qualifying prefix tha-, which stands
for thu, tho, "water, " and indicates that the verb is intransitive, because it
already contains in its make-up some sort of a complement. Who, with
ever so slight a tincture of Dene morphology, would dream of having, for
instance, tha-st-nan, "I drink"; tha-int-nan, "thou drinkest," etc., any
more than Montagnais esd-an, Hare el-on?
As to Dr. Sapir's No. 10, while he admits that his verbal stem -t^AC
(or -thcec) is genetically related to my -thoss, which in all the dialects that
I know of refers to the plural, he assures me that it is in Chasta Costa
really used in connection with singular subjects. This is very surprising
and well deserves deep investigation. Pending this, I accept his word
for it, merely suggesting that this root is possibly not the equivalent of
northern -thces, which is strictly plural in intent, but of singular -thih, the
desinential -h corresponding here to C. G. -c. Until I get incontrovertible
evidence of it I cannot conceive of a plural form coming to express a
singular concept.
With regard to Dr. Sapir's criticism under No. 11, he writes (p. 722):
"Father Morice is, in my opinion, quite right." What, then, is corrigen-
dum in this, his own statement or mine?
N. B. — I take this opportunity to remark that, in addition to those pointed out
by Dr. Sapir, the following misprints occurred in my appreciation of his paper
on the Chasta Costa language:
Appendix
511
144
Page
instead of
read
Page
instead of
read
560
t.okcet
tlokcst
567
hwenti
hwentil
561
theR
thea
567
ItsetuB-sos'a
Itsence-soe' a
564
cerael'en
CB Reel' en
570
yad<Bzquh
yadcezquh
564
nukweidintsi'
nuhkweidintsi'
570
yadcezqeh
yadcezqeh
564
verb -kessi'
verb -kessi'
570
yatse-saest.s
yalse- sees tits
567
tsit
tsit
571
(note 3) tse
tu
567
(note 2) tedceKul tedceKu
571
(do.) tsi
tsi'
A. G. MoRiCE, O.M.I.
Saint Boniface
Manitoba,
Canada.
* Just to show Dr. Sapir that, especially in material such as his and mine, an
author is not always responsible for the printed mistakes, I will point out the fact that
his fe".' does not correspond to my q, as he is made to say in note 16 of his first essay.
This, at least, is a real corrigendum — but, I am sure, imputable to the printer alone.
Editorial Note
Originally published in American Anthropologist 19, 132-144 (1917).
Reprinted by permission of the American Anthropological Association.
Corrigenda and Addenda to Takelma Texts
[After the publication of Takelma Texts (Sapir 1909c), Sapir published a
set of corrigenda and addenda to that work; these were bound with his Notes
on Chasta Costa Phonology and Morphology (1914c: 265 — 267). The correc-
tions made in the Takelma Texts as they appear in Volume VIII of The
Collected Works were based on Sapir's manuscript notes in his own copy of
the texts. The latter do not fully coincide with the published Corrigenda. The
items given below are noted in the 1914 publication, but were not in the
manuscript notes, and were inadvertently omitted from Volume VIII. The
remainder of the changes listed in 1914 appear in Volume VIII, either as
corrections in the text itself, or as editorial endnotes. The following items
should therefore be incorporated in the texts in volume VIII, which appeared
before the present volume. (Note that Sapir's line count refers to lines of
native text plus gloss, so that "1. 4 (interlinear)" is the gloss of the fourth
line of text; page numbers are those of the original.)
- Eds.]
p. 13, 1. 10 (interlinear): change "mourning" to "bereft of child"
p. 14, 1. 11 (text): change ahailhvili'" to ahailiwila"'
p. 22, 1. 8 (interlinear): change "Di'^lo^mi'" to "Di'^lo"mi'"
p. 22, 1. 10 (text): change xa'iyasgip.'Tlhf to xa'iyasgipTlhi
p. 22, 1. 10 (interlinear): omit ", it is said"
p. 23, 1. 2 (interlinear): insert quotes before "Strings"
p. 25, 1. 3 (text): change da-idamak' to deHdamcfk'
p. 31, 1. 14 (text): change fnu'"x-danhi' to mii"'xdcmhi
p. 46, 1. 1 (text): change t.'i'ldp'agit'gwa to t.'i'lap'igit'gna
p. 50, 1. 2 (text): change he'Tleme'k'wana' to he'Tleme'''k'nana'
p. 60, 11. 6, 7 (text): change nagaik'wd to nagaik'wa
p. 76, 1. 8 (interlinear): change "with it" to "thereby"
p. 78, 1. 4 (text): change he'dcdd to he'dadd
p. 105, 1. 14 (text): change ' to -
p. 122, 1. 21 (text): change dchu""k'i to dehu"k'i
p. 127, note 6, add: -t'git' 'i't'e' is very likely transformed Ironi -I'gnai' cJi'c' "1 am
provided with ...". See T. L., p. 261, footnote,
p. 142, 1. 7 (text): change k'ai'ldp'ak.'i to k'ai'ldp'ik.'i
p. 144, 1. 17 (text): change sgalduk, to sgalduk'
p. 188, 1. 9 (text): change wi'wdkdi' to u-i'ndk'dl
p. 194, 1. 10: change tlomomdn to t.'onionidn
p. 229, 1. 34: change iter, to usit.
p. 229, 1. 35: change usit. to iter.
p. 230, 1. 21: change lia'wi- to liawi-
514 Vf American Indian Languages 2
p. 238, s.v. yewei-, add:
ba''- revive, be cured
ba"-'^i-yewe'=-n-(i-) 2 III cure, bring to life
: -yeV-an-
p. 263, note 1: change yowo to yowo-
Phonetic Key to Publication^^
of Edward Sapir
Compiled by William Bright
The following list of symbols is based on the planned contents of The Col-
lected Works of Edward Sapir, volumes V and VI, plus selected monographic
works. It does not attempt to include symbols of unambiguous value (such as a
or k); or symbols used only in material quoted from other authors; or symbols
used in standard orthographies or Roman transliterations of familiar Old
World languages. Phonetic symbols proposed in the Report of the Committee
on Phonetic Transcription (1916), but not found in Sapir s publications, are also
omitted here.
Part 1 consists of diacritic modifications which are used with the entire class of
consonants (indicated by C) or of vowels (indicated by v). Part 2 consists of
letters or digraphs of the Roman alphabet which are used with special values,
with or without diacritics. Part 3 consists of Greek alphabetic symbols. Part 4
consists of miscellaneous segmental symbols.
Part 1.
C^ Glottalized or ejective stop or affricate; glottalized spirant or
sonorant. Also printed as C and 'C. In some earlier work, this
notation indicated a weaker degree of glottalization in stops than
C!
C! Glottalized or ejective stop or affricate (in earlier work). Note that
ts! tc! are glottalized affricates.
C Stop or affricate with strong aspiration.
C Syllabic consonant.
"C Nasalized consonant.
C",C^ Labialized consonant.
Cy Palatalized consonant.
C- Long or geminated consonant.
C.C Non-affricate consonantal sequence, e.g. I.s when it represents a
sequence of two independent consonants.
^ (Superscript) Weakly articulated consonant, echoed consonant, or
consonantal glide. In early work, whispered or voiceless sonorant
(later written with a small capital).
516 y^I American Indian Languages 2
V Primary stress; in Athabaskan, high pitch. In Takelma forms of
Sapir & Swadesh 1946, stress with high or rising tone.
V Secondary stress; in Athabaskan, low pitch. In Takelma forms of
Sapir & Swadesh 1946, "stress with falling pitch". In the traditional
orthography of Nahuatl, vowel with following glottal stop.
V Falling pitch in Athabaskan.
V Rising pitch in Athabaskan.
^ Intermediate pitch in Sarcee. (For intermediate falling and rising
pitches, see Sapir 1925f, Volume VI).
V Rising pitch in Takelma and Pawnee.
V Vowel length; in some early work, close vowel quality, not neces-
sarily with length. (See specific symbols in Part 2, below.)
V Nasalization of vowel.
if Glottalization of vowel.
v' Primary stress.
V Secondary stress.
V Length of vowel.
v: Length of vowel; when contrasted with v, indicates more than
ordinary length.
v' Unusual shortness of vowel.
(v) Glide value of vowel.
V. V Non-diphthongal sequence of vowels, e.g. a.i divided between two
syllables.
v'v or v'^ Glottal interruption of vowel; the latter alternative is recom-
mended when the post-glottal portion is weakly articulated.
yv or v-^ "Over-long" vowel with glide-like rearticulation, as in Takelma and
Sarcee.
X^ (Superscript vowel) Reduced prominence of vowel; e.g. murmured
or echoed quality, or vocalic resonance of a preceding consonant.
Also indicates whispered or voiceless vowels, later written with
small capitals.
Xv (Subscript vowel) In Indo-European, a reduced vowel.
Part 2.
a Open mid-back unrounded vowel, as in Eng. but; usually replaced
by small capital a or Greek alpha.
Phonetic Key 517
a Long low back rounded vowel, as in Eng. law.
a Low front unrounded vowel, as in Eng. hat.
a Open low back rounded vowel.
a In Sarcee, "a velarized, dark-timbred «".
A (small capital) Voiceless a, as in Southern Paiutc; in Wishram, a
mid back unrounded vowel, as in Eng. but.
A (large capital) Voiceless a; in Takelma and Chasta Costa, a mid
back unrounded vowel, as in Eng. but: in Indo-European, cover
symbol for a reconstructed iaryngeaP.
A In Indo-European, cover symbol for an a-coloring laryngeal, i.e.
one which induces a-timbre in an adjacent vowel.
b Voiced or "intermediate" labial stop; "intermediate" refers to
sounds heard sometimes voiced, sometimes voiceless — probably
voiceless lenis in most cases (esp. in Hittite.)
b (barred b) In Hebrew, spirant b.
B, B (Capital or small capital) "Intermediate" labial stop; see b above.
c In Nahuatl traditional orthography, used as in Spanish: s before /
or e, but k elsewhere. In Sapir s earlier works, a voiceless alveo-
palatal sibilant like Eng. sh — subsequently replaced by s; in later
works, a voiceless alveolar affricate, equivalent to earlier ts.
q In Nahuatl traditional orthography, equivalent to Mexican Spanish
z, i.e. Eng. s.
c Voiceless alveo-palatal affricate, English ch as in church, equiv-
alent to earlier tc or ts.
c In Esperanto, equivalent to c.
ch In Nahuatl traditional orthography, equivalent to c.
d Voiced or "intermediate" apical stop; see b above.
d Voiced "cerebral" or retroflex apical stop.
d In Hebrew, spirant d.
6 Voiced interdental spirant, as in English this.
D, D (capital or small capital) Voiced or "intermediate" apical slop; see
b above.
dj, dz Voiced or "intermediate" alveo-palalal affricate (see b, above):
replaced in later work by %.
e Long open mid front unrounded vowel, as in French fete.
518 VI American Indian Languages 2
e In earlier work, an open mid front unrounded vowel, as in Eng.
met; in Yana, Chasta Costa, and Nootka, the long vowel of this
quality, as in French fete.
e Close mid front vowel, as in French ete.
e Close mid front vowel, as in French ete.
e In Athabaskan reconstructions, a formula for "the reduced or
'pepet' vowel", i.e. 3.
E (Small capital) In early work, "short obscure vowel of undefined
quality", as in unaccented English the (replaced in later works by
3); occasionally also indicates voiceless e.
g Voiced or "intermediate" velar stop; see b above.
g,g,g,g-,gy Voiced or "intermediate" front-velar stop.
g In Semitic, a voiced velar spirant ( = 7).
g (barred g) in Haida, an "intermediate" velar stop.
g Voiced or "intermediate" back-velar stop.
G, G "Intermediate" velar stop; see b, above.
h In earlier work, represents "strong aspiration" as opposed to the
backward apostrophe; later, h is preferred whenever an indepen-
dent consonant is indicated, as opposed to aspiration of a stop.
h In Sanskrit (and stated to be in Semitic), a voiceless laryngeal spi-
rant; in Nootka (and perhaps properly in Semitic), a phar-
yngealized h. [Some of Sapir s publications seem to use "laryngeal"
improperly for "pharyngeal"; Lg. 10.276 (1934), fn. 4, is
bewildering.]
h (with semicircle below) In Hittite, a velar spirant.
hu In traditional Nahuatl orthography, equivalent to w.
n (small capital) In Nootka, pharyngealized h; later replaced by h.
i In some early work, / as in English it; also a long vowel of that
quality.
i High back unrounded vowel.
i In Tutelo, "as in French fini" (in this language, plain / is "as in
English /?//").
I (Small capital) In Nootka, a "short open i-vowel of rather unclear
quality"; in early work on Southern Paiute, an "obscure" i; later, in
Southern Paiute, voiceless i.
Phonetic Key 519
I (capital) In early work on Southern Paiute, an "obscure" i.
j In earlier work , an alveo-palatal sibilant, like French j in jour\ later
replaced by z.
J Voiced or "intermediate" alveo-palatal affricate.
R,k,k,k- Voiceless front-velar stop.
k Voiceless back-velar stop, =q.
k (with underbar) In Tutelo, an "intermediate" velar stop.
ky,ky Voiceless front-velar stop.
/ (Italic 1) In Yana, a voiceless lateral.
1 Voiceless lateral spirant of American Indian languages; in Wish-
ram, described as "voiceless palatal lateral".
1, i Voiceless lateral spirant as in Welsh //.
L (Small capital) In early work, a voiceless lateral spirant, replaced
later by 1.
L (Capital) In early work, sometimes a voiceless lateral spirant (1),
sometimes the corresponding affricate (equivalent to tL, tl, or K).
m (italic) In Yana, an unvoiced bilabial nasal.
M (small capital) Unvoiced labial nasal.
n (italic) In Yana, an unvoiced apical nasal.
n In earlier work, a velar nasal; later replaced by r).
n In earlier work, a back velar nasal.
V" (superscript) Nasalization of preceding vowel.
Velar nasal; varies typographically with Greek eta (t|).
N (small capital) Voiceless apical nasal.
N (small capital) Voiceless velar nasal.
6 In earlier work, an open mid back rounded vowel, as in German
voll\ in Southern Paiute and Nootka. a long low back rouiulcd
vowel, as in Eng. saw. Later replaced in both values by o.
o Mid front rounded vowel, as in German schon.
6 In Nootka, a close mid back rounded vowel, as in French chaud.
o In Indo-European, low back roundcci |;i]. produced by the
'laryngeal' 7 next to the e-type full-grade vowel.
520 ^f American Indian Languages 2
C (Subscript o) In Indo-European, cover symbol for shwa (murmur
vowel) or syllabic resonance.
ou In Takelma, "like o but with final u-vanish''.
Open mid or low back rounded vowel.
5 Low mid back rounded vowel, as in German Gotter.
p (with underbar) In Tutelo, an "intermediate" labial stop.
q Voiceless uvular (back-velar) stop.
qu In traditional Nahuatl orthography, a voiceless labiovelar stop
(kw,k^).
r (italic) In Yana, voiceless r; in Tsimshian, a voiced uvular r.
r Voiced uvular r.
•■ (superscript) In Yana, the combination ""t indicates a "peculiar
voiceless-r quality" oft.
R, R Voiceless vibrant.
R (small capital) Voiceless uvular r.
s Voiceless alveo-palatal sibilant, replacing earlier c.
s In Esperanto, equivalent to s.
s In Chasta Costa and Nootka, a "palatalized" c (i.e. s), "acoustically
midway between s and c".
s- In Yana, a sibilant "midway acoustically" between s and c (i.e. s).
t Voiceless "cerebral" or retroflex stop.
t (with underbar) in Tutelo, an "intermediate" apical stop.
tc In earlier work, a voiceless alveo-palatal affricate; later replaced
by ts, c.
tl In traditional Nahuatl orthography, a voiceless lateral affricate,
equivalent to tt or K.
tL, tl In earlier work, a voiceless lateral affricate, equivalent to X..
tl, tj Equivalent to X.
ts In earlier work, a voiceless alveolar affricate; later replaced by c.
ts- In Yana. an affricate corresponding to s-.
ts In earlier work, a voiceless alveo-palatal affricate, like English ch
in church; later replaced by c.
Phonetic Key 521
te In Chasta Costa and other Athabaskan, a voiceless interdental
affricate.
tz In traditional Nahuatl orthography, a voiceless alveolar affricate,
equivalent to phonetic ts.
u In early work on Southern Paiute, a long open high back rounded
vowel.
u High front rounded vowel, as in German kiihl; in early work on
Southern Paiute, used for a high back unrounded vowel (later writ-
ten as i).
uh In traditional Nahuatl orthography, a voiceless labiovelar semi-
vowel, equivalent to phonetic W.
U (capital) In Southern Paiute, "a duller variety of T.
u (small capital) In Southern Paiute, same as U; elsewhere, voiceless
u.
V In early work on Southern Paiute, a voiced bilabial spirant; later
written with Greek beta.
V (italic) In early work on Southern Paiute, a voiceless bilabial spi-
rant; later written with Greek phi.
vw In early work on Southern Paiute, a voiced bilabial spirant "with
inner rounding".
V (capital) In early work on Southern Paiute, a voiceless bilabial spi-
rant; later replaced by Greek phi.
w, ^y In Indo-European, systematic equivalent for u.
W (capital) Voiceless bilabial semivowel.
X In Nahuatl traditional orthography, a voiceless alveo-palatal sib-
ilant, equivalent to s; in phonetic transcription, a voiceless velar
spirant, like German ch in cich; but in Wishram, further back than
the German sound.
X, X, x' Voiceless front-velar fricative, like ch in German ich.
X Voiceless uvular (back-velar) fricative; in Wishram, a Iricative
"between ch of German ach and ch of (icrnian ich": in llitiite,
velar.
x' In Yana, "as in German ich^.
xy Voiceless front-velar fricative, as in German ich.
"C (superscript x) In earlier work on Scuithcrn Paintc. "weak x
developed from [aspiration] before . . . velar q ".
522 ^f American Indian Languages 2
5^, Y In Indo-European, systematic equivalent for i.
V (small capital) Voiceless palatal semivowel.
z Voiceless alveo-palatal sibilant, replacing earlier j.
3 Voiced or lenis alveolar affricate, replacing earlier dz.
3 Voiced or lenis alveo-palatal affricate, replacing earlier dj or dz.
Part 3 (Greek letters)
a. (alpha) Lower mid back unrounded vowel, as in Eng. but.
3 (beta) Voiced bilabial spirant.
7 (gamma) Voiced velar spirant ("North German g in Tage"" , Arabic
'ghain').
•y Voiced front-velar spirant.
-y Voiced back-velar spirant.
8 (delta) Voiced interdental spirant, as in Eng. this.
e (epsilon) Open mid front unrounded vowel, as in Eng. met.
^ (superscript) In earlier works, indicates a glottal stop; varies
typographically with an inverted superscript "3". Replaced in later
work by ^.
7] (eta) A typographical variant of o, the velar nasal.
TQ In Eskimo, the uvular (back-velar) nasal.
9 (theta) Voiceless interdental spirant, as in Eng. think.
I (iota) Open front unrounded vowel, as in Eng. pit.
L (iota with dot) In Southern Paiute, a high central unrounded vowel.
L (iota with dieresis) Open back unrounded vowel.
X (lambda) Voiced or lenis lateral affricate, equivalent of dl.
K (barred lambda) Voiceless lateral affricate, equivalent to tL or tl.
p (rho) Uvular r.
V (upsilon) Open high back rounded vowel, as in Eng. put.
V (upsilon with dieresis) Open high front rounded vowel, as in German
Miitze.
<(> (phi) Voiceless bilabial spirant.
X (chi) Voiceless velar fricative, equivalent to x.
X Voiceless front velar fricative, equivalent to x.
Phonetic Key 523
X Voiceless back velar fricative, equivalent to x.
(o (omega) Low back rounded vowel, as in Eng. law.
Part 4. Other symbols
^ (inverted superscript "3^ varies typographically with superscript
Greek epsilon). In earliest work, indicates glottal stop. Later
replaced by the apostrophe, then by ^.
^ In Nootka, "a peculiarly harsh and choky glottal stop," i.e. a phar-
yngealized glottal stop, in later work written as ! .
("smooth breathing") In earlier American Indian work (and as late
as 1938 for Indo-European), the glottal stop; sometimes indicates a
weak articulation, as opposed to a "true" '^. In later work, the apos-
trophe is retained only to mark glottalization of consonants; else-
where, ^ is used. When used for glottalization, the apostrophe is
placed before resonants ('m 'n 'w 'y), after stops (p' t' k'), and
internally for affricates (t's t's). — In Wishram, denotes elision of
final vowel. In Ugaritic, '1 '2 '3 indicate glottal stop with one of
three vowels.
(prime) A feature of "hardening" which imparts glottalization in
Nootka.
^ Glottal stop.
' ("rough breathing") In earlier work, the voiceless laryngeal spi-
rant; sometimes indicates a weak articulation, as opposed to h. In
later work, the rough breathing is retained only to mark aspiration
of stops; elsewhere, h is used. In Semitic, indicates the Arabic "uin,
inexactly stated to be a "voiced laryngeal spirant".
'y "Palatalized aspiration", equivalent to the voiceless front-palatal
spirant of German ich.
! In Nootka, a pharyngealized glottal stop; in Indo-European, a
glottal (stop) phoneme inducing velar or a-timbrc.
9 ("shwa") A weak mid central unrounded vowel, like the a in Eng.
idea.
+ Extra length of a preceding symbol.
= Between vowels in Wishram, denotes that they "are to be pro-
nounced separately".
References
Works appearing in The Collected Works of Edward Sapir are designated by a
roman numeral, referring to the appropriate volume, at the end of the entry.
Aoki, Haruo
1975 The East Plateau Linguistic Diffusion Area. InternationalJournal
of American Linguistics 41, 183-199.
Barrett, Samuel A.
1908 The Ethno-Geography of the Porno and Neighboring Indians.
University of California Publications in American Archaeology
and Ethnology 6(1), 1-332.
Berman, Howard
1983 Some California Penutian Morphological Elements. International
Journal of American Linguistics 49, 400-412.
Bibolotti, Benigno
1917 Moseteno Vocabulary and Treatises. Evanston and Chicago:
Northwestern University.
Boas, Franz
1890 The Indians of British Columbia: Tlingit, Haida, Tsimshian,
Kutonaqa. Fifth Report on North-Western Tribes of Canada,
1889. Report of the Meeting of the British Association for the
Advancement of Science, 1890, 797-893.
1891 The Indians of British Columbia: LkungEn, Nootka, Kwakiutl,
Shuswap. Sixth Report on the North-Western Tribes of Canada,
1890. Report of the Meeting of the British Association for the
Advancement of Science, 1891, 553-715.
1898 Linguistics. In Twelfth and Final Report on the North-Wcstcrn
Tribes of Canada, 1898. Report of the Meeting of the British Asso-
ciation for the Advancement of Science, 1898, 654-666.
1904 Notes on the Chinook Vocabulary. American Anthropologist n.s.
6, 118-147.
1910 Kwakiutl Tales. Columbia University Contributions to Anthro-
pology 2.
1911 Handbook of American Indian Languages, Pan 1. Bureau oi
American Ethnology, Bulletin 40. Washington, DC: Smithson-
ian Institution.
1912 Tsimshian Texts, New Series. Publications o\ the American Eth-
nological Society 3, 65-284. Leiden: E. J. Brill.
1917 Grammatical Notes on the Language of the Tlingit Indians.
University of Pennsylvania, University Museum, Anthropologi-
cal Publications 8(1), 1-179.
526 yf American Indian Languages 2
1920 The Classification of American Languages. American Anthro-
pologist 22, 367-376.
1922 Handbook of American Indian Languages, Part 2. (Bureau of
American Ethnology, Bulletin 40). Washington, D.C.: Smithson-
ian Institution.
1930 The Religion of the Kwakiiitl Indians. Part I: Texts. Columbia Uni-
versity Contributions to Anthropology 10, 1-284.
1966 Kwakiutl Ethnography. Edited by Helen Codere. Chicago: Uni-
versity of Chicago Press.
Boas, Franz, and Pliny E. Goddard
1924 Ts'ets'aut, an Athapascan Language from Portland Canal, British
Columbia. International Journal of American Linguistics 3, 1-35.
Boas, Franz, and George Hunt
1902-05 Kwakiutl Texts. Publications of the Jesup North Pacific Expedition
3. Memoirsof the American Museum of Natural History 5. Leiden
and New York: American Museum of Natural History.
1906 Kwakiutl Texts — Second Series. Publications of the Jesup North
Pacific Expedition 10(1). Memoirs of the American Museum of
Natural History 14, 1-269. Leiden and New York: American
Museum of Natural History.
Bommelyn, Loren, et al.
1989 The Tolowa Language. Second edition. Areata, Calif.: Center for
Community Development, Humboldt State University.
Bright, Jane O.
1964 The Phonology of Smith River Athapaskan (Tolowa).
International Journal of American Linguistics 30, 101-107.
Campbell, Lyle
1979 Middle American Languages. In Campbell and Mithun (eds.)
1979, 902-1000.
Campbell, Lyle, and Terrence Kaufman
1976 A Linguistic Look at the Olmecs. American Antiquity 4, 80-89.
Campbell, Lyle, and Marianne Mithun (eds.)
1979 The Languages of Native America: Historical and Comparative
Assessment. Austin and London: University of Texas Press.
Chapman, John Wight
1914 Ten'a Texts and Tales from Anvik, Alaska. Publications of the
American Ethnological Society 6. Leiden: E. J. Brill.
Cook, Eung-Do
1971 Vowels and Tones in Sarcee. Language 47, 164-179.
1984 A Sarcee Grammar Vancouver: University of British Columbia
Press.
References 527
Dallaire, Louise
1984 Edward Sapir's Correspondence: An Alphabetical and Chronolog-
ical Inventory, 1910-1925. Paper 97, National Museum of Man
Mercury Series, Canadian Ethnology Service, National Museums
of Canada. Ottawa.
Darnell, Regna, and Joel Sherzer
1971 Areal Linguistics in North America: A Historical Perspective.
International Journal of American Linguistics 30, 20-28.
Dixon, Roland B.
1909 Outlines of Wintun Grammar. In Boas, Franz (ed.), Putnam Anni-
versary Volume, 461-476. New York: G. E. Stechert.
1911 Maidu. In Boas (ed.) 1911, 679-734.
Dixon, Roland B., and Alfred L. Kroeber
1913 New Linguistic Families in California. American Anthropologist
15, 647-655.
1919 Linguistic Families of California. University of California Publica-
tions in American Archaeology and Ethnology 16(3), 47-118.
Dorsey, J. Owen
1890 The Gentile System oftheSiletz Tribes. 7owr/ifl/o//lmencfl/i Folk-
Lore 3, 227-237.
Dyk, Walter
1933 A Grammar of Wishram. Ph.D. dissertation, Yale University.
Enrico, John
1986 Word Order, Focus, and Topic in Haida. International Journal of
American Linguistics 52, 91-123.
Ferguson, Charles A.
1964 Baby Talk in Six Languages. American Anthropologist 66, part 2,
103-114.
Frachtenberg, Leo J.
1913a Contributions to a Tutelo Vocabulary. American Anthropologist
15, 477-479.
1913b Coos Texts. Columbia University Contributions to Anthropology
1.
1914a Coos. Extract from Boas (ed.) 1922, 297-429.
1914b Lower Umpqua Texts. Columbia University Contributions to
Anthropology 4.
1917 Siuslawan (Lower Umpqua). Extract from Boas (cd.) 1922,
431-629.
1918 Comparative Studies in Takelma, Kalapuyan. and Chinookan
Lexicography: A Preliminary Paper, hitcrnatiotial Journal of
American Linguistics 1, 175-182.
1920 Abnormal Types of Speech in Quileute. International Journal of
American Linguistics 1, 295-299.
528 VI American Indian Languages 2
Franciscan Fathers
1910 An Ethnologic Dictionary of the Navaho Language. St. Michaels,
Arizona.
1912 A Vocabulary of the Navaho Language. 2 volumes. St. Michaels,
Arizona.
Freeland, L. S.
1930 The Relationship of Mixe to the Penutian Family [with notes by
Edward Sapir]. International Journal of American Linguistics 6,
28-33.
Freeman, John E., and Murphy D. Smith
1966 A Guide to Manuscripts Relating to the American Indian in the
Library of the American Philosophical Society. Philadelphia:
American Philosophical Society.
Goddard, Pliny Earle
1905 The Morphology of the Hupa Language. University of California
Publications in American Archaeology and Ethnology 3, 1-344.
1907 The Phonology of the Hupa Language: Part I — The Individual
Sounds. University of California Publications in American
Archaeology and Ethnology 5(1), 1-19.
1909 Kato Texts. University of California Publications in American
Archaeology and Ethnology 5(3), 65-238.
1912a Analysis of Cold Lake Dialect, Chipewyan. Anthropological
Papers, American Museum of Natural History 10, part 2, 67-170.
1912b Elements of the Kato Language. University of California Publica-
tions in American Archaeology and Ethnology 11(1), 1-176.
1917 Beaver Dialect. Anthropological Papers, American Museum of
Natural History 10, part 6, 399-546.
1920 Has Tlingit a Genetic Relationship to Athapascan? International
Journal of American Linguistics 1, 266-279.
1928 Pitch Accent in Hupa. University of California Publications in
American Archaeology and Ethnology 23, 333-338.
Golla, Victor
1970 Hupa Grammar. Ph.D. dissertation. University of California,
Berkeley.
1976 Tututni (Oregon Athapaskan). International Journal of American
Linguistics 4, 217-227.
1984 (ed.) The Sapir-Kroeber Correspondence: Letters between Edward
Sapir and A. L. Kroeber, 1905-25. Survey of California and Other
Indian Languages, Report 6. Berkeley: Department of Lin-
guistics, University of California.
1986 Sapir, Kroeber, and North American Linguistic Classification. In
Cowan, William, et al. (eds.). New Perspectives in Language,
Culture, and Personality: Proceedings of the Edward Sapir Cente-
nary Conference (Ottawa. October 1-3, 1984) , 17-38. Amster-
dam: Benjamins.
References 529
Greenberg, Joseph
1956 General Classification of Central and South American Lan-
guages. In Wallace, Anthony F. C. (ed.). Men and Cultures:
Selected Papers, Fifth International Congress of Anthropological
and Ethnological Sciences, 791-794. Philadelphia: University of
Pennsylvania Press.
1987 Language in the Americas. Stanford: Stanford University Press.
Greenberg, Joseph, Christy G. Turner and Stephen L. Zegura
1986 The Settlement of the Americas: A Comparison of the Linguistic,
Dental and Genetic Evidence. Current Anthropology 27, 477-497.
Haas, Mary R.
1970 Diminutive Symbolism in Northern California: A Study in Diffu-
sion. In Swanson, Earl H. (ed.). Languages and Cultures of West-
ern North America, 86-96. Pocatello: Idaho State University
Press.
Haeberlin, Herman K.
1918 Types of Reduplication in the Salish Dialects. International Jour-
nal of American Linguistics 1, 154-174.
Haile, Berard
1926 A Manual of Navaho Grammar. Santa Fe: Santa Fe Publishing
Company.
1941-48 Learning Navaho. 4 vols. St. Michaels, Ariz.: St. Michaels Press.
1950-51 A stem vocabulary of the Navaho language. St. Michaels, Ariz.:
St. Michaels Press.
Hess, Thomas M.
1966 Snohomish Chameleon Morphology. International Journal of
American Linguistics 32, 350-356.
Hill-Tout, Charles
1902 Ethnological Studies of the Mainland Halkome'lEm, a Division of
the Salish of British Columbia. Report of the British Association
for the Advancement of Science, Ethnological Survey of Canada,
72, 355-490.
191 1 Report on the Ethnology of the Okanak-en of British Columbia.
Journal of the Royal Anthropological Institute of Great Britain and
Ireland 4\, 130-161.
Hoijer, Harry
1960 Athapaskan Languages of the Pacific Coast. In Diamond. Stanley
(ed.). Culture in History, 960-976. New York: C\)lunibia Univer-
sity Press.
1966 Galice Athapaskan: A Grammatical Sketch. Inicrnaiional Journal
of American Linguistics 32, 320-327.
1974 A Navajo Lexicon. University of California Publications in Lin-
guistics 78. Berkeley and Los Angeles: University of California
Press.
530 VI American Indian Languages 2
Hoijer, Harry, and Janet Joel
1963 Sarsi Nouns. In Hoijer, Harry (ed.). Studies in the Athapaskan
Languages (University of California Publications in Linguistics
29), 62-75. Berkeley and Los Angeles: University of California
Press.
Hymes, Dell
1955 Positional Analysis of Categories: A Frame for Reconstruction.
Word 11, 10-23.
1956 Na-Dene and the Positional Analysis of Categories. American
Anthropologist 58, 624-638.
1957 Some Penutian Elements and the Penutian Hypothesis.
Southwestern Journal of Anthropology 13, 69-87.
1979 How to Talk Like a Bear in Takelma. International Journal of
American Linguistics 45, 101-106.
1981 '7a7 vain I tried to tell you": Essays in Native American Eth-
nopoetics. Philadelphia: University of Pennsylvania Press.
Jacobsen, William H., Jr.
1979 Wakashan Comparative Studies. In Campbell and Mithun (eds.)
1979, 766-791.
Jenness, Diamond
1932 The Indians of Canada. Bulletin 65, Anthropological Series 15,
Geological Survey, Department of Mines, Canada. National
Museum of Canada. Ottawa.
Kendall, Daythal
1982 A Supplement to A Guide to Manuscripts Relating to the American
Indian in the Library of the American Philosophical Society.
Philadelphia: American Philosophical Society.
Kess, Joseph Francis, and Anita Copeland Kess
1986 Nootka Baby Talk. International Journal of American Linguistics
52, 201-211.
Krauss, Michael E.
1964 Proto-Athapaskan-Eyak and the Problem of Na-Dene I: The Pho-
nology. International Journal of American Linguistics 20, 118-131.
1965 Proto-Athapaskan-Eyak and the Problem of Na-Dene II: The
Morphology. International Journal of American Linguistics 3\,
18-28.
1968 Noun Classification Systems in Athapaskan, Eyak, Tlingit and
Haida Verbs. International Journal of American Linguistics 34,
194-203.
1969 On the Classifiers in the Athapaskan, Eyak, and Tlingit Verb.
Indiana University Publications in Anthropology and Linguistics
24. Supplement to International Journal of American Linguistics
35(4).
References 53 1
1973 Na-Dene. In Sebeok, Thomas A. (ed.). Current Trends in Lin-
guistics, Volume 10, Linguistics in North America, 903-979. The
Hague: Mouton.
1979 Na-Dene and Eskimo-Aleut. In Campbell and Mithun (eds.)
1979, 803-901.
1986 Edward Sapir and Athabaskan Linguistics. In Cowan, William,
Michael K. Foster and Konrad Koerner (eds. ), New Perspectives in
Language, Culture, and Personality: Proceedings of the Edward
Sapir Centenary Conference (Ottawa, 1-3 October 1984), 147-190.
Amsterdam and Philadelphia: Benjamins.
Krauss, Michael E., and Victor Golla
1981 Northern Athapaskan Languages. In Helm, June C. (ed.).
Handbook of North American Indians, Volume 6, Subarctic,
67-85. Washington, D. C: Smithsonian Institution.
Krauss, Michael E., and Jeff Leer
1981 Athabaskan, Eyak, and Tlingit Sonorants. Alaska Native Lan-
guage Center Research Papers 5. Fairbanks: Alaska Native Lan-
guage Center.
Kroeber, Alfred L.
1907 The Yokuts Language of South Central California. University of
California Publications in American Archaeology and Ethnology
2(5), 165-377.
1910 The Chumash and Costanoan Languages. University of California
Publications in American Archaeology and Ethnology 9, 237-271 .
1911 The Languages of the Coast of California North of San Francisco.
University of California Publications in American Archaeology
and Ethnology 9, 273-435.
1984 Reflections on Edward Sapir, Scholar and Man. In Koerner,
Konrad (ed.), Edward Sapir: Appraisals of his Life and Work,
131-139. Amsterdam and Philadelphia: Benjamins.
Kroeber, Paul D.
1988 Inceptive Reduplication in Comox. International .Journal of
American Linguistics 54, 141-167.
Kuipers, Aert H.
1978 On the Phonological Typology of Proto-Salish. Proceedings of the
Forty-second International Congress of Americanists, Paris. 1976,
4, 607-621.
Lawrence, Erma, and Jeff Leer
1977 Haida Dictionary. Fairbanks: Alaska Native Language Center.
Legoff, Laurent
1889 Grammaire de la langue montagnais. Montreal.
532 VI American Indian Languages 2
Lehmann, Walter
1920 Zentral-Amerika: 1. Die Sprachen Zentral-Amerikas in ihren
Beziehungen ziieinander sowie zu Siid-Amerika unci Mexiko.
Berlin: Museum fiir Volkerkunde zu Berlin, D. Reimer.
Levine, Robert D.
1977 The Skidegate Dialect of Haida. Ph.D. dissertation, Columbia
University.
1979 Haida and Na-Dene: A New Look at the Evidence. International
Journal of American Linguistics 45, 157-170.
Lewis, Albert B.
1906 Tribes of the Columbia Valley and the Coast of Washington and
Oregon. Memoirs of the American Anthropological Association 1,
part 2, 147-209.
Li, Fang-Kuei
1930 A Study of Sarcee Verb-Stems. InternationalJournal of American
Linguistics 6, 3-27. [XIII]
1933 A List of Chipewyan Stems. International Journal of American
Linguistics 7, 122-151.
Mason, John A.
1916 The Mutsun Dialect of Costanoan Based on the Vocabulary of De
la Cuesta. University of California Publications in American
Archaeology and Ethnology 11, 399-472.
Mills, Elaine (ed.)
1981 The Papers of John Peabody Harrington in the Smithsonian Institu-
tion, 1907-1957. A Guide to the Field Notes: Native American His-
tory, Language, and Culture of Alaska/Northwest Coast.
Millwood, N.Y. : Kraus International.
Mithun. Marianne
1983 The Last Tutelo a Century Later. Unpublished paper delivered at
the Third Annual Siouan Linguistics Conference, Rapid City,
South Dakota, May 20-22, 1983.
Morice, Adrien Gabriel
1889-90 The Dene Languages. Transactions of the Canadian Institute 1,
170-212.
1915a Review of Sapir, Notes on Chasta Costa Phonology and Mor-
phology. American Anthropologist 17, 347-350. VI.
1915b Chasta Costa and the Dene Languages of the North. American
Anthropologist 17, 559-572. VI.
1917 Misconceptions Concerning Dene Morphology: Remarks on Dr.
Sapir's Would-be Corrigenda. American Anthropologist 19,
132-144. VI.
1932 The Carrier Language (Dene Family): A Grammar and Dictionary
Combined. 2 volumes. St. Gabricl-Modling, Austria: Anthropos.
References 533
Murray, Stephen O. , and Wayne Dynes
1986 Edward Sapir's Coursework in Linguistics and Anthropology.
Historiographia Linguistica 13, 125-129.
Newman, Stanley
1944 Yakuts Language of California. Viking Fund Publications in
Anthropology 2. New York: The Viking Fund.
Nichols, Johanna
1971 Diminutive Consonant Symbolism in Western North America.
Language 47, 826-848.
1986 Head-marking and Dependent-marking Grammar. Language 62,
56-119.
Osgood, Cornelius
1936a Contributions to the Ethnography of the Kutchin. Yale University
Publications in Anthropology 14. New Haven: Yale University
Press.
1936b The Distribution of the Northern Athapaskan Indians. Yale Uni-
versity Publications in Anthropology 7. New Haven: Yale Univer-
sity Press. VI.
Petitot, Emile
1876 Dictionnaire de la langue Dene-Dindjie. Bibliotheque de Lin-
guistique et d'Ethnographie Americaines, 2. Paris: E. Leroux.
1891 Autour du Grand Lac des Esclaves. Paris: A. Savine.
Pinnow, Heinz-Jiirgen
1966 Grundziige einer Historischen Lautlehre des Tlingit: ein Versuch.
Wiesbaden: Harrassowitz.
1976 Geschichte der Na-Dene Forschung. Indiana, Beiheft 5.
1985a Sprachhistorische Untersuchung zur Stellung des Haida. Indiana
10, 25-76.
1985b Sprachhistorische Untersuchung einiger Tiernamen im Haida
(Fische, Stachelhauter, Weichtiere, Gliederfussc. u.a.).
Abhandlungen der Volkerkundlichen Arbeitsgemcinschafi 39.
Nortorf, West Germany: Volkerkundlichen Arbeitsgemeinschaft.
1985c Das Haida als Na-Dene-Sprache: Materialien zu den Wortfcldcrn
und zur Komparation des Verbs. I-IV. Abhandlungen der Vol-
kerkundlichen Arbeitsgemeinschaft 43-46. Nortorf: Vol-
kerkundlichen Arbeitsgemeinschaft.
1986a Die Zahlworter des Haida im sprachvcrglcichcndc Sicht.
Abhandlungen der Volkerkundlichen Arbeitsgcnicitischaft 47.
Nortorf: Volkerkundlichen Arbeitsgemeinschaft.
1986b Saugetiernamen des Haida und Tlingit: Materialien zu ihrer histo-
rischen Erforschung. Abhandlungen der Volkerkundlichen
Arbeitsgemeinschaft 50. Nortorf: Volkerkundlichen
Arbeitsgemeinschaft .
534 Vf American Indian Languages 2
1988 Verwandtschafts- und andere Personenbezeichnungen im Tlingit
und Haida: Versuch ihrer sprachhistorischen Deutung.
Ahhandlungen der Volkerkundlichen Arbeitsgemeinschaft 62.
Nortorf: Volkerkundlichen Arbeitsgemeinschaft.
1990 Die Na-Dene-Sprachen im Lichte der Greenberg-Klassifikation.
Ahhandlungen der Volkerkundlichen Arbeitsgemeinschaft 54.
Nortorf: Volkerkundlichen Arbeitsgemeinschaft.
Pitkin, Harvey
1984 Wintu Grammar. University of California Publications in Lin-
guistics 94. Berkeley: University of California Press.
Powell, John Wesley
1891 Indian Linguistic Families of America North of Mexico. In
Seventh Annual Report, Bureau of American Ethnology, 1-142.
Washington, D.C.: Smithsonian Institution.
Radin, Paul
1914 Some Myths and Tales of the Ojibwa of Southeastern Ontario.
Memoir 48, Anthropological Series 2, Geological Survey, Depart-
ment of Mines, Canada. Ottawa.
1916 On the Relationship of Huave and Mixe. American Anthro-
pologist 18, 411-421.
Rood, David S.
1979 Siouan. In Campbell and Mithun (eds.) 1979, 236-298.
Sapir, Edward
1906 The Rival Chiefs, a Kwakiutl Story Recorded by George Hunt
[edited, with synopsis, by Edward Sapir]. In Boas Anniversary
Volume, 108-136. New York: G.E. Stechert.
1907b Notes on the Takelma Indians of Southwestern Oregon. American
Anthropologist n.s. 9, 251-275. IV.
1907c Preliminary Report on the Language and Mythology of the Upper
Chinook. American Anthropologist n.s. 9, 533-544. IV and VI.
1907d Religious Ideas of the Takelma Indians of Southwestern Oregon.
Journal of American Folk-Lore 20, 33-49. IV.
1908a Luck-Stones among the Yana: Journal of American Folk-Lore 2\,
42. IV.
1909c Takelma Texts. University of Pennsylvania, Anthropological Pub-
lications 2(1). 1-263. VIII. Corr. /Addenda, 1914, 2(2), 265-267.
1909d Wishram Texts, together with Wasco Tales and Myths collected by
Jeremiah Curtin and edited by Edward Sapir. American Eth-
nological Society Publications 2. Leiden: E. J. Brill. VII.
1910d Song Recitative in Paiute Mythology. Journal of American Folk-
Lore 23, 455-472. IV.
1910h Yana Texts. University of California Publications in American
Archaeology and Ethnology 9, 1-235.
References 535
1911c The Problem of Noun Incorporation in American Languages. Amer-
ican Anthropologist 13, 250-282. V.
191 le Some Aspects of Nootka Language and Culture. American Anthro-
pologist 13, 15-28. IV and VI.
191 Ig Wishram Chinook [incorporated in Franz Boas, Chinook]. In Boas
(ed.) 1911, 578, 579, 625-627, 638-645, 650-654. 673-677. VL
1912a The Indians of the Province. In Boam, H. J. (compiler) and Ashley
G. Brown (ed.), British Columbia: Its History, People, Commerce,
Industries, and Resources, 135 — 140. London: Sells.
1912h The Takelma Language of Southwestern Oregon. Extract from Boas
(ed.)(1922), 1-296. VIII.
1913b A Girls' Puberty Ceremony among the Nootka Indians. Transactions.
Royal Society of Canada, Third Series, 7, 67 — 80. IV.
1913g A Tutelo Vocabulary. American Anthropologist 15, 295 — 297. VI.
1913h Wiyot and Yurok, Algonkin Languages of California. American An-
thropologist 17, 188-194. V.
1914b Indian Tribes of the Coast. In Shortt, Adam and Arthur G. Doughty
(eds.), Canada and Its Provinces, vol. 21, 315 — 346. Toronto: Glasgow,
Brook and Co. IV.
1914c Notes on Chasta Costa Phonology and Morphology. University of
Pennsylvania, Anthropological Publications, 2 (2), 271 — 340. VI.
1915a Abnormal Types of Speech in Nootka. Memoir 62, Anthropological
Series 5, Geological Survey, Department of Mines, Canada. Ottawa.
VI.
1915c Corrigenda to Father Morice's "Chasta Costa and the Dene Languages
of the North." American Anthropologist 17, 765 — 773. VI.
191 5d The Na-dene Languages, a Preliminary Report. American .Anthropol-
ogist 17, 534-558. VI.
191 5f Noun Reduplication in Comox, a Salish Language of Vancouver Island.
Memoir 63, Anthropological Series 6 , Geological Survey, Department
of Mines, Canada. Ottawa. VI.
191 5g A Sketch of the Social Organization of the Nass River Indians. Museum
Bulletin 19, Anthropological Series 65, Geological Survey, Department
of Mines, Canada. Ottawa. IV.
191 5h The Social Organization of the West Coast Tribes. Transactions. Roval
Society of Canada, Second Series, 9, 355 — 374. IV.
1915i Southern Paiute and Nahuatl. a Study in Uto-Aztekan. Part II. Amer-
ican Anthropologist 17, 98—120, 306-328; Journal de la Societe des
Americanistes de Paris 11 (1914- 1919). 443-488. V.
191 5j Canada. Summary Report of the Geological Survey. Department of
Mines, for the Calendar Year 1915. Anthropological Division. 168 —
177. Ottawa. IV.
1916c Nootka [phonetic orthography and notes]. In Boas, Franz (ed.). Vo-
cabularies from the Northwest Coast of America. Proceedings. Amer-
ican Antiquarian Society 26, 4—18. XII.
536 ^^ American Indian tankages 2
1916g Terms of Relationship and the Levirate. American Anthropologist 18,
327-337. IV.
1917e The Position of Yana in the Hokan Stock. University of California
Publications in American Archaeology and Ethnology 13, 1 —34. V.
1917k Review of C. C. Uhlenbeck, Het passieve karakter van het verbum
transitivum of van het verbum actionis in talen van Noord-Amerika.
International Journal of American Linguistics 1, 82 — 86. V.
19171 Review of C. C. Uhlenbeck, Het identificeerend karakter der possessieve
flexie in talen van Noord-Amerika. International Journal of American
Linguistics 1, 86-90. V.
191 8e Review of Benigno Bibolotti, Moseteno Vocabulary and Treatises. In-
ternational Journcd of American Linguistics 1, 183 — 184. VI.
19181 Tom. Canadian Courier (December 7), 7. IV.
191 8j Yana Terms of Relationship. University of California Publications in
American Archaeology and Ethnology 13, 153 — 173. IV.
1919e A Flood Legend of the Nootka Indians of Vancouver Island. Journal
of American Folk-Lore 32, 351-355. IV.
1920b The Hokan and Coahuiltecan Languages. International Journal of
American Linguistics 1, 280 — 290. V.
1920c Nass River Terms of Relationship. American Anthropologist 22,
261-271. IV.
1921a A Bird's-eye View of American Languages North of Mexico. Science
54, 408. V.
1921b A Characteristic Penutian Form of Stem. International Journal of
American Linguistics 2, 58 — 67. VI.
1921c A Haida Kinship Term among the Tsimshian. American Anhropologist
23, 233-234. IV.
1921d Language: An Introduction to the Study of Speech. New York: Har-
court, Brace and Co. II.
1921e The Life of a Nootka Indian. Queen's Quarterly 28, 232-243,
351-367. [Also published as 1922y.] IV.
1921o A Supplementary Note on Salinan and Washo. International Journal
of American Linguistics 2, 68 — 72. V.
1922a Athabaskan Tone. American Anthropologist 24, 390 — 391. VI.
1922d The Fundamental Elements of Northern Yana. University of California
Publications in American Archaeology and Ethnology 13, 215 — 234.
IX.
1922y Sayach'apis, a Nootka Trader. In Parsons, Elsie Clews (ed.), American
Indian Life. 297 — 323. New York: B. W. Huebsch. [Also published as
1921e]. IV.
1922aa Vancouver Island Indians. In Hastings, James (ed.). Encyclopaedia of
Religion and Ethics, Volume 12. 591—595. New York: C. Scribner's
Sons. IV.
1923c A Note on Sarcee Pottery. American Anthropologist 25, 247 — 253. IV.
1923d The Phonetics of Haida. International Journal of American Linguistics
2. 143-159. VI.
References
537
1923m Text Analyses of Three Yana Dialects. University of California
Publications in American Archaeology and Ethnology 20, 263 — 294.
IX.
1923n A Type of Athabaskan Relative. International Journal of American
Linguistics 2, 136- 142. VI.
1924d Personal Names among the Sarcee Indians. American Anthropologist
26, 108-119. IV.
1924g The Rival Whalers, a Nitinat Story (Nootka Text with Translation
and Grammatical Analysis). International Journal of American Lin-
guistics 3, 16- \02. VI.
1925b The Hokan Affinity of Subtiaba in Nicaragua. American Anthropolo-
gist 27, 402-435, 491-527. V.
1925c Indian Legends from Vancouver Island. Transactions. Women's Ca-
nadian Historical Society of Ottawa 9, 142-143. IV.
1925f Pitch Accent in Sarcee, an Athabaskan Language. Journal, Societe des
Americanistes de Paris 17, 185 — 205. VI.
1925o The Similarity of Chinese and Indian Languages. Science 62 (1607),
supplement of 16 October, xii. [Report of an interview.] VI.
1926a A Chinookan Phonetic Law. International Journal of American Lin-
guistics A^ 105-110. VI.
1926f Review of Berard Haile, A Mcmual of Navaho Grammar. American
Journal of Sociology 32, 511. VI.
1927b An Expedition to Ancient America. The University of Chicago Mag-
azine 20, 10-12. IV.
1928i A Summary Report of Field Work among the Hupa, Summer of 1927.
American Anthropologist 30, 359 — 361. VI.
1928J The Unconscious Patterning of Behavior in Society. In Child, C. M.,
et al. (eds.), The Unconscious, A Symposium, 114 — 142. New York:
Knopf. III.
1929a Central and North American Languages. Encyclopaedia Britannica.
Fourteenth Edition, 5, 138-141. London and New York. V.
1929c A Linguistic Trip among the Navaho Indians. The Gallup Independent
(Ceremonial Edifion, August 23, Gallup, N.M.), 1-2. IV.
1929d Male and Female Forms of Speech in Yana. In Teeuwen, St. W. J.
(ed.), Donum Natalicium Schrijnen. Nijmegen-Utrcchl: N. v. Dckkcr
& vande Vegt, 79-85. V
1929e Nootka Baby Words. International Journal oj American Linguistics 5.
118-119. VI.
1929m A Study in Phonetic Symbolism. Journal oJ L.\pcrimciii(d Psvchoiogy
12, 225-239. I.
1931b The Concept of Phonetic Law as Tested in Primitive Languages by
Leonard Bloomfield. In Rice, Stuart A. (cd.). Methods in Social Sci-
ence: A Case Book. 297 — 306. Chicago: Univcrsily of Cliicago Press.
1932d
Two Navaho Puns. Language 8, 217-219. VI.
538
VI American Indian Languages 2
1 933c La realite psychologique des phonemes. Journal de psychologic normale
ct pathologiquc (Paris) 30, 247 — 265. [Also published in the original
English version in Mandelbaum 1949.] I.
1935b A Navaho Sand Painting Basket. American Anthropologist 37,
609-616. IV.
1935c Review of A. G. Morice, The Carrier Language (Dene Family): A
Grammar and Dictionary Combined. American Anthropologist 37,
500-501. VI.
1936e Hupa Tattooing. In Lowie, R. H. (ed.). Essays in Anthropology Pre-
sented to Alfred Louis Kroeher, 273 — 277. Berkeley, Calif. IV.
1936f Internal Linguistic Evidence Suggestive of the Northern Origin of the
Navaho. American Anthropologist 38, 224 — 235. VI.
1936h Kutchin Relationship Terms. In Osgood, Cornelius, Contributions to
the Ethnography of the Kutchin (Yale University Publications in An-
thropology 4), 136 — 137. New Haven: Yale University Press. IV.
19361 Linguistic Classification within the Northern Athapaskan Area. In
Osgood, Cornelius, The Distribution of the Northern Athapaskan In-
dians (Yale University Publications in Anthropology 7), 21 —22. New
Haven: Yale University Press. VI.
1938b Glottalized Continuants in Navaho, Nootka, and Kwakiutl (with a
note on Indo-European). Language 14, 248 — 274. II and V.
1939e Songs for a Comox Dance Mask [edited by Leslie Spier]. Ethnos
(Stockholm) 4, 49-55. IV.
1942 Navaho Texts, with Supplementary Texts by Harry Hoijer. Edited by
Harry Hoijer. Philadelphia: Linguistic Society of America. XV.
Sapir, Edward, and Harry Hoijer
1967 The Phonology and Morphology of the Navaho Language. University
of California Publications in Linguistics 50. Berkeley and Los Angeles:
University of California Press. XV.
Sapir, Edward, and Albert G. Sandoval
1930 A Note on Navaho Pottery. American Anthropologist 32, 575 — 576.
IV.
Sapir. Edward, and Morris Swadesh
1939 Nootka Texts: Tales and Ethnological Narratives with Grammatical
Notes and Lexiccd Materials. William Dwight Whitney Linguistic Series
11. Philadelphia: Linguistic Society of America. XL
1953 Coos-Takclma-Penutian Comparisons. International Journal of Amer-
ican Linguistics 19, 132-137. VI.
1955 Native Accounts of Nootka Ethnography. Bloomington: Indiana Uni-
versity Research Center in Anthropology, Folklore, and Linguistics
(issued as International Journal of American Linguistics 21 (4)). XL
Schlichter, Alice
1981 Wintu Dictionary. Survey of California and Other Indian Languages,
Report 2. Berkeley: Department of Linguistics, University of Califor-
nia.
References 539
Seaburg, William
1982 Guide to Pacific Northwest Native American Materials in the Melville
Jacobs Collection and in Other Archival Collections in the University of
Washington. Seattle: University of Washington Libraries.
Shafer, Robert
1952 Athapaskan and Sino-Tibetan. International Journal of American Lin-
guistics \^A2-\9.
1957 A Note on Athapaskan and Sino-Tibetan. International Journal of
American Linguistics 23, 1 1 6 — 1 1 7.
Shipley, William F.
1980 Penutian among the Ruins: A Personal Assessment. Proceedings of the
Annual Meeting of the Berkeley Linguistic Society 6, 437 — 441.
Silverstein, Michael
1977 Person, Number, Gender in Chinook: Syntactic Rule and Morpholog-
ical Analogy. Proceedings of the Annual Meeting of the Berkeley Lin-
guistic Society 3, 143 — 156.
1979 Penutian: An Assessment. In Campbell and Mithun (eds.) 1979.
650-691.
Simeon, Remi
1885 Dictionnuire de la langue Nahuatl. Paris: Imprimerie Nationale.
Snow, Catherine, and Charles A. Ferguson (eds.)
1977 Talking to Children: Language Input and Acquisition. Cambridge: Cam-
bridge University Press.
Swadesh, Mary Haas, and Morris Swadesh
1933 A Visit to the Other World, a Nitinat Text. International Journal of
American Linguistics 7, 195 — 208.
Swadesh, Morris
1949 The Linguistic Approach to Salish Prehistory. In Smith, Marian W.
(ed.), Indians of the Urban Northwest, 161 - 173. New York: Columbia
University Press. VI.
1952 Discussion of Robert Shafer, Athapaskan and Sino-Tibelan. Interna-
tional Journal of American Linguistics 18, 178 — 181.
1953 Mosan I: A Problem of Remote Common Origin. International Journal
of American Linguistics 19, 26 — 44.
1954 On the Penutian Vocabulary Survey. International Journal of American
Linguistics 20, 1 23 — 1 33.
1964 Comparative Penutian Glosses of Sapir. In Bright. Wilham {Ci.\.),
Studies in Californian Linguistics (University of California Publications
in Linguistics 34), 182-191. Berkeley and Los Angeles: University o\~
California Press. VI.
Swanton, .John R.
1900 Morphology of the Chinook Verb. American Anthropologist n.s. 2.
199-237.
1908 Social Condition, Beliefs, and Linguistic Relationship of the Tlingit
Indians. Twenty-sixth Annual Report of the Bureau of American Eth-
540 ^f American Indian Languages 2
nologyfor the Years 1904-1905, 391 -485. Washington, D.C.: Smith-
sonian Institution.
1911 Haida. In Boas (ed.) 1911,205-282.
Thalbitzer. William
1904 A Phonetical Study of the Eskimo Language. Meddelelser am Gronland
31, 1-405.
Toelken, Barre
1969 The "Pretty Language(s)" of Yellowman: Genre, Mode, and Texture
in Navaho Coyote Tales. Genre 2, 211 —235.
Voegelin, Charles F.
1941 Internal Relationships of Siouan Languages. American Anthropologist
43, 246-249.
Whistler, Kenneth W.
1977 Winlun Prehistory: An Interpretation Based on Linguistic Reconstruc-
tion of Plant and Animal Nomenclature. Proceedings of the Annual
Meeting of the Berkeley Linguistic Society 3, 157 — 174.
Whorf, Benjamin L.
1935 The Comparative Linguistics of Uto-Aztecan. American Anthropologist
37, 600-608.
References to Unpublished Archival Materials
Haas, Mary R., and Victor GoUa
1962 [Kutchin stem file.] Personal possession of Prof. Mary R. Haas,
Department of Linguistics, University of California, Berkeley.
Harrington, John P.
1933 [Lower Rogue River Athabaskan field notes.] Smithsonian
Institution. See Mills 1981: 69-76.
Jacobs, Melville
1935-55 [Galice field notes.] Melville Jacobs Collection, University
Libraries, University of Washington, Seattle. See Seaburg 1982.
References 541
Sapir, Edward
1905 [Wishram grammatical notes.] Boas Collection, American Philo-
sophical Society Library, 497.3 B63c Pn4a.l(). See Kendall 1982:
108, item 4907.
1911 [Wakashan Comparative Notes.] Boas Collection, American Phil-
osophical Society Library. See Freeman and Smith 1966: 216, item
1954; 380, item 3836.
1915 [Comparative Na-Dene dictionary (ledger).] Boas Collection,
American Philosophical Society Library, 497.3 B63c Na20a.3,
vols. 1, 3, 4. See Kendall 1982: 28, item 4115.
1918 [Letter to Paul Radin, 17 July 1918.] Canadian Ethnology Service,
National Museum of Civilization, Ottawa. Cited in Krauss 1986:
156; Dallaire 1984: 169, item 263.
1920 [Sino-Dene comparative ledger.] Boas Collection, American Phil-
osophical Society Library, 497.3 B63c Na20a.3, vol. 2. See Ken-
dall 1982: 28, item 4115.
1923 [Kutchin and Anvik field notes. 6 notebooks.] Personal possession
of Prof. Mary R. Haas, Department of Linquistics, University of
California, Berkeley.
1926, [Navajo texts, field notes, and word lists.] Boas Collection, Amer-
1929 ican Philosophical Society Library, 479.3 B63c Na31.5. See Ken-
dall 1982: 74, item 4574.
1927 [Hupa texts and slip file.] Boas Collection, American Philosoph-
ical Society Library, 497.3 B63c Na20.a.4. See Kendall 1982: 56,
item 4369.
1932 Problems in Athapaskan Linguistics. Boas Collection, American
Philosophical Society Library, 497.3 B63c Na. 1 . See Freeman and
Smith 1966: 80, item 428.
Sapir, Edward, and Berard Haile
1929-38 [Correspondence.] Berard Haile Collection, University
Archives, University of Arizona, Tucson.
Sapir, Edward, and Alfred L. Kroeber
1925-38 [Correspondence.] Bancroft Library, University of California,
Berkeley.
Swadesh, Morris
1950-51 Wakashan Comparative Vocabulary. Boas Collection. American
Philosophical Society Library. See Freeman and Smith 1966: 380,
items 3837-3838.
Thomas, Alex
1910-1920 [Nootka texts.] Boas Collection, American Philosophical Society
Library. See Freeman and Smith 1966: 257, item 2410.
Index
Abenaki, V.452, 458-482, 493-528,
535-537
Aberle, David, V.449
Achomawi (= Achumawi), V.86, 98,
105, 177, 185-187, 189-229,
232-242, 248, 259, 261, 264-333
passim; VI. 153; see also Pit River
African languages, V.5
Agua Caliente (= Cupeiio), V.352
Aguacatec, V.102
Ahousaht (Nootka dialect), VI.466
Aht (Nootka). V.llO; VI. 353
Ahtena, VI.221
Alaska, V.63, 95-99; VI. 129, 166, 199,
201
Albanian, V.133
Alberni Canal, VI.321, 359, 373
Alberta (province), VI. 169
Aleut, V.93, 95; V.IOO
Aleutian Islands, V.95
Algic, V.490
Algonkian {— Algonkin, Algonquian)
(family), V.13, 15, 58, 62-63, 70, 73,
75-76, 79-81, 84-85, 90-91,
93-94, 95, 97, 451-482, 484-529,
531, 535, 542-543, 553-557; VI. 13,
15, 23-24, 133, 313
Algonkin (Ojibwa dialect), V.447. 477,
535
Algonkin-Ritwan, V.90-91, 97
Algonkin-Wakashan, V.81, 84-86, 90,
93, 97. 101, 181, 332, 490; VI. 23
Algonkin-Wiyot-Yurok, V.93
Algonquian, see Algonkian
Allegheny Mountains. V.95
Alsea, V.67, 81, 105; VI. 225, 227; see
also Yakonan
American Anthropological Association.
V.149, 177; VI.22
American Association for the Advance-
ment of Science. V23 — 24, 345
American Council of Learned Societies,
VI.206
Amusgo, V.102
Anglo-Saxon, V.128
Anthony, Rev. Anthony Seqaqkind.
V.493, 503
Antilles, V.105
Antoniano Sahnan, V.213, 248, 251,
272, 306, 321; see also San Antonio
Salinan
Anvik, VI.24-25, 96-102 passim,
106-129 passim. 142, 149, 196. 199.
201; see also Ingalik
Aoki, Haruo, VI. 379
Apache, V.102, 104, 242; VI.41,
101-102, 135, 169, 172, 199, 205.
209-220
Apachean, VI.189 .
Applegate Creek (Athabaskan). VI. 38,
47-48, 93-94
Arabic, V.112, 126, 242; VI. 207, 353,
377
Arapaho, V.447 -449, 451, 457-482
passim. 493-528 passim. 535-538,
542
Arawak, V.95, 103. 269
Arctic, V.104
Ankara, V.96
Arizona, V.96, 135. 223. 354-355;
VI. 199, 224
Arkansas. V.96
Armenian, V. 143
Arragon, Francisct^. V.264
Arroyo de la Cucsla. Felipe.
Vl.271-272
Aryan. V.549 — 550; see also Indo-Hurii-
pcan; Indo-Germanic
544
VI American Indian Languages 2
Asia, V.545; VI. 191
Assiniboine, see Stoney
Athabaskan ( = Athabascan. Athapas-
can. Alhapaskan), V.13, 15-17, 25,
52, 59, 62-63. 71, 90, 93, 95-98,
100, 102, 144, 179, 185, 223, 242,
261, 298; VI. 13, 15-17, 21-149,
158, 160, 163. 169-222, 314,
481 - 51 1 ; see also Dene
Atlantic Coast. V.95
Atsugewi. V.105. 177. 185-186,
190-221 passim. 226, 232-242 pas-
sim, 258, 266 — 333 passim; see also
Hat Creek
Attacapan ( = Atakapa), V.223
Avestan, V. 131, 452
Aztec-Tanoan ( = Aztek-Tanoan). V.84.
91, 99. 101. 103-104
Aztek(= Nahuatl), V.102
Babine. VI.213. 483, 499-511
Baja CaUfornia, see Lower California
Baltic, V.144
Bahimore, V.177
Bankalachi, V.352 — 443 passim
Bannock, V.352 — 443 passim
Bantu, VI.173
Baraga, Bishop Frederic, V.458, 466,
493,496, 512, 521-522, 559
Barbeau, Charles Marius, VI. 141, 171
Barkley Sound (Nootka), VI. 321.
373-374
Barrett. Samuel A.. V.190, 232; VI.267,
301
Basque, V.69, 72, 134, 545
Bear Lake (Athabaskan), VI.221
Beaver, VI. 172, 173, 199, 201, 205, 211.
213, 217-219, 221
Bella Bella, V.llO
Bella Coola (Salishan), V.96; VI. 382,
426. 467
Bengali. V.487
Beothuk, V.90. 94. 96-100
Berendt. C. H.. V.264
Berkeley (CaUfornia), V.177, 335
Berman, Howard, VI. 225
Bibolotti, Benigno, VI.475-476
Bill, Tommy, VI. 382
Biloxi, VI.474
Birket-Smith. Kaj, VI. 130
Blackfoot, V.22, 70, 75, 79-80, 447,
451, 457-482 passim, 487, 493-528
passim, 535-536; VI. 169
Bleek, D. F., VI. 367
Bloomfield, Leonard, V.26, 146, 447,
449, 535, 537; VI.24, 199
Boas, Franz, V.5, 15-17, 21-22,
25-26, 49, 52-54, 56-58, 61 -63,
68,82, 94, 105, 110, 114-115, 118,
141, 143, 151, 178, 356, 388-389,
402, 535, 550. 558; VI. 15- 17,
22-23, 26, 105, 114, 130, 141,
156-157, 165, 202, 203, 225, 231,
234, 241, 243, 275, 280, 301, 302,
319, 321, 323, 351-352, 356, 358,
366, 382-383, 386, 465,467
Bolivia, VI.475-477
Bommelyn, Loren, VI.94
Boston (Massachusetts), V.345
Bright. Jane. VI.94
Bright. William. V.13. 94. 178. 335;
VI. 13
Brinton, Daniel G., V.264, 302. 346.
351, 380,493, 503, 545
British Columbia, V.96, 136; VI. 129,
199
British Honduras. V.102- 103
Brown, Sam, VI. 195
Bureau of American Ethnology,
V.21 -22, 61 -68, 447, 477; VI.225
Buschmann, Johann Carl Eduard,
V.346, 351-353
Bushman, VI. 367
Byington, Cyrus, V.63
Cacaopera, V.102
Cacouna (Micmac dialect), V.494
Caddo(an), V.63, 82, 91, 93, 95,
99-100, 332
Index
545
Cahita, \.?>S2-A^7> passim
Cahuilla, V.305, 352-443 passim
Cakchiquel, V.102
Calgary (Alberta), VI. 169
California, V.15-16, 44, 63, 96-97,
103, 133, 135, 144, 177-178, 185,
187, 223, 245, 261, 264, 331, 335,
354, 448-449, 453-482, 485-492,
514, 524-525, 528, 553-561;
VI. 15-16, 195
California Penutian, V.81, 90, 93, 98;
VI. 225, 263-265, 267-272, 273
Campbell, Lyle, V.16, 94; VI. 16,
288-289
Canada, V.447, 477
Canixana, V.269
Cape Beale, VI.353, 435
Cape Breton Island. V.95
Cape Cook, VI.353
Cape Flattery, VI.353, 439
Carib, V.103
Carlisle Indian School, V.345
Carochi, Father Horacio, V.354, 366,
373, 375, 377, 384, 387, 392-393,
417, 420, 434, 444
Carrier, VI.22, 29-92 passim. 95-103,
141, 169, 171-172, 199, 201, 205,
207-208, 211-220 passim. 221,
481-511
Catawba, V.72
Cayuga, VI.471
Cayuse, V.90, 96, 98
Cegiha, V.62
Celtic, V.447, 452, 477, 487
Central America, V.24, 64, 84, 91 -92,
95-104, 180-181, 299
Chapman, John Wight, VI.96-103,
135, 196, 219
Chasta Costa, V.16, 106; VI. 16, 21, 24,
27-94, 95-103, 119, 143, 145, 154,
182-183, 205, 217, 481-511
dichalis, VI. 233
Chcmakum, V.29, 81, 90; VI.381; see
also Chimakuan
Chemehuevi, V.38, 347, 354
Cherokee, V.62, 95
Chetco-Tolowa, VI. 94
Cheyenne, V.63, 447, 451 -452, 458,
486-487, 500, 512, 535-536, 559
Chiapanec, V.102, 264, 293, 299
Chiapas, V.102
Chibchan, V.103
Chicago, V.15, 17, 23; VI. 15. 17, 24.
195
Chihuahua, V.96, 102, 104
Chilcat (THngit), VI. 114
Chilcotin, VI.205, 221, 482-483, 492,
499-511
Chimakuan, V.96-98, 105, 533;
VI. 264, 321; see also Chemakum
Chimariko, V.75-76, 78, 82, 86, 90,
96-97, 99, 105, 177, 179-180,
185-187, \^9 -229 passim,
231 - 242 passim, 245 - 249,
264-333 /Jfl.v.v//;/; VI. 25
Chinantec, V.102
Chinese, V.23, 52, 83, 485; VI.23-24,
133-138, 153-154, 173, 191
Chinook Jargon, VI. 246
Chinook(an), V.15, 28, 59. 62, 71-73,
77-79, 81, 85, 90-91, 96, 98. 101.
105, 134-135, 141, 186, 284, 486,
491, 557-558; VI. 15, 37, 225. 227.
231-261, 263-265, 267, 275-280,
283-288, 293-296, 299-315, 365,
368, 376
Chipewyan, V.144; VI.35-94 /j^/.v.v//?/.
96-103, 106-129 passim . 135, 141,
143-149, 169-170. 172-173.
182-183, 193, 199-201. 205-206,
21 1 - 220 passim. 22 1 . 504 - 505; see
also Montagnais
Chiricahua Apache. VI. 205. 225-220
Chiriqui, V.102
Chitimacha. V.82. 90. 96-97. 99, 105.
332, 448. 553, 556
Chocho, V.102
Choctaw. V.63. 332
546
VI American Indian Languages 2
Chomanes, see Moseteno
Chontal (of Oaxaca), V.82, 90, 97, 99,
102, 104, 179, \95- 229 passim, 261,
264 — 333 passim; see also Tequisis-
teco; Tequistlateco
Chorotega ( = Chorotiga), V.264, 269,
287, 293
Chukchi-Koryak, V.81
Chumano, see Moseteno
Chumash, V.78, 90, 92, 96-97, 99,
177, 1^9-222 passim, 223, 231-242
passim, 254, 257 — 261 passim; VI. 228
Circassian, V.133
Clayoquot, VI. 374
Cleveland (Ohio), V.149
Coahuila state (Mexico), V.102, 178
Coahuiltecan, V.15-16, 82, 90, 93,
96-103, 178-181, 223, 231-242,
254, 257-261; VI.15-16, 228; see
also Pakawan
Coahuilteco, V.82, 90, 99, 223,
231-242. 258, 259-261
Coast Penutian, V.90
Coast Salish, VI.382, 386-388, 402
Cochimi, V.190, 226, 232, 268, 274
Cocopa, V.190, 226, 232-242 passim,
274
Codere, Helen, VI. 352
Codrington, VI. 372
Cole, Daniel, V.14
Colorado (state), V.96, 347, 353
Colorado River, V.96, 135, 261
Columbia River, V.96, 135; VI.231
Columbia University, VI. 226, 351
Comanche, V.354
Comecrudo, V.91, 99, 223, 231 -242
passim, 254, 258-259, 261, 265-333
passim
Comox (Salishan), V.13; VI. 13, 154,
321, 381-433,467-468
Continental Na-Dene, V.90, 97
Cook, Eung-Do, VI. 189
Cook Inlet (Alaska), V.96
Coos, V.64, 71, 81, 90, 96-98, 105,
261, 553, 557; VI.225-226, 229, 263,
267, 272, 279, 286-288, 291-296,
299-315
Copper River, VI. 130, 205
Coquille, Upper, VI.27-28
Cora, V.104, 172, 352-443 pavi/m
Costa Rica, V.102- 104
Costanoan, V.75, 90-91, 96, 98, 177,
558; VI.225, 263-265, 270-272,
284-288, 293-296
Cotoname, V.91, 99, 223, 231-242 pas-
sim, 265 — 333 passim
Coulter, Thomas, V.353
Cowichan (Salish), VI.388, 426
Cree, V.58. 80, 264, 449, 451 -452,
457 - 482 passim , 486 - 487,
493-528/)fl.v.s7m. 559-560; VI. 169,
485-486
Cuicatec, V.102
Cuitlatec, V.102
Cumberland Inlet, V.62
Cuoq, Jean Andre, V.53
Cupefio. see Agua Caliente
Curtin, Jeremiah, V.198; VI.226
Dakota, V.29, 62, 70, 72-73, 78-79,
105, 541, 543; see also Sioux
Dall, William Healy, V.62
Dallaire, Louise, VI. 22
Danish, VI.151-152
Darnell, Regna, VI. 477
De Angulo, Jaime, VI.228, 283, 285
De Laguna, Frederica, VI. 130
Delaware, V.164, 451, 458-482,
493-533 passim, 535-538, 559
Dene, V.83; VI. 207, 481 -482,
499 — 511; see also Athabaskan
Diegueiio, V.190-221, 226-228,
234 - 242, 266 - 333 passim
Diguct, Leon, V.352, 354, 358-359,
363-364, 370-379, 381-383, 385,
420-421, 428-430, 436, 439
Dixon, Roland B., V.16, 44, 63,
177-179, 185-187, 190, 197, 215,
Index
547
223, 231-232, 245-247, 261,
264-265, 447, 477; VI. 16, 225-226,
228, 263, 268, 273, 283, 301
Dogrib, VI.205, 213, 215, 221
Dolores, Juan, V.399, 402, 409, 413,
416, 436, 438
Dorsey, James O., V.62-63, 160; VI.28
Dunn, John, VI.228, 300
Dyen, Isidore, V.449
Dyk, Walter, VI. 262
Dynes, Wayne, VI. 351
East Cape (Siberia), V.95
Eastern Algonki(a)n, V.482, 497-498,
528
Edgerton, Franklin, V.5
Eliot, John, V.451
English, V.30-31, 33, 57-58, 85, 100,
111, 134-136, 138-140, 151-169
passim, 171, 297-298, 351-352,
356-357, 401, 404, 489, 542-544,
564
Enrico, John, VI. 166
Eskimo, V.31, 59, 62, 64, 72, 82, 93, 95,
100, 135, 143, 162, 172, 542, 550;
VI.23, 108, 133, 163, 358, 370-372,
377; see also Greenlandic; Innuit
Eskimoan, V.95
Esselen, V.90, 96-97, 99, 177,
195-221 passim. 223, 226-229,
233-242, 252, 258, 264-333 passim
Euchre Creek (Athabaskan), VI.28, 94
Europe, V.133
Eyak, VI. 130, 141, 202
Ferguson, Charles A., VI. 466
Fernandeiio, V.352 — 443 passim
Finck, Franz Nikolaus, V.351
Finnish, V.134, 486
Finno-Ugrian, V.172
Fletcher, Alice C, V.64
Florida, V95-96, 103
Fort Totten Cree, V561
Foucaux, V.83; VI.23
Fox, V58, 63, 70, 264, 449, 451 -452,
486-487,493-528, 531-533,
535-536, 555, 557, 559
Frachtenberg, Leo J., V64, 553;
VI.227, 263-265, 267, 291, 301, 303,
309, 311, 321, 474
Franciscans, VI. 193
Fredson, John, VI. 25
Freeland, Lucy S., VI.228, 283-289,
299
Freeman, John E., VI. 319
French, V.25, 125, 134-136, 138-140,
159-160, 351-352, 354-355, 487,
489
GabrieHno, V.352 - 443 />a.s.s7>»
Galice Creek (Athabaskan), VI.27-28,
48, 93, 94, 205
Gamio, Manuel, VI.283
Gatschet, Albert S.. V.61 -63. 72, 190.
231, 233
Geological Survey of Canada, Anthro-
pological Division, 319
Georgia, V.96
Georgian, V.105
German, V.31, 39, 110, 133, 136,
152-165 passim, 300, 351-352, 357,
388-390,486, 550
Germanic, VI. 173
Giamina, V.352 — 442 passim
Gibbs. George, V.62 — 63
Gifford, Edward W., V. 190 -221 pas-
sim, 232-233, 492, 494-495, 497,
507-509, 513, 519, 521-522
Gitanemuk(= Kitanemuk), 352 — 442
passim
Gitkshan, VI. 171
Goddard, Ives, V.580
Goddard, PUny E., V.21 -22, 62, 82.
149, 151, 178; VI.21, 29-92 passim.
96-103, 105, 107, 130, 134- LVS,
141, 143-149, 154, 172, 195-196,
201, 202, 359, 500, 505-507, 508
Golla, Victor, V.13, 16, 22-25, 149,
1 77 - 1 79, 181. 345 346. 448 - 449;
548
VI American Indian Languages 2
VI. 13, 16, 22-23, 25, 94, 141, 189,
197, 227, 299, 321
Gothic, V.126, 162
Grand Canyon, V.135
Great Basin, V.96, 261, 491, 524
Great Bear Lake, V.96
Great Central Lake, VI.375
Great Lakes, V.95
Great Plains, VL169, 218, 220
Greater Antilles, V.103
Greek, V.25, 33, 100, 126-131, 133,
447, 452, 477, 488, 550
Green Bay (Wisconsin), V.96
Greenberg. Joseph H., V.94; VI.130,
139, 225, 288-289, 477
Greenland, V.95
Greenlandic (Eskimo), VI. 370 — 372; see
also Eskimo
Grinnell, George Bird, V.460, 462-463
Gros Ventre. V.460. 462-463,
493-528/;a.v.v/m, 535-536
Guatajiguala, V.264
Guatemala. V.102; VI.228
Guerrero state (Mexico), V.102 — 103,
264
Gulf languages, V.94
Gulf of Mexico, V.95, 223, 261
Haas, Mary, V.17, 94, 173, 490; VI. 17,
24 — 25, 262, 321; see also Swadesh,
Mary Haas
Haeberlin. Herman K.. VI. 433
Hagwelgate (Carrier dialect), VI. 141,
171
Haida. V.14, 16-17, 62-63, 71-72,
76. 78, 81, 85,90, 93, 96-98,
105-106, 125, 242, 254, 297-298;
VI. 14, 16-17,21, 105-129.
130-131, 133-138, 151-167, 170,
189, 191, 314
Haile. Father Berard. V.109. 141;
VI.24-26. 193
Haisla (Kwakiutlan). VI. 353; see also
Xaisla
Haka, V.138
Halkomelem, see KwantlEn
Hamitic-Semitic, V.143, 145; VI. 173
Han, VI. 205, 221
Hanis (Coos dialect), VI. 292, 301
Hare, VI.27-92 passim, 96-103 pas-
sim, 136, \43- 149 passim. 199, 201,
205-206, 210-221
Harrington. John P., VI. 94
Harvard University, V.447
Hat Creek, V.232; see also Atsugewi
Heepe, Martin, V.172
Heiltsuk (Kwakiutlan). VI.353. 467
Hernandez. Fortunato. V.355
Herzog, George, V.173
Hess, Thomas M., VI. 433
Hewitt. John Napoleon Brinton. V.62.
64, 190, 232, 259, 550, 553
Hidalgo state (Mexico), V.103
Hidatsa, V.72
Highland (Serrano dialect), V.352-441
passim
Hittite. V.126- 127. 131
Hockett. Charles F.. V.449. 535
Hoffman, Walter James, V.63
Hoijer, Harry, V.17, 24. 141; VI. 17.
24-26, 94, 210
Hokan, V. 13, 15-16, 82, 86. 90-91,
93,97-98, 101-104. 177-341;
VI.15-16, 226, 228. 263, 273,
286-287. 305. 313
Hokan-Coahuiltecan, V.16, 24, 82, 86,
90-91,93,97-98, 101-104,
231-242, 251-255 passim, 259, 265;
VI. 16
Hokan-Siouan. V.16. 81. 84. 90-91,
93, 98, 101, 151, 179. 181, 293,
331-333; VI. 16
Homeric Greek, V.128, 263
Honduras, V.102- 103; VI.228
Hopi. V.96. 354-441 passim
H'laam (Yuman), V.226, 232-241 pas-
sim
Huastec, V.102
Index
549
Huave, V.82, 98, 102-103; VI.228,
288-289
Hudson's Bay, V.96; VI. 199
Huichol, V.104, 366-441 passim
Humboldt Bay, V.453
Hungarian, V.133
Hunt, George, VI. 319, 323-352, 366
Hupa, V.13, 15, 17, 62, 106, 109, 144,
297, 524, 557-558; VI. 13, 15, 17,
24-25, 27-94, 96-103, 106-129
passim, 133-138, 141 -149, 154,
169, 172-173, 191, 195-201, 205,
2\\ -220 passim, 481, 506, 508
Hymes, Dell, VI. 130, 273, 379
Idaho, V.96
Iluano, V.91
Indie. V.131
Indo-Chinese, V.23, 83; VI.23,
133 — 138; see also Sino-Tibetan
Indo-European, V.5, 16, 24-26, 85,
101, 105-131 passim, 134, 143-145,
298, 345, 447-448, 477-478, 542,
544, 550, 558; VI. 16, 227; see also
Aryan; Indo-Germanic
Indo-Germanic, V.31, 33, 36, 56, 72,
487; see also Aryan; Indo-European
Indo-Iranian, V.131, 487; VI. 173
Indonesian, VI. 358
Ingalik, V.17; VI.24-25, 142, 196, 202,
205, 213, 217, 219, 221; see also An-
vik
Innuit, V.64; see also Eskimo
Interior Salish, VI.432
International Phonetic Association,
V.149
Inyo County (Mono dialect),
V.352-441 passim
Iranian, V.131
Irish, V.133, 297, 388-389
Iroquoian (Iroquois), V.35-36, 52,
55-58, 78, 82, 86,91, 93, 95, 99,
264, 448, 550, 553, 556-557; VI. 153
Iroquois-Caddoan, V.91 -93
Ishi (Yahi speaker), V.177, 183, 190
Iskoman, V.177- 178
Italian, VI 29, 165
Italic, V.131
Jabo (Liberia), VI.207
Jacobs, Melville, VI. 94
Jacobsen, William, V.335; VI.356
Jalisco state, Mexico, V.96, 355
Janambre, V.102
Jaschke, Heinrich A., V.83; VI.23
Jenness, Diamond, V.172
Jette, Jules, VI. 100
Jicaque, V.102; VI.228
Jicarilla Apache, VI.101-102, 135,
205, 219-220
Jiquilapa (Nicaragua), V.264
Joel, Janet, VI. 25
Jones, William, V.63, 458-469, 474,
493, 531
Joshuas, VI.28
Juaneiio, V.352 — 441 passim; see also
Netela
Kahclamat, Philip, VI. 262
Kaibab (Southern Paiute dialect), V.38
Kalapuya, V.81, 90, 96-98. 105, 553,
557; VI.225, 227, 263-265, 267, 286.
293
Kamia, V.174
Kansas, V.96
Karankawa, V.82, 90, 96-97, 99, 223,
231-242 passim. 258, 265-328 pas-
sim
Karesau-Papua (New Guinea). VI. 368
Karlgren. Bcrnhard, 136
Karok, V.24, 90, 96-97, 99, 105, 144,
177-178, 189-221 passim. 223.
225-229, 231-242 passim. 258. 261
264-333 /w.v.v//?/, 558
Kaska. VI. 213. 221
Kathlamet, V.62; VI. 231. 241
Kalo. VI.27-94. 96 103, 106-129
passim, 136, 145. 172. 205. 211, 213,
218
550
VI American Indian Languages 2
Kawaiisu, V.354, 416
Kekchi, V.102
Kelly. Peter R., VI. 151 -157
Kendall, Daythal, VI. 139, 197
Kentucky, V.96
Keres, V.82, 90, 93, 96, 99, 332
Kern River (California), V.354
Kess, Anita Copeland, VI. 466
Kess, Joseph Francis, VI. 466
Kickapoo, V.459, 477, 559
Kiliwa (= Kiliwi), V.190, 226,
232-242 passim, 268, 271 -272, 276,
280
Kiowa, V.63, 81, 91, 93, 96-97, 99,
105; VI. 169
Kiowa Apache, VI. 169, 205, 219-220
Kitamat, V.llO
Kitanemuk, see Gitanemuk
Kizh, V.354; see also Gabrielino
Klamath (language), V.62-63, 69,
90-91, 96, 98, 105; VI.226, 240
Klamath-Modoc, VI.225
Klamath River (Oregon), V.453
Kleinschmidt, Samuel R, V.162
Konomihu Shasta, V.203, 208
Kootenay (= Kutenai), V.67, 81, 90,
93, 96, 98, 105, 543
Koskimo (Kwakiutl dialect), V.120, 124
Koso, see Shikavayam
Koyukon, see Ten'a
Koyukukhotana, VI. 205
Krauss, Michael E., VI. 21 -22, 26, 130,
139, 141-142, 166, 189, 202
Kroeber, Alfred L., V.16, 21-24,
27-59, 81, 149, 151, 177-179, 181,
190, 199, 201, 203, 207, 223,
231-233, 249, 251, 257,261,
264-265, 311, 318. 347, 351-441
passim. 447-449, 451, 453-482 /7a.v-
sim, 488, 491 -497, 499-501,
504-505, 508-510, 513, 520,
541-557, 559-562; VI. 16, 21-23,
25, 133-138, 220, 225-229, 263,
269-270, 273, 299. 301, 359
Kroeber, Paul, VI. 433
Kuipers, Aert H., VI.433
Kutchan, V.232; see also Yuma
Kutchin, V.13, 15, 17, 100; VI. 13, 15,
17, 24-25, 181, 193, 195-196, 199,
201, 202, 205, 206, 207, 211-220
passim, 221; see also Loucheux
Kutenai, see Kootenay
Kwakiutl, V.24-25, 29-30, 59, 63, 77,
90, 96, 98, 105-131, 143, 316, 332;
VI.107-108, 157, 280, 312,
319-321, 323-351, 353-356,
366-368, 372-373, 377, 381, 388,
442-443, 455, 465, 467-468
Kwalhiokwa, VI.205, 213
KwantlEn (Salishan, = Halkomelem?),
VI.388, 396
Labrador, V.95
Lacombe, Albert, V.80, 458, 466, 474,
493, 521-524, 533
Laconian Greek, V.128
Ladusaw, William, V.150
Lake Erie, V.95
Lake Ontario, V.95
Langdon, Margaret, V.335
Lassik, VL205
Latin, V.33, 68, 100, 128-129, 131,
452, 486-487, 551
Laufer, Berthold, VI.23, 138
Lawrence, Erma, VI. 166
Leer, Jeff, VL130, 166
Legoff, Laurent, VI.69, 74, 78, 99, 128,
141, 143-149, 170-171, 193, 505
Lehmann, Walter, V.37, 180-181,
263-331 passim. 355, 369, 383, 441,
444, 548-562; VI. 228
Lenca, V.103, 264; VI.228
Leon (Nicaragua), V.263
Leon, Nicolas, V.264
Levine. Robert, VI. 130, 166
Levy-Bruhl, Lucien, V.78
Li, Fang-Kuei, V.17; VI.24-25,
196-197
Index
551
Liebe-Harkort, Marie- Louise, V.14
Lillooet (Salishan), VI.427-432, 467
Lipan Apache, VI. 199, 205, 219
Loucheux, M \ 35 -9A passim, 96—103,
106-129 passim, 135, 169-189 pas-
sim, 199, 201, 211-220; see also
Kutchin
Louisiana, V.96
Lower California ( = Baja California),
V.96, 102-103
Lower Chinook, VL225, 231-242, 247,
249, 275-280
Lower Umpqua, V.64; VL225, 311; see
also Siuslaw
Lowie, Robert, VI.21, 299
Luiserio, V.305, 352-441 passim. 492
Lumholtz, Carl, V.353
Lushootseed, VI.433
Lutuami, V.90-91, 94, 96-98
Mackenzie River, V.96, 166; V1.220
Maidu, V.63, 69, 85, 90, 96-98, 101,
144, 177, 186, 557; VI.225, 263-265,
268-269, 283-288, 293-296
Makah, V.llO; VI.435
Malecite, V.160, 458-482 passim,
493 - 528 passim
Mandelbaum, David, V.181
Mangue, V.102, 264, 287, 293
Maratino, V.236, 271
Maribio, V.263
Maricopa, V.190, 232-242 passim
Mason, John Alden, V.180, 190-221
passim, I'i'i-lAl passim. 247 — 249,
251-255, 257-261, 306, 310,
316-317, 324, 355, 410, 416-417,
422-426, 428, 430, 434-443
Masset (Haida dialect), VI.22-28,
1 06 - 1 29 passim. 1 52, 1 63 - 1 64, 1 66
Matagalpa, V.102- 103
Mattolc, VL25, 196, 205, 211,
213-214, 218
Maya(n), V.82, 91, 102-103, 105, 299,
333, 542, 545; VI.288
Mazatec, V.102, 264, 287, 293, 299
McGary, Jane, V.14, VL 15
McGee, William John, V.232
McGuff, Pete, VI. 352
Mechhng, WilHam H., V.397
Melanesia, V.78
Melanesians, VI.372-373
Mengarini, Gregory, V.63
Menomini, V.63, 535, 556
Mescalero Apache, VI. 205, 210-220
Meso-America, see Middle America
Mexican languages, V.lOl — 102
Mexican Penutian, V.98
Mexico, V.24, 64, 66, 81, 84, 93,
95-97, 101-104, 178-181, 223,
231, 264-265, 287, 299, 331, 333,
355, 545
Miao, VI. 136
Michelson, Truman, V.63 — 64, 70,
447-449, 451, 460, 462-463,
465-466, 477, 485-489, 493, 513,
531-532, 553, 558-561
Michoacan state (Mexico), V.102— 103
Micmac, V.457 — 483 passim, 487,
493 - 529 passim, 535 - 538, 559
Middle America ( = Meso-America),
V.24, 101, 103; VI. 228
Migueleiio Salinan, V.213, 248, 251,
271, 321; see also San Miguel Salinan
Mills, Elaine, VI.94
Miluk (Coos dialect), VI. 291 -296, 301
Miskito, V.102 -103
Mississippi (state), V.96
Mississippi River, V.95-96
Missouri River, V.95
Misteli, Franz, V.37, 44
Mithun, Marianne, V.16, 94; VI. 1 6, 474
Miwok, V.78, 90-91, 96-98, 177, 332,
448, 556-558; VI.153, 225,
263-265, 270, 279, 283-288,
293-296, 301
Miwok-Costanoan, V.90, 97 - 98
Mixe, V 1.228, 283-288, 299
Mixe-Zoquc. V.91, 98. 102 103; V1.228
552
VI American Indian Languages 2
Mixtec, V.102, 264, 287, 293
Mixtec-Zapotec, V.82, 91, 102-103
Mixtec-Zapotec-Otomi, V.91, 102-103
Modoc, V.82, 90, 96, 98, 105
Mohave (= Mojave), V.86, 190-221
passim, 225 — 228, 232 — 242 passim,
266-333 passim. 492; VI.367
Mohawk, V.52; VI.151-152
Mohegan, V.502, 504
Mohican, V.536
Mohineyam, V.352 — 441 passim
Mojave, sec Mohave
Molala (= Molale), V.82, 90, 96, 98,
553, 556, 558
Molale-Cayuse, VI.225
Mono, V.352 — 441 passim
Mono-Bannock, V.354
Mono-Paviotso, V.354
Montagnais (Athabaskan), VI. 35 — 94
passim, 96-103, 108-129,
143-149, 170; see also Chipewyan
Montagnais (Algonkian), V.458-482
passim. 493-528 passim; VI. 170
Montana, V.96
Monterey Costanoan, V.558
Mooney, James, V.63, 431, 462, 464
Morgan, Lewis Henry, V.504
Morice, Father Adrian Gabriel, V.16;
VI.21 - 23, 29 - 92 passim, 95 - 103,
128, 141, 143-149, 171,205,
207-208,481-511
Mosan, V.98, 101; VI. 321, 467
Moseteno (= Chumano, Chomane),
VI.475-477
Mota (Melanesian), VI.372-373
Mountain (Athabaskan group), VI. 221
Mukne (Costanoan), VI.288
Miiller, Friedrich, V.353
Munsi Delaware, V.493, 496
Murray, Stephen O., VI. 351
Muskogean (= Muskhogean, Musko-
gian), V.62, 72, 75, 78, 82, 91-92,
93, 95,99, 332-333; VI.286
Mutsun. V.75-78; VI. 271
Nabesna, VI.221
Na-Dene, V. 13, 15- 17, 81. 84, 90, 93,
98, 101, 104, 181, 242, 298, 332;
VI.13, 15-17, 22-28, 105-129,
130-131, 133-138, 141, 146, 154,
163, 169-170, 189, 191, 312
Naga, VI. 136
Nagrando, V.263
Nahani, VI.483
Nahuatl, V.15, 21, 37-38, 44, 46, 59,
91, 96, 99, 103-104, 264, 267, 346,
351-444, 542-543, 545, 548-562;
VI. 15, 358, 364-365, 371, 377
Nanaima (Salishan), VI.426-431
Narragansett, V.498, 506, 512-513,
515
Nass River (Tsimshian dialect), V.14,
284; VI. 14, 151-152, 157, 228, 302,
372
Natchez, V.82, 91 -92, 95, 99, 332
Natick, V.63, 451, 457 -4S2 passim.
487, 559-560
Navaho (= Navajo), V.13, 15, 17,
24-25, 95, 100, 105-131, 133,
135-141 passim. 144, 242; VI.13, 15,
17, 24-26,41, 89, 96-103,
1 06 — 1 29 passim, 133—138 passim.
143-149, 154, 158, 169, 172-173,
182, 189, 193, 195-196, 199-201,
203-204, 205-206, 207, 209-220,
221,481, 500, 507-509
Nebraska, V.96
Netela, V.354; see also Juaneiio
Nevada, V.96, 135
New Guinea, VI. 368
New Mexico, V.96, 135; VI. 199
Newman, Stanley S., V.17, 173; VI. 17,
292
Nicaragua, V.16, 102-103, 180,
263-332, 355; VI. 16
Nicarao, V.102, 264, 355
Nichols, Johanna, VI.227, 262, 379
Nicola, VI.221
Niskwalli, V.62
Index
553
Nisqa, VI.228
Nitinat (Southern Nootka), V.llO;
VI. 319-321, 353, 374-375. 435,
442, 450, 453
Nootka, V.13, 15, 24-25, 76-79,
90-91. 96, 98, 105-131, 134-141,
143, 162-163, 171, 316, 532; VI.13,
15, 107, 157, 164, 262, 280, 313,
353-378, 381-382,435-461,
465-466, 467-468; see also Nitinat
North America, V.81, 95-104
North Carolina, V.95
North Dakota, V.96
North Fork Mono, V.352 — 441 passim
Northern Hokan, V.90, 98, 331
Northwest (North America), V.62, 145
Northwestern University, VI. 475
Oaxaca, V.102, 104, 179, 231, 402, 412;
VI.228
Ofo, V.62; VI. 474
Ohio Valley Siouan, VI. 474
Ojibwa, V.58, 70, 447, 449, 451-452,
457 - 482 passim, 489, 493 - 528 pas-
sim, 535-536, 555-556, 559-561;
VI. 154, 361
Okanagan (Salishan), VI.426-431
Old Chinese, VI. 134 -138
Old Church Slavic, V.128
Old Irish, V.126, 388, 452, 488
Olive (Tamaulipas), V.102
Olmecs, VI.288
Olmos, Andres de, VI. 365
Omaha, V.62
Oneida, V.52- 54
Opata, V.355
Oregon, V.14, 16, 62-63, 103, 135,
261, 491, 524; VI.13. 16, 27. 226
Oregon Athabaskan, VI. 94
Oregon Penutian, V.90 -91, 93, 98;
VI. 225, 263-267
Orotina, V.263
Orton, Wolverton, VI.27, 37, 89, 94,
101
Osgood, Cornelius, VI.24-25,
221-222
Otomanguean, VI. 289
Otomi, V.82, 91, 102-103, 287, 293
Otomian, V.102 -103
Ottawa (city), V.17; VI.24-25, 319
Ottawa (language), V.496, 513
Pacific Coast, V.26, 95. 479. 492
Pacific Coast Athabaskan. VI.94. 199,
205
Paiute, V.14, 25, 63, 67, 73, 141. 549:
VI.13, \5A\ see also Kaibab;
Paviotso; Southern Paiute
Pakawan, V.172; see also Coahuiltecan
Panama, V.103
Panini, V.345
Panoan, VI.477
Papago, V.355, 399-443 passim
Paris, V.346
Parry, Sir William Edward, V.353
Passamaquoddy, V.538
Paviotso (= Northern Paiute), V.38,
354
Pawnee, V.35-36, 57-58, 100, 543
Paya, V.103; VI.228
Penobscot, V.477, 493-528 passim, 538
Pentlatch (Salishan), VI. 375. 388, 467
Penutian, V.13, 15-16, 81, 84, 90-91,
93, 97-98, 101, 103. 144, 177,
180-181, 261, 284, 292, 332; VI.13,
15-16, 263-272, 279, 283-288,
291-296, 299-315
Peoria (Algonkian), V.477, 555
Petitot, Emile, y\.25-9A passim.
96-103, 108, 135, 143-149. 170,
503 - 504
Petter. Rudolf, V.451, 458, 512
Philadelphia. V.345
Picrrcvillc (Abenaki dialect). V.494
Pima(n), V.62, 91, 99. 352-441 passim
Piman-Sonoran. V.102, 352-355; see
also Sonoran
Pimentel, Francisco. V.353
554
VI American Indian Languages 2
Pinnow. Heinz-Jiirgen, VI. 130— 131,
139
Pipil, V.102, 264, 355, 412
Pitkin, Harvey, VI. 301
Pit Riven V.232; see also Achomawi
Plains-Atlantic Algonki(a)n, V.482
Plains (culture area), V.97, 145, 492
Plateau (culture area), V.97
Plateau Penutian, V.90-91; VI.225
Plateau Shoshonean, V.353-354, 388,
407-408
Pochutla Nahuatl, V.102, 402
Pointe Bleue (Montagnais dialect),
V.494; see also Selish
Polish, V.133; VI. 360
Pomo(an), V.71, 75-78, 82, 86, 90, 97,
99, 105, 177, 189-221 passim. 223,
225-228, 231-242, 251-252,
254-255, 257-261, 264-333 pas-
. Sim, 448, 557; VI.285
Ponca (= Ponka), V.29, 62, 72, 105
Potawatomi, V535
Powell, John Wesley, V.15-16, 21-24,
61-68, 83-84, 93-94, 95, 97, 149,
231-242, 477; VI. 15-16, 319
Powers, Stephen, V.63
Prakrit, V.488
Preuss, Konrad Theodor, V.172, 352,
354-355, 358, 368, 370, 381, 388,
420, 427, 439
Primitive Central Algonkian,
V.535-538
Proto-American, V.81, 86
Proto-Athabaskan, VI. 202
Pucbla state (Mexico), V.102
Pueblo (culture), VI.209-220
Pueblo (languages), V.145, 261
Pullum, Geoffrey K., V.150
Punzie. Mrs., VI.93
Quebec, V.95
Quechua, V.105
Queen Charlotte Islands, VI. 166
Quiche, V.102
Quinault. VI. 241
Radin, Paul, V.23, 81-83, 232; VI.22,
228, 289, 361, 363
Rand, Silas Tertius, V.458, 460
Rasles, Sebastian, V.503
Reed, Thomas, VI. 25
Ricci, Seymour de, V.37
Riggs, Samuel R., V.62-63, 72
Rio Grande, V.102
Ritwan, V.90-91, 97-98, 101,
447-448,477-478,490
Rivet, Paul, V.172, 449
Riviere du Loup (Malecite dialect),
V.494
Rocky Mountains, V.349, 414, 424
Rogue River (Athabaskan), VI. 94
Rogue River (Oregon), VI.27-28, 94
Rogue River Jack, VI.93
Romance, VI. 173
Rood, David, VI.474
Rupert's House Cree, V.458, 467
Russell, Frank, V.62, 366 — 441 passim
Russian, V.llO, 136,487
Sacramento River, V.135, 177
Sadock, Jerrold M.. V.59
Sahapti(a)n, V.82, 90, 94, 97-98, 105;
VI.225, 226, 231, 240, 249
Saint Lawrence River, V.95
SaHnan, V.79, 82, 90, 97, 99, 177,
180-181, 189-221, 223, 225-229,
231-242, 248-249, 251-255,
257-261, 264-333, 558
Salish(an), V.13, 29-30, 59, 62, 81, 90,
96, 97-98, 105, 491, 533; VI. 13, 107,
233, 264, 319-321, 375, 381-433,
467-468
Salvador, El, V.102, 264, 355; VI.228
San Antonio Salinan, V.213, 251; see
also Antoniano
San Buenaventura Chumash,
V. 1 90 — 22 1 passim, 225 — 228 passim,
233 — 242 passim, 272
San Carlos Apache, VI. 205
San Francisco Solano (Coahuiltecan),
V.234
Index
555
San Juan Capistrano, see Juaneno
San Luis Obispo Chumash, V.190-221
passim, 225 — 228 passim, 265, 321
San Luis Potosi state (Mexico), V.102
San Miguel Salinan, V.213; see also
Migueleiio
Sandoval, Albert, V.13; VI.13, 26, 216,
352
Sanskrit, V.35, 128-129, 131, 345, 452;
VI. 360
Santa Barbara Chumash, V.190-221
passim, 225, 227, 233 — 242 passim,
321
Santa Catalina Yuman, V.226,
232-242 passim, 279
Santa Cruz (Island) Chumash,
V.190-221 passim. 226-228,
233-242 passim, 270
Santa Ynez Chumash, V.190-221 pas-
sim, 225 — 228, 233 — 242 passim
Santee Sioux, V.63
Sapir, Florence, VI. 24, 138-139
Sarcee (= Sarsi), V.13, 15, 17, 133, 142;
VI. 15, 17, 23-25, 141, 169-189,
191, 193, 195-196, 199-201, 206,
2\\ -220 passim, 221
Saskatchewan River, V.95
Saxon, V.487
Sayach'apis, Tom, VI.319-320, 435
Schlichter, Alice, VI. 301
Schmidt, Father Wilhelm, V.158;
VI. 173, 368
Schoolcraft, Henry R., V.353
Schuller, Rudolf, VI 72; Vl.475-476
Scouler, John, V.367
Sechell (= Sicatl; Salishan), VI. 388
Seaburg, William, VI. 94
Sekani (= Sekanais), VI. 221, 483, 497,
500 — 51 1 passim
Selish, V.543; see also Salish
Semitic, V.5, 26, 85, 145, 545; VI. 173,
377
Seri, V.64, 82, 90, 92, 97, 99, 102, 106,
177, 187-188, 223, 226-228,
233-242 passim, 254, 258-259,
264 — 333 passim
Serrano, V.352 — 441 passim
Shafer, Robert, VI. 139, 292-293
Shasta, V.82, 105, 144, 177, 185-187,
225-228, 233-242 passim.
264 — 333 passim
Shasta-Achomawi, V.90, 100, 189, 223,
231; see also Shastan
Shastan, V.97, 99, 185-187, 189,
232-242 passim, 251, 258, 260-261;
see also Shasta-Achomawi
Shawnee, V.535
Sheep Indian, VI.211
Sherzer, Joel, V.25, 94; VI. 477
Shikaviyam (= Koso?), V.352 -441
passim
Shipley, William F., VI.225
Shoshone (= Shoshoni), V.353, 372,
405, 407, 409, 433, 533, 558
Shoshonean, V.29, 34, 36, 38, 44, 59,
72, 78, 91 -92, 96, 99, 101, 105, 135,
144, 159, 186, 223, 242, 261, 305,
312, 345, 347, 353-441 passim, 449,
533, 545, 549
Shoshoni-Comanche, V.154
Shuswap (Salishan), VI. 151 - 152,
426-431, 467
Siamese, VI. 191
Siberia, V.95
Sicatl, see Sechelt
Sierra Nevada Mountains, V. 180
Sigua, V.102
Siletz, VI. 27, 98
Siletz Reservation, VI. 226
Silverstein, Michael, VI.225, 273. 281
Simeon, Remi, VI. 364
Sino-Dene, VI.23-24, 133-138
Sino-Tibetan, V.17; VI.22-24; see also
Indo-Chinese
Siouan, V.29, 59, 62, 72, 75 76, 78,
82, 86, 91 -92, 93, 95, 99, 101. 105,
186, 254, 259, 332, 542; VI. 471, 474
Siouan-Yuchi. V.82
556
VI American Indian Languages 2
Siouan-Yuchi-Muskogian, V.93
Sioux, V.63, 78, 86, 95, 293; see also
Dakota
Sitjar. Father Buenaventura, V.197,
199-200, 251,258
Siuslaw, V.64, 67, 81, 90, 97-98, 105;
VI.227, 263-267, 272, 291-296,
301, 311; see also Lower Umpqua
Six Nations Reserve (Ontario), V.471,
474, 493
Skeena River (Tsimshian dialect),
VI. 372
Skidegate (Haida dialect), V.62; VI. 152,
163-164, 166
Slave, VI.205, 213, 215, 221
Slavic, V.144, 161, 163, 172, 487;
VI. 173
Smith, Marian W., VI.468
Smithsonian Institution, V.14, 21,
61-68
Snohomish, VI. 433
Snow, Catherine, VI.466
Somass River, VI. 375
Sonora state (Mexico), V.102, 231
Sonoran,.V.102, 104, 240, 351, 376,
379. 383. 399-441 passim; see also
Piman-Sonoran
South America, V.95, 102-103
Southern Paiute, V.15, 21, 29-31,
37-38, 46, 72, 76-77, 79, 105,
135-141 passim, 164, 172, 264, 345,
349. 351-443; VI. 13, 15, 108, 313,
320, 364-365, 367-368, 371, 383
Southern Ute, V.407
Southwest United States. V.97. 491,
524; VI.209-220
Spanish, V.138, 265, 286, 313, 352, 389,
401-402. 423. 443
Speck, Frank G., V.493, 497, 499,
503-504, 510, 514, 521, 523; VI.477
Spier. Leslie. V.13, 183. 494. 504
Sprague. Andrew, VI. 471
Sproat Lake, VI. 375
Squier. Ephraim. V.264. 290, 301
Stevenson, Matilda C, V.64
Stoney (= Assiniboine), VI. 169
Straits Salish, VI. 321
Subtiaba, V.16, 24, 91, 99, 102-103,
180-181, 263-333; VI. 16
Subtiaba-Tlappanec, V.91-92, 180
Sumo, V.102- 103
Survey of California Indian Languages,
V.490
Swabian. V.487
Swadesh, Mary Haas, V.173; see also
Haas, Mary
Swadesh, Morris, V.17, 25, 105,
133-141, 143, 173, 183, 535; VI. 17,
206, 227-229, 291-296, 299-301,
319-321, 356,467-468
Swanton, John R., V.29, 62-64, 179,
223, 231, 234, 236, 271, 355, 553;
VI.22, 105, 107, 112, 130, 133,
151-166, 231
Swedish, V.160, 487
Sweet, Henry, V. 159 -160
Tabasco state (Mexico), V.102
Tacanan, VI.477
Tahltan, VI.205, 221
Takelma, V.14- 16, 21-22, 28, 31, 34,
36, 49-51, 57, 64, 71, 73, 75-79,
81, 85, 90, 97-98, 105, 135-141
passim, 144, 158, 185-186, 192, 261,
284, 491, 558; VI. 13- 16, 21, 27, 141,
153-154, 178, 225-229, 263-265,
272, 279, 284-288, 291-296,
299-315, 379, 383, 513-514
TamauHpas state (Mexico). V.102
Tanaina, VI.205
Tanana, VI.205, 221
Tanoan, V.59, 91, 93, 97. 99, 105
Tanoan-Kiowa, V.86, 91, 93, 97, 99
Taos. V.105
Tapachuhec. V.102
Tarahumare (= Tarahumara), V.104,
352 — 441 passim
Tarascan. V.82. 103. 543
Index
557
Tcil'Qeuk (Salishan), VI.428-431
Ten'a (= Koyukon), VI. 100, 136-137,
196, 219
Tepecano, V.355, 411, 416, 418,
422-428, 430, 433-435, 436-443
Tepehuane, V.104, 352-441 passim
Tequisisteco, V.102, 303, 308; see also
Chontal; Tequistlateco
Tequistlatec(an), V.90, 99, 179, 231
Tequistlateco, V.264; see also Chontal;
Tequisisteco
Teton Sioux, V.63
Tewa, V.59, 64, 81
Texas, V. 178 -179, 231, 265, 329
Thai, VI. 136; see also Siamese
Thalbitzer, William, V.64, 135; VI. 370
The Dalles (Oregon), V.135
Thomas, Alex, VI. 319, 321, 352, 359,
369-370, 374
Thomas, Cyrus, V.64, 355
Thompson River (Salishan),
VI.151-152, 426-432, 467
Tibetan, V.83; VI.23-24, 133-138,
191
Tibeto-Burman, VI. 136, 138
Tillamook, V.96; VI. 233, 241
Tillohash, Tony, V.345; VI. 352
Timucua, V.91, 99, 332
Tiwa, V.59
Tlappanec (= Tlapanec), V.91, 99,
102-103, 180, 263-264, 331
Tlatskanai, VI.205
THngit, V.16, 62-64, 78, 81, 90, 93,
97-98, 100, 105-106, 242, 254, 298,
557; VI. 16, 21-28, 105-129,
133-138, 141, 148, 154- 158 ;?a.v.s7m.
169, 189, 191
Tlingit-Athabaskan, V.93; VI.22
Tocharian, V.126, 129
Toelken, Barre, VI. 379
Tolmie, William Frascr, V.63
Tolowa, VI. 94
Tom (Nootka speaker), VI.319, 435
Tonkawa, V.82, 97, 99, 105, 223,
231-242 passim. 252, 254, 257-261
passim. 265 — 333 passim
Tonto (Yuman), V. 190 -221 passim.
226, 232-242 passim, 273-274
Totonac, V.103; VI. 288
Trager, George L., V.17, 105; VI. 17
Trinity River (California), V.181;
VI. 195
Trique, V.102, 287
Trombetti, Alfredo, V.269
Trumbull, James Hammond, V.63, 458,
491, 500-501, 508-509, 511
Ts'ets'aut (= Tsetsaut), VI. 202, 217,
221
Ts'otsMl, V.172
Tsimshian, V.14, 36, 58, 63, 71-72, 75,
78, 81, 85, 90, 93, 98, 101, 105-106,
125, 134, 143, 284, 556-557; VI. 14,
107, 151-167, 225, 227-229,
264-265, 267, 279, 286-288, 293,
300-303, 312, 372, 381, 383
Tsisha'ath (Nootka dialect), V.119;
VI.435, 465-466
Tiibatulabal, V.172, 352-441 passim
Tulkepaya (Yavapai, Yuman),
V.232 - 242 passim
Tunica, V.82, 90, 92-93, 97, 99, 261,
332, 553, 557
Turkish, V.IOO, 133, 172
Tusayan, V.354
Tutchone, VI.22 1
Tutelo, V.72; VI.47 1-474
Tututni, VI.94
Tuxtepec Nahuatl, V.402
Twain, Mark, V.485
Tyee, Bob, VI. 375
Uchucklcsil. VI. 373
Ucluelcl (Nootka dialect). V.l 19;
VI. 369
Uhlenbeck, Christianus Cornelius, V. 15.
22, 58. 69-80. 447. 457-460. 472.
492; V1.15
558
VI American Indian Languages 2
Uintah (Ute dialect), V.347
Ulua, V.102
Upernavik, VI.370-371
Upper Chinook, VI.231 -242
Upper Coquille, VI.27-28, 94
Upper Umpqua, VI.27-28, 94, 205
Utah, V.345, 347, 353
Ute, V.30, 38. 43, 264, 345, 347-349,
351. 353-354, 356-357, 362, 400,
407
Ute-Chemehuevi, V.347, 353-354, 394,
416
Uto-Aztecan (= Uto-Aztekan), V.13,
15,24,44, 49, 81, 93, 99, 102, 104,
223. 242, 345-346, 351-441 passim,
545. 549-550; VI. 13, 15, 21, 288,
314, 364-365
Vancouver Island, V.14, 135; VI. 13, 353
Vedic Sanskrit. V.129
Vera Cruz state (Mexico), V.102- 103
Virginia, V.95
Voegeiin. Charles F, V.17, 94. 143. 173;
VI. 17, 474
Voegeiin, Erminie Wheeler, V.94
Waicuri, V.102- 103
Waiilatpuan, V.90-91, 94
Waiilatpuan-Lutuami-Sahaptin, V.94,
97-98
Wailaki, VI.25, 196, 205
Wakashan, V.13, 25, 62, 81, 84-85, 90,
93. 97-98. 101. 110-113, 116,
118-119, 123, 125, 532, 543; VI. 13,
23, 133, 163, 264. 280, 319-321,
353, 368, 442, 465, 467-468
Wakashan-Salish, V.90, 93
Walapai, V. 1 90 - 22 1 passim. 232 - 242
passim, 261, 266 — 333 passim; VI. 209
Wappo, V.293, 557
Wasco. V.27; VI. 226, 231
Washington (state), V.63; VI.435
Washo, V.79, 90-91, 97, 99, 105,
178-181, 223, 225-228, 249,
252-255, 257-261, 264-333 pas-
sim
Waterman, Thomas Talbot, V.407, 453,
498, 506
Watts, Dan, VI.359-360, 369
Weitspekan (Yurok dialect), V.453
West Indies, V.95, 103
Westermann. Dietrich, VI. 173
Whistler, Kenneth W., VI.225
Whorf, Benjamin L., V.17; VI.288-289
Wikchamni (Yokuts dialect), VI.292
Williams, Frank, VI. 319
Williams, Roger, V.498, 504, 510-511,
515
Winnebago, V.63, 72, 105; VI.363
Wintun (= Wintu), V.82, 85, 90,
97-98, 105, 134, 177, 185-186, 332;
VI.225, 263-265, 267-269,
285-288, 291-296, 301-302, 313
Wishosk (Wiyot), V.453
Wishram, V.15, 77, 135-142 passim,
184,486; VI. 15, 225-226, 231-261,
267,275-280, 320, 376
Wissler, Clark, V.494, 496, 502, 511
Wiyot, V.13, 15, 81. 90-91, 93,
97-98, 447-449, 453-482,
477-498, 553-561; VI. 13, 15; see
also Wishosk
Wobonuch. V.352 — 441 passim
Wyandot, V.160
Xaisla (Kwakiutlan), VI. 353; see also
Haisla
Xinca, V.91, 103; VI.228
Yahi. V.177. 184. 190. 232. 267. 292,
295, 328, 335
Yakima, VI.231, 250-252
Yakonan. V.90. 97-98; VI.263-265,
267; see also Alsea
Yale University, V.17, 149; VI. 17. 24,
222, 321
Yana, V.15, 21, 31, 34, 36, 44-49, 55,
71, 73, 77-78, 83, 86, 90-91,
97-98, 105, 135-141, 177-183,
Index
559
186, 189-222, 225-229, 231-242,
249, 251, 254-255, 257-261,
264-333 passim, 335-341, 449, 532;
VI.14, 15, 108, 153-154, 306, 320,
358-359, 365
Yaqui, V.355
Yavapai, V.226, 279; see also Tulkepaya
Yawelmani (Yokuts dialect), VI.292
Yellowknife, VI.221
Yokuts, V.59, 97-98, 105-107, 125,
134, 137, 177, 561; VI.207, 225-227,
228, 263-265, 269-270, 283-288,
291-296, 301, 311
Yopi, V.263
Yucatan, V.102
Yuchi, V.82, 91,93,99, 332
Yuki, V.75-78, 82, 90, 93, 99, 105,
178-179, 223, 293, 332, 557,
560-561
Yukon, VI. 129
Yuma, V.190-221, I'il-lAl passim.
266 — 333 passim; see also Kutchan
Yuman, V.64, 90-92, 97-99, 102, 104,
177, 179, 189, 223, 226, 231-233.
253-254, 260-261, 266; VI. 209
Yungas, VI.475
Yurok, V.13, 15, 81,90-91,93,
97-98, 105, 447-449, 477-478.
453-528, 553-561; VI. 13, 15, 25
Zapotec, V.91, 102, 287, 293, 299
Zeisberger, David, V.493, 513
Zoque, V.102- 103; VI.228, 283
Zoque-Mixe, V.82
Zuni, V.82, 91-92, 94, 99
UNIVERSITY OF CALIFORNIA LIBRARY
Los Angdes
This book is DUE on the last date stamped below.
OCTl''^^^
REC'D
ftCNOVO
SEP 3 (f ^^
■"^^ ' 4 1996
l.^^''
YRL JAN 2 12003
315
L 006 331 529 5
