These texts were recorded, chiefly at Numbulwar Mission (Northern Territory) in the period 1973 to 1977. As a Research Fellow of the Australian Institute of Aboriginal Studies I was there working on several languages, but with primary emphasis on Nunggubuyu.
I have previously published or completed grammar-text-dictionary volumes on other languages (Ngandi, Ritharngu, Warndarang, Mara). My Nunggubuyu work will be of similar structure but in three volumes.
It may seem unusual that the texts are first to appear, but this is no accident. I believe that a substantial text collection is the publication of greatest long-range value which a field linguist can produce, for several reasons.
First, despite claims sometimes made to the contrary, a dictionary or grammar by itself will tell us little about the people who speak the language, while a text collection can tell us quite a bit in conjunction with ethnographic reports. Secondly, a large textual corpus is needed to illustrate and substantiate claims made by the linguist on lexical and grammatical points.
Pursuing the first argument, I would claim that Australianist linguists have not, in most cases, done even a minimally adequate job to date in obtaining and publishing myths and ethnographic texts of significant value to (social and linguistic) anthropologists, or indeed to the present-day Aboriginals and their descendants. Such neglect would be less serious were it not for the fact that most nonlinguistic social anthropologists in Australia have, to a greater extent than anthropologists working in many other regions, labored under serious linguistic difficulties which have made it very difficult for them to achieve a full understanding of native cultural systems. Of course, the publication of myths and ethnographic texts is only part of an adequate ethnography, but it certainly is one part of it, so if social anthropologists have typically been unable to record and publish sufficient texts in Aboriginal languages it behooves linguists to provide assistance. I do not think there is a single anthropologist working on traditional Aboriginals who could not profit from this kind of linguistic assistance.
As for the second point, even if our goal is limited to the usual kind of grammatical and lexical descriptions, I would argue strongly that a large textual corpus is a sine qua non of an adequate description of a language. For one thing, when reading other linguists' grammars and dictionaries, we have a right to expect documentation of the points made. If we simply "trust" the author, we will end up taking on blind faith many points which may turn out to be false, or at least susceptible to alternative interpretations not mentioned by the original author. In essentially every case I know of where a second linguist undertook a restudy of a language previously studied by the first linguist, important and basic factual and interpretative discrepancies have arisen. Both for the benefit of subsequent fieldworkers and library readers wishing to achieve a thorough knowledge of this language, and demanding knowledge of how I arrived at particular judgements or conclusions, this text collection can go a long way toward resolving doubts--or raising them, No fieldworker has the right to expect that his/her judgements will be upheld by posterity; hence every fieldworker has an obligation to present original data.
In the dictionary and grammar volumes (soon to appear), numerous cross-references will be made to words and passages in this collection. By sending readers back to the texts, they will be encouraged (and hopefully forced) to study word meanings and grammatical points in context, and in the flow of real discourse. I hope in some grammatical sections simply to list all relevant textual passages and suggest that the reader examine those passages and come to his/her own conclusions. In this fashion, the reader can (and indeed must) acquire a relatively good direct feeling for the data, and can hopefully proceed to analyse or reanalyse whatever points are of personal interest. Since the reader will find the material already partly processed (through transcriptions, word glosses, etc.), he or she will not have to repeat the original linguist's fumbling first steps, and may be able in a reasonable amount of time to arrive at the same level of understanding achieved by the original linguist in nearly two years of work.