A full account of which relatives were ‘taboo’ (and in whose presence the use of Dyalŋuy was obligatory) would require a more complete understanding of the Dyirbal kinship system than the writer has. For a woman, father-in-law was taboo; similarly for a man, mother-in-law, and also father’s sister’s daughter and mother’s brother’s daughter but not mother’s sister’s daughter. The relation was symmetrical: if X was taboo to Y then Y was taboo to X.¹

The name ‘Dyirbal language’, as used here, covers the ‘languages’ of three tribes: Mamu, Dyirbal and Giramay.² These are grammatically almost identical and have over seventy per cent common vocabulary so that it is convenient to refer to them as three dialects of a single language, and to give a single overall description of their grammar (Dixon 1968 a). Each of these tribes had its own Guwal and Dyalŋuy (and similarly for at least half-a-dozen surrounding tribes). Although the Guwals are still actively spoken, the Dyalpuys are not. Mamu and Dyirbal Dyalpuys, for instance, ceased being used about 1930; since then just the Guwals have been used, even in the presence of taboo relatives. However, older speakers remember a great deal of their Dyalpuy, and speak it readily when asked. The writer gathered a large amount of Dyalpuy data independently from a number of different Dyirbal and Mamu informants; the consistency of this data indicated that the language was being accurately remembered.

Each particular dialect had no lexical words common to its Guwal and its Dyalŋuy. However in a number of cases a word in the Dyalŋuy of one tribe occurs, with a similar meaning, in the Guwal of a neighboring tribe. For instance, the word for ‘back’ is mambu in Dyirbal Guwal, d^yud^ya in Dyirbal Dyalpuy, d^yud^ya again in Mamu Guwal and nabil in Mamu Dyalŋuy.

English loan words were taken directly into Guwal, suitably cast into the mould of Dyirbal phonology: thus, bigi ‘pig’, bulugi ‘cattle’, d^yuga ‘sugar’, d^yarud^ya ‘trousers’, mani ‘money’ and so on. We have seen that everything must have a different name in Dyalŋuy from that which it has in Guwal; and we have also noted that Dyalŋuy is extremely parsimonious, keeping its vocabulary as small as possible. Consistent with this, Dyalŋuy most often dealt with loans by extending the range of meaning of an existing Dyalŋuy term; only in one or two isolated cases did Dyalŋuy take on a new word to correspond to loans in Guwal. Thus bigi ‘ pig’ is translated into Dyalŋuy by ginga, which was already the Dyalŋuy correspondent of Guwal gumbiyan ‘porcupine (echidna)’: pig is assigned to the same taxonomic group as porcupine. Cattle are named in Dyalŋuy according to their main functional characteristic, ŋunŋun ‘ breast ’. Sugar is, by its texture, identified in Dyalŋuy with ‘ sand ’, warun^y; whilst trousers are given the same name as the part of the body they cover, Dyalŋuy dabara, ‘ thigh’.³ However, no existing Dyalŋuy item could be extended to cover ‘ money ’ and in this case a new word was introduced into Dyalŋuy: walba.

¹ Children were promised in marriage at an early age, thus acquiring a full set of taboo relatives; they seem to have learnt Dyalŋuy in the same way that children normally learn a language.

² Dyiru and Gulqay — both now pretty well extinct - could probably have been considered other dialects of this language. Gulqay was spoken by the Malanpara tribe, referred to a great deal by Roth (1901-5, 1907-10).

³ Loans were taken into Dyirbal as nouns, adjectives or time words, never as verbs. Thus ‘work’ becomes an adjective, wagi; however this normally occurs in verbalized form wagibin.