Westfälische Wilhelms-Universität Münster
Institut für Allgemeine Sprachwissenschaft
(Department of General Linguistics)
Hauptseminar: Grammatiken außerindoeuropäischer Sprachen - Dyirbal
Sommersemester 2000, Wintersemester 2001


Dyirbal in assimilation —

social surroundings and grammatical change of present-day Dyirbal

Or: What did actually become of the language DIXON described in 1968?


"A language is the emblem of its speakers.
Each language determines a unique way of viewing the world.
It encapsulates the laws and traditions and beliefs of its ethnic group"

(DIXON (1997), p. 135)

1. Introduction

For two semesters in the 2000/2001 academic year, I was - in the context of a main seminar "grammars of Non-Indo-European languages: Dyirbal" - working with one of the probably most well-known languages of Australia. Above all, the special features of its grammar were the centre of attention - as for example the ergative, the nominal classes and the so-called "mother-in-law language". Special features, which partly were described in detail for the first time on the basis of Dyirbal, by Robert "Bob" M.W. DIXON in his 1972 PhD-thesis "The Dyirbal Language Of North Queensland” (Univ. London, 1968), by which both the phenomenon of ergativity as well as the language Dyirbal itself achieved reputation in linguistic circles. Since then, in examples about ergativity one gladly uses phrases like: "in Dyirbal the following construction is being used..."

With this paper, I would like to investigate whether this present tense is still justified at all, and to explain which development took part in Dyirbal since DIXON described it for the first time more than thirty years ago, and which social factors are or were the reasons for that development. I will found my theses upon the investigation of Annette SCHMIDT (published as SCHMIDT 1985a and 1985b) and on some papers by other authors which can throw a picture on the development since 1985. Using some recent publications that describe the general mechanisms of language death, I would then like to illustrate the dying of Dyirbal and the reasons for it within a larger context. First, however, a general view of Dyirbal, as it is presented particularly by DIXON (1972) and on the most obvious symptom of the fall of this language - its significantly declining number of speakers.

[top of page]

2. General remarks on Dyirbal

2.1 The Language

By the designation Dyirbal (also written Jirrbal, Djirubal or Dyirrbal)(1), linguists usually summaris a continuum of dialects, whose largest and dominant one (often written Jirrbal for distinction) gives the name for the whole set (2). The other dialects are called: Giramay, Mamu, Dyiru, Gulay, and Ngadyan (cf. DIXON (1972), p. 24). They all are or were spoken in the north-east of the Australian Federal State Queensland, in the region south of Cairns, between Innisfail and Ravenshoe. There is no name for the totality of the tribes that speak these languages. On the contrary: the groups call themselves after their individual dialects Dyirbalan, Giramaygan, etc. (cf. op. cit., pp. 23f.).

An analysis of the age of certain geological phenomena (tracks of volcanism; changes in the coastal shoreline caused by ice-age changes of the sea level) with events in traditional myths suggest that the tribes specified above already have been residing in the region for more than 10,000 years (cf. op. cit., p. 29). The abovementioned and some neighbouring dialects are ranked among the Dyirbalic languages in a sub-group of the Pama-Nyungan languages (cf. YALLOP, pp. 47 and 52). The predominant part of the Australian Aboriginal languages belongs to this language family which extends from the west coast of Australia to Cape York in the north-east of the continent (cf. op. cit., p. 42).

Dyirbal is characterised by DIXON as a "typical Australian language" (DIXON (1972), p. 22). It has a small phoneme inventory (3), a system of four noun classes (genders) marked at the pronoun or demonstrative, three numbers (singular, dual, plural), a rich case system with Split Ergativity, an opulent system of derivational and aspectual verbal suffixes, two tenses (future and non-future), as well as a morphologically marked differentiation of transitive and intransitive verbs (cf. op. cit., pp. 22f.). A remarkable sociolinguistic feature of Dyirbal (and some other Aboriginial languages) is the "mother in law language” (such labelled by DIXON's main informant CHLOE GRANT; cf. DIXON (1991), p. 191). This system, in principle, is a parallel (generalising) lexicon with the same phonology and grammar as the basic language, which serves for concealing the speech in the presence of certain persons who are considered taboo with regard to the speaker. In Dyirbal, the system of the everyday life language is called Guwal; the avoidance language calls itself Dyaluy (cf. DIXON (1972), p. 32). According to DIXON, the "mother-in-law language" has not been in use since the 1930'es (cf. DIXON (1983); p. 168), a first indicator for the loss of the traditional culture and language.

[top of page]

2.2 The Language Death of Dyirbal

The number of Dyirbal speakers has been reduced considerably since the first contact with white settlers approximately 150 years ago. The estimates in the relevant sources deviate quite significantly from each other. Before the first contact (1848), the tribes in the region probably each had up to 500 members (cf. DIXON (1972), p.34). They were diminished, however, within only a few years after the first settlement (1864):

"There must have been at least 5,000 people speaking dialects of Dyirbal during the first part of the 19th century [...]. Many Aborigines died from catching European diseases [...] to which they had no immunity. And many were shot as they tried to fight - with spears and boomerangs against guns - for their heritage. There was a native police consisting of Aborigines brought in from southern Queensland, armed with guns and commanded by white officers to seek out local Aborigines and shoot on sight. [...] The Aboriginal population of the Cairns rain forest region was reduced to perhaps 10 percent of its precontact level within 50 years of the advent of Europeans."
(DIXON (1991), p. 186).

According to DIXON (1972; p. 37), only approximately 30 speakers of Dyirbal, about 10 speakers of Giramay, 5 or 6 for Mamu and Ngadyan and 1 or 2 each for Dyiru and Gulay remained by 1970. In different editions of the Ethnologue, the present Dyirbal is indicated as "nearly extinct" with a number of 40-50 speakers and the vague source "(1983 R.M.W. Dixon)"(4) (cf. GRIMES (1984), p. 559 and GRIMES (1996), p. 818). It is DIXON himself who 1997 supplied the most recent numbers available in the linguistic literature:

"Thirty years later [= 1993,J.W.] only those over about sixty-five can speak the language — there are just half-a-dozen of them."
(DIXON (1997), p. 105).

This specification still seems to be correct at present. Upon request, DIXON marked Dyirbal and Giramay with: "about nearly extinct, 6 speakers now " (DIXON, personal correspondence, 07 May 2001), Mamu and the other dialects he simply characterised as "gone" (ibid.). In the abovementioned figures, one thus can note a strong decrease of speakers - a phenomenon, which is generally called the dying of a language. In the introduction to his essay of 1991, DIXON describes the rapid fall of Dyirbal as follows:

"In 1963, when I began work on Dyirbal, there were several score fluent speakers, including a fair number of children. The language appeared to be in a reasonably healthy state. [...] I have worked fairly steadily on [Dyirbal] for just over a quarter of a century. During this period, I have seen the language decline from a state in which there was an abundance of speakers who could supply the information I sought to one in which there is just one good consultant left for each of three dialects, with no one to go for a second opinion. The language has died at a faster rate than I could record it."
(DIXON (1991), S. 183; emphases by me).

Annette SCHMIDT, in her thesis "Young people's Dyirbal. An example of language death from Australia", points out the agony of the language. In addition to SCHMIDT, other authors (among them DIXON again, of course) have commented the degeneration of Dyirbal. It will be the subject of chapter 3 to describe this decline in more detail.

[top of page]

3. The investigation of A. SCHMIDT

3.1 Generally

Annette SCHMIDT examined in her Master's thesis the situation of Dyirbal in the first half-year of 1982, 15 years after DIXON had described the language for the first time, thus half a generation later. Contrasting her findings with DIXON's grammar, SCHMIDT explained, in what respect the Dyirbal of the recent generation (short: YD) deviates from the traditional Dyirbal (short: TD).

3.1.1 Method

SCHMIDT's observations partly come from an open direct questioning, partly from "blind" studies. First, she took up informal discussions and conducted TD understanding tests, in which she also operated with intentional control errors at assumed weak points. Beyond that, SCHMIDT interviewed the young people with regard to language use and language prestige of Dyirbal and English. For the insusceptibility of her findings and for avoidance of the "observer effect" SCHMIDT additionally employed informal investigations, in which she unnoticed examined the language use of the young Dyirbalan outside of the overt questioning situation (cf. SCHMIDT (1985a), p. 7). Group of reference are the speakers aged between 15 and 35. According to SCHMIDT's estimate, in 1982 these were approximately 20 persons, of whom 12 agreed to participate in the tests and interviews (cf. op. cit., p. 44). Since the "semi-speakers" have a very negative attitude to their linguistic variety, they tried to refer SCHMIDT to "good traditional speakers", since they regarded their own simplified Dyirbal unworthy of linguistic investigation (cf. op. cit., p. 7). As soon as the juvenile speakers felt observed, they no longer spoke frankly, but either English or an artificial, "more correct" Dyirbal with more intended attention (cf. ibid.).

3.1.2 Setting

The Investigation was executed by Annette SCHMIDT from January to June 1982 in the Jambun Aboriginal Society in the municipality of Murray Upper. This is the last speaker community, in which (1982) there still were a considerable number of speakers of Dyirbal. By the time of her visit, approx. 250 whites, 15 Torres Strait Islanders and 110 Aboriginals of different origin lived in Murray Upper (cf. op. cit., pp. 6 and 13). The social situation and the linguistic conditions specified in the rest of chapter 3 always (if not indicated otherwise) refer to this period and place of investigation.

[top of page]

3.2 The Social Situation

Since the founding of the Jambun Aboriginal Society 1977 by the government of Queensland, the interaction between whites and Aboriginals is almost exclusively limited to official contexts such as school, the grocery store, and work (cf. op. cit., p. 13). The (self-)isolation of the Aboriginals is being kept upright in "mixed" contexts (e.g. in the school) as well. If possible, contact to whites is being avoided as far as possible (cf. op. cit. pp. 13f.). From the part of the whites, a strong linguistic pressure is exerted on the Aboriginals: Many employers sanction the use of Dyirbal at works, there is only purely English- media (print and radio), and English is the sole official language (cf. op. cit., p. 17). Since English is also used as exclusive language at school, the Aboriginals often already employ English within the family, in order to minimise the scholar and social disadvantages of their children (caused by their skin colour and culture), or at least not to add to it by a linguistic heterogeneity (cf. op. cit., p. 20). Dyirbal therefore is being avoided and regarded as low according to the value system obtained from the whites.

The speakers carefully use the language in selected contexts. The language, in particular the one only being incompletely acquired Dyirbal by the young people, has a bad prestige with the Aboriginals as well, and is thus neglected in favour of English. This opens the door to an extended cultural and linguistic remodelling (cf. op. cit. pp. 18f.). The following linguistic varieties are being spoken in Jambun: Traditional Dyirbal (TD), Young Dyirbal (YD), Jambun English (JE, a Pidgin) and Standard Australian English (SAE) (cf. op. cit., pp. 26-28). SCHMIDT's investigation focuses on YD and the deviations from TD as described by DIXON. The speakers of YD are so-called semi-speakers of Dyirbal: their language skills deviate from the standards of the "traditional" language (cf. SCHMIDT (1985b), p. 381):

"At first, my impression of 'imperfect' Dyirbal was of a dismal patchwork of inconsistencies and (from the point of view of TD) mistakes, haphazardly distributed over speakers and situations. It was easy to suppose that such picture reflected a sporadically disrupted stage in the decay of TD. However, I gradually became aware that the apparent 'mistakes' of the YD speakers were not random errors: rather, each individual had his own grammatical system for Dyirbal communication, involving simplification of the traditional grammatical norm to a greater or lesser degree."

The general differences between Young Dyirbal and Traditional Dyirbal will be the subject of the following sections. I prefer, however, not to follow SCHMIDT's sequence of chapters, but organise the sections of this paper in accordance with the "classical" arrangement of linguistic disciplines. Thus, I shall begin with the phonology, then turn to the morphology, the syntax, the semantics, and the vocabulary, and finally look at the pragmatics of Young Dyirbal.

[top of page]

3.3 Phonology

As mentioned above, YD is influenced by English. Usually, the phonological systems of Dyirbal and English are still being differentiated, but there are clear influences of SAE. In loan words or Pidgin forms in Dyirbal, English sounds unknown to Dyirbal emerge, e.g. the fricatives /f/ and /h/ as with the pronoun yufela 'you two' or in helicopter (cf. SCHMIDT (1985a), p. 197). English intermissions into Dyirbal speech are likewise pronounced in accordance with the English standards:

"anaji happen to buran bayi helicopter
  we-PL            see   he                         
We happened to see a helicopter."
(op. cit., p. 192; Emphases by the author).

Sounds or sound combinations unknown to English are particularly susceptible to change: The speakers of YD are apparently already so familiar with the sound inventory of SAE that they do not implement certain contrasts any longer and try to map the "strange" sounds of Dyirbal on the more familiar phonemes of English. Thus, the contrast between the two liquids /r/ and // is only retained in minimal pairs as /yaa/ 'man' vs. /yara/ 'fishing line'. In non-contrastive positions (with words not part of a minimal pair) the difference is blurred, and both sounds are used like allophones (cf. op. cit. pp. 192f.). The youngest speakers often produce /r/ and /d/ (especially intervocalic) equally as []. These sounds finally collapse into /d/. In approximation to English, // is often produced as [] or [l], so that the rhotic liquids are being adapted to those of the English phoneme system (cf. op. cit., p. 196). In traditional Dyirbal, the velar nasal // can also appear word-initial - an occurrence that is impossible in English. The speakers of YD now tend to use [n] here instead of []. Furthermore, the non-homorganic consonant sequence /ng/ is assimilated to /g/ (cf. op. cit., S.196f.), so that with the nasal consonants as well an adjustment to English can be observed. This needs not inevitably to be explained with English influence. Such assimilation can occur without external influences as well and would even then still mean no indication for the decay of a language.
Altogether it can nevertheless be noted that the sound system of TD is not strictly maintained any longer, but in YD approximates the superstrate English.

[top of page]

3.4 Morphology

The reshaping of a language is not only apparent within its phonology. If the speakers of a moribund language have only incomplete knowledge, this mostly also quite clearly shows in the use of grammatical forms. The more complex a form or a construction is, the longer it takes to learn it in language acquisition until a speaker learned its use, and the more easily such a form is also forgotten again, if it is not actively utilised any longer (cf. NETTLE/ROMAINE, p. 55).

3.4.1 Case

Dyirbal (TD) has a rich system of 9 cases (nominative, ergative, instrumental, (actual) genitive 1, locative, (general) genitive 2, dative, allative, ablative) which are marked by suffixes on nouns and adjectives. Five of these cases show phonologically conditioned allomorphs (cf. DIXON (1972), pp. 42f.). Ergative

TD is a language with split ergativity: nouns, Adjectives and the markers of the 3rd Person are inflected according to an absolutive/ergative pattern, whereas pronouns of the 1st and 2nd person are inflected according to a nominative/accusative pattern. In TD, six allomorphs for the ergative exist: {- gu, -gu, -ju, - ru, -bu, -du} (cf. SCHMIDT (1985a), p. 46.). With the speakers of YD examined by SCHMIDT, this set is reduced as follows:

Table 1: Ergative Allomorphs in TD and YD

systemTDYD (speakers)
environment "EM" "MJ" "BM", "EJ", "ED" the remaining 7
2 syllables + V-gu -()gu -()gu -()gu -gu ..>.. Ø
>2 syllables + V-gu
-y-ju -ju
-rr, -r, -l-ru (+drop of stem-final liquid)-ru -du -du
-m -bu -bu -bu
-n-du -du -du
-ny-ju -ju -ju
(after DIXON (1972), p. 42; and SCHMIDT (1985a), p. 47, tab. 4.1). (5)

As seen above, the variety of the ergative markers is diminished. With most YD speakers - if at all - only one form {-gu} is available, with a few still {-du} appears as reflex of a former assimilation to the stem-final consonant. Thus for example for the phrase [the/a] woman saw [the/a] dog instead of TD jugumbiru guda buran would render jugumbilgu buran guda in YD. (cf. SCHMIDT (1985b), p. 385). Frequently, however, a YD speaker would switch to a nominative/accusative construction or to a purely syntactic marking as in English: Subjects of transitive and intransitive are put before the verb, objects are placed behind it (cf. SCHMIDT (1985a), pp. 47f.). The abovementioned example thus would be: jugumbil buran guda. Some of the YD speakers who do not use the ergative any longer apply the spectrum of its allomorphs to the dative/allative, which was marked by {-gu} only in TD (cf. op. cit., p. 59).

What is remarkable about this reduction of the allomorphs is that they permit a conclusion on underlying representations, which would not be possible like that by looking at TD alone, since each allomorph only occurs in one environment. Yidiny, the northern neighbour language of Dyirbal, uses {-gu} after vowels and {-du} after consonants as ergative markers. After nasals and /y/, {-du} is assimilated, but it remains unchanged after liquids. Thus one can conclude that {-du} must also be the underlying post-consonantic form. These two underlying forms {-gu, -du } of Yidiny are the very same that result of the allomorphic simplification by three speakers in YD (see Table 1). (cf. SCHMIDT (1985b), p. 385). Other Cases

The TD locative was just as rich in suffix allomorphy as the ergative and differed from this only by final /a/ instead of /u/. In YD, the forms of the locative are reduced similarly to those of the ergative to {-ga} or {-da} (cf. SCHMIDT (1985a), p. 53f.). Two speakers of YD exclusively use, the other occasionally use English prepositions instead of the locative:

"jugumbil    nyina-nyu     on     yugu.
woman       sit-NOFUT	         log
The woman sat on the log." 
(ibid.; emphasis by the author).

The instrumental case, which in TD had the same forms as the ergative, is in YD also (or exclusively) marked by the suffix {-bila}, which in TD was a comitative marker only. The present interference can be explained from English, where both functions are expressed by the preposition with (cf. op. cit. p. 55 and 57).

The markers of the genitives collapse into {-u}, whereby the difference between GEN1 and GEN2 (actual vs. general possession) is neutralised. Also, the inalienable possession is increasingly being marked with this form whereas in TD it was not marked morphologically (cf. op. cit., pp. 60f.). The result of this development is displayed in the following table:

Table 2: Genitive markers in TD and YD

GEN1 (after nasal)-u-u
GEN1 (elswhere)-u
inalienable possession
(after SCHMIDT (1985a), p. 61).

Recapitulating, it can be stated for the case system that with the "core cases" (op. cit., p. 46) the allomorphs collapse into one or two forms, while the peripheral cases in YD mostly are being replaced by prepositions or Ø, so that the cases concerned finally fade completely and YD in its tendency thus becomes more isolating (cf. ibid.).

3.4.2 Other Morphologic Phenomena The Verb

The verbal inflection likewise is being reduced in YD. The affix for the marked tense Future {-n(ja)y, - l(ja)y} is dropped, a posterior action is marked by lexical means only, e.g. by ulga ('tomorrow'). The affix for the less marked nonfuture {-nyu, -n} remains as only and constant verb affix (cf. op. cit., p. 64f). Something similar is the case with the other inflectional suffixes (e.g. aspect, reflexivity, etc.; cf. op. cit., pp. 69-76 and 82ff.), thus I shall not to go into further detail about these here any further.

The -urra-construction for the connection of two verbs with same agent (of simultaneous or immediately subsequent actions) is omitted almost completely. In TD, it worked as follows:

DEM-ABSwood-ABSDEM-ERGman-ERGthrow-NOFUTgo uphill-urra
The man threw the stick and [immediately afterwards] went uphill.
(after DIXON (1972), p. 77).

In cases like the aforementioned, where a NP with the agens in the ergative and another NP with the agens in the nominative (here eliminated by the construction) are co-ordinated, the -urra construction is mandatory in TD, a direct co-ordination would be considered ungrammatical (cf. op. cit., p. 78). The speakers of YD examined by SCHMIDT no longer used this construction but for one exception, while linking the underlying phrases (bala yugu bagul yaragu madan and bayi yara waynyjin) of above example with English conjunctions like an' or then instead of the transformation (cf. SCHMIDT (1985a), p. 67). Other Word-Classes

The pronominal system maintains the three numbers. Whilst the pronouns of the singular remain unchanged in YD, in dual and plural pidgin-type forms appear, bearing a striking similarity with the forms of Tok Pisin and similar English-based Pidgins (cf. op. cit., p. 87ff):

Table 3: Pidgin-type Personal Pronouns

 YDTok Pisin
1.Pers.min-dubalawi-felayumitupelamipela (excl.)
yumi (incl.)
(after SCHMIDT (1985a), p. 88; and VERHAAR, p. 19).

In TD, there were actually no forms of the 3rd Person dual and plural, but the demonstratives balagara and balamagan, just like the demonstratives bayi (MASK Sg.) and balan (fem. sg.) partly fulfil the functions, for which in other languages pronouns would to be used (cf. SCHMIDT (1985a), p. 86).

The Demonstratives are likewise submitted to a change: the distinction of ergative/absolutive and nominative forms are omitted, S, A and O are in the unmarked Nominative, the allomorphy of the remaining cases is overthrown - thus yielding the same results as already shown in section 3.4.1. In addition to this, the nominal classes that are morphologically marked only by the demonstrative are reduced (see also chapter 3.6 for this), as the classes III and IV collapse (cf. op. cit., pp. 91ff.).

In the set of the interrogative pronouns a strong reduction of the forms is to be observed as well, further affirming the observations made above. (cf. op. cit., pp. 96f.). The morphology of YD is thus simplified in all areas, as paradigms are reduced in their extent and allomorphies are diminished. Complicated forms are avoided while the constructions and the categories expressed by them are approximated to English. A fundamental change, which also entails consequences for the syntax of YD.

[top of page]

3.5 syntax

3.5.1 Basic Order

The dropping of morphological forms leads to the situation that certain grammatical functions must be implemented in other ways. Especially the reduction of the ergative/absolutive system results in an ambiguous indication of the agent or subject function and the object function, unless it is marked by means of syntax. This is achieved by a more rigid arrangement of the parts of a sentence, following English patterns. In TD, the order of elements always has been excessively free, with a preference for S-V in intransitive sentences and O-A-V (if A is a noun) or A-O-V (if A is a pronoun) in the transitive sentence. The sequence in the intransitive sentence is not changed in YD, probably because in English the same sequence is used. In the transitive sentence however A-V-O or A-O-V (both with pronominal and nominal A) prevail. Thus these basic types likewise approach to the English S-V-O, where the subject (no A/S distinction) has to be initial. (cf. op. cit., p. 105f.):

TD:baunjugumbi-rumargi-gumirrany - Øbabi-n
YD: baunmargijugumbil-dubabinmirrany
SAE:The thin woman sliced the blackbeans.
(after SCHMIDT (1985a), P. 102).

3.2.5 The S/O Pivot

The so-called S/O pivot is a salient item of the syntactic ergativity of TD. If two sentences are to be co-ordinated, they must have a common NP in S or O function. If the common NP in one of one phrase is S or O and A in the other one, then a transformation of the NP of the latter must be executed by means of the dative case and the anti-passive in the verb of the second sentence, in order to be able to co-ordinate the two sentences (cf. op. cit., pp. 111f.). In YD, this transformation using the anti-passive and dative is used less frequently and almost only in constructions with a purposive (cf. op. cit., p. 114).

SCHMIDTs generalisations and interpretations based on the analysis of expressions of two of the twelve YD speakers are in my opinion somewhat far fetched. If with two speakers the allomorphs of the anti-passive suffix are split up on two new functions, then this is a remarkable innovation which indeed is against the general trend towards a reduction of complexity, but nevertheless it is only one minor exception (made by two out of 12 speakers) and no trend of its own. Finally it is undeniable that the larger part of the YD speakers regards the suffix {-laygu} as a final marker (cf. op. cit., p. 116). Originally, {-l()ay} was the suffix for the anti-passive and {-gu} the marker for purposive. These speakers re-analysed this sequence as one morpheme, because the {-l()ay} derivative and the co-ordination of two sentences by means of the anti-passive occurs almost exclusively with purposive sentences in YD (cf. op. cit., p. 117f.). Thus, the {-laygu} suffix can also be joined with intransitive verbs -- the original {-l()ay} could (logically and by its function) only occur after transitive verbs (cf. ibid.).

3.5.3 Sentence Co-ordination

Most YD speakers no longer used the S/O pivot to co-ordinate or subordinate sentences, but connected them by simple juxtaposition or by {-u} that is attached to the verb of the relative clause (cf. SCHMIDT (1985b), p. 391), or by (English or dyirbalised) conjunctions like an', then or baum 'then' (cf. op. cit., p. 388). Altogether, the tendency towards simplification and approximation to English structures already determined in the preceding paragraph is obvious again, both which regard to the basic sentence structure, and with the linkage of two sentences in the syntax of YD. The grammar of Dyirbal thus simplified and converted in the course of the language's decline. In addition to this, remarkable modifications within the range of semantics and dictionary took place, as we will see in the following chapter.

[top of page]

3.6 Semantics

3.6.1 The Nominal Classes

As already stated above in paragraph, the system of the nominal classes of TD shifted to a completely different pattern in YD. In TD, each noun was assigned to one of the four classes I to IV and was linked with the appropriate demonstrative which marked the class affiliation:

Table 4: Nominal classes in TD

ClassExampleSemantic Class
Ibayi yara 'the man'
bayi gagara 'the moon'
animate, male (human);
mythological association
IIbalan jugumbil 'the woman'
balan buni 'the fire'
female (human);
water, fire, war, danger
IIIbalam wuju 'the vegetable'
balam mirrany 'the blackbean'
edible (plant)
IVbala yugu 'the tree'anything else
(after SCHMIDT (1985a), p. 154, tab. 6.1).

The affiliation of the words to the respective classes is not arbitrary. For TD native speakers it actually was quite motivated since the affiliation was fully explained by cultural and mythological connections. The remarkable irregularities or "deviations" are explicable, for example by mythological or conceptional associations: Most birds, which presumably should be in class I due to their animatedness are in fact assigned to class II in TD, because one assumed the souls of deceased women in them; The words for 'fishing line' and 'fish spear' are (although the denotate is definitely inanimate) in class I, since they are associated with the (animated) item 'fish'. Some aggressive or poisonous animals are assigned to class II instead of class I for similar reasons (cf. op. cit., p. 151). This system of class allocations is absolutely logical for the Dyirbal speaker who is familiar with the cultural heritage, the traditions, and the mythology of his tribe. For outsiders it is complicated and hardly transparent. Young Dyirbalan who did not grow up in the traditional way any more (and no generation has done so after 1940)(6) therefore have difficulties with this system (cf. op. cit. p. 152f.). In their language use, they shift to a more regular and more predictable system which is based on two basic dichotomies: animate/inanimate and male/female, whereby only the animated ones are being differentiated for sexus. Thence, the following three classes result:

Table 5: Nominal classes in YD

ClassExampleSemantic Class
Ibayi yara 'the man'
bayi yuri 'the kangaroo (male or indefinite)'
+animate +male
+animate (not marked for sexus)
IIbalan jugumbil 'the woman'
balan yuri 'the kangaroo (female)'
+animate +female
IVbala wuju 'the vegetable'
bala yugu
'the tree'
bala gagara 'the moon'
bala buni'the fire'
bala mirrany 'the blackbean'
(after SCHMIDT (1985a), p. 154, tab. 6.1).

The allocation of lexical items to the new classes thus takes place more strictly and according to substantially more transparent criteria and by this simplifies the system substantially. This occurs most clearly with the lees fluent YD speakers. Those who had a larger Dyirbal competence, used scrambled systems or were uncertain about class affiliations of certain lexemes (cf. op. cit., p. 159).
Despite all inhomogenity, these mixed systems show some principles by which the restructuring of the nominal class allocation took place. Certain inexplicable exceptions (the words for 'dog', 'platypus' and some other animals in TD belonged to class II instead of I) are dropped, just as the exceptions justified by the culture e.g. the conceptually associated (like 'fish spear', cf. above) and those marked for the semantic component 'danger' (cf. above) are being reassigned to the classes one would intuitively expect them (cf. op. cit., p. 163). Rather "peripheral" members of a nominal class are shifted into a new and substantially expanded remainder class, so that finally a system develops, which differentiates exclusively by the categories 'male - female - inanimate' -- a system, which resembles those not only of English (and other Indo-European languages), but also of many other languages of the world (cf. op. cit., pp. 166f).
This change can thus be also explained quite easily even without direct influence of English by taking into account the language shift stimulated by the loss of a cultural background.

3.6.2 Generalisation and Extension of Meaning

Traditional Dyirbal has lexical units for a multitude of special terms. These often collapse into only one YD item that unites all meanings in itself. If available, this is normally is the hyperonym or the central term of the semantic field, e.g. with the word for 'large'. The generic term in TD is yugi, and that is the word that finds use in YD. The hyponyms in TD were among others wayja, yuguy, magara and some more, defining 'large' regarding certain animals respectively (cf. op. cit., p. 185).
Something similar occurs also within areas not having hyperonyms, e.g. with the names for different kinds of eels. Here, the connotation of the word for the most common species, jaban, was expanded to refer to all other kinds as well. It is remarkable that with this process similar criteria apply as with the "mother-in-law language" Jaluy. In TD also generalisations and generic terms as well as synonyms for the respective generic terms were used (cf. op. cit., pp. 186ff.).

[top of page]

3.7 Vocabulary

3.7.1 General

Even in the dictionary of Dyirbal it can easily be determined that the traditional culture of the Dyirbalan, like that of all Aboriginals, has been under-estimated or even regarded inexistent for centuries, and that it has strongly been over-formed by the "winner culture" of the European colonists. Many white ones still the believe the prejudice that the native languages were primitive and would have no considerable vocabulary. DIXON reports of a broadcast interview, which he gave in 1963 about his field studies:

"In our interview, [the reporter] asked the sort of questions his listeners might have put. 'You really mean the Aborigines have a language? I thought it was just a few grunts and groans.' Then, 'But they don't have more than about two hundred words, surely?' I replied that I had collected over five hundred names for animals and plants from the Stewarts in a single morning, and that could only be a fraction of the lexicon."
(DIXON (1983), P. 63).

Even without the terms for the "blessings" of western culture, the dictionary of Dyirbal was probably already very sumptuous and multilayered -- just in other semantic areas than one is used to from European ways of thinking. The vocabulary and the grammatical features of a language are coined/shaped by the ways of life and the history of its speakers:

"Through being used by a particular group of people over generations for particular purposes, each of the world's languages has come to express the things important to that culture in a distinctive and efficient way by naming them individually."

In the course of time, the DyirbalNan (involuntarily, in most cases) gave up traditional ways of life giving way to the cultural assimilation and made English the main medium of interaction in many areas of life


3.7.2 Vocabulary Loss

This aspect is shown very clearly in the vocabulary being at the YD speakers' disposal. In a vocabulary test conducted by SCHMIDT with six of the twelve speakers, correct responses were obtained within the range of 34 to max. 72% (170 - 356 of 498 of the words asked for). This is a clear reduction in both the active and passive vocabulary (cf. SCHMIDT (1985a), p. 169f.). A comparison sample with speakers from the age group of the 5 to 15-year-old resulted in an even more startling picture:

"Although no Jambun children in the 5-15 year age group could construct a Dyirbal sentence, several of them could recall some Dyirbal words (I estimate seven/ten of 40 children)."
(op. cit., p. 173f.; emphasis by me, J.W.).

Altogether, the tested children achieved between 19 and 83 of the 498 words, thus between 4% and 16%. These "hits" however were situated almost exclusively in the core area of the high frequency words e.g. parts of the body, well-known animals, everyday activities (cf. op. cit., p. 174). Words from the smaller, closed word-classes were not known by any of the children any more.
It is to be assumed that in the passive vocabulary of the YD speakers and also the younger children there is still more knowledge of vocabulary, than could be elicited by direct questioning (cf. op. cit., pp. 174f.).

The lost vocabulary and the decreasing use of Dyirbal might have mutual negative impact. A person having difficulties to find (the correct) words in a language rather changes over to a system, of which he has better control. As particularly can be seen from the vocabulary knowledge of the youngest ones, the dictionary of YD does not diminish evenly, but different semantic areas seem to be more or less completely gone whereas some "lexical islands", as SCHMIDT calls them, remain (op. cit., p. 175). SCHMIDT briefly summarises the decrease according to groups of words,:

"[...] in a vocabulary loss situation, it is the noun items with real-world referents that are most resistant. In contrast, YD speakers have a much smaller number of verbs and adjectives, which describe actions, states and qualities." (op. cit., p. 176).

The "lexical islands" of YD are situated mostly in the area of the everyday life vocabulary, in particular with the words for humans ('man', 'woman', 'Aboriginal', 'child', 'ancestor, spirit'), some common or very well-known animals ('dog', 'Kangaroo', 'fish', 'Taipan'(7)) and for the human body parts(8) (cf. ibid.). On the other hand, all the words for inanimate entities as well as designations of the non-ordinary, and the overthrown culture (the vastly differentiated system of kinship terms, the names for ceremonies, artefacts and artistic techniques) disappear just as well as the formerly very broad inventory of names for animals, plants, weather conditions and landscape forms (cf. op. cit., p. 177).

3.7.3 Restructuring of the Lexicon

A further factor regarding the scope of the YD vocabulary is the intrusion of pidgin-type forms. Not only with the pronouns (cf. chapter, tab. 3) but also with the particles such means are increasingly being used, such forms as nomo (< no more), gen (< again), bin (< been; as lexical marker for past), orait (< all right) intrude into YD (cf. op. cit., pp. 180f.). Apart from the quantitative modification of the vocabulary's scope also structural modifications can be observed. Some cases affixes of TD no longer are treated as such but as fixed constituents of the word root. Some speakers of YD fuse for example the iterative suffix {-gani} with certain verb roots, so that e.g. TD nyina-y 'sit' becomes nyina+gani+nyu in YD. All YD speakers fuse the false reflexive affix {-marri} with the root guni-y 'look up' to guni+marri+nyu. The latter, however, seems to have originated in TD already (cf. op. cit., p. 179).

3.7.4 Lexical Adjustment

The western culture, its concepts and goods, tremendously penetrated into the everyday life of the Aboriginals. Dyirbal did not have words for many of these new concepts. These new "lexical gaps" were closed in different ways. On the one hand, already in TD the (metaphorical or metonymic) connotations of certain words were expanded, for example in transferring the meaning of bulmban 'scattered grass for the sleeping place' onto 'bed of European style' or that of bugu 'knee, wave' onto 'wheel of a motor vehicle' (cf. op. cit., p. 181). On the other hand English words were taken over (under phonological adjustment), e.g. juwa (< store), bujigan (< pussycat), dagida (< doctor), modaga (< motorcar) or biba (< paper). Most of the words specified by SCHMIDT were already taken over into TD (cf. op. cit., pp. 182 and 190). The by far most frequent case however is the bare substitution: if in TD no word for a certain item or circumstance exists, or if the appropriate Dyirbal word is forgotten by the YD speaker, an English word is used instead and easily integrated into the YD phrase, as the shown by the following example:

"anaji happen to buran bayi helicopter
  we-PL            see   he                         
We happened to see a helicopter."
(op. cit., p. 192; Emphases by the author).

Here, it becomes exceptionally obvious that a displacement of Dyirbal takes place. The YD speakers increasingly rely on English means, or even switch to English completely.

[top of page]

3.8 Pragmatics

Annette SCHMIDT acquired most of her results by direct questionings and tests. In order to avoid the observer paradox and to investigate the actual YD language use she also performed informal observations (cf. op. cit., pp. 127-130).
"[...] the very presence of a stranger will influence the speech of the group under observation. In the case of the Jambun study, physiological difference in skin colour was a constant reminder of the presence of an outsider. At first, this presented a real problem, as my peers would constantly switch to English in my presence." (op. cit., p. 129).

From above quotation one can see that the use of YD is restricted to certain situations. If foreigners, particularly white ones are present, the YD speakers switch to English (SAE or JE, depending upon skills). Dyirbal (YD) is only then used in presence of white ones, if one wants to disassociate oneself from these, or if a very good bond of trust exists to these (cf. op. cit., pp. 31f. and 130). There were two groups of girls, who separated themselves from each other and the other young people among the YD speakers whom SCHMIDT interviewed. Linguistically they did this by talking their (slightly deviant) variety of YD within the group, and English (JE) with members of the other group (cf. op. cit., pp. 128f.). The language thus also serves as defining factor even within the community. I would not like to deal en détail with the subtle differences between these varieties in the context of this paper (cf. op. cit., chapter 5 for this).

Generally, however, YD was also less consistent outside of the questioning situation and morphologically significantly less complex than the style applied by the speakers in the tests (cf. op. cit., pp. 145f.). Dyirbal is no longer the primary language in Jambun, since Jambun English is used as primary medium of communication. Occasionally Dyirbal is used to address members of the own peer group or some older persons. The latter use, however, diminishes because the TD speakers often criticise and correct the YD as inappropriate, which leads to the situation that some YD speakers resign and rather use English in all contexts (cf. op. cit., p. 38).

[top of page]

3.9 Summarising

We have seen that YD on all levels differs from TD. As a rule, the structures in phonology, morphology, syntax, semantics and dictionary were simplified, exceptions were eliminated and complicated constructions done away with.
A simplification of complex systems also is quite a normal process within "healthy" languages. Tendencies towards more language economy basically are factors that promote the change and evolution of each human language. Nevertheless, the substantial difference to the processes observed in Dyirbal is that the reduction of complexity in certain parts of the system is not being compensated for in other positions. The language consequently went out of its equilibrium and loses more than it wins again by the changes occurring. The developing gaps get filled mostly by English than by means from Dyirbal itself. Here, a vicious downward spiral develops: unsatisfactory language competence (from bad language acquisition) leads to weaknesses and uncertainties in the performance, which finally leads to frustration and resignation and to the avoidance of communication situations in the language concerned. Lack of practise resulting from this has negative impact on the competence again, since more complex constructions and infrequent means are being forgotten. This again leads to errors and uncertainties, which finally result in more frustration and avoidance, so that the language is being used in fewer contexts, until it is completely being dropped in favour of another system (here: English).

In the following chapter, I would like to try to explain how far this process advanced, and how the situation of the Dyirbal itself and its (former?) speaker appears today, some nineteen years after SCHMIDT's inquiry and thirty-eight years after the beginning of DIXONs research.

[top of page]

4. Dyirbal and the Situation Today

In addition to the numbers of speakers specified in section 2.2, there are other sources also yielding information on the present status of Dyirbal and the situation of the Aboriginals in the region of SCHMIDT's investigation. On the home page of the Tully State High School (situated in the municipality of Tully, scarcely 20 kilometres from Murray Upper) one can e.g. find the following statement:

"The community [municipality Tully; J.W.] is predominantly rural (based on sugarcane, bananas and cattle) with associated services and tourism as the other major areas. Unemployment is low. The geographic location (two hours drive to Cairns or Townsville) enforces a marked cultural isolation. The community consists of a diversity of socio-economic, ethnic and cultural backgrounds, including a significant Aboriginal and Torres Straight [sic!] Islander population, some living in communities. Some Aboriginal students have English as a second language."

Only for few of the Aboriginal children English is not the primary language of their socialisation, but second language, and that is apparently already worth mentioning. From this it can be concluded that the predominant part has English as first language -- the young people already being linguistically assimilated completely.

The fate of Dyirbal is not an exception but rather one example for the decline of most Aboriginal languages of Australia, which are all together in a threatened position, as a parliamentary commission writes in its 1992 report:

"Only about one tenth of the original languages survive today in a relatively healthy state. About a third of the original languages continue to be spoken but are under considerable threat, often being spoken by only a handful of elderly speakers. Much of the language loss that has occurred is irretrievable. However, language maintenance activities can do much to maintain strong language [sic!] and assist weakened languages. As it is not possible to revive dead languages it is necessary to assist languages before they reach a severely threatened state. The committee acknowledges that many severely weakened languages do not have good prospects for survival as a comprehensive living language."
(Standing Committee).

This is an awareness that definitely comes too late for the predominant part of the languages of Australia. They either vanished already or are facing their immediate extinction, since they have been neglected and suppressed to for a much too long time.

There are projects for the stabilisation of native languages since a few years ago, but nevertheless English remains the sole official language of Australia(9), also being used exclusively in school teaching:

"The present curriculum is oriented only toward white society and its values. In other words, rather than enhancing or enriching cultural identity, education replaces Dyirbal with English and an impression that Dyirbal is unimportant."
(SCHMIDT (1985b), p. 381).

Compared to that, the status of the Maori language in the neighbouring State of New Zealand is remarkably good:

"Unlike Australia's Aborigines, New Zealand's 100,000 indigenes all shared a common language [...]. Thus Maori have a single resource under which they can unite. Maori is still spoken by at least 50,000 people. There are over 400 Maori schools, and the language is widely written and used in the law. Indeed, Maori took its place alongside English as an official language of New Zealand in 1987."

The Aboriginals and their languages are still very far from such a situation. There has been a reorientation in the policy and in the society of Australia, yet in Australia lies far behind the achievements made in New Zealand. Certainly it is more difficult to maintain some (formerly) several hundred languages than only one, but without the state's suitable basic attitude towards its minorities even a relatively strong language hardly has a chance of survival in the long run.

In the course of the political reorientation at the beginning of the 1990es, the commission mentioned above has been formed to compile suggestions how to tackle the problem of cultural and linguistic assimilation and the shifting to the English as the main system of communication, like SCHMIDT described it. Apart from the establishment of court and civil authority interpreters, a special attention of the recommended measures is drawn to the area of training both Aboriginal and white teachers for native languages and the reorganisation of curricula (cf. Standing Committee).

Instruction in native languages is however only to be offered, where trained teachers are available, where the official demand (by the local resident population) exists, and where the language concerned still has a sufficient number of speakers (cf. ibid.), this is mostly not the case so far:

"In recent years there has been growing pressure for all students to learn a language other than English (LOTE) before leaving school. These changes emphasise the neglect of ATSI [= Aboriginal and Torres Strait Islander; d.A.] languages within schools and in the wider community. During the course of the inquiry no ATSI language had a curriculum to year 12 level. In fact it was extremely rare to find an ATSI language offered at high school. The committee finds it intolerable that while most migrant children with a first language other than English have been able to study that language up to matriculation level, ATSI children cannot study their language at high school." (ibid.)

Likewise, at the Tully State High School there has been instruction about the Aboriginal languages for some years, but not in them:

"Year 8 English students are introduced to the culture and language of the local Jirrbal and Girramay people. This unit of work develops understanding of local Aboriginal people and English language through comparing the different language structures." (TSHS).

Moreover, according to the heading "student support" of the quoted paragraph (cf. ibid.), this appears to be a simple type of additional supportive lessons or a working group in the context of English instruction, which is offered in only one class level.

In Murray Upper - which would have been best suitable to keep Dyirbal alive due to the size of the speaker community -- the native language instruction has been quite ill-fated: before the replacement of the head master in 1976, Aboriginal children faced spanking, if they were overheard using anything else than English on the school grounds (cf. DIXON (1983), p. 67). The new head master, who only stayed for four years, had lively interest in the Aboriginal languages and e.g. let his pupils create small visual dictionaries on the basis of own collections on traditional ways of life (cf. ibid.). His successor officially also had plans to introduce native Dyirbal instruction and has been equipped for this by DIXON with material (among other things ready-made language lessons and small vocabulary lists). However, he left those unused (cf. DIXON (1991), p. 193). DIXON reports the development during the 1980es as follows:

"Things could have been so much better. [...] Marcia Jerry [...] is a trained schoolteacher who also speaks Dyirbal pretty well. She was at the Murray Upper school in the mid-1980s and did some valuable teaching of Dyirbal, to both Aboriginal and white children. But Marcia did not like the school principal and left [...]. The whole Aboriginal community dislikes the principal and asked me to try to get him moved. [...] But the principal is still there, alienating another generation of Aboriginal children and effectively standing in the way of any teaching of Dyirbal in the school."
(op. cit., p. 199).

Thus, due to personal resentments of a few people, the chance to strengthen Dyirbal in the following generations had been missed. This added to the powerful process of replacing Dyirbal by English even at one of the few outposts at which it sensibly could have been opposed.

Native instruction would be an important factor for the adolescent Aboriginals to identify themselves with their culture and language and find out that their origins are not so primitive and unworthy as it had been publicised for decades in schools and media.

Beyond that, in the remaining areas of life an environment must be created, which promotes the use of the native languages, instead of obstructing it, and thus effectively fights the stigma of these languages. For some of the Aboriginal languages of Australia this seems to be working quite well already: There are regional radio stations, newspapers, school books and whole collections of internet documents for languages such as Western Desert or Pitjantjatjara, covering culture, language, community life etc...

Furthermore, there are some projects which offer a broad spectrum of information and documentation on and for the Aboriginals, e.g. the Koori Network(10) (http://www.koori.net/) and the Koori Centre of the University of Sydney (http://www.koori.usyd.edu.au/).

For Dyirbal, however, every assistance and initiative like a creation of Dyirbal media might already come too late. There is no speaker community any longer, which still sees sense in using Dyirbal in the everyday life and passing it on to following generations. Like DIXON already reports for the situation back in 1968:

"My first teacher of Dyirbal, Chloe Grant, was fluent in her own language and in English. But Chloe and her husband spoke only in English to their children, since they felt it would give them a better chance in the world. This undoubtedly helped the children, but the side effect was that they cannot speak in their parents' language (and regret that they can't)."
(DIXON (1997), p. 109).

Extrapolating the data from SCHMIDT's investigation, the speakers of YD today should be between 34 and 54 years old, the last generation of TD speakers accordingly older. In the meantime, at least one if not two generations of Aboriginal children grew up who did not have an opportunity to acquire Dyirbal as native language.
Now it is merely a question of time when the last speakers or semi-speakers of Dyirbal completely give up, to speak the language of their ancestors and switch to all-powerful English instead. Finally, when no member of the generation of the old TD speakers is alive any more, the last reason for the active use of the Dyirbal will have faded with them.

[top of page]

5. Mechanisms of Language Death

Languages at all times influenced and displaced each other. It is nothing unnatural that they develop and perish, and nothing that has occured for the first time in the recent past. Yet it is frightening at which rate languages stopped to exist during the last decades. The observation that the number of the languages on earth continues to decrease tremendously, has induced many linguists to develop models to explain this phenomenon. I would like to briefly present some of these models here, if they can be referred to the descend of Dyirbal.

The present dying of the Australian languages is for example placed into a systematic context by Daniel NETTLE: The absolute number of languages in Australia remained constant over thousands if not tens of thousands of years. Nevertheless languages were exchanged by language death and the developing of new idioms. Since the speaker communities of these languages always were very small (on average approx. 2000 speakers/language), and the entirety of the Aboriginals (approx. 750,000 back in 1788) share more an above average number of languages (around 1788 at least 266), language change has always been possible substantially faster, even without the effect of white settlers. However, the system of the Aboriginal languages and their absolute figure was at equilibrium for a long time (cf. NETTLE 1999, pp. 93 and 98).
It is new that now not different Aboriginal languages replace each other any longer, but that solely English takes their place, so that the old equilibrium becomes punctuated and the total of the different languages is being reduced, since the original languages are being deprived of their foundations:

"The European expansion had a devastating effect on the indigenous languages of the settled areas. Expanding European populations [...] had little interest in sharing the lands they were appropriating with anyone, and either murdered indigenous peoples, forcibly assimilated them, or drove them into marginal habitats they could not use. Their replacement of the indigenous peoples was aided by [...] infectious diseases. [...] Where indigenous populations did not die out, they were forcibly disrupted, and both disruption and reduction of numbers prevented them from continuing as viable cultural units."
(cf. op. cit., pp. 107f.).

An additional factor leading to the disappearance of these languages is the voluntary language shift of the speakers to other languages, which in their eyes possess a higher prestige.
According to NETTLE, there is a classical sample for this, which already occurred everywhere in the world at different points in time of history:

"It begins with a monolingual or nearly monolingual community. An ambitious younger generation that has contact with, or at least aspiration to, the more developed economy becomes bilingual in the language of that economy and the vernacular. [...] the third generation, then becomes monolingual in the dominant language. For them, not only has the vernacular lost its prestige in relation to the larger language, but much of the functional need for it had been attenuated by the fact that the generation above them is all bilingual. [...] the dominant language becomes primary and the vernacular is forgotten." (op. cit., p. 110).

What SCHMIDT observed for YD, thus seems no to be such an exotic individual case but rather usual when two cultures with different degrees of resources and prosperity come into contact. Material obligations due to changing life circumstances play a further important role in this transition. In regions which until recently climatically or geographically appeared too unfavourable to the European colonists, the indigenenous languages could survive much longer (cf. op. cit., p. 107).

David CRYSTAL similarly explains the fast replacement of the traditional language in the colonialisation and displacement situation as one of the principal reasons for dying languages:

"In cases where a community has been displaced, many of the survivors, unwilling or unable to remain in their habitat, find their way to population centres where they slowly lose their cultural identity within a milieu of poverty. To survive, they acquire as much as they can of a new language [...]. The ethnic language tends not to outlast a generation - if the members of that generation survive at all."
(CRYSTAL, p. 75).

As it can clearly be seen here, the shifting from one language to another is not at all that voluntary as the term "voluntary language shift” used by DIXON (1997) and NETTLE (cf. above) suggests.

The fact that cultural assimilation and linguistic displacement are not the only reasons for language death should be self-evident. Epidemic diseases, famines, military conflicts and natural disasters can particularly harm languages with extremely small speaker communities by driving the speakers out of the region or by killing them (cf. CRYSTAL, pp. 72-74).
These factors are, however, not relevant in the case of Dyirbal, therefore I shall not go into further detail about them here.

There always are two reasons why a language ceases to exist: either the language loses its speakers (by the events already mentioned according to CRYSTAL), or the speakers lose their language - by force or voluntary decision (cf. DIXON pp.107-110). The (forced) cultural assimilation and the strong linguistic pressure of English as the exclusive official language have been decisive for the change and fall of Dyirbal. When the language of the "superior community”(11) enjoys a much higher prestige it is assumed the key to success and prosperity, so that parents decide to arrange for their children the more "valuable" language.
The fact that the loss of the language and culture of the ancestors often is not even perceived as such demonstrates how deeply the thought of the white man' superiority and the lack of civilisation of the natives has been indoctrinated into the heads of the Aboriginals.

[top of page]

6. Summary

The case of Dyirbal and its development, as it was described by Annette SCHMIDT, exemplifies the extinction a language in the course of a gradual exchange of a language by another. The colonial language contact situation opened descending vicious spiral of minor language prestige, which leads to reduced usage of the language, which contributes to gradual forgetting, expiry and thus reduction of complicated forms, which causes "errors" in terms of the traditional grammar. This again has negative effects on the prestige of the language again, etc. pp.
The formal reduction occurs on all levels of linguistic description, ranging from the phonology by morphologic and syntactic reduction up to semantic and lexical phenomena, all of which can be regarded as indicators of the decay of Dyirbal in its traditional form.

There have hardly been any attempts to save Dyirbal and reinforce its use again. The few initiatives in native language school teaching failed due to personal resentments of some persons in key positions.
Many languages of the world have a fate which is quite similar to that of Dyirbal: cultural infiltration of all areas of life and displacement of the indigenenous languages are processes that have led and will lead to further extinction of languages at many points of our planet.

I would like to add a remark concerning the term language death. Actually one would many cases have to talk in more correctly about speaker death. For actually it occurs that there simply is no-one left who has the complete competence of a native speaker. What can be found thereafter, are more or less extensive and authentic fragments with semi-speakers who have a reduced to insufficient competence.
The Dyirbal language, of which we say here that is dying has not completely perished yet. It still continues to exist for some time as a mixture language and in few fragments, being more and more regarded as useless and not being passed on to a following generation of speakers any more. The point of time can thus be foreseen in the near future at which the last reflexes of the language will also have been lost.

But when in this long period of time did the language death occur now?

As with the discussion concerning the human death, one can come to many different statements, depending upon the applied criteria. And similar to the belief of many religions that there is something immortal in every human, one can also say of a language that it did not completely vanish, if it has been thoroughly analysed, described, and documented. And thus did not disappear without any trace.

Unfortunately, the present situation of many languages dying seems not to be stoppable or reversible. Most of the over 6000 languages spoken on this planet will not survive the millennium that just started. NETTLE and ROMAINE even assume that more than half of the languages known today will be extinct by the end of the 21st century (NETTLE/ROMAINE, p. 7).

If this process is already inexorable, then it should be one of the most urgent duties of linguistics to document the languages of the world in order to preserve them for future generations. The hypotheses, assumptions, and theses that are made in theoretical linguistics about language and languages are generally based on the comparison of the existing languages. Many special features of human language, which do not correspond to universal belief, are found in the "small" languages, which often still are insufficiently documented. It would be an inconceivable loss, if the many still unknown secrets and the not yet documented knowledge of so many human languages remained undiscovered and were to disappear with them:

"There are 2,000 or 3,000 languages, for which we have no decent description, that will pass into disuse within the next few generations. Trained linguists are urgently needed to document them."
(DIXON (1997), p. 138).

There is a high demand for thorough documentation and description of the (still seizable) languages of the world, based on a solid theoretical and practical linguistic education

"[..] if every linguistics student (and faculty member) in the world today worked on just one language that is in need of study, the prospects for full documentation of endangered languages (before they fade away) would be rosy. "
(op. cit., p. 137).

From his perspective as field researcher and theoretician, DIXON condemns linguistics, which set up (usually short-lived) theories for the theories' sake and which do not care to actually work with the human languages and to verify the theories by practice and vice versa (cf. op. cit., pp. 130-134).(12)

As the fast decline of Dyirbal and its neighbouring languages reveals, it is inexcusable to neglect the "vanishing voices" (NETTLE/ROMAINE, title) of the world and not to describe them, thus preserving at least some of them reasonably, if their entirety can not to be protected from extinction any more.

"If this work is not done soon it can never be done. Future generations will then look back at the people who call themselves 'linguists' at the close of the twentieth and beginning of the twenty-first century, with bewilderment and disdain."
(DIXON (1997), p. 138).

Warendorf, Spring of 2001,
Jan Wohlgemuth

[top of page]

7. Appendix

7.1 Bibliography

[top of page]

7.2 Abbreviations

A Subject of a transitive sentence (Agens)
ATSI Aboriginal and Torres Strait Islander
C Consonant
DEM Demonstrative (pronoun / article)
ERG Ergative
FUT Future tense marker
GEN1 Genitive 1 (actual)
GEN2 Genitive 2 (general)
JE Jambun English
NOFUT Non-future tense marker
NOM Nominative
NP Nominal phrase
O Object
PREP (English) preposition
S Subject of the intransitive sentence
SAE Australian Standard English
TD Traditional Dyirbal
V Verb
V Vowel
YD Young Dyirbal

[top of page]

7.3 Notes

1) In the different articles the problem is found, to reflect the lamino-alveolo-palatal plosive [] <dy, dj, j> and the semi-retroflex liquid [] <r, , rr> in a definite and yet practical and simple spelling. [back to the text passage]

2) To the Aboriginals themselves, such a summary is completely inappropriate. For them, the approx. 20% differences between the individual dialects are far more important than the things in common, since the different tribes identify and almost constitute themselves by means of the differences of their respective "languages" (cf. DIXON (1991), p. 184). [back to the text passage]

3) The phoneme inventory consists of exactly 17 distinctive units: 12 consonants (/b/, /d/, /dj/, /g/, /m/, /n/, //, //, /l/, /r/, //), 2 semi-vowels (/w/, /j/) and 3 vowels (/i/, /o/, /u/) (cf. DIXON (1972), pp. 37f). [back to the text passage]

4) In the bibliography of the Ethnologue (cf. GRIMES (1996), p. 933) however, there is not one entry under DIXONs name between "Dimmendaal" and "Doerfer", so that the origin of this figure remains obscure. According to the Bibliographie Linguistique of 1983, there have been altogether three publications of DIXON in the year of reference. Besides "Where have all the Adjectives gone?” (originally published in 1982) these are "Searching for Aboriginal languages - memoirs of a field worker” and the third volume of the "Handbook of Australian languages", ed. by R.M.W. DIXON and Barry J. Blake. In that third volume, however, no information about Dyirbal can be found. Probably, the specification refers thus to DIXONs memoirs "Searching for Aboriginal languages”, in which there are some vague statements is in styles of "several score speakers" relating to the number of Dyirbal speakers in the 1960's and 70's. [back to the text passage]

5) To obtain a screen display as problem-free as possible and to increase the legibility of the HTML text, I follow the Dyirbal orthography used by SCHMIDT in her works (1985a and 1985b), since she gets along without special characters. The characters used by DIXON (1972) are being replaced as follows:
<dj> by <j>, <> by <ny>, <> by <r>, and <r> by <rr>
To replace by a Greek Eta would cause problems with HTML display as well, therefore I chose to use a .gif file instead.
Besides, this alteration alleviates the comparison between the two sources. For the HTML version, I sometimes have adapted the orthography of the respective author also in the literal quotations based on abovementioned convention. [back to the text passage]

6) Around 1940, the last free group of Dyirbalan, which had lived undisturbed in the almost inaccessible rain forest mountains along the Tully River up to then, moved to Murray Upper. Only one woman from this group was still monolingual in 1963 - the only monolingual speaker DIXON ever encountered for Dyirbal. (cf. DIXON (1991), pp. 186f.). [back to the text passage]

7) An extremely poisonous and therefore well-known species of snake. [back to the text passage]

8) One should bear in mind that A. SCHMIDT as a woman and as a non-Aboriginal could not investigate into this area of the vocabulary completely. It can be deduced from her informal observations however that the vocabulary of this area is still quite used (also in JE), in order to conceal offensive topics from outsiders (cf. SCHMIDT (1985a), pp. 169f. and 209f.). Apparently thus rather social taboos and personal shyness than lacking knowledge were the reasons for not responding to the questions about the names for certain body parts and - functions (cf. op. cit., pp. 129 and 169f.). [back to the text passage]

9) As far as I could find out from relevant legal text collections, there is not even a regulation or a definition which language is official and national language, let alone that English should be the only state language of the Commonwealth of Australia. Here, British customary law and colonialist arrogance clearly show. [back to the text passage]

10) Koori is a self-designation of the Aboriginals from Southeast Australia that nowadays is being used in all Australia (also in SAE) as a politically correct synonym for Aboriginal. [back to the text passage]

11) This superiority actually is very doubtful -- above all because the Europeans have awarded it to themselves. The white ones have still to prove that they are able to survive on that continent in agreement with the environment for tens of thousands of years!! [back to the text passage]

12) This is an explicit reproach particularly to the schools of GTG and dependency grammar (cf. DIXON (1997), pp. 130f., footnote 9). [back to the text passage]

13) These texts are internet documents, which are not divided into pages or unique paragraphs, so that the source specification unfortunately is not possible more precisely. [back to the text passage]

[start of page] | table of contents | [www.linguist.de]