Aboriginal Child Language Acquisition Project (ACLA1)

Typology

Gurindji is a suffixing Pama-Nyungan language spoken in the north-west of Australia, particularly in Kalkaringi and Dagaragu. It is a member of the Ngumbin subgroup of languages which includes Ngarinyman, Bilinara, Malngin, Nyininy, Mudburra, Jaru and Warlmatjarri. Gurindji is an endangered language, with only 60 speakers remaining in 2003. Gurindji Kriol is the language transmitted to the new generation at present.

Phonologically, Gurindji is a fairly typical Pama-Nyungan language. It contains stops and nasals which have five corresponding places of articulation (bilabial, apico-alveolar, retroflex, palatal and velar), three laterals (apico-alveolar, retroflex, palatal), two rhotics (trill/flap and retroflex continuant), two semivowels (bilabial and palatal) and three vowels (a, i, u). Combinations of semivowels and vowels produce diphthong-like sounds. Like most Pama-Nyungan languages, Gurindji is notable because it contains no fricatives or a voicing contrast between stops. Stress is word initial, and syllables pattern CV, CVC or CVCC.

Gurindji is a dependent marking language. Word order is relatively free, though constrained by discourse functions. The verb phrase is made up of a free coverb and an inflecting verb which contains information about tense, mood, modality. Bound pronouns also attach to the inflecting verb to cross reference subjects and objects for person and number. These pronouns inflect for nominative and accusative case, unlike free pronouns whose form only changes for dative case.

The noun phrase may contain nouns, adjectives, demonstratives and free pronouns. Case marking for nouns is ergatively patterned, and generally other elements in the noun phrase must agree with noun's case.

References

• Lee, Jason and Dickson, Greg. State of indigenous languages of the Katherine region. Katherine: Diwurruwurru-jaru Aboriginal Corporation, 2002
• McConvell, P. "Changing places: European and Aboriginal styles," in Hercus, L., Hodges, F. and Simpson, J. (eds.,). The land is a map: Placenames of indigenous origin in Australia. Canberra: Pacific Linguistics and Pandanus Press, 2002, pp. 50-61
• McConvell, P. "Linguistic stratigraphy and native title: the case of ethnonyms," in Henderson, J. and Nash, D. (eds.,). Language in native title [Native Title Research Series]. Canberra: Aboriginal Studies Press, 2002, pp. 259-290
• McConvell, P. "Mix-im-up speech and emergent mixed languages in Indigenous Australia," in Texas Linguistic Forum [Proceeedings from the Ninth Annual Symposium about Language and Society - Austin April 20-22, 2001] 44, 2001, pp. 328-349
• McConvell, P. "Gurindji," in Dixon, R. M. W. and Blake, B. J. (eds.,). The handbook of Australian Languages Vol. 5. Oxford: Oxford University Press, 2000
• McConvell, P. "Ergativity and the scope of 'again' in English and Gurindji." Copenhagen, Functional Grammar Conference, 1990
• McConvell, P. "Mix-im-up: Aboriginal code-switching, old and new," in Heller, M. (ed.,). Codeswitching: anthropological and sociolinguistic perspectives. Berlin: Mouton de Gruyter, 1988, pp. 97-150
• McConvell, P. "Nasal Cluster Dissimilation and constraints on phonological variables in Gurindji and related languages," in Evans, N. and Johnson, S. (eds.,). Aboriginal Linguistics 1. Armidale: Department of Linguistics, University of New England, 1988, pp. 135-187
• McConvell, P. "The Origins of Subsections in Northern Australia," in Oceania 56, 1985, pp. 1-33
• McConvell, P. "Domains and codeswitching among bilingual Aborigines," in Clyne, Michael (ed.,). Australia, meeting place of languages. Canberra: Pacific Linguistics. Series C - No. 92, 1985, pp. 95-125
• McConvell, P. "Hierarchical variation in pronominal clitic attachment in the Eastern Ngumbin languages," in Rigsby, B. and Sutton, P. (eds.,). Papers in Australian Linguistics No.13: Contributions to Australian Linguistics. Canberra: Pacific Linguistics, 1980, pp. 31-117

Language change and formation

(excerpt from McConvell, Patrick and Meakins, Felicity. "Gurindji Kriol: A Mixed Language emerges from Code-switching," in Australian Journal of Linguistics 25(1), 2005, pp. 9-30)

Gurindji Kriol is the main language of Kalkaringi and Dagaragu, twin communities situated 460km south west of Katherine in the north of Australia. It arose from contact between white pastoralists who spoke English, and the Gurindji, the traditional owners of the country the pastoralists colonised. After the initial conflict period in the late nineteenth and early twentieth centuries, many Gurindji people worked on the cattle stations as kitchen hands and stockman. The lingua franca between the two groups was an English-creole, Kriol. The Gurindji, who already spoke a number of the related neighbouring languages added Kriol to this repertoire and their code-switching practices. Nowadays all Gurindji people speak Gurindji Kriol, older people also speak Gurindji and younger speakers have a reasonable passive knowledge of Gurindji. Gurindji is an endangered language, with only 60 speakers remaining in 2003. Gurindji Kriol is the language transmitted to the new generation at present.

Some socio-historical evidence might be relevant as to why full language shift did not take place, as it has done in other areas in northern Australia. In 1966 the Gurindji went on strike from the cattle stations where they had worked and the long-standing dispute over wages and conditions revealed itself as a struggle for land rights. 1975 saw the hand back of traditional lands to the Gurindji by the then Australian Prime Minister, Gough Whitlam, a highly significant step for post-colonial law and history and for the Gurindji themselves. It is possible that the pride associated with these momentous events and the resultant desire to mark Gurindji identity linguistically may have affected the course of language shift and motivated the maintenance of a mixed language.

Typology

(excerpt from McConvell, Patrick and Meakins, Felicity. "Gurindji Kriol: A Mixed Language emerges from Code-switching," in Australian Journal of Linguistics 25(1), 2005, pp. 9-30)

Typologically, Gurindji Kriol exhibits a split between the verbal and nominal systems as do other mixed languages like Michif. In Gurindji Kriol, basic verbs such as go and sit, the tense-aspect-mood system and transitive morphology are derived from Kriol, whereas emphatic pronouns, possessive pronouns, case markers and nominal derivational morphology have been transplanted from Gurindji relatively intact, but with some innovations . Demonstratives, nouns, verbs and adpositions are adopted from both languages, however some generalisations can be made about their distribution. A short excerpt of a GK story which demonstrates some of these features is below (1). Gurindji elements are in italics:

nyawa-ma wan karu bin plei-bat pak-ta nyanuny warlaku-yawung-ma.
this-TOP one child PST play-CONT park-LOC 3sg.DAT dog-HAVING-TOP
'This one kid was playing at the park with his dog.'

tubala bin pleibat. i bin tokin la im
2pl PST PST play. 3sg PST talk PREP 3sg
'The two of them were playing and the kid said to him:'

"kamon warlaku partaj ngayiny leg-ta ..."
come.on dog go.up 1sg.DAT leg-LOC
'Come on dog jump up on my leg …'

"ngali pleibat nyawa-ngka."
1plu.inc play this-LOC.
"We'll play here."

References

• McConvell, P. and Meakins, F. "Gurindji Kriol: A Mixed Language emerges from Code-switching," in Australian Journal of Linguistics 25.1, 2005, pp. 9-30
• Meakins, F. and O'Shannessy, C. "Possessing variation: Age and inalienability related variables in the possessive constructions of two Australian mixed languages," in Monash Papers in Linguistics 4.2, 2005, pp. 43-63
• Charola, Erika. The Verb Phrase Structure of Gurindji Kriol. Unpublished honours thesis. The University of Melbourne, 2002
• McConvell, P. "Mix-im-up speech and emergent mixed languages in Indigenous Australia," in Texas Linguistic Forum [Proceeedings from the Ninth Annual Symposium about Language and Society - Austin April 20-22, 2001] 44, 2001, pp. 328-349
• Dalton, L., Edwards, S., Farquharson, R., Oscar, S. and McConvell, P. "Gurindji children's language and language maintenance," in International Journal of the Sociology of Language 113, 1998, pp. 83-96
• McConvell, Patrick. "Discourse frame analysis of code-switching," in Gorter, D. and Piebenga, A. (eds.,). Code-switching: papers from the Leeuwarden Summer School 1994. Network on Language Contact and Codeswitching, 1994
• McConvell, P. "Mix-im-up: Aboriginal code-switching, old and new," in Heller. M. Codeswitching: anthropological and sociolinguistic perspectives. Berlin: Mouton de Gruyter, 1988, pp. 97-150
• McConvell, P. "Domains and codeswitching among bilingual Aborigines," in Clyne, Michael. Australia, meeting place of languages. Canberra: Pacific Linguistics. Series C - No. 92, 1985, pp. 95-125

Language situation

The varieties of English spoken in Tennant Creek can best be characterised on a continuum, ranging from heavily creolised English (CE) at one end, to light Aboriginal English (AE), close to standard Australian English (SAE) at the other. The variety a speaker uses depends on the situation, interlocutors present and the varieties that a speaker has in their repertoire. Speakers generally use very light Aboriginal English in interactions with Non-Indigenous people. Standard Australian English (SAE) is the language of education and institutions. Through media such as television, videos and popular music, people hear SAE and other varieties of English.

Some speakers use Aboriginal English most of the time, but include some features of creolised English in interactions with speakers whose style is generally heavier.   In some speech networks, heavily creolised English is the code of in group comunication. Given that these varieties exist in the speech community of Aboriginal people in Tennant Creek, there is great variablity. A single stretch of discourse may include features from lighter and heavier ends of the continuum. These sub-systems do not appear to operate independently in speaker useage.

Speakers also use items from Warumungu, such as lexicon, some verbs and nominal morphology. Older people tend to be full speakers of Warumungu, while most people under 40 years of age are partial speakers and so their use of Warumungu is always within otherwise CE or AE discourse.

Typology

Sound system

The phonology of heavily creolised English is influenced by Indigenous languages. Affricates and fricatives, which appear in English only, are generally replaced by corresponding stops. Consonant clusters are separated with a weak vowel. Dipthongs alternate with short clear vowels. The phonology of Aboriginal English is closer to English, though features mentioned in heavily creolised English appear.

Verb phrase

Word order in Aboriginal English and creolised English is SVO, though the object may be moved to clause initial position, for topicalisation.

An object pronoun referent may also be topicalised with a postposed referential noun phrase.

Transitivity is marked on verbs with the suffix -im.

1) Jangay dei gatim.

Shanghay(s) they have.

2) Dei bin gitim im, nunuwan babiwan.

They got it, (the) little baby (bird)

If reference is made to an event that occurred prior to the time of speaking verbal auxillary 'bin' can be used with a verbal element . For future or potential state, garra, gata tends to be used with verbal element.

3) An nyili garra pokim im na.

And ( the) prickle is going to poke him (he going to step on the prickle)

Progressive aspect is indicated by the suffix - in, ing.

4) Oni wanbala running na.

Only one (person) is running now

Durative and iterative aspect is indicated by the suffix -bat, or -abat, which Sandefur (1991), discussing Kriol spoken to the north of the Tennant Creek region, claims has an overlap in meaning with the progresssive, and an interweaving of distribution and co-occurrence. The iterative meaning of -bat is more common than the durative meaning and can either refer to repetition of an action or plurality of participants. The iterative aspect is associated with creolised English, as it tends to occur more frequently in this speech style.

5) Tubla bin jasimbat dat julaka.

They (2) were chasing that bird

6) "Maami, dei bin bildimabat mi".

"Mummy, they were beating me"

Verbless clauses are used to describe states perceived as existing at the time of speaking.

7) Dis imkay karnanti.

This (is) his mother

8) Dei brabli cruel.

They are very cruel.

9) Weya im na? weya im kina?

Where (is) he now? where (is) he (going)

Noun phrase Nouns need not be marked for number or definiteness in AE and CE, (see examples 1,2,3, 5) but determiners can be used to express these.

(10) Tribala kartti bin jasimbat.

Adjectives generally appear with the nominalising suffix -wan (example 2), and in the case of numerals -bala (example 10)

References about Kriol in general

• Munro, J. Substrate language influence in Kriol: The application of transfer constraints to language contact in northern Australia. Unpublished PhD, University of New England, Armidale, 2005
• Harris, J. "Kriol - The creation of a new language," in Romaine, S. (ed.,). Language in Australia. Cambridge: Cambridge University Press, 2004
• Mühlhaüsler, P. "Overview of the pidgin and creole languages of Australia," in Romaine, S. (ed.,). Language in Australia. Cambridge: Cambridge University Press, 2004
• Sandefur, J. "A sketch of the structure of Kriol," in Romaine, S. (ed.,). Language in Australia. Cambridge: Cambridge University Press, 2004
• Meyerhoff, M. "Transitive marking in contact Englishes," in Australian Journal of Linguistics 16, 1996, pp. 57-80
• Rhydwen, M. Writing on the backs of blacks: Voice, Literacy and community in Kriol fieldwork. Brisbane: University of Queensland Press, 1996
• Harris, J. "Losing and gaining a language: The story of Kriol in the Northern Territory," in Walsh, M. and Yallop, C. (eds.,). Language and culture in Aboriginal Australia. Canberra: Aboriginal Studies Press, 1993
• Graber, P. "Kriol in the Barkly Tableland," in Australian Aboriginal Studies 2, 1987, pp. 14-19
• Graber, P. "The Kriol particle 'na'," in Working papers in language and linguistics 21, 1987, pp. 1-21
• Marret, M. "Kriol and literacy," in Australian Aboriginal Studies 2, 1987, pp. 69-70
• Sandefur, J. "Mission life, mission education and the rise of creole language," in Journal of Christian Education 85, 1986, pp. 23-34
• Harris, J. Northern Territory pidgins and the origin of Kriol. Canberra: Pacific Linguistics, 1986
• Sandefur, J. and Harris, J. "Variation in Australian Kriol," in Fishman, J. (ed.,). The Fergusonian impact. Berlin: Mouton de Gruyter, 1986, pp. 180-190
• Harris, J. and Sandefur, J. "Kriol and multilingualism," in Clyne, M. (ed.,). Australia, meeting place of languages Vol. 92. Canberra: Pacific Linguistics, 1985, pp. 257-264
• Sandefur, J. "Dynamics of an Australian creole system," in Pacific Linguistics A 72, 1985, pp. 195-214
• Sandefur, J. "Aspects of the socio-political history of Ngukurr (Roper River) and its effects on language change," in Aboriginal History 1-2, 1985, pp. 205-219
• Sandefur, J. A language coming of age: Kriol of North Australia. Unpublished MA Thesis, University of Western Australia, Perth, 1984
• Harris, J. and Sandefur, J. "The Creole language debate and the use of creoles in Australian schools," in The Aboriginal Child at School 12(1), 1984, pp. 8-29
• Sandefur, J. "A resource guide to Kriol," in Sandefur, J. (ed.,). Papers on Kriol Vol. 10. Darwin: SIL, 1984, pp. 107-140
• Hudson, J. Grammatical and semantic aspects of Fitzroy Valley Kriol. Darwin: SIL, 1983
• Harris, J. and Sandefur, J. "Creole languages and the use of Kriol in Northern Territory schools," in Unicorn 9(3), 1983, pp. 249-264
• Hudson, J. "Transitivity and aspect in the Kriol verb," in Papers in pidgin and creole linguistics: Pacific Linguistics A-65 3, 1983, pp. 161-175
• Sandefur, J. "Kriol and the question of decreolisation," in International Journal of the Sociology of Language 36, 1982, pp. 5-13
• Sandefur, J. "When will Kriol die out?" in MacKay, G. A. and Sommer, B. A. (eds.,). Application of linguistics to Australian Aboriginal contexts (ALAA Occasional Papers). Melbourne: Melbourne University, 1982, pp. 34-43
• Sandefur, J. "Kriol: An Aboriginal language," in Hemisphere 25, 1981, pp. 252-256
• Sandefur, J. "Kriol: Language with a history," in Northern Perspective 4(1), 1981, pp. 3-7
• Sandefur, J. "The stepchild who began Cinderella: Pidgin English comes into its own," in On Being 8(8), 1981, pp. 43-45
• Sandefur, J. "An introduction to conversational Kriol," in Working papers of SIL - AAIB Series B. Darwin: SIL, 1981
• Hudson, J. Fitzroy Valley Kriol wordlist. Unpublished manuscript, 1981
• Rumsey, A. On some syntactico-semantic consequences of homophony in northwest Australian Pidgin/Creole English. Unpublished manuscript, 1981
• Meehan, D. Kriol literacy: Why and how ... Notes on Kriol and the Bamyili school bilingual education program. Barunga, N. T.: Bamyili Press, 1980
• Capell, A. "Languages and creoles in Australia," in Sociologia Interationalis 17, 1979
• Murtagh, E. Creole and English Used as Languages of Instruction with Aboriginal Australians. Stanford University, 1979
• Sandefur, J. An Australian creole in the Northern Territory: A description of Ngukurr-Bamyili dialects (Part 1). Darwin: SIL, 1979
• Steffensen, M. "Reduplication in Bamyili Creole," in Papers in Pidgin and Creole Linguistics No. 2. Pacific Linguistics A57, 1979, pp. 119-133
• Davidson, G. A preliminary report on traditional culture learning and Aboriginal pidgin as part of the school's bilingual program at Bamyili, N.T. Unpublished manuscript, Canberra, 1977
• Fraser, J. "A phonological analysis of Fitzroy Crossingchildren's pidgin," in Hudson, J. (ed.,). Five Papers in Australian Phonologies: Work papers SIL-AAB A.1. Darwin, 1977, pp. 145-204
• Fraser, J. A tentative short dictionary of Fitzroy Crossing children's pidgin. From data collected Oct-Nov 1974; revised March 1977.Unpublished manuscript, Darwin, 1977
• Hudson, J. Some common features in Fitzroy Valley Kriol, 1977
• Sharpe, M. and Sandefur, J. "A brief description of Roper Creole," in Brumby, E. and Vaszolyi, E. (eds.,). Language problems and Aboriginal education. Mt Lawley: College of Advanced Education, 1977, pp. 51-60
• Steffensen, M. A description of Bamyili creole.Unpublished manuscript, Urbana, Illinois, 1977
• Steffensen, M. "Double talk: When it means something and when it doesn't," in Papers from the 13th Regional Meeting, Chicago Linguistic Society, 1977, pp. 603-611
• Thompson, H. Creole as the vernacular language in a bilingual program at Bamyili school in the Northern Territory. Torrens College of Advanced Education, Adelaide, 1976
• Steffensen, M. Bamyili creole. Unpublished manuscript, Madison, Wisconsin, 1975
• Richards, E. and Fraser, J. "A comparison and contrasting of the noun phrases of Walmatjari with the noun phrases of Fitzroy Crossing children's pidgin," Paper presented at the ALS 7, Sydney, 1975
• Sharpe, M. "Report on Roper Pidgin and the possibility of its use in a bilingual program," in Report on the Third Meeting of the Bilingual Education Consultative Committee held in Darwin on 27-29 November 1974. Darwin: Department of Education, 1974
• Hall, R. A. "Notes on Australian Pidgin English," in Language, 19, 1943, pp. 263-267

Language situation

(the following sketch of the phonology and grammar of the language is from the Walmajarri - English Dictionary compiled by Eirlys Richards and Joyce Hudson)

The Walmajarri people traditionally lived in the Great Sandy Desert to the south of the Kimberley. Subsequent events took them to cattle stations, towns and missions scattered over a wide area. Today, communities with substantial Walmajarri population are:

Bayulu (Gogo), Bidyadanga (La Grange), Djugerari (Cherrabun), Junjuwa (Fitzroy Crossing), Looma, Millijidee, Mindibungu (Bililuna), Mindi Rardi (Fitzroy Crossing), Mulan (Lake Gregory), Ngumban (Pinnacles), Ngalapita, Ngurtawarta, Wangkajungka (Christmas Creek),Yagga, Yakanarra (Old Cherrabun) - the redrawn boundary now means Yakanarra is located on Gogo, Yungngora (Noonkanbah).

Typology

There are seventeen consonants and six vowels, three long and three short. The charts below show how the various sounds are made:

Walmajarri consonants

Walmajarri vowels

Verbal auxiliary

There is no equivalent in English to the Walmajarri verbal auxiliary. The two purposes of the verbal auxiliary are:

• to show which person or thing or how many persons or things were involved in the action. There is no distinction between masculine and feminie so where English distinguishes between he, she and it, Walmajarri does not
• to show the mood of the sentence. For instance, if the boy (1 person) does something TO the girl (1 person), the verbal auxiliary has the form pa. If the boy (1 person) does something FOR the girl (1person) the verbal auxiliary is parla. If the boy (1 person) ACCOMPANIES the girl (1 person) the verbal auxiliary is manyanta. If other people are involved the verbal auxiliary changes to indicate this.

With so much information contained in the verbal auxiliary, the sentence can be reduced to two words. The verb tells what the action was and the verbal auxiliary tells who or what was involved in the action. These two words form a mini-sentence.

Cases

In a Walmajarri sentence the words can be moved about fairly freely, unlike English which requires a fixed word order to maintain the same meaning, eg in the sentence 'The girl saw the boy', a change of order makes a new sentence with a different meaning, 'The boy saw the girl'. Walmajarri word order is more free because where English uses word order and prepositions to indicate relationships between words in a sentence, Walmajarri uses suffixes.

There are two types of cases in Walmajarri. One group is shown on the verbal auxiliary as well as on the nominal, ie they are cross referenced [Ergative, Accessory, Dative]. The other group is only shown on the nominal (Locative, Purposive, Preventative, Allative, Ablative, Consequent, Manner).

Nominals

The term nominal is used Walmajarri for word that potentially take the case suffixes. Most of these words would be clearly classified in English as nouns or adjectives. In Walmajarri this distinction is not clear because adjectives often take the place of nouns.

Manga purlka           [kirta(E)]           pirriyani

girl                          big                    came

'The big girl came.'

Purlka                     [kirta(E)]           pirriyani

'The big (girl) came.'

The form without the noun is used unless there is need to be specific about the person involved.

Verbs

The verb consists of a stem and four orders of suffixes. Stems may be monomorphemic or compound and are divided into five conjugation classes. Before tense suffixes can be added to the verb, the mood of the sentence has to be known. This is because there are two sets of suffixes, based on the Realis-Irrealis distinction. When the mood is indicative, interrogative or hortatory, the Realis Tense System applies. When the mood is intentive, admonitive, imperative, negative, prohibitive or inabilitative the Irrealis Tense System applies. In finite verbs the Realis Tense System makes four distinctions: past, customary and future, with present only in the repetiitive/continuous aspect. The Irrealis tense System distinguishes only two, past and non-past. The two infinite forms are non-past (infinitive) and past. The past expresses an action prior to that of the main verb.

Kamparnu-rla marna ngarni.

cook-past   I   ate

'After cooking it I ate it.'

The non-past is the infinitive to which the suffixes are added.

Kiy [Kuyu(E)] kamparnu-jangka   marna ngarni.

meat    cook-from    I   ate

'I ate the cooked meat.'

(Richards, Eirlys and Hudson, Joyce. Walmajarri-English dictionary. Darwin: SIL, 1990)

References

• Richards, Eirlys and Hudson, Joyce. Walmajarri-English dictionary. Darwin: SIL, 1990
• Richards, Eirlys. "The Walmajarri noun phrase," in Work Papers of SIL-AAB, A:3, 1979, pp. 93-127
• Hudson, Joyce. The core of Walmatjari grammar. Canberra: A.I.A.S, 1978
• Hudson, Joyce and Richards, Eirlys et al. "The Walmatjari: an introduction to the language and culture," in Work Papers of SIL-AAB, B:1, 1978
• Hudson, Joyce and Richards, Eirlys. "The phonology of Walmatjari," in Oceanic Linguistics 8:2, 1969, pp. 171-89

Language situation

Warumungu speakers live at Tennant Creek, to the north and east at Elliott, Marlinja, Kurnturlpara, Ngurrara, Wogayala, and Alroy Downs, and on several outstations such as Likarrapartta, Jurntu Jungu, Pingala, and to the south at Kurraya, Alekarenge, Karlinjarangi, and Jungkkaji. In most of these communities other languages are spoken.

In the mid 1990s Robert Hoogenraad surveyed the languages of the Barkly area, and estimated that there were then about 700 people who could speak some Warumungu. While older people can speak Warumungu fluently, very few children and teenagers have spoken it regularly since at least the early 1980s.

Warumungu was profiled in The Federation of Aboriginal & Torres Strait Islander Languages and Culture (FATSIL) newsletter "Voice of the Land"

Language work on Warumungu is carried on in Tennant Creek through the Papulu Apparr-kari Aboriginal Corporation, as well as in the Nyinkka Nyunyu Aboriginal Art and Cultural Centre.

Typology

Warumungu has been classified as a subgroup on its own within the Western Desert type of Pama-Nyungan languages by Oates (1975: I, 145). According to this classification, Warumungu would be more closely related to Warlpiri and Warlmanpa, than, say, to Arandic. However,in the classification of O'Grady, Voegelin and Voegelin (1966, p.42), Warumungu is coordinate with the Wakayic group, the Arandic group and the South-West group to which Mudbura, Warlpiri and Warlmanpa belong. Its exact genetic relationship to these languages is still uncertain.

Phonologically, the presence of two stop series sets Warumungu apart from neighbouring languages.

There are five places of articulation for stops and nasals: bilabial, apico-alveolar, apico-postalveolar (retroflex), lamino-palatal and velar. There are three laterals, one tap r and three semivowels (including a retroflex glide r). There are three vowels. Vowel length is distinctive in initial syllables. Primary stress occurs word-initially. The most striking phonological rules are alternations in consonant length and voicing, and vowel deletion, which deletes the first of two adjacent vowels across word boundaries.

Warumungu is a suffixing language. Grammatical functions are expressed by Ergative-Absolutive case-marking. Word order is used for information structure purposes. Initial and second position are particularly salient. Pronouns appear in clusters, marking either one pronominal argument (intransitive subject) or two (transitive subject and direct, reflexive or indirect object). They normally appear in initial position, or encliticised to a constituent in initial position. Verb roots are a small closed class. Complex verbs are created by compounding these verb roots with preverbs, which appear to be an open class. There is no evidence for a surface verb phrase, but nouns and their modifiers do appear to form phrases.