PROGRAMME
AUSTRALIAN LANGUAGES WORKSHOP,
PEARL BEACH, 2007
updated 30/3/07

Following Noel Pearson's manifesto for saving Indigenous languages published in The Australian (10/3/07), we had a discussion of where this manifesto could be taken.  Abstract here.

TIMETABLE

ABSTRACTS




FRIDAY 16 MARCH
11.00-11.40
Nicholas Evans,
University of Melbourne

Polyglot narratives: two case studies
11.40-12.20 Sarah Cutfield,
Monash University
Discourse functions of nominal demonstratives in a Dalabon narrative

12.20-1.00
Joe Blythe,
University of Sydney
Prosodic Person Reference in Murriny Patha (Reported) Conversation
1-2.00

LUNCH

2.00-2.40
Felicity Meakins,
University of Melbourne
Alternational code-switching and mixed language genesis
2.40-3.20 Eleonora Deak, Wangka Maya Pilbara Aboriginal Language Centre
Contemporary Thalanyji: How a Highly Endangered Language is Being Used Today
3.20-4.00
John Giacon,
Australian National University

Linguistics and language revival in SE Australia - some examples and principles and questions
4.00

LEAVE FOR PEARL BEACH





SATURDAY 17 MARCH
9.00-9.40
Brett Baker,
University of New England
Geminates and NC clusters as onsets: the Australian evidence
9.40-10.20 Rachel Nordlinger,
University of Melbourne
Reciprocal Constructions in the Daly
10.20-10.40
MORNING TEA
10.40-11.20
Hywel Stoakes,
University of Melbourne
Articulatory evidence regarding glottalization of stops in Bininj Gun-Wok
11.20-12.00 Jeffrey Chapman,
University of Queensland
Some Aspects of Intonation in Warlpiri
12.00-12.40
Bruce Birch,
University of Melbourne
Some Diagnostics Of the Iwaidja Word
12.40-2.00

LUNCH
2.00-2.40
Mark Harvey,
University of Newcastle
The Origin of Conjugation Markers
2.40-3.20 Patrick McConvell, AIATSIS Discussion of Noel Pearson's manifesto
3.20-3.40
AFTERNOON TEA
3.40-4.20
Michael Walsh,
University of Sydney
Australian Aboriginal song language: so many questions, so little  to work with
4.20-5.00
Aaron Corn,
University of Sydney
and
Joe Gumbula,
University of Sydney
Following the voices of ancestors

WALK.. SWIM.. DINNER ETC






SUNDAY 18 MARCH
9.00-9.40
Neil Bell
Idea Words in Pitjantjatjara
9.40-10.20 Kazuko Obata, AIATSIS
Harold Koch, ANU
Towards AIATSIS Indigenous languages/dialects classification - with a Yuin-Kuri example

You can download the classification as a .pdf file here.
10.20-10.40
MORNING TEA
10.40-11.20
Patrick McConvell, AIATSIS Loanwords in Gurindji: identification, stratigraphy and loan-paths
11.20-12.00 Mary Laughren,
University of Queensland
Another foray into Warlpiri complex verbs; which bit determines argument structure?
12.00-12.40
David Nash,
ANU and AIATSIS
Where do Warlpiri simple verbs come from?
12.40-2.00

LUNCH
2.00-2.40
Tom Honeyman,
University of Sydney
A broad search for cognates across 40+ Northern Australian languages
2.40-3.20
Jane Simpson,
University of Sydney

Reconstructing Warumungu pronominals

3.20-3.40
AFTERNOON TEA
PACKING UP






ABSTRACTS [arranged in alphabetical order by surname]


Indigenous languages policy and regional archives

a discussion theme, prepared by Patrick McConvell

Noel Pearson has recently spoken out about the severely endangered state of Australia's indigenous languages and proposed measures to meet the situation, in an  opinion piece in The Australian. A  national report  in 2005 also drew attention to the desperate situation of Indigenous languages  and made a number of recommendations. The  National Indigenous Language Survey report (compiled by the Australian Institute of Aboriginal and Torres Strait Islander Studies together with the Federation of Aboriginal and Torres Strait Islander Languages was released by the Department of Communications, Information Technology and the Arts. Among the prime recommendations were support for Language Nests, an early childhood program that has been successful in New Zealand and elsewhere.; community language teams; regional language centres, and a National Indigenous Language Centre, which could carry out many of the functions which Mr. Pearson views as essential. Other strong themes of the report  are Indigenous languages as a key part of our national heritage; the importance of detailed documentation of languages; and the use of new technology, all advocated by Mr. Pearson. Opportunities exist for taking up these  initiatives  but as Mr. Pearson comments, there continues to be lack of adequate funding.

There are also points of difference between Mr. Pearson and many indigenous people and linguists over his insistence that school education should be an English domain. Yet there may be sufficient points of commonality to allow a joint approach with him and it is timely for us to discuss what such a strategy may entail.

One possible area which we may be able to advance without entering what may be murky waters around the national centre question, is providing access to at least some sets of digital documentation, and ability to annotate such material, for remote sites, and this is precisely the kind of technological development Mr. Pearson is also advocating. An offer has recently been made by the Max-Planck Institute in Nijmegen to assist AIATSIS to establish a DOBES regional archive which might house not only DOBES collections but also others such as a Cape York Peninsula collection including both ELDP and other material. AIATSIS does not have funds in its budget to support the infrastructure for such an initiative this year or next year; DCITA may have such funds but the logistics of application are difficult. If stakeholders and other interested parties think this is a good idea then we should discuss how it can be achieved.



Return to timetable
Return to workshop



Brett Baker, University of New England
Geminates and homorganic nasal-stop clusters as onsets: the Australian evidence
 
Geminates and homorganic nasal-stop clusters are both 'monogestural': involving a single articulatory stricture location and type in the oral cavity, but having the duration of a cluster. The ambivalent nature of monogestural clusters leads to conflicting analyses by linguists, according to their phonological and phonetic characteristics. I examine evidence from quantity-sensitive stress, reduplication patterns, phonotactics of the syllable, and speaker parsing which point towards these clusters being treated as onsets in many Australian languages, at least for some phonological patterns. The evidence leads us to think that syllabification in Australian languages may work on the basis of gesture, rather than timing positions as in most other languages. This has implications for the universality of the 'mora' as a unit of timing, as well as weight.




Return to timetable
Return to workshop



Bruce Birch, University of Melbourne
Some Diagnostics Of the Iwaidja Word

This paper examines aspects of the intonation system of Iwaidja which make reference to the boundaries of both monomorphemic and polymorphemic words. The evidence suggests that studies which diagnose wordhood on the basis of broad notions of accentability may be missing out on diagnostics which emerge from a finer-grained analysis of intonational features such as accent distribution. It is also claimed that both function words and content words in Iwaidja exhibit prosodic alternations, requiring in some cases adjustments in foot structure.


Return to timetable
Return to workshop



Joe Blythe, University of Sydney
Prosodic Person Reference in Murriny Patha (Reported) Conversation

Recent research within interactional linguistics and CA on the pragmatic roles of prosody within talk-in-interaction has concentrated on areas such as mimicry, affect, reported speech, repair and turn-taking; all of which can have a bearing on the interpretation of referents. However there has not been a lot of research that specifically deals with the pragmatic role of prosody in deixis. This may be because better understood referring strategies (i.e., morphosyntactic strategies) can mask the deictic role of prosody when speakers have employed more than one strategy at a given time. In Murriny Patha conversations, person references may be multilayered constructions that span across more than one turn-at-talk.

In the case of reported speech, hearers interpret the bundling of prosodic features across passages of talk (global marking) as hailing from an ‘alien’ world of discourse. This ‘alien’ world of discourse has its own referential frame, where all second person references are understood to not be personally addressing current members of the audience. On the other hand, local prosodic marking also plays an important role in person reference. Distinctive marking of referential items can be demonstrated to mark both coreference and non-coreference. When the capacity of polysynthetic reference is pushed to the limit, the use of this local prosodic marking gives the hearer important cues as to the correct readings of referents.



Return to timetable
Return to workshop

Jeff Chapman, University of Queensland Australia
Some Aspects of Intonation in Warlpiri

This paper presents initial findings of an investigation of the intonational contours and prosodic structure of the indigenous Australian language Warlpiri.  The study was undertaken because Warlpiri segmental syntax and phonology has been well described, providing the basis for a detailed analysis of intonation.  Warlpiri is a non-configurational language (Hale, 1983) with free word order, discontinuous constituents and zero anaphora for clause arguments. Some aspects of word order are determined by information structure, and these features make discourse and syntactically related phenomena more salient.

A detailed analysis of a limited corpus was undertaken to identify significant elements of Warlpiri prosody, and provide a basis from which to develop hypotheses about prosodic structure.  An acoustic analysis was made of two studio quality recordings of two narrative monologues, spoken by two female speakers of the south-western (Yuendumu) dialect. Performance narratives provide a text with few disfluencies and display discourse relations. The recordings were annotated in an  autosegmental-metrical framework using a tones and break indices (ToBI) style notation.

The principal findings which I will discuss were:
- Many words conformed to a standard intonation contour, with a high initial tone and a gradual fall in intonation to a low final tone.
- Verb compounds and the auxiliary complex show a single falling contour, leading to the conclusion that these expressions are a single accentual phrase (Beckman and Pierrehumbert, 1986), bounded by initial high and final low tones.
- No evidence was found for a culminative accent on words with either narrow or broad focus.
Warlpiri uses a final rising intonation in questions, and final lowering at the end of a number of declarative utterances. 
- Some declarative utterances conclude with a rising final intonation which occurs where the following sentence continues the action with the same participants. 
- Final rising intonation associated with pauses, indicates the presence of intonational phrase boundaries related to syntactic constituent structure, marking off left dislocated topics, adjuncts, and words and phrases used in apposition, a common device in Warlpiri. 
- Initial words receive a greater prominence when they start direct speech within the narrative. 
- A stylized high sustained intonation is used in an iconic way to indicate duration, as in other Australian languages, e.g., Bininj Gun-wok (Bishop, 2003).

Future research will analyse a larger corpus of narrative texts to confirm (or disprove) current hypotheses about Warlpiri prosodic structure; and identify how Warlpiri speakers use the placement of intonational phrase boundaries, combined with word order changes, to signal the precise pragmatic prominence required.

References

Beckman, M. E. and Pierrehumbert, J. B. 1986. Intonational Structure in English and
 Japanese. Phonology Yearbook, 3, 255-310.
Bishop, J. B. 2003. Aspects of intonation and prosody in Bininj Gun-wok: an
 autosegmental-metrical analysis, Department of Linguistics and Applied
 Linguistics, University of Melbourne: Ph D thesis.
Hale, K. 1983. Warlpiri and the grammar of non-configurational languages. Natural
 Language & Linguistic Theory, 1, 5-47.


Return to timetable
Return to workshop

Sarah Cutfield, Monash University
Discourse functions of nominal demonstratives in a Dalabon narrative

In this paper I describe the discourse uses and functions of the nominal demonstratives found in a Dalabon narrative. Of particular interest is the use of demonstratives where underspecification of referents and lack of definiteness marking on nominals is the unmarked referencing strategy. I also reflect on the usefulness of discourse measures in characterising the data. For example, Himmelmann (1996) proposes four typological usage types for nominal demonstratives in discourse: situational, discourse deictic, tracking and recognitional. The recognitional use involves the speaker appealing to the hearer’s shared knowledge and is widely reported in Australian languages. Looking at the Dalabon data, it is possible to give a refined description of this use as it occurs in Dalabon.


Return to timetable
Return to workshop



Aaron Corn and Joe Gumbula, University of Sydney
Following the voices of ancestors

Names are the most important hereditary property held by the Yolngu (People) of NE Arnhem Land other than country itself. They mark the identities of individuals and land-owning groups, record the original observations of ancestors, and largely comprise the esoteric languages traditionally deployed by appointed Yolngu Elders in ceremonial and diplomatic contexts.

This presentation will explore the role of hereditary Yolngu Manikay (Song) as the root medium through which Names and their esoteric meanings are expressed and taught using Gupapuyngu Yolngu recordings directed by Neparrnga Gumbula at Djiliwirri in 2004 and 2005. It will demonstrate the vital role of Yolngu Elders in maintaining traditional knowledge and esoteric discourses that can only be taught through Manikay, and the continuing centrality of such endeavours to Yolngu cultural survival and social cohesion.



Return to timetable
Return to workshop



Eleonora Deak, Wangka Maya
Contemporary Thalanyji: How a Highly Endangered Language is Being Used Today

Thalanyji is one of the highly endangered languages of the Pilbara region in North Western Australia. The six fluent speakers and the ten to twenty partial speakers of Thalanyji mostly live in the town of Onslow. This informal study of contemporary Thalanyji has taken place within the context of Wangka Maya Language Centre's current work with Thalanyji: database and dictionary updating, a book on local plants and the Onslow town signage project. Historically, the Thalanyji speakers' strongest geographic, economic, social, cultural relationships have been with the speakers of the other Ganyara languages: Binigura and Burduna which are now no longer spoken at all and Bayungu which has a handful of partial speakers. These languages are structurally almost identical and  lexically very similar. There is also evidence of some social and linguistic interaction with other neighbouring language communities. Insofar as its possible, all of these relationships continue today. These factors all suggest that certain aspects of contemporary Thalanyji linguistic forms and even a few Thalanyji structural elements may now differ from previous descriptions of the language.



Return to timetable
Return to workshop



Nicholas Evans, University of Melbourne
Polyglot narratives: two case studies.

Many commentators mention that narratives in Australian languages may move between more than one language according to character, setting and other motives, but there have been few actual case studies. In this paper I analyse two traditional narratives from different parts of northern Australia: one from Croker Island (in Marrku, Iwaidja and Kunwinjku) and one from Mornington Island (in Lardil and Yangkaal), showing rather different bases for language choice by a polyglot narrator.


Return to timetable
Return to workshop


Mark Harvey, University of Newcastle
The Origin of Conjugation Markers

This paper compares two hypotheses as to the origin of ‘conjugation markers’ in the verbal paradigms of Australian languages. Conjugation markers are segments in verbal paradigms which cannot be straightforwardly assigned to either the root or to the set of Tense, Aspect, Mood suffixes. Dixon (1980:409, 2002:215) proposes that the conjugation markers originated as the final segment of the verb root. Over time, various types of phonological interactions across the boundary between the verb root and TAM suffixes have resulted in roots no longer ending in their original final segment in all cases. In the remaining cases where they do, the original final segment is analysed as a conjugation marker.

Alpher, Evans & Harvey (2003:347-349) present an alternative hypothesis. They propose that conjugation markers originated in the use of inflected verbal forms as stems.

[[verb root + TAM suffix]stem + suffix A / + suffix B /+ suffix C]word 

Over time, developments have obscured the status of the [verb root + TAM suffix] form as a stem, leading to a re-analysis of the TAM suffix as a conjugation marker.

[[verb root + conjugation marker]stem + suffix A / + suffix B /+ suffix C]word

I show that the evidence supports the second hypothesis and not the first.



Return to timetable
Return to workshop

Tom Honeyman, PARADISEC, University of Sydney
A broad search for cognates across 40+ Northern Australian languages

In this presentation I will describe the methodology used on a project to collect cognates from roughly 40 neighbouring non Pama-Nyungan languages. This was done using a mix of computational analysis and human annotation. The aim was to collect a broad collection of cognates rather than a deep one, and as such the method was geared towards assisting a human reader in discovering cognates as quickly as possible while also grading their likelihood of being cognates. This is not an exhaustive method for collecting cognates.

The lexicon for each language was normalised to a single orthography and then all entries in one lexicon are compared to all entries in the other. They are compared for exact matches, matches with lenition rules applied, and then fuzzier sub-string matches. This produces a summary file which must then be checked.

The idea was to produce a summary that could include almost all possible matches to save a researcher from jumping around a dictionary looking for cognates themselves. The algorithm depends however on a human to find the actual cognates, The output is sorted so that, based on the shape of the lexical entry, the more likely matches appear closer to the top of the list. Where there are simply too many matches the list is culled in various ways.

Because the matches are indexed across languages, matches across languages pairs that have been checked can be used to predict matches across unchecked languages pairs by transitivity. Longer chains of cognates across languages can then be extracted and examined more closely. Additionally it may be useful for suggesting matches that were missed in the first pass.



Return to timetable
Return to workshop

Mary Laughren, The University of Queensland

 Another foray into Warlpiri complex verbs; which bit determines argument structure?

Warlpiri verbs can be extremely complex in their morphological structure and internal syntactic behaviour. They are headed by an inflectional category: tense/mood if finite, versus non-finite. The inflectional category is a verbal suffix whose stem is a member of a relatively limited class of verbal stems. This stem may be 'thematic' or 'aspectual'. If the latter, then the thematic verb must be inflected as non-finite. Thematic stems may (some must) be 'modified' by a 'lexical' preverb or a non-finite verb. This complex may be further modified by more 'outer' preverbal elements: adverbial, adjunct dative and quantificational (following Nash's (1982) classification of Warlpiri preverbs). This paper will reexamine the 'inner thematic core' of Warlpiri verbs [(PVlexical])=V], with a view to establishing which elements determine the verb's predicate-argument-structure.

In (1a) karri- 'stand, stop' hosts a finite inflection –mi and determines the PAS of the verb as intransitive. In (1b-e) karri hosts the non-finite –nja and precedes the verb stem which hosts the finite inflection. In (1b,c) karri does not determine the PAS – the finite verb does.  In (1d,e), however the non-finite karri does determine the PAS while the finite verb expresses aspectual values.

(1)          a. [karri-mi] (stand-npast) 'stand, stop' (x)

               b. karri-nja-[pardi-mi] (stand-inf-rise-npast) 'stand up, arise' (x)

               c. karri-nja-[yirra-rni] (stand-inf-put-npast) 'put in standing position' (x,y)

               d. [karri-nja]-parnka-mi (stand-inf-run-npast) 'go to stand' (x)

               e. [karri-nja]-ya-ni (stand-inf-go-npast) 'stand along' (x)

While the semantic relationship between the non-finite verb and the finite verb in (1b-e) is fairly transparent and these combinations are productive, the relationship between the lexical PV pata in (2) and the following verb is somewhat less transparent and productive. The bracketed verb determines the PAS of the complex

(2)          a. pata-[karri]-mi 'fall, drop' (x) (synonymous with wanti-mi)

               b. pata-[kiji]-rni 'throw down; make drop' (x,y)

               c. pata-[karri-nja]-ya-ni 'fall as going along' (x)

               d. pata-[kiji-rninja]-ya-ni 'throw down as going along' (x,y)

The complex verbs in (3a,b), however, are semantically opaque, made up of synchcronically non-productive PV-V combinations. (3a) and (3b) express meaning which is the same as that expressed by the simplex verb in (3c).

(3)          a. [tarda-ya-ni] 'sit/perch/land' (x)

               b. [pirri-ma-ni] 'sit/perch/land' (x)

               c. [nyina-mi] 'sit/perch' (x)

In the non-productive (4a) the finite verb ya-ni is part of the thematic core, whereas it is the non-finite ya-ninja which is part of the thematic core while the finite –ya-ni is aspectual.

(4)          a. [nyunjurl-ya-ni] 'spit curse at/on' (x,y)

               b. [nyunjurl-ya-ninja]-ya-ni 'spit curse at/on along' (x,y)

The verb ‑ma-ni in (3a) typically indicates a two place predicate, but this combination is monadic, whereas -ya-ni in (4a) (and ya-ninja in (4b)) typically indicates a one place predicate, but the PV-V combinations in (4) are dyadic. While the PV-V combinations in (2-4) must be lexically stored, elements which appear to be generated syntactically can combine with these complex verbs in ways which suggest that syntactically generated elements can intervene between lexically generated elements.

This paper will propose that part of processing the inner core of Warlpiri verbs is mapping semantic features onto a phrasal syntactic structure headed by categories assocated with rather abstract features; simplex verbs like karri- in (1a) and nyina- in (3c) may be mapped onto the same underlying structure as complex thematic cores.

Reference:

Nash, David. 1982. Warlpiri preverbs and verb roots. Swartz, S. (ed.),. Papers in Warlpiri grammar: in memory of Lothar Jagst Work Papers of SIL-AAB, Series A Volume 6, 165-216. Berrimah, N.T.: SIL-AAB.




Return to timetable
Return to workshop


Patrick McConvell, AIATSIS

 Loanwords in Gurindji: identification, stratigraphy and loan-paths


Gurindji is a Pama-Nyungan language of the Ngumpin-Yapa subgroup, spoken in the Victoria River District of the Northern Territory, Australia. Its vocabulary consists of at least 30% loanwords, mostly from the Jaminjungan branch of the non-Pama-Nyungan Mirndi family, to its north. The paper surveys the sources of the loans and their distribution in different grammatical and semantic categories * for instance Jaminjungan loans are particularly numerous in environmental vocabulary and coverbs. The paper focuses on three aspects of the analysis of loanwords:

(1) identification: signatures of loanword pnonology (eg absence of sound-changes like intervocalic lenition) , tracing of sources^;

(2) stratigraphy: chronological layering of different loanwords based on phonological and morphological characteristics, and its relationship to culture contacts;

(3) loan-paths: some loanwords travel far in many directions (Wanderwoerter); some follow more circumscribed routes and may be predominantly in one direction. From a typology of loan-paths inferences can be drawn about cultural prehistory.


Return to timetable
Return to workshop

Felicity Meakins, University of Melbourne
Alternational Code-switching and mixed language genesis

The extent to which code-switching is a factor in the formation and resulting character of mixed languages is debated extensively. Although some maintain that code-switching plays no role in mixed language genesis
(Bakker, 2003), a body of work is growing which supports the contribution of code-switching (Auer, 1999). Within this work the different structures of mixed languages are compared with insertional and alternational
code-switching, with insertional code-switching favoured as the predecessor to mixed languages. Insertional code-switching constraints are then considered to be responsible for the resultant shape of the mixed language (Myers-Scotton, 2002). Alternational code-switching is deemed too unconstrained and structurally variable to grammaticalise with any consistency (Backus, 2003).
 
One of the problems with this discussion is the lack of empirical evidence. Such evidence exists in one case of a mixed language, Gurindji Kriol (Australia). In the mid 1970s, code-switching between Gurindji and Kriol was a very common style of communication (McConvell, 1985). Thirty years later Gurindji people speak a mixed language, Gurindji Kriol, which exhibits a split along NP-VP lines. Gurindji contributes the NP structure including case and derivational morphology, and the VP structure including TAM morphology originates from Kriol. McConvell and Meakins (2005) show that code-switching not only preceded the formation of
Gurindji Kriol, but that the split of the mixed language corresponds with the pattern of code-switching. In this paper I will build on this work and demonstrate that, contrary to expectations, the code-switching of the 1970s exhibits both insertional and alternational structures and the alternational structures are responsible for the presence of case-marking in Gurindji Kriol.

Auer, P. (1999). From codeswitching via language mixing to fused lects:
Toward a dynamic typology of bilingual speech. Journal of Bilingualism,3(4), 309-332.

Backus, A. (2003). Can a mixed language be conventionalized alternational codeswitching? In Y. Matras & P. Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 237-270). Berlin: Mouton
de Gruyter.

Bakker, P. (2003). Mixed languages as autonomous systems. In Y. Matras & P. Bakker (Eds.), The mixed language debate: Theoretical and empirical advances (pp. 107-150). Berlin: Mouton de Gruyter.

McConvell, P. (1985). Domains and codeswitching among bilingual Aborigines. In M. Clyne (Ed.), Australia, meeting place of languages (Vol. C-92, pp. 95-125). Canberra: Pacific Linguistics.

McConvell, P., & Meakins, F. (2005). Gurindji Kriol: A mixed language emerges from code-switching. Australian Journal of Linguistics, 25(1), 9-30.

Myers-Scotton, C. (2002). Contact linguistics : bilingual encounters and grammatical outcomes. Oxford ; New York: Oxford University Press.

O'Shannessy, C. (2006). Language contact and children's bilingual language acquisition: Learning a mixed language and Warlpiri in northern Australia. Unpublished PhD, University of Sydney, Sydney.



Return to timetable
Return to workshop

David Nash, ANU and AIATSIS
Where do Warlpiri simple verbs come from?

The Ngumbin-Yapa subgroup of Pama-Nyungan (McConvell and Laughren 2004) comprises half a dozen languages whose verb inventory is at most 45 or so simple verbs -- except for Warlpiri which has more than twice as many (over 118 at http://www.anu.edu.au/linguistics/nash/aust/wlp/wlp.verb-roots.html). The verb roots common across the Ngumbin-Yapa subgroup have clear correspondences also in Warlpiri, but what of the 70 or so "extra" Warlpiri verb roots?  Is Warlpiri the sole language of the subgroup to preserve scores of verbs lost from all its sister languages? Alternatively, has Warlpiri acquired (by loan or internal formation) scores of verb roots since it separated from its sister languages?  Dixon 2001 interprets Nash 1982 to propose the latter.  I present evidence for the partial truth of both histories.

References

1982. Warlpiri preverbs and verb roots, pp.165-216 in Work Papers of SIL-AAB, Series A Volume 6 in Memory of Lothar Jagst, ed. by Stephen M. Swartz. Berrimah, NT: SIL-AAB. http://www.anu.edu.au/linguistics/nash/papers/1982-Wlp-preverbs.pdf

McConvell, P. and M. Laughren (2004). The Ngumpin-Yapa subgroup, pp.151-178 in Australian languages: classification and the comparative method, ed. by C. Bowern and H. Koch. Amsterdam, John Benjamins.

Dixon, RMW. 2001. The Australian linguistic area, pp.64-104 In Areal diffusion and genetic inheritance: Problems in comparative linguistics, Alexandra Y. Aikhenvald and RMW Dixon. OUP.  §3.1 Verbal organization, pp.71-75; slightly revised version in Dixon 2002.



Return to timetable
Return to workshop


Kazuko Obata,  AIATSIS, and Harold Koch, ANU
Towards AIATSIS Indigenous languages/dialects classification - with a Yuin-Kuri example

AIATSIS is currently doing research on Australian Indigenous languages/dialects classification. So far, we have compiled information from the past major classifications, Wurm (1994 and 1971), Ethnologue (2005), Oates (1975), Dixon (2002), Atlas (1981) and O'Grady et al (1966) as well as information from a selection of recent publications. We have then put together a tentative classification with problematic groups and languages/dialects identified. One of these is the Yuin-Kuri group and Harold Koch will present his analysis on the internal relationship of this group.


Return to timetable
Return to workshop

Jane Simpson, University of Sydney
Warumungu pronouns: another go

At first glance the Warumungu pronoun system diverges markedly from its neighbours, both in syntactic structure and in morphological shape.  In part this can be attributed to paradigmatic reshaping  - the Warumungu pronouns show considerable  regularity in structure,  although  there are some  irregularities, especially in the  non-singular portmanteau Subject-Non-subject forms.   There is one system of pronouns, which are restricted to first or second position, and which are portmanteau  (subject and object pronouns do not appear separately) and may include a tense inflection.   Their restricted position and portmanteau structure makes them more like the bound pronouns in dual pronoun systems than like free pronouns.  Among Warumungu's neighbours on the west,  north and north-east  (Ngumbin-Yapa languages,  Mirndi languages and the Ngarna language Wakaya)  dual systems of bound pronouns and free pronouns are found (although the free pronouns in Warlmanpa are reduced to just two forms for first and second person). To the south and east are Arandic  languages with basically a single series of pronouns, which are usually  treated as free pronouns.  I show that, while Warumungu pronouns show some resemblances to most of the neighbours,  there are more similarities with pronouns in Arandic languages.



Return to timetable
Return to workshop



Michael Walsh, University of Sydney
Australian Aboriginal song language: so many questions, so little to work with

Song language traditions in Aboriginal Australia exhibit a considerable range of variation. Some traditions include material far removed from 'everyday' spoken language: for example, there are ghost or spirit languages which have little, or no, connection with the host language. There are also situations in which the song language is quite similar to the everyday spoken language, even if showing a considerable degree of ellipsis, economy of expression and allusion. In this overview I will seek to delineate the range of variation found in Aboriginal song language traditions. In doing so, a number of questions will arise: does the structure of the everyday language determine or influence the nature of the song language? as many Aboriginal languages are losing speakers, what effect does this have on song language? are there regional differences in the deployment of expressive language?

This overview will draw on more detailed knowledge from a particular set of song traditions hosted by the Murriny Patha language, mainly spoken at Wadeye, about 250 kms south west of Darwin in the Northern Territory. This knowledge derives from an ongoing collaborative research project (involving a team of linguists and musicologists) running from 2004 to 2008 (http://azoulay.arts.usyd.edu.au/mpsong/). But the overview will also survey the research carried out across Aboriginal Australia up to the present day.


Return to timetable
Return to workshop