Word Order in Australian Languages

In 1840 the authors of an early grammar of Kaurna, an Australian Abo-riginal language spoken around Adelaide (Teichelmann and Schürmann 1840), noticed that the order of words is determined in part by how the speaker wants to present information to the hearer, that is, information packaging concerns, and in part by a tendency for the Object to precede the verb. Their material also showed question words and negators occurring sentence-initially. Pragmatic considerations of this kind have been claimed by gram-marians to determine word order in many Australian languages (Mushin and Simpson submitted 2004). Teichelmann and Schürmann's data also shows the means for indicating the discourse status or referential status of partici-pants, especially the use of clitic pronouns and zero anaphora. These means of indicating discourse status and information packaging are quite common in Australian languages. However, work on the representation of these as-pects of the grammar has lagged behind work on non-configurationality. In this paper I chart the development in our understanding of what is happen-ing in word order in one Australian language, Warlpiri, with respect to in-formation packaging.

Word order in Warlpiri is constrained by the interaction of phrase struc-ture with pragmatic considerations. The auxiliary (AUX), a complex cluster of pronominal clitics and tense/aspect markers, occurs in second position or occasionally third position. As in Kaurna, in Warlpiri newsworthy informa-tion, question words and the negative marker occur in initial position (Swartz 1991, Hale, 1992 #3184: 67). Austin and Bresnan (Austin and Bresnan 1996) proposed to express the second position constraint on the placement of the AUX by expanding the left periphery of the flat structure which represents non-configurationality, to include an initial XP linked to a grammaticised discourse function followed by the AUX. This proposal not only captured the second position of the AUX, but also the fact that Subjects tend to precede verbs ? since Subject is a grammaticised discourse function, it will be attracted to initial position.

I argue that the articulation of the left periphery proposed by Austin and Bresnan can easily be modified to accommodate data about the position of Focus and Topic. The main modification consists of adding a C position which allows for the appearance of verbs in initial position, and for their complementary distribution with the negative complementiser kula. I dis-cuss recent extensions of the left periphery by Laughren and Legate (Laughren 2002, Laughren, in press #2985, Legate 2002:280) to account for the placement of Topic and Focus NPs, and show that if this data is to be accounted for by phrase structure positions, it results in structural redun-dancy. In particular I argue that a system where Focus is defined by con-figurational position provides no natural interpretation in the phenomenon of split Focus, as in (1).

1.	Karnta	ka=lu	nyajangu	jilimi-rla=ju	nyina?
	woman	PRES=333S	how.many	single.women's.camp-LOC=KN	be.NPST
	How many women are in the single women's camp?

Instead, I suggest that split Focus provides evidence, firstly for Hale's (1992) proposal that new information precedes the verb, secondly for Choi's proposal (Choi 1999) of classifying discourse functions by binary features +/- Prominent and +/- New (+ New information precedes the verb, + Promi-nent information precedes the AUX), and thirdly for Mohanan's and Choi's proposal for an independent information structure dimension (Choi 1999, Mohanan and Mohanan 1994).

References

• Austin, P. and Bresnan, J. 1996. Non-configurationality in Australian Aboriginal languages. Natural Language and Linguistic Theory 14:2:215-268.
• Choi, H.-W. 1999. Optimizing structure in context: scrambling and in-formation structure. Stanford CA: CSLI Publications.
• Laughren, M. 2002. Syntactic constraints in a "free word order" lan-guage. Language universals and variation, ed. M. Amberber and P. Collins. 83-130. Westport CT: Praeger Publishers.
• Legate, J. A. 2002. Warlpiri: Theoretical Implications. Phd, Massachu-setts Institute of Technology.
• Mohanan, K. P. and Mohanan, T. 1994. Issues in word order in South Asian languages: enriched phrase structure or multidimensionality? Theo-retical perspectives on word order in South Asian languages, ed. M. Butt, M., T. H. King, and G. Ramchand. 153-184. Stanford, Calif.: CSLI Publica-tions.
• Mushin, I. and Simpson, J. submitted 2004. Word order in Australian languages. In a book edited by P. Austin and W. McGregor.
• Swartz, S. M. 1991. Constraints on zero anaphora and word order in Warlpiri narrative text. Darwin, Northern Territory: Summer Institute of Linguistics.
• Teichelmann, C. G. and Schürmann, C. W. 1840. Outlines of a gram-mar, vocabulary, and phraseology, of the aboriginal language of South Australia, spoken by the natives in and for some distance around Adelaide. Adelaide: Published by the authors, at the native location.