ALS99 ABSTRACTS: Plenaries

The Morphosyntax of Clitics in Kwakw'ala

Stephen R. Anderson   Yale University

Since the publication of Anderson (1984), Kwakw'ala has been the prototypical example of a language in which the syntactic and phonological affiliations of (certain) clitics are quite different. In this language, NP-initial clitics represent morphosyntactic properties of a particular NP (case, determiner, possessor) but are phonologically attached to a preceding word which does not form part of the NP whose properties they mark.

This talk will survey the range of cliticization phenomena in Kwakw'ala on the basis of the description in Boas (1947 and various collections of text material) together with my own fieldwork, suggesting ways in which the morphosyntactic regularities of the language interact with its phonology to yield the full distribution of clitic elements. Apart from the NP-initial clitics referred to above, these include (a) complementizer elements marking subordinate clauses; (b) subject and object pronominals; and (c) NP-internal, second position clitics. A number of general points follow from the analysis of Kwakw'ala clitics, including the fact that neither syntactic nor phonological requirements alone will not suffice to describe second position phenomena.

Some comparison with the clitic systems of related Northern Wakashan languages (Haisla and Heiltsukw) will be provided to yield further perspective on this unusual language.

Women at Work: analysing women's talk in New Zealand workplaces

Janet Holmes   Victoria University of Wellington

Over the last ten years, evidence of gender difference in the use of language has been re-examined from a social constructionist perspective. This approach emphasises the extent to which we "perform" or "construct" aspects of our social identity in interaction with others. How do women use language to construct their professional identities at work? Are successful managers, for instance, those who construct a stereotypically masculine identity or is there a place for the use of more feminine discourse strategies in senior management positions? These are some of the issues that will be addressed in this paper which draws on an extensive database of workplace interactions in a wide range of New Zealand workplaces.

Exploring metatypy: how does contact-induced typological change come about?

Malcolm D. Ross   RSPAS, Australian National University   (

Since the publication of Thomason & Kaufman (1988), there has been an increasing understanding that contact-induced change falls into sociolinguistically differentiated types.In this paper, I focus on the kind of gradual contact-induced change which occurs when a community of mostly bilingual speakers has a language which is emblematic of its identity, yet uses an inter-community language intensively for communication with outsiders. In such communities in Papua New Guinea, emblematicity is seen as residing in lexicon and there is little norm enforcement to constrain changes in language structure . There is thus a tendency for the language's semantic patterns to be reorganised on the model of the inter-community language, and there is usually also a restructuring of the syntax of the emblematic language so that it more closely resembles the syntax of the inter-community language. There is evidence that this process has occurred in traditional societies in many parts of the world. The process often results in a change in the morphosyntactic type of the emblematic language, and I have dubbed this change 'metatypy' (Ross 1996).

Metatypy has several components. The reorganisation of semantic patterns seems to begin before the restructuring of syntax. Syntactic restructuring apparently proceeds from larger units to smaller, from the sentence to the clause to the phrase to the word. In Takia (Austronesian, Papua New Guinea) semantic reorganisation, apparently on the model of a neighbouring Papuan language, is near-total, and sentence, clause and phrase structures have been remodelled. Metatypy here includes shifts from verb-medial to verb-final and from prepositional to postpositional constituent orders. Often metatypy is accompanied by very little lexical borrowing, but it is commonplace for discourse particles or inter-clausal conjunctions to be copied from the inter-community language.

Previous work (Grace 1987, Ross 1996) has tended to emphasise semantic reorganisation, but this leaves some of the components of metatypy less than well accounted for. How do such radical changes in syntactic type come about? Why does metatypy proceed from larger units to smaller? Why do discourse particles and inter-clausal conjunctions get borrowed, but not other lexical items?

We can achieve a coherent account of metatypy by seeing it as the outcome of speakers' labour-saving assimilation of the discourse strategies of their emblematic language to those of their inter-community language, so that in the end they use a single discourse strategy for both languages.

In interactive discourse, particles and inter-clausal conjunctions serve as boundary-marking speech acts in their own right or as markers of particular kinds of act, and their borrowing seems to reflect the borrowing of ways of structuring discourse. Similarly, in monologue discourse, the assimilation of interclausal structures to those of the inter-community language represents the borrowing of ways of stringing speech acts together, as the clause is the most common manifestation of the speech act. However, changes in interclausal syntax may lead willy-nilly to changes in clause-internal syntax, e.g. verb-medial to verb-final, and this may in turn foster phrase-internal changes (prepositional to postpositional).

A speech act often entails the operations of reference, predication, and attribution, and these represent the intersection of propositional content (what the speaker wants to say) and information structure (the structuring of the proposition into presupposition and assertion) which is encoded as grammar. There is good evidence that the assimilation of discourse strategies includes the assimilation of the ways that information structure is encoded.

As Pawley and Lane (1998) have shown from Kalam, syntax develops in tandem with conventionalised ways of reporting situations and events. There is ample evidence that metatypy entails borrowing the conventionalised ways of reporting that are in use in the inter-community language. This has the consequence not only that the semantics of the emblematic language are reorganised, but that syntax may begin to be reorganised in tandem with semantics. The tandem nature of these changes is more obvious when we remember that much of the lexicon of a language consists of conventionalised phrases and clause fragments.

This informal sketch of the way metatypy occurs points to some equally informal observations about linguistic theory. Grace (1981) observed that what differs between the two languages of a bilingual after metatypy has occurred is not their lexicons, but the 'lexifications' (phonological shapes) of the lemma store (the semantic and syntactic components of lexical items) shared by the two languages. That is, a lexical item consists two separate cognitive units: the lemma and the lexification, as Levelt (1989) also notes. Most models of speaking separate 'conceptualisation' (planning the speech act) and 'formulation' (encoding the conceptualised message by accessing lemmas). In Levelt's (1989) model the 'conceptualiser' feeds the 'formulator', but not vice versa. But given the phrasal and clausal size of many lemmas as confirmed in bilingual vocabularies, it is difficult to see how a speech act could be planned without direct access to the lemma store.

Youth Language Development

Shirley Brice Heath   Stanford University

Applied linguists, particularly those in bilingual studies and the teaching of language arts, have benefitted from the research of scholars in early language development. Little help, however, has come forward on later language development and the roles of language models, learning environments, and relations between the printed and spoken word. Hoping to help fill this void, research on the spoken and written language patterns of older children and youth has over the past decade revealed much about learning and especially the acquisition of advanced language skills in hypothetical reasoning, future scenario-building, and critique. This address will present these findings with implications for their use in learning environments from community organizations to comic books.