The consequences of talking to strangers: sociocultural influences on the lexical unit Alison Wray & George Grace Cardiff University/University of Hawaii wraya@cf.ac.uk We propose to reconcile a number of contradictory proposals about language and how it evolved, by demonstrating that they are consistent with a model featuring flexible lexical storage. One debate concerns whether any language can be (fully) characterised in terms of "words and computational procedures ('rules') for constructing expressions from them" (Hauser, Chomsky & Fitch 2002:1576), or whether words and rules are more a feature of how we describe language when we externalise it than of how we actually process it (Grace 1998:69f). A second thread relates to longstanding evidence that languages have fundamentally different characteristics depending on whether they are commonly learned by adult outsiders (exoteric languages) or only by children (esoteric) (Kay 1977, Chafe 1985, Thurston 1989, Trudgill 1989). Esoteric languages tolerate complexity, irregularity and semantic opacity, and natives may perceive them as a collection of large formulaic chunks (Laycock 1979, Thurston 1987). The third theme is how writing impacts on the complexity of language by supporting (a) autonomous (context-free) expression (Kay 1977), (b) the transition from clause-stringing to clause-embedding (Kalmár 1985), and (c) in consequence, the potential for Subjacency to apply for the first time (Newmeyer 2002). These issues can be accommodated within a single account, if we eschew the 'word' as the recombinable unit, in favour of the more inclusive 'lexical unit' (see below). We propose that humans naturally apply a pattern-recognition procedure to linguistic input, but are not naturally predisposed to select a consistent unit size (Peters 1983). They home in on phonological forms associated with effects that they need to achieve, e.g. object-naming, expressing a feeling, manipulating someone, carrying out a social function, conveying a nuance of meaning, narrating a traditional story. The units in their lexicon are, thus, variously morpheme-, word-, phrase-, clause-, and text-sized (Wray 2002a). A language that increases its autonomy (e.g. under pressure to be comprehensible to outsiders, and learnable by adult non- native speakers) will promote in its users and, indirectly, child learners, smaller units overall, altering the lexical balance and creating new opportunities for novel expression. But it is non-autonomous, isolated, esoteric languages that we may assume most closely resemble those of our early modern ancestors (Newmeyer 2002), and research here is consistent with the individual's lexicon typically containing a relatively small inventory of individual words for common objects and a lot of complete phrases with social functions (Laycock 1979:91), often impenetrable analytically to the point of being unlearnable other than by rote (Grace 1998:71; Laycock 1979, Thurston 1987, 1989). Three things follow: (1) The lexicons of the first language users may have contained fewer discrete words and morphemes, and required fewer grammatical rules to combine them, than most languages today. (2) Since forms acquired holistically need not be logical in construction (even etymologically) there need not have been an original inventory of morphemes and combinatory rules from which the first lexical units derived. Rather, the first fully human language(s) could have been fed by pre-existing sound sequences holistically associated with semantically complex messages (Wray 1998, 2000, 2002b; Kirby 2000). (3) Full compositionality is not a property that we have to account for at the dawn of language: it has developed in response to particular social, political and cultural applications of language over time (Grace 2002a,b). References Chafe, W. 1985. Linguistic differences produced by differences between speaking and writing. In Olson, D.R., Torrance, N. & Hildyard, A. (eds.) Literacy, language and learning. Cambridge: Cambridge University Press, 105-123. Grace, G. 1998. Some puzzles that arise from the assumption that to learn a language is to construct a grammar. In Janse, M. (ed.) Productivity and creativity: studies in general and descriptive linguistics in honor of E.M.Uhlenbeck. Berlin: Mouton de Gruyter, 69-81. Grace, G. 2002a. Collateral damage from linguistics? 1: The post-Chomskyan paradigm and its underlying assumptions. Ethnolinguistic Notes 4, 21. http://www2.hawaii.edu~grace/elniv21.html Grace, G. 2002b. Collateral damage from linguistics? 4: Do we really know what kind of language the language acquisition device prepares us to acquire? Ethnolinguistic Notes 4, 21. http://www2.hawaii.edu~grace/elniv21.html Hauser, M.D., Chomsky, N. & Fitch, W.T. 2002. The faculty of language: what is it, who has it, and how did it evolve? Science 298: 1569-79. Kalmár, I. 1985. Are there really no primitive languages? In Olson, D.R., Torrance, N. & Hildyard, A. (eds.) Literacy, language and learning. Cambridge: Cambridge University Press, 148-166. Kay, P. 1977. Language evolution and speech style. In Blount, B.G. & Sanches, M. (eds.) Sociocultural dimensions of language change. New York: Ac. Press, 21-33. Kirby, S. 2000. Syntax without natural selection: how compositionality emerges from vocabulary in a population of learners. In Knight, C., Studdert-Kennedy, M. & Hurford, J. (eds.) The evolutionary emergence of language: social function and the origins of linguistic form. Cambridge: Cambridge University Press, 99-119. Laycock, D. 1979. Multilingualism: linguistic boundaries and unsolved problems in Papua New Guinea. In Wurm, S.A. (ed.) New Guinea and neighboring areas: a sociolinguistic laboratory. The Hague: Mouton, 81-99. Newmeyer, F.J. 2002. Uniformitarian assumptions and language evolution research. In Wray, A. (ed.) The transition to language. Oxford: Oxford Univ. Press, 359-375. Peters, A.M.1983. Units of language acquisition. Cambridge: Cambridge Univ. Press. Thurston, W. 1987. Processes of change in the languages of north-western New Britain. Pacific Linguistics B99. Canberra: The Australian National University. Thurston, W.R. 1989. How exoteric languages build a lexicon: esoterogeny in West New Britain. In Harlow, R. & Hooper, R. (eds.) VICAL 1: Oceanic languages. Papers from the Fifth International Conference on Austronesian Linguistics, Auckland, New Zealand, January 1988. Auckland, NZ: Linguistic Society of New Zealand, 555-579. Trudgill, P. 1989. Contact and isolation in linguistic change. In Breivik, L.E. & Jahr, E.H. (eds.) Language change: contributions to the study of its causes. Berlin: Mouton de Gruyter, 227-237. Wray, A. 1998. Protolanguage as a holistic system for social interaction. Language and Communication 18: 47-67. Wray, A. 2000. Holistic utterances in protolanguage: the link from primates to humans. In Knight, C., Studdert-Kennedy, M. & Hurford, J. (eds.) The evolutionary emergence of language: social function and the origins of linguistic form. Cambridge: Cambridge University Press, 285-302. Wray, A. 2002a. Formulaic language and the lexicon. Cambridge: Cambridge University Press. Wray, A. 2002b. Dual processing in protolanguage: performance without competence. In Wray, A. (ed.) The transition to language. Oxford: Oxford Univ. Press, 113-137.