Thursday - November 09th, 2006
5:00pm – 6:30pm
UNM Student Union Ballroom B
Keynote Speaker: William Croft
University of New Mexico
A Radical Construction Grammar analysis of noun phrase structure
Contemporary syntax, not only generative grammar but its nontransformational offshoots and much non-formalist syntax, retains many basic concepts and argumentation from American structuralism (and earlier approaches). In particular, it is assumed that (1) sentences are made up of constituents belonging to global categories such as N and NP (independent of specific constructions), and (2) that one must use distributional tests (occurrence in different constructions) to identify those categories and constituents. Radical Construction Grammar challenges these assumptions on the basis of grammatical variation within and across languages. Distributional tests for categories and constituents do not match, within or across languages. Hence categories are defined only in terms of the constructions in which they occur. Likewise, constituents represent formal groupings (Langacker 1997) that are specific to particular constructions, and are therefore variable within and across languages.
Here I investigate noun phrase structure by examining the claim by some linguists that what appear to be dependent modifiers of a noun in certain languages are actually appositional phrases. This claim implies a syntactic dichotomy between a noun phrase with dependents and sequences of appositional noun phrases. The claim is supported by the employment of distributional tests, in particular discontinuity, freedom of word order, 'headless' modifier phrases, prosody, and coding of gender, number and/or case on modifiers as well as nouns ('agreement'). These claims are problematic in two respects. First, they assume that the structures found in contiguous modifier+head structures are identical to those in 'headless' and discontinuous phrases. Second, they ignore crosslinguistic variation in the occurrence of the various constructions invoked to justify the apposition-dependent dichotomy. Based on a 45 language database of attributive constructions assembled with Mark Donohue, there appears to be an implicational universal hierarchy such that discontinuity ---> 'agreement' ---> 'headless' phrases. However, freedom of word order does not appear to fit into this implicational hierarchy (though it does appear to imply 'headlessness'). In other words, there is no syntactic dichotomy between apposition and head+dependent phrases. However, the implicational hierarchy has more to do with the relationship between the modifier and its head than with degrees of formal integration.
