Jiang colloquium

February 28
Rosenwald 011
Harvard University
Articles, classifiers, and the theory of argument formation

This talk is about nominal argument formation in classifier languages (ClLs), languages with a typically extensive inventory of ‘measure words’ that must be used in combining a numeral with any noun, and its consequences on nominal argument formation in general. A universal property of nouns across languages is their ability to serve as the arguments of predicates. How nouns become arguments, however, has been subject to much debate. Some authors claim that a noun must co-occur with a determiner in order to serve as an argument (e.g. Longobardi 1994 et seq; Borer 2005). Other authors claim that whether or not nouns require determiners when serving as arguments is a language-specific property (e.g. Chierchia 1998; Dayal 2004; Bošcović 2005 et seq). Another way to conceive of this debate is as a difference in whether or not determiner phrases (DPs) are universally projected across languages. This topic has been widely investigated in number marking languages (NMLs), languages where number morphology is obligatory when a numeral combines with any count noun. Unfortunately, focusing only on NMLs leaves the landscape of nominal arguments incomplete, and worse, ignores facts relevant to the universal DP hypothesis. In this talk, I consider how the facts from ClLs fit into and inform the debate about the universal DP hypothesis.

My discussion focuses on two types of ClLs. The first type, exemplified by Mandarin, does not have any overt evidence of the syntactic category D. The second type, exemplified by Yi (a Sino-Tibetan language with SOV word order), is previously unattested and in fact predicted to not exist (e.g. Chierchia 1998): this is a ClL with an overt article-like definite determiner, distinct from other members of the determiner family like that or this. The discovery of a ClL with an overt article-like determiner may at first seem to tilt the balance in favor of the universal DP hypothesis: having found overt Ds in one ClL, one could argue by analogy in favor of (null) determiners everywhere. However, I will argue that the opposite is the case. I will, in particular, argue for the following three points. First, numeral constructions have identical syntax and semantics in ClLs and NMLs (possibly universally). Second, language variation in the nominal domain is due primarily to two interrelated factors: what nouns denote (kinds or properties) and what low functional heads (i.e. number morphology and classifiers) denote. Third, determiners in ClLs are in fact expected, contrary to the standard view, but while they can combine with numeral-classifier phrases (ClPs) and numeral-less ClPs, they can never combine with bare nouns. The proposal is that bare nouns in ClLs are always argumental regardless of whether or not there are overt article-like determiners. In the end, I show that the developed analysis of nominal arguments and language variation yields an updated language typology of argument formation.