The morphosyntax of numerals ʥi33/ʥĩ35 'one' in Shuhi and implications for the semantics of numerals.

This paper argues that numerals as single morphemes can be formed from morphological fusion of more basic numerals and classifiers. It reports that such morphological fusion is found in Shuhi, a Qiangic language of the Tibeto-Burman branch, and discusses its theoretical implications for the ontological semantics of numerals. Specifically, we argue that there are two kinds of numerals, deficient numerals and full-fledged numerals, each demonstrating distinctive properties in syntax and semantics, and that data regarding Shuhi supports Krifka's (1995) syntax-semantics interface analysis of numeral-classifier-noun expressions, in which the numerals and classifiers form constituents excluding the nouns. We present empirical data from this language to support the Fregean view that numerals are number-referring terms (deficient numerals), while ʥi 33 'one' in Shuhi and possibly liǎ 'two-Cl' and sā 'three-Cl' in Mandarin Chinese are full-fledged numerals that should be best treated as denoting properties of cardinality. [ABSTRACT FROM AUTHOR]