Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
935492 | Lingua | 2012 | 24 Pages |
Based on the insight that a measure word (M) in [Num (Adj)-C/M N] is semantically substantive, while a classifier (C) is redundant and does not block modification or quantification to N (Her and Hsieh, 2010), this paper proposes a distinction of C/M from a mathematical perspective. Synthesizing the concepts of parceler (Landman, 2004), divider (Borer, 2005), and multiplicand (Au Yeung, 2005 and Au Yeung, 2007), I follow Her (2010) and contend that while C/M both function as a multiplicand mathematically, C's value is necessarily 1 and M's is not, thus ¬1. This offers a natural explanation to the semantic tests developed in Her and Hsieh (2010). Implications are discussed for these areas: typology of classifiers and classifier languages, correlations between numeral systems and the employment of C/M, the universal count/mass distinction at the lexical level, and first language acquisition of classifiers and numbers.
► A distinction between classifiers (C) and measure words (M) from a mathematical perspective is proposed. ► C/M both function as a multiplicand in [Num C/M N], where Num is the multiplier. ► C/M differ in value: C is necessarily 1, and M is not and can thus in principle be any mathematical value. ► Implications are discussed for typology of classifier languages and numeral systems, universal count/mass distinction, and language acquisition.