Re-examining the mass-count distinction
This paper argues that the mass-count distinction does not represent a fundamental division between the world's languages. We demonstrate that such a distinction, as commonly defined within the linguistic literature, often conflates two facts: the semantic fact, found in all languages, that some words have atomic denotations and some do not, and the morphosyntactic fact, found in languages with contrasting singular-plural morphology, that some nouns have both singular and plural forms while others have only one such form. By comparing English with Mandarin Chinese, we discuss whether this morphosyntactic distinction might correlate with the presence or absence of a rich classifier system (as well as other types of quantification). This potential correlation has greatly influenced how linguists have investigated nominal systems across languages and it has even led some to hypothesize that morphosyntactic subcategories might determine the ways in which a grammar can “count” and “quantify.” We outline some important exceptions to this proposed correlation in languages such as Ch’ol, Mi’gmaq and Western Armenian. The paper concludes by arguing not only that there is no such correlation, but that linguists should rethink how they investigate nominal systems, focusing more on lexical variation (even within a single language) than on parametric variations across languages.
Article outline
- 1.Introduction
- 2.Prototypical mass-count and classifier languages
- 2.1English: A mass-count language
- 2.2Mandarin: A classifier language
- 3.
Moving away from the prototypes
- 3.1Western Armenian
- 3.2Ch’ol and Mi’gmaq
- 3.3The case against parameters
- 4.
Conclusion
-
Acknowledgment
-
Notes
-
References