The Linguistic Association of Korea Journal, 21(1), 29-45. Chierchia(1998) develops an influential semantic theory called Nominal Mappong Parameter, which predicts that a language with a generalized classifier system does not have plural making. He claims that common nouns in classifier languages do not differentiate singular entities from plural entities in the syntax just like English mass nouns and, as such, they are bound to have obligatory classifier systems and no plural making. In this article, I argue against Cheirchia`s view and claim that a generalized classifier system in a language should not be taken to show that the language has a mass lexicon. My argument is based on data drawn from Korean in which a generalized classifer system coexists with fairly productive plural marking.