Master of Arts

thesisThere has long been an assumption of a bilateral divide of languages into nonclassifier languages and classifier languages, i.e., those languages that do not use numeral classifiers when counting nouns and those that do use numeral classifiers. While this assumption holds for most languages, it cannot account for languages such as Armenian, which optionally allow for numeral classifiers to appear when nouns are counted, and also cannot account for languages such as Paiwan, a so-called poor-classifier language, which has numeral classifiers for certain noun classes and not for others. I propose that by adopting Borer's method of dealing with numeral classifiers and plural morphology as two expressions of the same underlying phenomenon, and by spelling out a specific syntactic mechanism for achieving the process outlined by Borer, we find a theory that is able to account for the full range of possibilities (i.e., nonclassifier, classifier, classifier-optional, and poor-classifier languages)