Affixes, stems and allomorphic conditioning in paradigm function morphology

Abstract

Commentary on Gregory T. Stump's Inflectional Morphology: A Theory of Paradigm Structure (Cambridge: Cambridge U Press, 2001) focuses on three empirical problems that are argued to arise from the unforeseen interaction of hypotheses in Stump's inferential-realizational model named paradigm function morphology, particularly hypotheses based on two central assumptions: all phenomena of inflection are to be described in terms of their contribution to the expression of morphosyntactic properties, & optimal morphological theories are either entirely lexical or entirely inferential. (1) Carstairs's (1987) notion of principal vs secondary exponence is defended against Stump's putative counterexample to Carstairs's peripherality constraint in an analysis of tense & person markers in Bulgarian verb inflection. (2) A distributional uniformity analysis is preferred over Stump's morphosyntactic realization analysis of the arbitrary distribution of special stems across paradigmatic category combinations in relatively large classes of Sanskrit consonant-stem nouns & Italian verbs with stem allomorphs ending in a velar stop. (3) Stump's proposed Paninian determinism hypothesis, fortified by notions of expanded rule schemata, is argued to be inadequate to account for complex combinatorial assymetries of subject & object marking in Georgian verb inflection. Tables. J. Hitchcock