Remarks on Nominal Inflection in German

Any theory of nominal inflection has to fulfil two tasks for a given language. First, the inventory of inflection markers has to be determined; and second, the syntactic distribution of these markers has to be accounted for. Typically, the latter task is handled within the syntactic component of a grammar, whereas the former task is addressed in a morphological (or lexical) component that is preor postsyntactic, and that relies on the notion of paradigm as an inventory of inflection markers that exists independently of syntax. The analysis of nominal inflection in German that I want to develop in this article deviates from standard approaches in that it is based on the assumption that both the form of inflection markers, and their distribution, is exclusively determined in the syntactic component of grammar. Consequently, the notion of paradigm emerges as an epiphenomenon without theoretical significance. Much recent work on nominal inflection in German incorporates the idea that competition plays an important role in nominal inflection, in the guise of the Specificity Principle that forces the choice of the most specific form among a class of competing candidates that are all otherwise well formed.1 The Specificity Principle has been employed both for the determination of the inventory of inflection markers (see Blevins 1995, Wunderlich 1996; 1997a; 1997b, and Wiese 1996; 1999), and for the distribution of inflection markers in syntax (see Gallmann 1996; 1998). I think that a competition-based approach is indeed on the right track. However, I will move from a specificity-based to an optimality-theoretic approach (see Prince & Smolensky 1993), according to which a candidate qualifies as optimal (i.e., grammatical) if it best satisfies a set of violable and ranked constraints. The optimality-theoretic analysis of German nominal inflection will be shown to have three desirable consequences. First, paradigmatic syncretism among nominal inflection markers can be explained in its full generality. Second, by adopting the sonority hierarchy as a set of violable constraints with a fixed internal ranking, a partial motivation can be given for the make-up of the German nominal inflection marker inventory. And third, by optimizing the form and the distribution of inflection markers simultaneously, cases where the syntactic context seems to directly determine the choice of inflection marker can be accounted for. In addition to the simultaneous determination of the form and the distribution of inflection markers and the optimality-theoretic perspective, the present approach has a third conspicuous property that separates it from most existing analyses: Standardly, nominal inflection is brought about by positive constraints that demand a certain inflection marker for a given