Extended exponence by enrichment: Argument encoding in German, Archi, and Timucua

This paper is concerned with a principled approach to cases of extended (or multiple) exponence, as postulated in Matthews (1972, 82) for Greek verb inflection (perfect forms in particular), and in Matthews (1974, 149) for English verb inflection (among other phenomena). Extended exponence characterizes those cases of morphological realization where a single morpho-syntactic property seems to be expressed by more than one exponent (i.e., inflection marker, in the cases to be considered here). Extended exponence raises a problem in Distributed Morphology. The solution that I propose centers around a new type of rule that applies to syntactic operations before morphological realization takes place, and that I call enrichment. Enrichment rules are complementary to well-established impoverishment rules. They make it possible to account for extended exponence without invoking a concept of secondary exponence via contextual features. The empirical evidence that I discuss comes mainly from the interaction of argument encoding and number/person marking in German, Archi, and Timucua (with case-marking of nouns in the first two cases, and agreement morphology on verbs in the latter), but I also address verb inflection in Tamazight Berber. Let me begin by introducing the phenomenon of extended exponence on the basis of case-marked plural forms of nouns in German. As shown in (1), plural can be marked twice on nouns in dative (dat) contexts in German (see, e.g., Eisenberg (2000), Wiese (2000)). The inflected nouns in (1-ac) involve the cooccurrence of a plural marker (er in (1-a), which is common for neuter stems such as Kind (‘child’), and e in (1-c), which is the unmarked plural marker for masculine stems like Tisch (‘table’)), and a dat plural marker n. This n must