Pro-Drop and Impoverishment

It is often assumed that some notion of morphological richness plays a central role in the theory of pro-drop: In languages with sufficiently rich verbal φ-feature (person, number, gender) agreement morphology, pronominal arguments can (and, in some contexts, must) remain without phonological realization; in languages without such a rich verbal agreement morphology, pronominal arguments must be overtly realized. Focussing on subject pro-drop here and in what follows, this difference can be illustrated with evidence from Spanish and Italian on the one hand (languages with rich subject agreement morphology), and English and German on the other (languages without rich subject agreement morphology); see (1-ab) vs. (1-cd). Following Chomsky (1982), Rizzi (1986), and Grewendorf (1989), among many others, I assume that the pro-drop phenomenon does not involve post-syntactic deletion (cf. Perlmutter (1971)), but an empty category pro. Such a non-overt pronoun pro is merged in the canonical position for subjects (in Spec of vP or within the VP, depending on its status as an external or internal argument), and undergoes Agree with T in the languages under consideration here (which rely on a nominative/accusative system of argument encoding), thereby ensuring nominative case and subject agreement; cf. Chomsky (1995; 2001).1 (1) a. [TP Hemos have-3.PL [vP pro trabajado worked todo all el the día ]] day b. [TP Ha has-3.SG [vP pro cantato ]] sung c. *I think [TP have [vP pro worked all day ]] d. *Ich I denke, think dass that [TP [vP pro gesungen sung habe ]] have-1.SG Even though the hypothesis that morphological richness is involved in the licensing of argumental pro seems to be a natural one, and is widely accepted, it has proven extremely difficult to pin down. In fact, it seems that the notion of morphological richness relevant here is left somewhat vague in most of the relevant literature. Jaeggli & Safir (1989, 29-30) assume that pro-drop is possible in languages with morphologically uniform inflectional paradigms, where an inflectional paradigm counts as uniform iff it “has either only underived inflectional forms or only derived inflectional forms”: The former option covers pro-drop languages like Spanish and Italian; the latter option covers pro-drop languages like Japanese or Chinese, which lack verbal φ-feature agreement completely; non-pro drop languages like English and French, which have bare-stem inflectional forms in their verbal paradigms, are correctly excluded. As noted by Rohrbacher (1999, ch. 5), this approach ignores the fact that base-stem forms can act as fully distinctive members of morphological paradigms (and are accordingly often analyzed as involving null morphemes in morphology that are accorded the same formal status as