The present paper treats the expansion of the Preterit in the Spanish spoken in Buenos Aires, Argentina (henceforth Rioplatense). Using both primary sources (interviews) and secondary sources (literature), I present evidence which suggest that the perfect/preterit distinction was alive in Rioplatense ca. 1900, and that it has gradually decreased in the course of four generations. Today, the Perfect is almost completely absent in young speakers of Rioplatense.
These findings are particularly interesting in the light of Bybee et al.’s (1994) Theory of Paths, which predicts that “the source meaning uniquely determines the grammaticization path the gram will follow in its semantic development” (1994:12). For instance, the source meaning ‘resultative’ is expected to generalize to perfective or simple past. Whereas this is the case for most Romance languages, it is not the case for Rioplatense. In fact, the Preterit seems to be generalizing to perfective, this way erasing the distinction between the preterit and the perfect.
On this basis, I suggest that a major morphosyntactic change has occurred in Rioplatense. Also, I concentrate on two additional findings in my data. One is the possible explanation of the phenomenon as the expansion of a highly frequent construction (viste, ‘see’ 2SG.PRET). I argue, with foundation in usage-based grammar, that this single construction may have been the starting point for the expansion of the entire preterit category. I also study the possible grammaticalization of the adverb ya (‘already’). Cross-linguistically, this is a common source of perfect markers, and I present evidence that suggest that a parallel change may be taking place in Rioplatense.