Caribbean Spanish = Brazilian Portuguese? Some comparative thoughts on the loss ofpro-drop

Abstract

Abstract The present contribution compares Caribbean Spanish (CS), especially the Dominican variant (DS), and Brazilian Portuguese (BP). These dialects, which show several similarities with respect to the loss ofpro-drop(and related grammatical features), are described in a framework based on Toribio (1996, 2000): both in CS/DS and in BP the inflectional head seems to partly have lost not only its capacity of identifying the contents of a null pronoun, but also the possibility to assign nominative Case in a rightward direction. Speakers of BP have replaced the morphology-based identification mechanism by a special kind of antecedent identification, and it is argued that their competence is expanded by a “grammatical bolt-on” along the lines of Rowlett (2013). The situation in DS is much more complex due to the fact that null subjects are still possible in virtually all contexts where General Spanish has them. Although antecedent identification is a plausible mechanism (attested in other languages, for instance Icelandic and Russian), there is no conclusive evidence that it is at work in DS. Furthermore, and in contrast to BP, contact with, and exposure to, an educated variety of the language can hardly be made responsible for the actual variation in DS, with the exception of the use ofelloas an overt expletive, which is disfavoured by more educated, urban speakers. The paper discusses also alternative proposals to the “competing grammars model” (cf. Toribio 2000) for DS, especially Holmberg et al.’s (2009) explanation for the partial null subject status of BP and other languages. Apart from exhibiting theoretical problems, it fails to do justice to the actual language variation in BP and is incompatible with the DS data. On the other hand, Sessarego and Gutiérrez-Rexach’s (in press) model for Chinchano Spanish (an Afro-Hispanic dialect spoken in Peru), which is based on the idea that input consisting of variable L2 outputs is responsible for variation in L1, may be promising, if similar acquisition processes can be established to have taken place in the dialects under scrutiny. For the moment, however, there is still some mystery about the partialpro-dropnature of CS/DS.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 18 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.