To the Right of the Verb: An Investigation of Clitic Doubling and Right Dislocation in Three Spanish Dialects by Valeria A. Belloro (review)

To the Right of the Verb draws from pragmatics, corpus linguistics, psycholinguistics, and syntax to build an empirically based account of clitic doubling and right dislocation in contemporary Argentine, Mexican, and Peninsular Spanish. While these syntactic phenomena have been addressed in both theoretical (Chafe, Lambrecht, Suñer) and corpus studies (Silva-Corvalán, Weissenrieder), Belloro's approach builds upon prior research in two ways: 1) by considering data from three dialects in order to control for and document dialectical change; and 2) by arguing that cognitive accessibility drives the observed variation. While the book draws from Lambrecht's work on identifiability to provide a detailed taxonomy of doubling and right dislocation structures for both dative and accusative clitics, this review will focus on accusative clitic doubling, as this case provides a representative example of the empirical methods used, in addition to serving as the basis for the book's key theoretical claims.

As many readers will already know, clitic doubling refers to sentences in which both clitic and referent are explicitly stated in the same clause. In Spanish, this is common with dative clitics, as in "Se lo di a Juan." Right dislocation, on the other hand, refers to cases in which both clitic and referent are explicit, but where the referent appears after the right bound of the clause containing the clitic, as in 'La llamaron enseguida a una enfermera.' Some readers will also be aware that accusative clitic doubling containing a full NP (e.g., "Lo empujé a mi hermano") is far more common in certain dialects than in others. According to the eight corpora used in the study, this structure represents just 14% of doubled constructions in Peninsular Spanish, but accounts for 63% in Mexican Spanish and fully 82% in Argentine Spanish.

In order to account for this cross-dialectical variation, Belloro proposes that the three dialects represent different stages of grammaticalization of the clitic. Following Chafe, she argues that cognitive accessibility, or the degree of activation of the referent for the speaker, is the determining factor behind which form is produced. Belloro presents numerous examples suggesting that Argentine Spanish is not restricted to a dichotomous expression of topicality that selects either clitic or full NP; instead, this dialect offers speakers a third option in which both forms are present. In this case, the clitic indicates that the referent is topical while the NP serves to disambiguate it from other, more discourse-recent, potential referents. In this light, accusative doubling provides a syntactic construction that indexes an intermediate gradient of cognitive accessibility.

Because this study presents its arguments with corpus data and syntactic analysis, perhaps it is appropriate to applaud and critique it along those same two axes. Regarding the first, the use of ample empirical data that is representative of different dialects adds significant weight to Belloro's claims and helps to explain some of the discrepancies between previous studies on the subject. However, perhaps because Belloro explicitly criticizes the problems inherent with focusing on a single dialect, it quickly becomes clear that the Argentine corpus is at the center of the book, and that the Mexican and Peninsular data receive less attention. Although this may be justifiable on theoretical grounds because the phenomenon described above is most pronounced in the Argentine corpus, this focus on a single regional dialect draws attention to another fact about the study: of the eight corpora consulted, only one is from Argentina. Furthermore, it is titled "El habla culta de la Ciudad de Buenos Aires: materiales para su estudio," which suggests that it may not be representative of Argentine Spanish in general since it is restricted both geographically and socio-economically. While this does not mean that the corpus is flawed, it does introduce the possibility of bias in the study due to restriction of...