- Bilingualism in the community: Code-switching and grammars in contact by Rena Torres Cacoullos and Catherine Travis
Linguists have long asked whether one finds in bilinguals, in addition to lexical transfer, crosslinguistic influences in morphosyntax. The familiar answer given in most places is: yes (see e.g. Thomason & Kaufman 1988). But in this book the answer, in a revealing dissent, is pursued by Rena Torres Cacoullos and Catherine E. Travis (TC&T) with such seriousness and rigor as to make this substantial volume indispensable. Their answer to the question is that there is no morphosyntactic convergence in bilinguals. Not, at least, in this painstakingly carried out case study of the use of subject personal pronouns by fluent English-Spanish bilinguals in the state of New Mexico in the United States.
Subject personal pronouns in Spanish (yo, tú, ella, él, etc. 'I, you, she, he', etc.) and especially the difference between expressed versus unexpressed pronominal subjects in finite clauses (ella camina ∼ camina, both 'she walks') represent a mature research topic (Carvalho et al. 2015). Most studies, like the authors', have been conducted on the basis of speech data under the quantitative variationist paradigm first established by William Labov (1966). Also comprising a large variationist literature is the related matter of the possible influence of English on the use of these pronouns in the Spanish of bilinguals. But scholars plowing this field have mostly concentrated on the presumed target of contact influence, Spanish, and not enough on the presumed source, English. This is because while the expression of subjects is obviously variable in Spanish, it is typically assumed to be essentially categorical in English. Yet TC&T make a point that is no less crucial for its being obvious: unexpressed subjects are also found in English (e.g. He .. took off for .. Big Bear. // You're kidding. // 0 Had no idea where he was going (112); the '0' indicates an unexpressed pronoun).
There are other important differences between this book and most work on English-Spanish bilingualism in the US. First, the speakers here are almost all US-born and belong to a 150-year-old bilingual community, thus differing from those of the relatively recent immigrant settings studied in most US bilingual research. This speaker profile facilitates conclusions about long-term linguistic change that are harder to reach when a time depth of only two or three generations is considered. Second, the very frequent code-switching that marks the normal speech mode of many bilingual settings is especially noticeable in the corpus utilized here. This allows TC&T to analyze pronominal use in stretches of both languages produced by the same speakers. In this way the authors take into account more directly than other investigators the familiar dictum that the true locus of crosslinguistic influence is the bilingual individual. And since, more than in [End Page 575] other studies, English here thrives in the mind of a single speaker as much as does Spanish, the authors can expose their central claim of no contact effects to the highest risk of falsification, increasing the value of their conclusions. Moreover, the authors can dismiss the claim that the incidence of crosslinguistic influence of English on Spanish is greater in stretches of Spanish located in the vicinity of English stretches. And of special interest to the wider field of bilingual studies, the characteristics of the corpus lead TC&T to note that our confidence is misplaced in such notions as 'the native language', 'the matrix language', 'the second language', or more generally in what they correctly discount as the belief in 'the regime of ordered first and second language acquisition' (66).
A third distinctive trait of the study is that the corpus comes with a prosodic transcription, crucially incorporating into the analysis intonational units, or IUs, that are not always found in this literature. A fourth and final differentiating feature is that the English and Spanish stretches of the bilingual corpus are contrasted with two different English...