Bilingualism in the Community: Code-switching and Grammars in Contact by Rena Torres Cacoullos and Catherine E. Travis (review)

This volume—consisting of eleven chapters, two appendices and an index—explores the role of codeswitching in a Spanish-English language contact situation. The first four chapters of Bilingualism in the Community: Code-switching and Grammars in Contact contextualize the foundation for the large-scale variationist study that comprises the core of the book and tests the hypothesis of contact-induced grammatical change.

The authors open by addressing the intersection between language variation and language contact, as they discuss their use of subject pronoun expression empirical analyses to probe whether grammatical change is occurring between English and Spanish. These languages provide an excellent testing ground; pronominal subjects are variably expressed in Spanish whereas English features overwhelming pronominal expression. Chapter 2 depicts the language contact setting of northern New Mexico, the oldest Spanish-speaking speech community in the continental United States, where Spanish has been spoken for over 400 years and English for over 150. This chapter also situates the bilingual speakers, descendants of the original Spanish-speaking settlers who arrived in the 1500s, in their social context and describes New Mexico Spanish.

Chapter 3—in itself a remarkable corpus development manual—describes in detail the constitution of the 300,000-word New Mexico Spanish-English Bilingual (NMSEB) Corpus. This corpus, which masterfully captures unelicited codeswitching in spontaneous speech, is unparalleled in both the amounts of the bilinguals' languages and codeswitching it contains. Chapter 4 provides a composite sociolinguistic profile of the northern New Mexico bilingual speaker resulting from several information sources including speakers' language preference self reports, their bilingual experience, and NMSEB corpus recordings analyses. This composite profile offers a reliable alternative to traditional measures such as questionnaire responses or proficiency tests.

As they explore Spanish first and third person singular subject pronoun expression (SPE) in chapter 5, Torres Cacoullos and Travis reconsider and reinterpret in cross-linguistic perspective the main SPE constraints. Their groundbreaking study finds broader variation principles behind familiar probabilistic constraints. Subject continuity and priming constitute the strongest SPE predictors. Pronominal expression is favored at greater distances from the previous mention of the subject referent and by previous mention in pronominal form. SPE is also conditioned by clause type, Tense-Mood Aspect (TMA), verb class, and grammatical person. Non-coordinate main clauses favor overt subjects whereas perfective aspect and dynamic verbs favor null subjects. 1sg subjects occur more often with greater distance between them while 3sg subjects cluster close together. Further, the analysis uncovers interaction between subject continuity and priming as well as between verb class and TMA.

Searching for interlingual differences, the SPE analysis extends to English in chapter 6. Although null subjects are rare in English, parallels with Spanish include a coordination effect. Higher rates of unexpressed subjects occur with prosodic and syntactic linking between coreferential-subject clauses in both English and Spanish, demonstrating a similarity across the languages in what was considered an English-specific feature. Findings suggest that primary loci of cross-language differences are language-specific restrictions to the variable context, relative magnitude of probabilistic constraints, and lexically particular constructions. The languages are distinguished by their envelopes of variation with unexpressed subjects outside of coordinate clauses limited to prosodic-initial position only in English. Moreover, linking between clauses constitutes the strongest probabilistic predictor in English but not in Spanish.

Chapter 7 explores whether the grammar of Spanish has changed by being in contact with English. Two sets of comparisons with the NMSEB data—one with an earlier stage New Mexico variety and the other with a monolingual variety—yield positive evidence for grammar stability rather than change by revealing the same direction of effects. Subject continuity impacts tense, priming, and genre effects in the Spanish of both monolinguals and bilinguals. Thus, findings reveal no evidence of (contact-induced) change in New Mexico Spanish. Chapter 8 tests two sets of convergence predictions by comparing contact and non-contact varieties of...