Vol. 3. No. 3 R-16 September 1998
Return to Table of Contents Return to Main Page

Puerto Rican Discourse: A Sociolinguistic Study of a New York Suburb

Lourdes Torres (1997)
Mahwah, NJ: Lawrence Erlbaum Associates, Publishers
Pp. xx + 138
ISBN 0-8058-1931-2 (paper)
US $16.95
(also available in cloth: ISBN 0-8058-1930-4; US $32.50)

During the past twenty years, there has been a growing body of sociolinguistic research in Puerto Rican speech communities in the U.S., most dealing with New York City neighborhoods. In Puerto Rican Discourse, Lourdes Torres takes us beyond the city with her in-depth study of the teenagers and adults of the Puerto Rican community in suburban Brentwood on Long Island, NY.

Torres, a Puerto Rican sociolinguist currently teaching at the University of Kentucky, was once a resident of Brentwood and possesses extensive first-hand knowledge of the linguistically dynamic town. Her slender but packed volume (apparently a condensation of her 1988 Ph.D. thesis) takes a methodologically integrated approach to ascertaining the degree of language maintenance and shift that exists in this stable middle- and working-class minority community.

Chapters 1 and 2 provide us with a detailed ethnographic description of language practices and attitudes, with data obtained via three years of participant observation and a 50-item language attitude survey completed by 371 residents. Chapter 3 examines the discourse structure of oral narratives in Spanish produced by 30 speakers (evenly divided among three different generations) during interviews with the researcher. The verb morphology and syntactic complexity used by the speakers are carefully scrutinized and compared across generations and with other Hispanic groups. In chapter 4, these same narratives are submitted to quantitative analysis of the mixing of English and Spanish elements. Chapter 5 considers the ideological content of the narratives with a focus upon the speakers' internalization of and resistance to racism and sexism. Finally, chapter 6 speculates upon the linguistic and sociopolitical implications of the findings.

Many of the linguistic studies of Hispanic communities in the U.S. operate under the assumption that contact with English has deformed the Spanish spoken there. Torres informs us at the outset that "I approach the study of U.S. Puerto Rican Spanish with the understanding that what I am examining is not a corrupted language, but rather a variety that is evolving in a restricted context" (p. xii). Her goal is to go beyond deficit theories to seek out the "innovative and inventive strategies Puerto Ricans and other Latinos use to express themselves in the codes of their linguistic [-1-] repertoire" (p. 123). A secondary goal is to debunk the myth that Hispanic communities in the U.S. are all the same. Her study points to important differences between the speakers of Brentwood and those of other Latino enclaves.

Among Torres' most salient findings are the following:

Spanish is widely used in the community, particularly in shops, social service organizations, churches, bilingual programs, and homes. Most individuals have at least a passive knowledge of the language. It is used more in working class families, while English is more frequent in middle class families. It is the preferred language for speaking to parents and spouses. There is strong emotional loyalty to Spanish, although it is not seen as the defining feature of being Puerto Rican.

English is highly regarded for instrumental reasons (education, employment, etc.), and is used in all formal public meetings. It is the preferred language for speaking with friends, workmates, and job superiors. Students use English almost exclusively when speaking to their siblings.

Most residents (61% of adults and 81% of teenagers) report using both languages rather than either Spanish or English exclusively, and both languages coexist across social domains. The great majority oppose English-only legislation (which was proposed and defeated in Brentwood) and support bilingual education (which exists only in a very limited form in the community at present).

First generation speakers (seniors born in Puerto Rico who came to New York as teenagers) use Spanish extensively, and their English skills tend to be weak. Second generation speakers (middle aged individuals who came to New York before age five or were born in New York) are bilingual and function competently in both Spanish and English. Third generation speakers (teenagers or young adults born in Brentwood) are English-dominant and vary considerably in their Spanish skills. Nevertheless, the traditional immigrant pattern of mother tongue loss by the third generation is not operative. Young Puerto Ricans still use Spanish in the home, especially with older relatives, although they use it exclusively less frequently than their elders. Many reactivate passive Spanish skills when they become adults and are integrated into Spanish-speaking social networks.

Code-mixing, or the alternation of the two languages, is very common; 82% of the adults and 77% of the students claim to do it, but many feel it is a bad habit. They identify the practice as one of the reasons their speech is stigmatized.

Analysis of the structure of the Spanish narratives reveals that despite the fact that they live in a setting where Spanish is used [-2-] less often than in a monolingual context, the narratives of the English-dominant generation are as well developed and syntactically complex as those of the Spanish-dominant and bilingual generations. There are slight differences in the tenses favored in different sections of the narratives; however, all speakers share similar rules for producing the narratives. Contrary to evidence from studies of other Hispanic communities, there is no pattern of declining subjunctive use across generations. The English-dominant generation has some variation in the conditional structures, but so do the Spanish-dominant speakers. Examination of other structures (periphrastic verbs, subordinate clauses) does not reveal the simplification and reduction found in some Mexican American communities. In short, the most striking difference in the Spanish narratives of the three generations is the somewhat increased use of English words or phrases by the English-dominant group. Otherwise, they are quite similar.

Analysis of code-mixing practices reveals that despite the categorical perception that Puerto Ricans are constantly confusing their two languages and speaking "Spanglish," English single-word borrowings account for only 1.9% of the total word count in the narratives, and at the clause level, no group switches to English for more than 10% of the clauses. Phonologically or morphologically unintegrated loanwords are the most common type of borrowing for all speakers but especially for the Spanish-dominant generation. The bilingual generation favors unintegrated loanwords along with phrasal calques (where the semantic components of a given phrase are translated literally into the borrowing language). The English-dominant generation prefers phrasal calques, unintegrated loanwords, and integrated loanwords. Contrary to expectations, the English-dominant speakers are seven times more likely than Spanish-dominant speakers to integrate loanwords into Spanish. The percentage of switched discourse markers like so and y'know increases as the English proficiency of the speaker increases.

Torres feels that the English-dominant speakers are the most innovative in their integration of English into Spanish. As she puts it, "From a prescriptive point of view, one might argue that Brentwood Puerto Rican bilingual and English-dominant speakers are corrupting Spanish. However, a perspective that accepts U.S. Puerto Rican Spanish as a variety that is flourishing in a new environment interprets the described changes as an expected result of natural language evolution" (p. 71).

In terms of the discourse function of the code-mixing, Torres' data reveal that while Spanish-dominant speakers and English-dominant speakers often switch to English to fill in lexical gaps they may have in Spanish (especially technical lexical items learned in English), fully bilingual speakers utilize code-mixing as an optional discourse strategy to enhance and enliven their narratives. The points in the narratives containing the most mixing are the [-3-] evaluation portions and the quoting of direct speech, both of which make stories more interesting and convincing.

In her analysis of the ideological aspects of the narratives, Torres demonstrates how the stories reveal the values, goals, and identities of the speakers, including in some cases, the internalized prejudices of the majority society which creates feelings of self-hatred among minority members. She cites extensively from the narratives to illustrate how her informants sometimes buy into negative stereotypes regarding Puerto Rican language or culture, castigating themselves in the process, and how they resist or fight back at other times. Similarly, she reveals how Puerto Rican women deal with conflicting feelings about gender relations and how the youngest generation is the most articulate in voicing opposition to sexist practices and attitudes.

Overall, Torres' study is impressive in its comprehensiveness, attention to detail, and fidelity to both her intellectual and cultural roots. She has single-handedly carried out research that is normally done by a team of fieldworkers. For the most part, her analysis is compelling and well-reasoned. As a researcher who has also worked in Puerto Rican neighborhoods, I find her work to be solid and thought-provoking. It certainly confirms what I have long believed about the inherent variability within and among Hispanic speech communities.

There are some areas in which I disagree with her approach. For example, in the chapter on code-mixing, her unintegrated loanwords are to me code-switches, not loans, and I am not totally convinced by the data which show the Spanish-dominant speakers to be using high levels of unintegrated loanwords. I question her criteria for integration, since in my experience in North Philadelphia, East Harlem, and now in Puerto Rico, Spanish-dominant speakers have always been the ones who have phonologically and morphologically integrated English loanwords into their native system due to their very limited access to English phonology and morphology. It should be noted that in this same section, there is one rather jarring error which should be corrected in any future edition of the book. On page 67, table 4.2 is missing an entire line (the results for unintegrated loanwords excluding discourse markers). The line for the integrated loanwords reports some of the numbers which should be on the missing line. As it stands now, the table makes no sense.

In the chapter on the verb structures within the Spanish narratives, I can't help but wonder why Torres did not compare her findings to the results Shana Poplack and I obtained in our 1981 study of the Puerto Rican verb system in East Harlem (listed in her bibliography). Finally, in the chapter on ideology, I missed any reference to Catherine Walsh's excellent analysis of Puerto Rican teenage discourse structures (1991) or to Bonnie Urciuoli's analysis [-4-] of race and class ideologies in the Puerto Rican community of the Lower East Side (1996).

These points aside, I would heartily recommend the book to anyone interested in language contact situations, or to readers involved in comparative studies of U.S. Hispanic communities.


Pousada, A. & Poplack, S. (1981). No case for convergence: The Puerto Rican verb system in a language contact situation. In G. Keller & J. Fishman (Eds.). Bilingual education for Hispanic students in the United States (pp. 207-237). New York: Teachers College Press.

Urciuoli, Bonnie. (1996). Exposing prejudice: Puerto Rican experiences of language, race, and class. Boulder, CO: Westview Press.

Walsh, Catherine. (1991). Pedagogy and the struggle for voice: Issues of language, power, and schooling for Puerto Ricans. Westport, CT: Greenwood Publishing.

Alicia Pousada
University of Puerto Rico

© Copyright rests with authors. Please cite TESL-EJ appropriately.

Editor's Note: Dashed numbers in square brackets indicate the end of each page in the paginated ASCII version of this article, which is the definitive edition. Please use these page numbers when citing this work.

Return to Table of Contents Return to Top Return to Main Page