The Routledge Handbook of Second Language Acquisition and Sociolinguistics

Kimberly Geeslin

• Publisher: Routledge
• Year: 2022
• ISBN:
  9781000549775
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

This state-of-the-art volume offers a comprehensive, accessible, and uniquely interdisciplinary examination of social factors' role in second language acquisition (SLA) through different theoretical paradigms, methodological traditions, populations, contexts, and language groups. Top scholars from around the world synthesize current and past work, contextualize the central issues, and set the future research agenda on second language variation, including languages studied or taught less commonly. This will be an indispensable resource to scholars and advanced students of SLA, applied linguistics, education, and other fields interested in the social aspects of language learning in research practice and instruction.

PART I Theoretical and Conceptual Foundations in the Study of Second Language Acquisition and Sociolinguistics

1 VARIATIONIST APPROACHES TO SECOND LANGUAGE ACQUISITION

DOI: 10.4324/9781003017325-2

Robert Bayley and Chelsea Escalante

Introduction

The study of language variation and change in the modem sense, or variationist sociolinguistics, and the study of second language acquisition (SLA) developed in the 1960s and 1970s. Research in both fields was concerned with demonstrating the systematicity of language varieties, often varieties spoken in marginalized communities, where variation had previously been considered random or the result of error. In sociolinguistics, a great deal of pioneering research focused on demonstrating the systematicity of working-class and minority speech, especially the speech of African Americans (e.g., Labov, 1972; Wolfram, 1969). In SLA, a considerable amount of early research also focused on speakers who were marginalized, including immigrants in Boston (e.g., Schumann, 1978) and “guest workers” in Germany (e.g., Klein & Dittmar, 1979).

The variationist paradigm was soon applied to both classroom and community-based studies of second language speech in order to examine the variation that occurs in learners’ interlanguage¹ as they move to different situations or carry out different tasks (James, 2007). Early work included Dickerson’s (1974, 1975) studies of variation in the pronunciation of Japanese adult learners of English, Gatbonton’s (1978) study of the pronunciation of English interdental fricatives by French Canadian adults. Wolfram and Hatfields’s (1984) community-based study of the acquisition of verbal morphology by Vietnamese speakers of English in Virginia, and Young’s (1991) study of the acquisition of plural marking by Chinese learners, to name just a few. While these early studies of variation in SLA were relatively rare, and most focused on English as a target language (e.g., Bayley & Preston, 1996), the situation has changed in recent decades and numerous studies of a range of languages have appeared. Studies include work on the acquisition of both obligatory and variable features in Canadian and hexagonal French (e.g., Howard, 2012; Mougeon et al., 2010; Regan et al., 2009), Chinese (Li, 2010), Norwegian (Lybeck, 2002), Spanish dialects in a number of different countries (e.g., Escalante, 2018a; Geeslin, 2002; Geeslin & Gudmestad, 2010; Pozzi & Bayley, 2021; Solon et al., 2018), and Vietnamese (Nguyen & Macken, 2008), among other languages. The earlier studies aimed to assess the numerous factors, both linguistic and social, that influence learners’ acquisition of particular structures that are usually considered obligatory in the target language, e.g., /s/ plural marking in English (Young, 1991). Although research on variation in the use of obligatory fomrs has continued, later studies began to examine the acquisition of target language patterns of variation, e.g., variation between the velar and alveolar fomrs of ING (Adamson & Regan, 1991).

This chapter outlines the theoretical and methodological basis of the sociolinguistic study of language variation. Then, using representative studies of a variety of second languages (L2s), the chapter illustrates how this approach can inform SLA theory and research methods. Finally, the chapter offers suggestions for future research, with emphasis on studies that focus on perception as well as production, studies of populations that have received relatively little attention, longitudinal studies that combine traditional variationist methods with ethnographic observation, and studies of heritage language learners.

Variationist Linguistics: Development and Key Assumptions

In the early 1960s, William Labov established the discipline of variationist sociolinguistics both in method and agenda. He made his initial observations during a visit to Martha’s Vineyard in Massachusetts, where he became curious about diphthongs in the islanders’ speech. To interpret them, he dealt with the question of where language change exists and how to investigate and describe it, the social life of language as it is spoken throughout the community, and the fundamental fact that variation in speech occurs in meaningfully systematic ways. Labov’s (1963) foundational study on Martha’s Vineyard began what has become a well-developed subdiscipline of linguistics, i.e., the study of language variation and change, the goal of which is to understand the mechanisms that link social and cultural phenomena with the internal, variable, system of language (Sankoff, 1988, p. 157). Labov’s dissertation on New York City English (1966), as well as subsequent work in Harlem on African American English (Labov, 1972), dramatically increased the size of the community studied as well as the number of variables studied and key issues examined.

The theoretical implications of the work Labov initiated are set forth most clearly in Weinreich et al. (1968). In this work, Weinreich et al. develop the concept of “orderly heterogeneity" (p. 100), which maintains that variation in language is not a consequence of random performance errors or dialect mixing, but rather an essential component of linguistic competence. That is, the variable components of language, e.g., whether a New Yorker pronounces the /r/ in words like “park”, are not random but vary systematically, subject to constraints of the linguistic environment, the social characteristics of the speaker, and the nature of the interaction. They further argue that the absence of variation would be “dysfunctional” (p. 101), since all language change is preceded by variation, although not all variation leads to language change.

The study of variation requires a clear definition of the sociolinguistic variable, the variants, and the envelope of variation. Put simply, a sociolinguistic variable is two or more ways of saying the same thing or expressing the same function. The variants are the possible forms that appear at the surface level, and the envelope of variation includes all the contexts in which the variable occurs. Consider the variable of final consonant cluster reduction (CCR), or -t,d deletion, in English. In the phrase I missed my bus, a speaker may produce the variants "ai mist mai bΛs" on one occasion and "ai mis mai bΛs" on another. Although word internal clusters may also be reduced, studies normally focus on clusters ending with /t/ or /d/ because such clusters are at the interface of phonology and morphology. Hence, all final /t/ or /d/ clusters constitute the envelope of variation. The next step in a variationist study is to determine the influences of a set of linguistic and social factors, known as the independent variables, on the dependent variable in question. In the case of CCR, the influencing factors include whether the following segment is a consonant or a vowel, whether the final /t/ or /d/ is a separate morpheme (as in the earlier example), and whether the cluster consists of two or three consonants, as in nust or æskt. CCR is also influenced by stylistic and social factors. For example, final clusters are more likely to be reduced in casual speech than in careful speech, and speakers of African American English (AAE) are more likely to reduce clusters than speakers of many other dialects (Labov, 1972; Thomas, 2007).

Many studies of language variation focus on phonology, as in the case of CCR. However, variation exists at all linguistic levels. In morphology, for example, French exhibits morphological variation in subjunctive constructions with croire “to believe", whereby speakers variably extend regular subjunctive conjugations from other paradigms (suivre, vivre), to this irregular verb:

1. Il est nécessaire qu’elle croie/croive. It is necessary that she believe.

At the level of syntax, speakers of many languages, including Chinese and Spanish, sometimes use an overt subject pronoun and sometimes omit it (Li & Bayley, 2018; Otheguy & Zentella, 2012), as shown in examples 2 from Spanish and 3 from Chinese:

• 2. Yo/Ø quiero ir a la playa. I want to go to the beach.
• 3. 他/Ø 轻快 地 下了 脚踏车。. Tā/Ø qīngkuài de xià le jiǎotàchē. He got off his bicycle easily.

The quotatives go, be all, and be like - present in several varieties of English, including Chicano English - provide an example of the use of different forms to express the same function, as in the following passage from a sociolinguistic interview with a 12-year-old Latina in south Texas:

1. Then some girl goes "eh they jumped you right?" And I was like, "Oh, my god, you had to say that!" And I was like, "No they didn’t". And she (the speaker’s mother) was all "what, what happened?" I was like, "uh nothing". She’s all, "J., you better tell me". And I had to tell her. And I go "well don’t, don’t go to my school. If I have to fight then I’ll take care of it. I’ll fight them by myself". And she goes, “Well they gave you a ticket J”. And I, she goes, "Does Miss A. (the principal) know?" So I was like, "Yes". Source: Bayley and Santa Ana (2008, p. 578)

Given the variability illustrated in the previous examples, some obvious questions arise:

1. What are the causes of the variation we observe?
2. What kinds of data are needed to understand the patterns of a community?
3. How can we model the observed variation?

Four principles have been developed to guide researchers in answering these questions. The first two are “the principle of quantitative modeling" and “the principle of multiple causes” (Young & Bayley, 1996, p. 253). The “principle of quantitative modeling” means that we can closely observe what variants surface and what features of the context co-occur with those variants. Context includes both the linguistic environment and social factors such as gender, class, ethnicity, etc. With sufficient data, we can state the likelihood of the co-occurrence of a particular form and any one of the factors of interest. These statements express in quantitative terms the strength of the relationships and allow us to assess the relative strength of the various influences on a speaker’s use of a particular variant.

The “principle of multiple causes” means that it is unlikely that variation in natural language use can be explained by a single cause. For example, the choice between an overt and a null pronoun in Spanish is constrained by whether the subject is co-referential with the subject of the previous tensed verb, as well as by person and number and the speaker’s regional origin, among other factors (Otheguy & Zentella, 2012).

Guy (1991) summarized two additional principles that characterize variationist analysis. First, individual speakers may differ in their basic rate of use of one or another variant of a sociolinguistic variable. Second, speakers who belong to the same speech community should be "similar or identical" in the factor values on the linguistic constraints on the rale (p. 5).² For example, in the case of variable subject expression, while speakers may differ in the rate at which they choose an overt pronoun, all members of the same speech community will choose the overt option more frequently when the subject differs from the subject of the preceding tensed verb than when it does not.

Methodological Issues

Data

Typically, studies of sociolinguistic variation involve the analysis of many hours of relatively informal speech collected by means of sociolinguistic interviews in the speech community. The goal of such interviews is to approximate an informal conversation so that the speaker will pay less attention to speech (Schilling, 2013). To achieve that goal, interviewees may be asked about childhood games or, most famously, if they have ever been in a situation when they thought they might die (Labov, 1984). Often, the interview also includes a number of tasks designed to promote shifts to more formal styles, such as reading a continuous passage, a word list, and a list of minimal pairs. Sociolinguistic interviews provide the data from which all examples of the variable in question are extracted.

Quantitative Analysis

In addition to being subject to multiple constraints, the type of data used in studies of sociolinguistic variation is usually unbalanced. For example, English monomorphemes ending in consonant clusters, e.g., mist, are far more common in conversation than regular past tense forms, e.g., missed. Given multiple possible independent variables and highly unbalanced data, the types of statistical procedures commonly used in experimental studies are unsuitable for studies of linguistic variation. Therefore, for many years, variationists adopted one or another version of the programs known as VARBRUL, a specialized application of logistic regression (for details, see Bay ley, 2013). VARBRUL. however, has a number of limi...

Table of contents

• Cover
• Half Title
• Series
• Title
• Copyright
• Dedication
• Contents
• List of Illustrations
• About the Contributors
• Acknowledgements
• Introduction
• PART I Theoretical and Conceptual Foundations in the Study of Second Language Acquisition and Sociolinguistics
• PART II Sociolinguistic Factors and Their Role in Second Language Acquisition
• PART III Tools for Research on Second Language Sociolinguistics
• PART IV Learner Populations and Learning Contexts
• PART V Language-Specific Research on Second Language Sociolinguistics
• Index