The Handbook of Language Variation and Change

J. K. Chambers, Natalie Schilling

• Publisher: Wiley-Blackwell
• Year: 2013
• ISBN:
  9781118335512
• Edition: 2
• Topic: Languages & Linguistics
• Subtopic: Sociolinguistics
• Index: Languages & Linguistics

Reflecting a multitude of developments in the study of language change and variation over the last ten years, this extensively updated second edition features a number of new chapters and remains the authoritative reference volume on a core research area in linguistics.

• A fully revised and expanded edition of this acclaimed reference work, which has established its reputation based on its unrivalled scope and depth of analysis in this interdisciplinary field
• Includes seven new chapters, while the remainder have undergone thorough revision and updating to incorporate the latest research and reflect numerous developments in the field
• Accessibly structured by theme, covering topics including data collection and evaluation, linguistic structure, language and time, language contact, language domains, and social differentiation
• Brings together an experienced, international editorial and contributor team to provides an unrivalled learning, teaching and reference tool for researchers and students in sociolinguistics

Part I

Data Collection

1

Entering the Community

Fieldwork

Crawford Feagin

While the ultimate goal of sociolinguistic research is to resolve questions of linguistic importance, such as how language change comes about, nothing of that sort can be accomplished without first entering a community in order to collect data which will help provide the basis for any such answers. The central problem in collecting sociolinguistic data has been described by Labov as the Observer's Paradox: “our goal is to observe the way people use language when they are not being observed” (1972a: 61). Sociolinguistic fieldwork of all kinds, whether recorded interviews, participant observations or street-corner quizzes, must be geared to overcome this problem. In this chapter, I consider several well-established methods. I begin with a section on “Planning the Project,” dealing with preliminary considerations for designing and conducting a sociolinguistic survey. The heart of the chapter, as indeed of field research, is the second section on the “Sociolinguistic Interview,” the Labovian protocol for selecting informants and eliciting different styles of speech. I then consider some other elicitation methods used in sociolinguistics: participant observation and rapid and anonymous observations. While telephone surveys have been fruitful in the past (see Labov et al. 2006; Ash 2000), today they have limited use because of the general shift to cell phones, eliminating the use of area codes or telephone directories in identifying likely participants. Long-distance surveys today can utilize internet phone services such as Skype or other internet technologies (e.g. web-based surveys for gathering self-reports of linguistic production and/or information on linguistic perceptions and attitudes; see Schilling 2013); they will not be discussed here. Instead, I will focus on face-to-face methods.

1 Planning the Project

Although the methods involved are presented here as if they were sequential, in practice the various phases of fieldwork and other aspects of research are cyclical, or perhaps spiral. Investigation in one area will influence what can be done in another. An interview might provide insights about the community that can be incorporated into the protocol and produce a much better interview with subsequent informants. For instance, in my work in Anniston, Alabama (Feagin 1979), one teenager mentioned a recent snowstorm, an unexpected and exciting phenomenon in that part of the world, and so in later interviews I asked the rest of the teenagers about it. As a result, I came away with excited accounts of sledding on garbage-can tops and cookie sheets, wearing improvised boots made from plastic bags, and skidding dangerously over slippery roads. My interview protocol for the older people already included questions about a tornado that had hit Anniston 20 years before; the snowstorm provided similarly dramatic stories from an incident in the recent memories of the teenagers.

Similarly, sometimes in the course of an interview, investigators might discover an unexpected grammatical form or phonological realization. They must be attentive and flexible in order to pursue the newly discovered linguistic feature for that community.

As an aid to planning, a small-scale pilot project along the general lines of the main research will indicate more precisely what might be feasible goals and procedures. A larger consideration is that collecting data is only an intermediate goal. The ultimate goal is linguistic.

The hypothesis that motivates the project will influence how to go about collecting the data. Again, in my own work in Anniston, I hypothesized that over the three-and-a-half centuries of close contact, African-American speech would have influenced European-American grammar in the South. I therefore set out to elicit data from the white community that was parallel to Labov's African-American data from Harlem (Labov et al. 1968; Labov 1972b). Even though it turned out in large part that my hypothesis was not correct, nonetheless it was important to try to get parallel data so that a comparison would be possible.

An important guideline for fieldworkers at the planning stage is that a close analysis of a small amount of data is better than an unfinished grandiose project. With that in mind, I concentrated on the extreme generations (teenagers and grandparents) and extreme social classes (local working class and upper class), and the older rural working class (with no younger counterpart). More than that I could not handle, though ideally I would have liked to include the middle class and the middle aged, not to mention the local African-American community. However, examining only the two urban classes plus the older rural working class, using adolescents and grandparents in the city and elderly people from the country, and keeping the sample balanced in terms of gender, I was able to see change progressing through the community.

A rule of thumb in disciplines that require fieldwork is that one third of the project time will be spent in fieldwork, one third in analysis, and the final third in writing up the work. Though far from scientific, this rule provides an effective reminder of the point that time required for analysis and writing increases in a ratio of about 2:1 for each hour of data elicitation.

Competent fieldworkers have included a wide range of personality types. Because fieldwork requires face-to-face interaction, it is usually assumed that gregarious persons do best, and it seems likely that they would have an advantage, at least in getting started. Shy people might find this sort of work excruciating, especially in the beginning. However, shy people have sometimes proven highly successful in conducting interviews and obtaining data, for the simple reason that people often open up when talking to quiet people, perhaps because they find them unthreatening and perhaps because the lack of interruptions encourages them to speak at length (Schilling, personal communication).

1.1 Library research

Once the community has been selected for research, the next step is to get a perspective on the community itself – linguistic, demographic, and historical. Information on local speech, major industries, labor, religious institutions, communications, movement of peoples, and the historical development of the area can aid in understanding local society.

A survey of previous linguistic work must be carried out, both on the linguistic aspects you intend to study and on any previous research concerning the local language variety. Earlier work on the local variety, regardless of its quality, can be useful for time depth or for pinpointing interesting problems.

First-hand accounts of fieldwork can be found in Labov (1966), Feagin (1979), Milroy (1980), Dayton (1996), and Eckert (2000) for linguistics, and in Whyte (1943, 1984) and Liebow (1967) for ethnography. Such personal accounts are rarely published, but dissertations often include them in chapters on methodology. More general discussions may be found in Labov (1972a, 1984), Wolfram and Fasold (1974), Milroy (1987), Romaine (1980), Baugh (1993), Milroy and Gordon (2003), Di Paolo and Yaeger-Dror (2011), and Schilling (2013). For sociolinguistic fieldwork in non-Western societies where the investigator is clearly an outsider, see Albó (1970), Harvey (1992), Wald (1973), and Bowern (2008). Obviously, a different set of problems arises when the fieldworker is a foreigner, of different ethnicity, and not a native speaker of the language. While addressed to researchers doing basic linguistic fieldwork (rather than sociolinguistic research) in non-Western languages (frequently in remote areas), Samarin (1967) provides an overview of linguistic fieldwork, though now somewhat dated. Bowern (2008) is a more recent resource.

1.2 Ethnography

Along with gathering linguistic data, it is important to study the community itself in situ. While material collected from library research must not be overlooked if it is available, the researcher in the field must begin by observing the physical layout of the place, who lives where, who associates with whom, and in what situations particular people associate with each other. While this type of research can be seen in Fischer (1958) and more elaborately in Labov (1963), subsequent studies have become more sophisticated and more detailed, culminating in Eckert's intricate study of a suburban Detroit high school (Eckert 2000). It is through a thorough knowledge of both the structure and dynamics of the local community that the patterning and social meanings of language variation and change in the speech community can be fully understood. While some linguists have criticized sociocultural investigations as outside the competence of linguists who are not specialists in sociology or anthropology (Bailey 1996), the only way some aspects of language behavior can be understood and analyzed is through such an undertaking.

It was through such a study that Labov was able to show that younger people on the island of Martha's Vineyard who had decided to remain on the island after their high school years were picking up the fishermen's pronunciation of (ay) and (aw), regardless of their social class, while those who had decided to leave the island for further education and employment were shifting toward mainland speech norms (Labov 1963). Similarly, Eckert (2000) was able to show that the social division between “jocks” (middle class) and “burnouts” (working class) in suburban high schools played a role in transmitting urban Detroit features into suburban teenage speech. See Eckert (2000: Chapter 3) for a valuable account of the process of studying the ethnography of a community.

1.3 Linguistic variables

In a quantitative study of linguistic variation, acquaintance with previous work and perhaps a pilot study should help to narrow the focus of the project. In practical terms, however, this does not always take place right at the beginning. What needs to be isolated before analysis can begin, and preferably before data-gathering begins, is a selection of linguistic variables to be studied. As with fieldwork more generally, though, the process is iterative, and it may turn out that the variables one originally sets out to study are not of great sociolinguistic interest, and more important features may be revealed as fieldwork progresses.

The linguistic variable, a concept originating with Labov (1963, 1966), is a linguistic entity which varies according to social parameters (age, sex, social class, ethnicity), stylistic parameters (casual, careful, formal), and/or linguistic parameters (segmental, suprasegmental). Usually the social and stylistic variation will be coordinated in some way, so that the casual speech of an accountant will be similar to the formal speech of a plumber – though that remains to be seen in the course of the investigation.

The linguistic variable can be found at all linguistic levels: most common are phonological, such as, for example, (r) might be realized as [ɹ] or as [ə] in a community which has been r-less and is becoming r-ful; morphophonological as in (ing), the English present participle marker which has two common pronunciations, standard [ɪŋ] and casual [ɪn]; morphological as in the realization of the past tense form of dive either as dived or as dove; syntactic as in the realization of negated be variously as ain't, isn't, 's not, is not; or lexical as in the use of either hero or grinder as the word to designate a particular kind of sandwich. The most frequently studied variables are phonological and morphological.

The main criterion for determining the set of variants of a single variable is that the referential meaning must be unchanged regardless of which variant occurs. (This can present a problem when dealing with grammar, as pointed out by Lavander...

Table of contents

• Cover
• Blackwell Handbooks in Linguistics
• Title page
• Copyright page
• Dedication
• Illustrations
• Contributors
• Preface to the Second Edition
• Studying Language Variation: An Informal Epistemology
• Part I: Data Collection
• Part II: Evaluation
• Part III: Linguistic Structure
• Part IV: Language and Time
• Part V: Social Differentiation
• Part VI: Domains
• Part VII: Contact
• Part VIII: Sociolinguists and Their Communities
• Postscript
• Index