The Routledge Companion to Sociolinguistics

Carmen Llamas, Louise Mullany, Peter Stockwell, Carmen Llamas, Louise Mullany, Peter Stockwell

• Publisher: Routledge
• Year: 2006
• ISBN:
  9781134303489
• Edition: 1
• Topic: Filología
• Subtopic: Lingüística

Have you ever noticed an accent or puzzled over a dialect phrase? Language can be a powerful tool with which one can create a persona; it can be a common ground between people or can be used as a divide between social groups. This Companion is for anyone who is interested in how and why people speak and write with such diversity.

The Routledge Companion to Sociolinguistics includes articles by leading scholars in the field on:

• methods of observation and analysis
• social correlates
• socio-psychological factors
• socio-political factors
• language change.

With a substantial A-Z glossary of key terms and concepts, directions for further study, and detailed cross-referencing with links to the glossary, The Routledge Companion to Sociolinguistics is both an essential broad-based introduction for those new to the field, and a highly useful reference for the more advanced linguist.

Part I

METHODS OF OBSERVATION AND ANALYSIS

1

VARIATION AND THE VARIABLE

DOMINIC WATT

DEFINITION AND EXAMPLES

In all human languages, spoken and signed, we can find examples of cases in which speakers have multiple ways of saying the same thing. Some variation is accidental and transitory; it may arise from the mechanical limitations of the speech organs, for instance, and may not be fully under the speaker’s control. Other, more systematic variations represent options speakers may consciously or unconsciously choose (Coulmas 2005). A choice between two or more distinct but linguistically equivalent variants represents the existence of a linguistic variable. Speakers in Aberdeen, north-east Scotland, for instance, may choose between the terms boy, loon, loonie, lad or laddie when referring to a young male person, or between quine, quinie, lass, lassie, or girl in reference to a young female. These sets exemplify lexical variables, and, following the convention of labelling variables in parentheses, we might refer to them as (boy) and (girl), respectively.

Variables are also found at all other levels of linguistic structure. Speakers may exploit phonological variables by choosing from different pronunciations of the same word or phrase. For example, Aberdonians may pronounce what using either the Scottish standard [M] or the (stereotyped) local form [f] (thus [fitsa?] what’s that?). Though alternation in (wh) is typically treated as binary, other pronunciations such as [w] can also be heard in the accent. As discussed in Chapter 3, phonological variables may additionally be continuous rather than having discrete, clearly distinguishable variants.

Discourse variables are used as a means of structuring discourse, such as when organizing conversational turns. Markers in English such as you know, you see, like and I mean, tags (e.g. or something, and that), or tag questions (innit, right, know what I mean, etc.) have, however, been under-researched compared with lexical and, in particular, phonological variables. The study of discourse variation is still at an early stage, and while it presents challenging problems – in what sense, for example, is an utterance ending in the tag you know ‘equivalent’ to the same utterance which lacks the tag? – the fact that such variation has been found to be systematic indicates that a full understanding of how speakers construct conversations will necessitate a good deal of further research to establish more explicitly the forms, functions and uses of discourse variables (see Schiffrin 1987, 1994; Ochs et al. 1996; Couper-Kuhlen and Selting 2001; Macaulay 2002a; Cheshire 2005a, b; and Chapter 5).

Grammatical (morphological and syntactic) variables have, on the other hand, received much more attention in the sociolinguistics literature over the last four decades, focusing on the notion of the variable rule (Cedergren and Sankoff 1974; Sankoff 1978, 1988; Sankoff and Labov 1979; Wolfram 1991). Lack of space prevents fuller discussion of the hotly debated issue of the extent to which syntactic forms claimed to be functionally equivalent are in fact (or even can be) exactly synonymous; see instead Lavandera (1978), Labov (1972b, 1978); Romaine (1982); Cheshire (1987, 2005a); Cheshire et al. (2005); and Chapter 4. Unambiguous synonymy can none the less be found. While, for instance, Aberdonian speakers very frequently use the distal demonstrative that with plural noun phrases – as in example (1) – they can also use standard those alongside the other non-standard alternatives given in (3)–(6) without any difference in linguistic meaning intended or implied (McRae 2004; Beal 1997; Smith 2005).

(1) This is enough to feed all that rabbits.

(2) This is enough to feed all those rabbits.

(3) This is enough to feed all them rabbits.

(4) This is enough to feed all thae rabbits.

(5) This is enough to feed all thon rabbits.

(6) This is enough to feed all yon rabbits.

It is of course not true that all Aberdeen speakers would necessarily use all the forms at (1)–(6): only (2) is likely if Scottish Standard English is being used, and forms like (1) and (3) might be avoided in ‘polite’ speech owing to their perception as ‘bad English’. To this extent a speaker’s choice of variant may be constrained by non-linguistic, ‘external’ factors such as the social situation (an interview in a doctor’s surgery, say, versus an argument at home), or the speaker’s educational and economic background, age, etc., these being powerful predictors of non-standard variant usage. Alternatively, a variant’s use may be constrained by an internal, linguistic factor: in Aberdeen (wh), lexical distributional constraints favour [f] in function words like what, why, where and who more highly than in content words like white, whittle or whale (see further Jones 1997: 331; Johnston 1997: 507; Smith 2005). In certain infrequent words such as whippet, whimsical, wherewithal, etc., [f] appears never to occur. When investigating alternations the domain of variability is circumscribed by eliminating those contexts in which variability is absent. Structural factors may assist. If, for example, a London English speaker uses the labiodental approximant [v] as a pronunciation of (r), s/he will obviously only do so where phonological constraints allow (r) to occur, namely in pre-vocalic or intervocalic positions in words like red, brown, string, around, marry, soaring and sawing, across word boundaries in sequences like soar above and saw it up, and, as a consequence of H-dropping in the variety, also sore head and saw himself (Wells 1982; Foulkes and Docherty 2000, 2001; Altendorf and Watt 2005; Hughes et al. 2005). Whether the constraints are linguistic or non-linguistic, the fundamental premise is the same: that the distribution of the different surface forms of a dependent variable (the linguistic feature under scrutiny) can be correlated with bi- or multivalent independent variables (speaker characteristics, speech style, linguistic context, and so forth).

Identifying the social and linguistic constraints that prevent or disfavour a particular form from occurring in a given language variety and that license the use of another form instead is the central empirical preoccupation of variationist sociolinguistics. In this way, the social meaning of each of a variable’s variants can be deduced, and their distribution within the system circumscribed. This is done by correlating patterns of variation in a community’s language with the social and demographic characteristics of its speakers and the social networks and/or more generic categories to which they can be assigned (social class, gender, ethnicity, etc.), and by noting those linguistic contexts in which certain variants are always, frequently, seldom or never found. It should be emphasized that the distribution of variants is not held to be ‘either/or’, but rather probabilistic. Categorical distribution of linguistic forms is clearly of secondary interest to researchers aiming to account for patterns of variation in language data.

THE HISTORY AND UTILITY OF THE (SOCIO)LINGUISTIC VARIABLE

The sociolinguistic variable was first systematically used for quantification of language variation in Labov’s Martha’s Vineyard study (1963). While in this guise it is a relatively new addition to the toolkit used by linguists for describing, analysing and modelling language structure and use, the (at least tacit) notion of the linguistic variable is as old as language study itself. Paān.ini’s grammar of Sanskrit (?350 BC) incorporates variable rules that allow for differing outputs (Kiparsky 1979), and in the dialect geography and historical linguistics of more recent centuries the establishment of sets of ‘equivalent’ dialect terms and historical cognates entails identifying direct lexical and structural correspondences within and between languages. This is not at all surprising if, instead of assuming – as many modern linguists do – that variation is of only marginal significance to ‘language proper’, we take a more socially and historically realistic view of language structure, development and function. It hardly needs to be said that knowing that there are different ways of expressing the same idea in a given language is a fundamental element of people’s everyday linguistic awareness – as Sapir (1921: 147) remarked, ‘everyone knows that language is variable’. Despite this, and the fact that modern linguistics has its roots in the work of scholars who sought to provide a model of language structure and evolution to account for historical and contemporary intra- and interlinguistic differences, variability was generally marginalized or ignored by practitioners of the dominant schools of linguistics during the twentieth century, not least those working in the Chomskyan generativist tradition which continues to hold sway over large areas of the discipline. Intralinguistic variation is seen by many of the more conservative researchers in the generativist tradition to be irrelevant to an understanding of the nature of language beyond the most trivial level because, they argue, variability of the sort that interests sociolinguists is an epiphenomenon arising from the vagaries of language in use rather than a property of grammars at a deeper level (Chomsky 1986; Guy 1997; Henry 2002, 2005; Chambers 2003). But assuming, as seems reasonable, that o...