Sociolinguistics and The Sociolinguistic Development of the Child

I defined sociolinguistics as 'the study of language in relation to society', implying (intentionally) that sociolinguistics is part of the study of language. Thus, the value of sociolinguistics is the light which it throws on the nature of language in general, or on the characteristics of some particular language. As we might expect, students of society have found that facts about language can illuminate their understanding - after all, it is hard to think of any characteristic of a society which is as distinctive as its language, or as important for its functioning. 'The study of society in relation to language' (the converse of our definition of sociolinguistics) defines what is generally called THE SOCIOLOGY OF LANGUAGE.

The difference between sociolinguistics and the sociology of language is very much one of emphasis, according to whether the investigator is more interested in language or society, and also according to whether they have more skill in analyzing linguistic or social structures. There is a very large area of overlap between the two, and it seems pointless to try to divide the disciplines more clearly than at present. Much of what follows in this book could equally well have been written in a textbook on the sociology of language. On the other hand, there are some issues that such a textbook ought to include, which this one will not, notably most of what is called 'macro' sociology of language, dealing with the relations between society and languages as wholes. This is an important area of research from the point of view of sociology (and politics), since it raises issues such as the effects of multilingualism on economic development and the possible language policies a government may adopt- However, such 'macro' studies generally throw less light on the nature of language than the more 'micro' ones described in this book, because the notion of 'language X' is usually left unanalyzed. (There is a good discussion of the relations between sociolinguistics and the sociology of language in the introduction to Trudgill 1978.) For more information on the sociology of language, see Gibbons 1992 (a brief overview) and Fasold 1984 (the main textbook).

The sociolinguistic development of the child

Although we may assume that each speaker has a unique experience of language, and on this basis develops a unique grammar, a number of generalizations can be made about the stages through which people may be expected to pass in their sociolinguistic development. However, it should be emphasized from the outset that the following generalizations must all be treated as tentative hypotheses rather than established research findings, since they are based on a very small body of research supported by anecdotal evidence.

The first generalization concerns the linguistic models that the child follows. For many children, the pattern is as follows: first parents, then peers, then adults. Within the 'peers as models' phase, we can distinguish between childhood and adolescence, giving four life-phases: babyhood, childhood, adolescence, and adulthood. The sociolinguistic evidence for these distinctions is summarized in Chambers (1995:151ff.). The following is a very rough summary that obviously ignores many important subtleties and details.

Babyhood. The models are parents and other carers, who use 'baby-talk' (words like gee-gee, 'horse', and constructions like Baby go night-night now).

Childhood. The models are other children of the same age or somewhat older, and if these children speak differently from the parents, the children's model generally is the one that is actually adopted. At this stage, children are extremely conservative in their language; their main concern seems to be the same as that of older children, and some attempt to imitate teenagers as well. This leads to what is called 'AGE-GRADING' (Hockett 1950), a pattern of use in which linguistic items are used by people of a particular age, who then stop using it when they grow older. The language used by primary-school children is full of archaic forms which were in use by children of the same age centuries ago - ancient counting rhymes, skipping songs, truce-terms (for calling a truce in a 'battle'), names for playground games, taunts, challenges, and so on. (Opie and Opie, 1959, is a classic survey of this language in England.) Another example of age-grading is found in the speech of black American children, who use archaic forms which were part of the creole out of which the English of black Americans is widely believed to have developed (Dillard 1976). Perhaps because of this obsessive conformism with the local models, children can learn a new dialect or language perfectly (i.e., just as though it had been their first language).

Two other facts about this stage are important. Firstly, children tend very strongly to prefer members of their own sex, so it is very easy for girls and boys to develop differently. As we shall see in 4:5, men and women tend to perpetuate these differences in the way they use language. And secondly, children are also learning to recognize the social significance of different linguistic norms, as we shall see below.

Adolescence. The models now are other adolescents, but the foundations of language have already been laid - for most people. (perhaps all) it is too late to learn a new language or dialect perfectly(Chambers 1995: 160). This is the stage at which children prepare to be the next generation of adults. Unlike children, adolescents aim to be different from all previous adolescents, which gives rise to the constantly changing picture of teenage slang (Chambers 1995: 171). It is essential to the self-respect of a fourteen-year-old to be different from the ex-teenage twenty-year-olds (not to mention ex-ex-teenage fifty-year-olds!); but of course it is also essential to be in step with other 'relevant' adolescents, identified in terms of alternative lifestyle models. Maybe teenage gangs and social types are a preparation for the complexities of adult life, but whatever their reason, they have a profound effect on the speech of adolescents (as we shall see in a case-study in 5.2.6).

Adulthood. Our models are other adults, with current adolescents as a potential source of inspiration (or offence). Work, parenthood, and other social activities bring us into contact with other adults who offer competing models which we may either avoid or copy. for reasons that we shall explore in later chapters. There is still scope for change, and in particular for learning to use more or less standard speech for work purposes under the pressures of the 'linguistic market-place' (Bourdieu and Boltanski 1975. Explained well in Chambers 1995: 177ff). However, by now we are all more or less stable linguistically, with a personal language that defines our place in the social world in terms of region, age, sex, social class, and other characteristics.

After this overview of the main stages, we can return to the earlier years to ask how children adapt to the very varied linguistic world into which they are born. First, at what age do children become aware of the social significance of different speech forms? They appear to be aware of different speech forms and the fact that there are social differences between them from an early age. Children brought up in a bilingual environment have been reported as being aware that two separate language systems were being used even at eighteen months (see Romaine 1989, chapter 5, for a convenient summary). Some anecdotes suggest that this may happen even earlier, though others put the age later. For instance, Robbins Burling reports (1959) that his son learned Garo (a tribal language of north-eastern India) from his nanny at the same time as he learned English from his parents, and that he was about two years three months before he realized that different people spoke different languages; only then would he work out who was likely to understand his Garo before speaking. Before this, by eighteen months, he had noticed that many things had more than one word to express them, such as English ‘milk’ and Garo ‘dut’, but he had not yet made the considerable abstraction to the existence of two separate systems. As for dialect differences, there is little evidence relevant to young children, but it seems a fair assumption that they are at least capable of being aware of such differences by the time they start to model themselves on their peers, and will be aware of dialect differences to the extent that the speech of their parents and of their peers is different.

Assuming that a child has learned that two different languages or varieties are different systems, each used by a different range of people, how long does it take to become aware of the positive and negative prejudices that grown-ups have towards some of these varieties? And how long does it take children to adopt these prejudices themselves? Again, the evidence is sketchy and, to some extent, contradictory, but we shall see (6.2.5) that there is some evidence at least which suggests strongly that there are communities in which many children, as early as age three, have not only already become aware of such prejudices but have adopted them themselves. On the other hand, this is clearly not the same thing as claiming that Tour-year-olds have fully developed adult prejudices, and we may safely assume on the basis of other evidence that their prejudices go on developing throughout childhood and adolescence. Indeed, there seems to be no reason for thinking that the process ever stops completely.

What about the child's own speech? How does this develop in relation to the social environment? It is clear that children from an extremely early age adapt their speech to its social context (Andersen 1990). As soon as they start to speak, they speak differently to different people (Giles and Powesland 1975: 139), which is hardly surprising if we think of their speech as just one aspect of social behavior, and remember that they behave differently towards different people from very soon after birth. Moreover, from a very early age - in the first year, before they have learned any of the adult forms - they use different noises for different purposes, such as asking for something or saying the equivalent of 'I say, just look at that!' (Halliday 1975). Similarly, a child of twenty-three months was reported as deliberately separating her syllables off to make them clearer when she was being misunderstood (Weeks 1971).

By age three, the children of bilingual parents will probably be reasonably efficient at keeping the two languages separate from each other in their own speech, and any three-year-old may have started practicing a range of roles such as baby, doctor, or cowboy (Weeks 1971). The role of 'baby' is a particularly interesting one, as children get better at playing it, rather than worse, as one might expect from a naive view of baby-talk as a leftover from their own babyhood (Berko Gleason 1973, Sachs and Devin 1976). A four-year-old is already remarkably versatile. As Jean Berko Gleason puts it (1973), 'Four-year-olds may whine at their mothers, engage in intricate verbal play with their peers, and reserve their narrative, discursive tales for their grown-up friends.' There is no reason to think that there is any endpoint in the process of acquiring new styles of speaking, or of becoming more sophisticated in the use of the styles we already have.

We are all used to thinking of languages like English, French, and Japanese as relatively independent of the societies that speak them; so, for instance, we could, in principle, learn about French vocabulary and grammar without learning anything at all about research, social questions are hard to ignore. It would be wrong to imply that such questions are always center-stage. For some kinds of work on language, they can safely be kept in the background, and in fact, this has probably been true of most work on language over the last two thousand years (since linguistics started in Ancient Greece and India). This is done simply by loading all the social information into the language label, Instead of saying 'The people who live in England typically say cat for "cat"' and 'The people who live in England typically say dog for "dog"', we say "The people who live in England typically speak English' and 'The English for "cat" is cat' and 'The English for "dog" is dog'. It is easy to see how enormously useful these language labels are for any work on language, but at the same time, it is important to bear in mind that they are socially problematic. Where exactly are the boundaries of English? For example, what about Afro-Caribbean pidgins and creoles? (The following sentence is part of Nigerian Pidgin (Todd 1994): 'A bin kam, kariam go', meaning 'I came and carried it away'; is it English? And all about all the variation within English? Sometimes these questions hardly matter, and it's important not to get sidetracked by them from the matter at hand. This kind of work is what we can call 'non-social linguistics'.

In sociolinguistics, in contrast, the social questions are in full focus, though they share the stage with the same kinds of questions that linguistics studies. Sociolinguists flourish where linguists flounder, wherever there is variability within a community, or fuzzy boundaries between communities (or between languages), or words that are tied to particular social situations or to particular cultural beliefs. Over the few decades of its existence, sociolinguistics has developed ways of analyzing and thinking about the links between such things, and there are even the beginnings of some theoretical frameworks. In short, sociolinguistics and non-social linguistics complement one another – sociolinguistics takes over where non-social linguistics gives up, and vice versa. However, the findings of sociolinguistics, as described in this book, challenge a number of widely held views. Chapter 2 proposes reasons for questioning the assumption that languages are discrete, identifiable entities, consisting of dialects which can further be subdivided until the individual is reached, as the locus of the 'smallest dialect '. It would be wrong to imply that such questions are always center stage.