• Introduction
  • knowledge of Language
  • Variations
  • Scientific Investigation
  • Language and Society
  • Sociolinguistics and the Sociology of Language
  • Methodological Concerns
  • Language and Communitites

  • Introduction

    Any discussion of the relationship between language and society, or of the various functions of language in society, should begin with some attempt to define each of these terms: Language and Society.

    Let us say that a society is any group of people who are drawn together for a certain purpose or purposes. By such a definition ‘society’ becomes a very comprehensive concept, but we will soon see how useful such a comprehensive view is because of the very different kinds of societies we must consider in the course of the various discussions that follow.

    We may attempt an equally comprehensive definition of language: a language is what the members of a particular society speak. However, as we will see, speech in almost any society can take many very different forms, and just what forms we should choose to discuss when we attempt to describe the language of a society may prove to be a contentious matter.

    Sometimes too a society may be plurilingual; that is, many speakers may use more than one language, however we define language. We should also note that our definitions of language and society are not independent: the definition of language includes in it a reference to society. I will return to this matter from time to time.

    (Back to the Top)

    Knowledge of Language

    When two or more people communicate with each other in speech, we can call the system of communication that they employ a code. In most cases that code will be something we may also want to call a language. We should also note that two speakers who are bilingual, that is, who have access to two codes, and who for one reason or another shift back and forth between the two languages as they converse by code-switching are actually using a third code, one which draws on those two languages.

    The system (or the grammar, to use a well-known technical term) is something that each speaker ‘knows,’ but two very important issues for linguists are just what that knowledge is knowledge of and how it may best be characterized.

    In practice, linguists do not find it at all easy to write grammars because the knowledge that people have of the languages they speak is extremely hard to describe. It is certainly something different from, and is much more considerable than, the kinds of knowledge we see described in most of the grammars we find on library shelves, no matter how good those grammars may be. Anyone who knows a language knows much more about that language than is contained in any grammar book that attempts to describe the language.

    What is also interesting is that this knowledge is both something which every individual who speaks the language possesses (since we must assume that each individual knows the grammar of his or her language by the simple reason that he or she readily uses that language) and also some kind of shared knowledge, that is, knowledge possessed by all those who speak the language. It is also possible to talk about ‘dead’ languages, e.g., Latin or Sanskrit. However, in such cases we should note that it is the speakers who are dead, not the languages themselves, for these may still exist, at least in part. We may even be tempted to claim an existence for English, French, or Swahili independent of the existence of those who speak these languages.

    Today, most linguists agree that the knowledge speakers have of the language or languages they speak is knowledge of something quite abstract. It is a knowledge of rules and principles and of the ways of saying and doing things with sounds, words, and sentences, rather than just knowledge of specific sounds, words, and sentences. It is knowing what is in the language and what is not; it is knowing the possibilities the language offers and what is impossible. This knowledge explains how it is we can understand sentences we have not heard before and reject others as being ungrammatical, in the sense of not being possible in the language.

    Communication among people who speak the same language is possible because they share such knowledge, although how it is shared – or even how it is acquired – is not well understood. Certainly, psychological and social factors are important, and genetic ones too. Language is a communal possession, although admittedly an abstract one. Individuals have access to it and constantly show that they do so by using it properly. As we will see, a wide range of skills and activities is subsumed under this concept of ‘proper use.’

    Confronted with the task of trying to describe the grammar of a language like English, many linguists follow the approach which is associated with Chomsky, undoubtedly the most influential figure in late twentieth-century linguistics. Chomsky has argued on many occasions that, in order to make meaningful discoveries about language, linguists must try to distinguish between what is important and what is unimportant about language and linguistic behavior.

    The important matters, sometimes referred to as language universals, concern the learnability of all languages, the characteristics they share, and the rules and principles that speakers apparently follow in constructing and interpreting sentences; the less important matters have to do with how individual speakers use specific utterances in a variety of ways as they find themselves in this situation or that.

    Chomsky has also distinguished between what he has called competence and performance. He claims that it is the linguist’s task to characterize what speakers know about their language, i.e., their competence, not what they do with their language, i.e., their performance.

    The best-known characterization of this distinction comes from Chomsky himself (1965, pp. 3–4) in words which have been extensively quoted:

    “Linguistic theory is concerned primarily with an ideal speaker–listener, in a completely homogeneous speech-community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, dis- tractions, shifts of attention and interest, and errors (random or characteristic) in applying his knowledge of the language in actual performance. This seems to me to have been the position of the founders of modern general linguistics, and no cogent reason for modifying it has been offered. To study actual linguistic performance, we must consider the interaction of a variety of factors, of which the under- lying competence of the speaker–hearer is only one. In this respect, study of language is no different from empirical investigation of other complex phenomena.”

    From time to time we will return to this distinction between competence and performance. However, the knowledge we will seek to explain involves more than knowledge of the grammar of the language for it will become apparent that speakers know, or are in agreement about, more than that. Moreover, in their performance they behave systematically: their actions are not random; there is order. Knowing a language also means knowing how to use that language since speakers know not only how to form sentences but also how to use them appropriately. There is therefore another kind of competence, sometimes called communicative competence, and the social aspects of that competence will be our concern here.

    Discussion 1

    Hymes (1964b, p. 16) presents the following two instances of behavior which the participants, speakers of Ojibwa, an American Indian language, describe as language behavior:

    “An informant told me that many years before he was sitting in a tent one afternoon during a storm, together with an old man and his wife. There was one clap of thunder after another. Suddenly the old man turned to his wife and asked, ‘Did you hear what was said?’ ‘No,’ she replied, ‘I didn’t catch it.’ My informant, an acculturated Indian, told me he did not at first know what the old man and his wife referred to. It was, of course, the thunder. The old man thought that one of the Thunder Birds had said something to him. He was react- ing to this sound in the same way as he would respond to a human being, whose words he did not understand. The casualness of the remark and even the trivial character of the anecdote demonstrate the psychological depth of the ‘social relations’ with other-than-human beings that becomes explicit in the behavior of the Ojibwa as a consequence of the cognitive ‘set’ induced by their culture.”
    “A white trader, digging in his potato patch, unearthed a large stone similar to the one just referred to. He sent for John Duck, an Indian who was the leader of the wábano, a contemporary ceremony that is held in a structure something like that used for the Midewiwin (a major ceremony during which stones occasionally had animate properties such as movement and opening of a mouth). The trader called his attention to the stone, saying that it must belong to his pavilion. John Duck did not seem pleased at this. He bent down and spoke to the boulder in a low voice, inquiring whether it had ever been in his pavilion. According to John the stone replied in the negative.”

    “It is obvious that John Duck spontaneously structured the situation in terms that are intelligible within the context of Ojibwa language and culture. . . . I regret that my field notes contain no information about the use of direct verbal address in the other cases mentioned (movement of stone, opening of a mouth). But it may well have taken place. In the anecdote describing John Duck’s behavior, however, his use of speech as a mode of communication raises the animate status of the boulder to the level of social interaction common to human beings. Simply as a matter of observation we can say that the stone was treated as if it were a ‘person,’ not a ‘thing,’ without inferring that objects of this class are, for the Ojibwa, necessarily conceptualized as persons.”

    Hymes argues that ‘in general, no phenomenon can be defined in advance as never to be counted as constituting a message.’ How does this observation apply to the above examples? Can you think of possible examples drawn from your own experience? Note that a basic assumption here is that ‘messages,’ whatever they are, require a ‘language.’ Should every ‘language’ in which you can send ‘messages’ be of equal interest to us as sociolinguists, e.g., the ‘language’ of flowers, semaphore signaling, dress codes, and road signs? If not, what principles should guide us in an attempt to constrain our interests? And how do you view the ‘languages’ of logic, mathematics, and computers?

    What obstacles do you see in an attempt to define English as a language when you consider that such a definition must cover all of the following (and much more): both Cockney and Jamaican English; the speech of two- year-olds; fast colloquial speech; the language of formal written documents such as real estate transfers; formulaic expressions such as How do you do? and It never rains but it pours; completely novel sentences, i.e., sentences you have not heard or seen before (e.g., just about any sentence in this book); and slips of the tongue, e.g., queer dean for dear Queen? What kind of abilities must you yourself have in order even to consider attempting such a task?
    (Back to the Top)


    The competence–performance distinction just mentioned is one that holds intriguing possibilities for work in linguistics, but it is one that has also proved to be quite troublesome, particularly when much of the variety we experience within language is labeled ‘performance’ and then put to one side by those who consider ‘competence’ to be the only valid concern of linguists. The language we use in everyday living is remarkably varied. Some investigators believe that this variety throws up serious obstacles to all attempts to demonstrate that each language is truly a homogeneous entity, and that it is possible to write a complete grammar for a language which makes use of categorical rules, i.e., rules which specify exactly what is – and therefore what is not – possible in the language.

    Everywhere we turn we seem to find at least a new wrinkle or a small inconsistency with regard to any rule we might propose. When we look closely at any language, we will discover time and time again that there is considerable internal variation and that speakers make constant use of the many different possibilities offered to them. No one speaks the same way all the time and people constantly exploit the nuances of the languages they speak for a wide variety of purposes. The consequence is a kind of paradox: while many linguists would like to view any language as a homogeneous entity and each speaker of that language as controlling only a single style, so that they can make the strongest possible theoretical generalizations, in actual fact that language will exhibit considerable internal variation, and single-style speakers will not be found (or, if found, will appear to be quite ‘abnormal’ in that respect, if in no other!)

    A recognition of variation implies that we must recognize that a language is not just some kind of abstract object of study. It is also something that people use. Can we really set aside, at any point in our study of language, this fact of use? It is not surprising therefore that a recurring issue in linguistics in recent years has been the possible value of a linguistics that deliberately separates itself from any concern with the use, and the users, of language. Following Chomsky’s example, many linguists have argued that we should not study a language in use, or even how the language is learned, without first acquiring an adequate knowledge of what language itself is. In this view, linguistic investigations should focus on developing this latter knowledge. The linguist’s task should be to write grammars that will help us develop our understanding of language: what it is, how it is learnable, and what it tells us about the human mind. This kind of linguistics is sometimes referred to as ‘theoretical linguistics’ and it has claimed a privileged position for itself within the overall discipline of linguistics. Investigations of language use have little to offer us in such a view.

    Many sociolinguists have disagreed, arguing that an asocial linguistics is scarcely worthwhile and that meaningful insights into language can be gained only if such matters as use and variation are included as part of the data which must be explained in a comprehensive theory of language; such a theory of language must have something to say about the uses of language. This is the view I will adopt here. However, while doing so, from time to time I will voice some skepticism about the claims of other investigators that we should pursue certain ideological ends in investigating such use. Detachment and objectivity are essential requirements of serious scientific inquiry

    We will see that there is considerable variation in the speech of any one individual, but there are also definite bounds to that variation: no individual is free to do just exactly what he or she pleases so far as language is concerned. You cannot pronounce words any way you please, inflect or not inflect words such as nouns and verbs arbitrarily, or make drastic alterations in word order in sentences as the mood suits you. If you do any or all of these things, the results will be unacceptable, even gibberish. The variation you are permitted has limits and these limits can be described with considerable accuracy. Individuals know the various limits (or norms), and that knowledge is both very precise and at the same time almost entirely unconscious. It is also difficult to explain how individual speakers acquire knowledge of these norms of linguistic behavior, for they appear to be much more subtle than the norms that apply to such matters as social behavior, dress, and table manners. This is another issue to which we will return from time to time. Our task will be one of trying to specify the norms of linguistic behavior that exist in particular groups and then trying to account for individual behavior in terms of these norms. This task is particularly interesting because most people have no conscious awareness that we can account for much of their linguistic behavior in this way.

    People have also learned such behavior. We must be concerned with that learning. Why does speaker X behave this way but speaker Y behave that way? To answer that question we must look at such issues as identity, group membership, power, and socialization.

    Each of us has an identity (or, perhaps more accurately, a set of identities). That identity has been constructed from interaction with others and it is the sense of self each of us has achieved, the result of our socialization, i.e., our experiences with the outside world as we have dealt with that world in all its complexity. Consequently, any of many factors might have affected it: race, ethnicity, gender, religion, occupation, physical location, social class, kinship, leisure activities, etc. Identity is created in dealing with such factors and in dealing with members of groups for whom these factors are their identifying characteristics. An identity may also change for identities can sometimes be quite malleable, but, of course, it may also stay fixed if change is not allowed or if a fixed identity is to be maintained at all costs.

    Identity is very important: individual identity and group identity. It will be a recurrent theme in the pages that follow. Much of what we find in linguistic behavior will be explicable in terms of people seeking to negotiate, realize, or even reject identities through the use of language. In fact, as we will see, language is a profound indicator of identity, more potent by far than cultural artifacts such as dress, food choices, and table manners.

    Groups, too, have identities, their ways of achieving a sense of solidarity among members, so we will be interested in the linguistic characteristics of both individuals and groups. Concepts such as ‘community’, ‘social network’, and ‘community of practice’ will be found in the pages that follow. These are useful in referring to groups of various kinds, for it is within groups that individuals form relationships or reject such a possibility. However, groups, like individuals, are complex entities so we must never forget that any reference made in the following pages to ‘middle class,’ ‘women,’ ‘speakers of Haitian Creole,’ ‘teenagers,’ etc. in reality subsumes a variety of individual identities each in its own way just as complex as the whole.

    Finally, in all the above we must recognize that ‘power’ plays a significant role in everything that happens. Some forces in society are stronger than others and produce real effects, among them linguistic effects that have consequences for the lives we live.

    Bourdieu (1991) conceives of languages as symbolic market- places in which some people have more control of the goods than others because certain languages or varieties have been endowed with more symbolic power than others and have therefore been given a greater value, e.g., standard languages, certain accents, a particular gendered style of speaking, a specific type of discourse.


    I have said that languages contain a great deal of variety. What evidence can you cite to show some of the variety? Consider, for example, how many different ways you can ask someone to open a window or seek permission to open the window yourself because the room you are in is too warm. How many ways can you pronounce variants of and, have, do, of, and for? When might Did you eat yet? sound like Jeechet? What did you do with the words and sounds? Do you speak the same way to a younger sibling at home over the breakfast table as you would to a distinguished public figure you meet at a ceremonial dinner? If you do not, and it is almost certain that you do not, what are the differences in the linguistic choices you make? Why do you make them?

    An individual can use language in a variety of ways and for many different purposes. What might cause a speaker to say each of the following? When would each be quite inappropriate?

    Do you know of any grammar book that tells you when to use (or not to use) each of the above? Would you describe your knowledge of when to use (or not to use) each as a matter of competence or of performance? (In thinking about this you might consult just about any discussion of Chomsky’s work on linguistic theory.)

    Do you always agree with people you know about the ‘correct’ choice to make of certain linguistic forms? What do you, and they, regard as the correct completions of the tag questions found in the following examples? (The first is done for you.

    What kinds of difficulties did you find in completing this task? What kinds of agreements and disagreements do you find when you compare your responses to those of others? What do the standard grammars have to say about correctness here? How would you advise an adult learning English as a foreign language concerning this particular problem?

    Describe some aspects of your own speech which show how it varies from the speech of certain other people you know. Do you pronounce words differently, use different word forms, choose different words, or use different grammatical structures? How do you view, i.e., judge, the speech of those who speak differently from you?

    Hudson (1996, p. 12) says that we may be impressed by the amount of agreement that is often found among speakers. This agreement goes well beyond what is needed for efficient communication. He particularly points out the conformity we exhibit in using irregular forms, e.g., went for the past tense of go, men as the plural of man, and best as the superlative of good. This irregular morphology is somewhat inefficient; all it shows is our conformity to rules established by others. How conformist do you consider yourself to be so far as language is concerned? What ‘rules’ do you obey? When do you ‘flout the rules,’ if you ever do?
    (Back to the Top)

    Scientific Investigation

    The scientific study of language, its uses, and the linguistic norms that people observe poses a number of problems. Such a study must go a long way beyond merely devising schemes for classifying the various bits and pieces of linguistic data you might happen to observe. That would be a rather uninteresting activity, a kind of butterfly collecting. A more profound kind of theorizing is called for: some attempt to arrive at an understanding of the general principles of organization that surely must exist in both language and the uses of language

    It is just such an attempt that led Saussure (1959) to distinguish between langue (group knowledge of language) and parole (individual use of language); Bloomfield (1933) to stress the importance of contrastive distribution (since pin and bin are different words in English, /p/ and /b/ must be contrastive units in the structure of English); Pike (1967) to distinguish between emic and etic features in language (/p/ and /b/ are contrastive, therefore emic, units, but the two pronunciations of p in pin and spin are not contrastive, therefore etic); and Sapir (1921) and, much later, Chomsky (1965) to stress the distinction between the ‘surface’ characteristics of utterances and the ‘deep’ realities of linguistic form behind these surface characteristics. A major current linguistic concern is with matters such as language universals, i.e., the essential properties and various typologies of languages (see Comrie, 1989, and Cook and Newson, 1996), the factors that make languages learnable by humans but not by non-humans (see Pinker, 1994), and the conditions that govern such matters as linguistic change (see Labov, 1994, and McMahon, 1994).

    There is not just one way to do linguistics, although it is true to say that some linguists occasionally behave as though their way is the only way. It is actually quite possible for two linguists to adopt radically different approaches to both language and linguistic theorizing in their work while still doing something that many consider to be genuine linguistics. Perhaps nowhere can such differences of approach be better observed than in attempts to study the relationship of language to society.

    Such attempts cover a very wide range of issues and reveal the diversity of approaches: different theories about what language is; different views of what constitute the data that are relevant to a specific issue; different formulations of research problems; different conceptions of what are ‘good’ answers, the ‘significance’ or ‘interest’ of certain findings, and the generalizability of conclusions; and different interpretations of both the theoretical and ‘real-world’ consequences of particular pieces of research, i.e., what they tell us about the nature of language or indicate we might do to change or improve the human condition.


    Find out what you can about Saussure’s distinction between langue and parole and about Pike’s etic–emic distinction. How might these distinctions relate to any study of language use in society?

    Bloomfield’s views on contrastive distribution are very important. Be sure you know what is meant by the concept of ‘contrast’ in linguistics. You might test out your knowledge of the concept by trying to find out how many contrastive consonant and vowel sounds you have in the variety of English you speak. If you find the number of consonant sounds to be any other than 24 and the number of vowel sounds to be far different from 14, you may be on the wrong track.
    (Back to the Top)

    language and society

    If we look back at the history of linguistics it is rare to find investigations of any language which are entirely cut off from concurrent investigations of the history of that language, or of its regional and/or social distributions, or of its relationship to objects, ideas, events, and actual speakers and listeners in the ‘real’ world. That is one of the reasons why a number of linguists have found Chomsky’s asocial view of linguistic theorizing to be a rather sterile type of activity, since it explicitly rejects any concern for the relationship between a language and those who use it.

    We must acknowledge that a language is essentially a set of items, what Hudson (1996, p. 21) calls ‘linguistic items,’ such entities as sounds, words, grammatical structures, and so on. It is these items, their status, and their arrangements that language theorists such as Chomsky concern themselves with. On the other hand, social theorists, particularly sociologists, attempt to understand how societies are structured and how people manage to live together. To do so, they use such concepts as ‘identity,’ ‘power,’ ‘class,’ ‘status,’ ‘solidarity,’ ‘accommodation,’ ‘face,’ ‘gender,’ ‘politeness,’ etc. A major concern of this book is to examine possible relationships between ‘linguistic items’ on the one hand and concepts such as ‘power,’ ‘solidarity,’ etc. on the other.

    We should note that in doing so we are trying to relate two different kinds of entities in order to see what light they throw on each other. That is not an easy task. Linguistic items are difficult to define. Try, for example, to define exactly what linguistic items such as sounds, syllables, words, and sentences are. Then try to define precisely what you understand by such concepts as ‘social class,’ ‘solidarity,’ ‘identity,’ ‘face,’ and ‘politeness.’ Finally, try to relate the two sets of definitions within some kind of theory so as to draw conclusions about how items in these two very different classes relate to each other. Do all this while keeping in mind that languages and societies are constantly changing. The difficulties we confront are both legion and profound.

    There are several possible relationships between language and society. One is that social structure may either influence or determine linguistic structure and/or behavior. Certain evidence may be adduced to support this view: the age-grading phenomenon whereby young children speak differently from older children and, in turn, children speak differently from mature adults; studies which show that the varieties of language that speakers use reflect such matters as their regional, social, or ethnic origin and possibly even their gender; and other studies which show that particular ways of speaking, choices of words, and even rules for conversing are in fact highly determined by certain social requirements.

    A second possible relationship is directly opposed to the first: linguistic structure and/or behavior may either influence or determine social structure. This is the view that is behind the Whorfian hypothesis, the claims of Bernstein, and many of those who argue that languages rather than speakers of these languages can be ‘sexist’ . A third possible relationship is that the influence is bidirectional: language and society may influence each other. One variant of this approach is that this influence is dialectical in nature, a Marxist view put forward by Dittmar (1976), who argues that ‘speech behaviour and social behaviour are in a state of constant interaction’ and that ‘material living conditions’ are an important factor in the relationship

    A fourth possibility is to assume that there is no relationship at all between linguistic structure and social structure and that each is independent of the other. A variant of this possibility would be to say that, although there might be some such relationship, present attempts to characterize it are essentially premature, given what we know about both language and society. Actually, this variant view appears to be the one that Chomsky himself holds: he prefers to develop an asocial linguistics as a preliminary to any other kind of linguistics, such an asocial approach being, in his view, logically prior.

    We must therefore be prepared to look into various aspects of the possible relationships between language and society. It will be quite obvious from doing so that correlational studies must form a significant part of sociolinguistic work. Gumperz (1971, p. 223) has observed that sociolinguistics is an attempt to find correlations between social structure and linguistic structure and to observe any changes that occur. Chambers (2002, p. 3) is even more direct: ‘Sociolinguistics is the study of the social uses of language, and the most productive studies in the four decades of sociolinguistic research have emanated from determining the social evaluation of linguistic variants. These are also the areas most susceptible to scientific methods such as hypothesis-formation, logical inference, and statistical testing.’

    However, as Gumperz and others have been quick to indicate, such studies do not exhaust sociolinguistic investigation, nor do they always prove to be as enlightening as one might hope. It is a well-known fact that a correlation shows only a relationship between two variables; it does not show ultimate causation. To find that X and Y are related is not necessarily to discover that X causes Y (or Y causes X), for it is also quite possible that some third factor, Z, may cause both X and Y (or even that some far more subtle combination of factors is involved). We must always exercise caution when we attempt to draw conclusions from such relationships.

    A worthwhile sociolinguistics, however, must be something more than just a simple mixing of linguistics and sociology which takes concepts and findings from the two disciplines and attempts to relate them in simple ways. It certainly must go beyond Horvath’s view (1998, p. 448) that sociolinguists should just pick and choose freely from sociology: ‘What my kind of sociolinguists do is go periodically to sociology and find “social networks” or “the linguistic market place”...and we find [these concepts] terribly useful in understanding the patterns that emerge from our data. However, we are not engaged in the sociologists’ struggles over the importance of social networks vis-à-vis other ways of dealing with the structure of society and may remain blissfully unaware of whether or not these models have become contentious within the home discipline.’

    A serious scientific approach is incompatible with ‘blissful unawareness’ in an essential part of its underpinnings. Hymes (1974, p. 76) has pointed out that even a mechanical amalgamation of standard linguistics and standard sociology is not likely to suffice in that in adding a speechless sociology to a sociology-free linguistics we may miss what is important in the relationship between language and society. Specific points of connection between language and society must be discovered, and these must be related within theories that throw light on how linguistic and social structures interact.

    Holmes (1992, p. 16) says that ‘the sociolinguist’s aim is to move towards a theory which provides a motivated account of the way language is used in a community, and of the choices people make when they use language.’ For example, when we observe how varied language use is we must search for the causes. ‘Upon observing variability, we seek its social correlates. What is the purpose of the variation? How is it evaluated in the community? What do its variants symbolize?’ (Chambers, 2003, p. 226). For Chambers these questions ‘are the central questions of sociolinguistics.’ Chambers is not alone in holding such views. Others too believe that sociolinguistics is the study of language variation and that the purpose of such study is to find out what variation tells us about language and speakers’ ‘knowledge’ of language, in this case their unconscious knowledge of subtle linguistic differences.

    We will also see that there is some opposition to this idea that sociolinguistic investigations should be confined to fairly straightforward correlational studies of this kind. Critics such as Cameron (1997) claim that these studies do not provide very satisfactory explanations for linguistic behavior because of inadequacies with social theory – sometimes there is none at all – and failure to appreciate the difficulties in using social concepts. Any conclusions are likely to be suspect. What is needed, according to Cameron (p. 62), is more social engagement so that sociolinguistics would ‘deal with such matters as the production and reproduction of linguistic norms by institutions and socializing practices; how these norms are apprehended, accepted, resisted and subverted by individual actors and what their relation is to the construction of identity.’

    Milroy (2001, pp. 554–5) makes a somewhat similar claim in discussing the processes of standardization and change: ‘Social patterns are adduced only in so far as they may elucidate pat- terns of language by exhibiting co-variation with linguistic variables . . . and as long as internal analyses are quite strongly biased in favor of linguistic, rather than social, phenomena, the quantitative paradigm will be to that extent impeded in its attempts to explain the social “life” of language and the social origins of language change.’ I have already mentioned this idea of necessary social engagement and I will return to it later. However, one point is clear in the above disagreement: sociolinguistics, whatever it is, is about asking important questions concerning the relationship of language to society.


    To convince yourself that there are some real issues here with regard to the possible relationships between language and society, consider your responses to the following questions and compare them with those of others.

    Does an Inuit ‘see’ a snowscape differently from a native of Chad visiting the cold north for the first time because the Inuit must be using a language developed to deal with the surrounding snowscape?

    If men and women speak differently, is it because the common language they share has a gender bias, because boys and girls are brought up differently, or because part of ‘gender marking’ is the linguistic choices one can – indeed, must – make?

    Is language just another cultural artifact, like property, possessions, or money, which is used for the expression of power and/or as a medium of exchange? b.If language is an essential human attribute and humans are necessarily social beings, what problems and paradoxes do you see for theoretical work in sociolinguistics if the latter is to grapple with the relationships between linguistic and social factors?

    One aspect of the power of professionals is said to be the way they are able to use language to control others. How do physicians, psychiatrists, lawyers, social workers, teachers, priests, police officers, etc. use language to control others? Does this same power principle apply to parents (in relation to children), men (in relation to women), upper social classes (in relation to lower social classes), speakers of standard languages (in relation to speakers of nonstandard varieties of those languages), and so on?

    (Back to the Top)

    Sociolinguistics and the Sociology of Language

    Some investigators have found it appropriate to try to introduce a distinction between sociolinguistics or micro-sociolinguistics and the sociology of language or macro-sociolinguistics.

    In this distinction, sociolinguistics is concerned with investigating the relationships between language and society with the goal being a better understanding of the structure of language and of how languages function in communication; the equivalent goal in the sociology of language is trying to discover how social structure can be better understood through the study of language, e.g., how certain linguistic features serve to characterize particular social arrangements

    Hudson (1996, p. 4) has described the difference as follows: sociolinguistics is ‘the study of language in relation to society,’ whereas the sociology of language is ‘the study of society in relation to language.’ In other words, in sociolinguistics we study language and society in order to find out as much as we can about what kind of thing language is, and in the sociology of language we reverse the direction of our interest. Using the alternative terms given above, Coulmas (1997, p. 2) says that ‘micro-sociolingustics investigates how social structure influences the way people talk and how language varieties and patterns of use correlate with social attributes such as class, sex, and age. Macro-sociolinguistics, on the other hand, studies what societies do with their languages, that is, attitudes and attachments that account for the functional distribution of speech forms in society, language shift, maintenance, and replacement, the delimitation and interaction of speech communities.’

    The view I will take here is that both sociolinguistics and the sociology of language require a systematic study of language and society if they are to be successful. Moreover, a sociolinguistics that deliberately refrains from drawing conclusions about society seems to be unnecessarily restrictive, just as restrictive indeed as a sociology of language that deliberately ignores discoveries about language made in the course of sociological research. So while it is possible to do either kind of work to the exclusion of the other, I will be concerned with looking at both kinds. My own views are essentially in agreement with those of Coulmas (1997, p. 3), expressed as follows:

    “There is no sharp dividing line between the two, but a large area of common concern. Although sociolinguistic research centers about a number of different key issues, any rigid micro–macro compartmentalization seems quite contrived and unnecessary in the present state of knowledge about the complex interrelationships between linguistic and social structures. Contributions to a better understanding of language as a necessary condition and product of social life will continue to come from both quarters.”

    Consequently, I will not attempt to make the kinds of distinctions found in Trudgill (1978). He tries to differentiate those studies that he considers to be clearly sociolinguistic in nature from those that clearly are not, for, as he says, ‘while everybody would agree that sociolinguistics has something to do with language and society, it is clearly also not concerned with everything that could be considered “language and society”.’ The problem, therefore, lies in the draw- ing of the line between language and society and sociolinguistics. Different scholars draw the line in different places (p. 1). Trudgill argues that certain types of language studies are almost entirely sociological in their objectives and seem to fall outside even the sociology of language. Included in this category are ethno- methodological studies and work by such people as Bernstein. For Trudgill, such work is definitely not sociolinguistics, however defined, since it apparently has no linguistic objectives.

    According to Trudgill, certain kinds of work combine insights from sociology and linguistics. Examples of such work are attempts to deal with the structure of discourse and conversation, speech acts, studies in the ethnography of speaking, investigations of such matters as kinship systems, studies in the sociology of language, e.g., bilingualism, code-switching, and diglossia, and certain ‘practical’ concerns such as various aspects of teaching and language behavior in classrooms. While Trudgill considers all such topics to be genuinely sociolinguistic, he prefers, however, to use that term in a rather different and somewhat narrower sense. Elsewhere (1995, p. 21), he says that such concerns are perhaps better subsumed under anthropological linguistics, geolinguistics, the social psychology of language, and so on.

    For Trudgill there is still another category of studies in which investigators show a concern for both linguistic and social matters. This category consists of studies which have a linguistic intent. ‘Studies of this type are based on empirical work on language as it is spoken in its social context, and are intended to answer questions and deal with topics of central interest to linguists’ (1978, p. 11). These studies are just another way of doing linguistics. Included in this category are studies of variation and linguistic change, and the seminal figure is Labov. According to Trudgill, Labov has addressed himself to issues such as the relationship between language and social class, with his main objective not to learn more about a particular society or to examine correlations between linguistic and social phenomena, but to learn more about language and to investigate topics such as the mechanisms of linguistic change, the nature of linguistic variability, and the structure of linguistic systems.

    Trudgill’s view is that ‘all work in this category is aimed ultimately at improving linguistic theory and at developing our understanding of the nature of language’ (1978, p. 11). For him this is genuine sociolinguistics. Chambers (2002, 2003) voices a similar view and Downes (1998, p. 9) echoes it: ‘sociolinguistics is that branch of linguistics which studies just those properties of language and languages which require reference to social, including contextual, factors in their explanation.’ However, in reviewing research on language and society, Downes’ reach far exceeds that of Trudgill, even that of his glossary of terms (2003, p. 123), where he characterizes sociolinguistic research as ‘work which is intended to achieve a better understanding of the nature of human language by studying language in its social context and/or to achieve a better understanding of the nature of the relationship and interaction between language and society.’

    This work has been called ‘linguistics with a conscience and a cause, one which seeks to reveal how language is used and abused in the exercise of power and the suppression of human rights’ (Widdowson, 1998, p. 136). Two of its main exponents are Fairclough (1995, 2001) and van Dijk (1993), who champion an approach called ‘critical discourse analysis.’ This work focuses on how language is used to exercise and preserve power and privilege in society, how it buttresses social institutions, and how even those who suffer as a consequence fail to realize how many things that appear to be ‘natural’ and ‘normal’ are not at all so. They are not so because it is power relations in society that determine who gets to say what and who gets to write what. The claim is that politics, medicine, religion, eduation, law, race, gender, and academia can only be understood for what they really are within the framework of critical discourse analysis: as systems that maintain an unequal distribution of wealth, income, status, group membership, education, and so on. Fairclough (2001, p. 6) expresses what he sees as the failure of sociolinguistics to deal with such matters as follows: ‘Sociolinguistics is strong on “what?” questions (what are the facts of variation?) but weak on “why?” and “how?” questions (why are the facts as they are?; how – in terms of the development of social relationships of power – was the existing sociolinguistic order brought into being?; how is it sustained?; and how might it be changed to the advantage of those who are dominated by it?).’

    This is very much an ideological view. Its proponents maintain that all language use is ideological as are all investigations, i.e., that there is no hope of an ‘objective’ or ‘neutral’ sociolinguistics. Consequently, critical discourse an- alysis is ideological and judgmental. It claims the high ground on issues; it is ‘a resource for people who are struggling against domination and oppression in its linguistic forms’ (Fairclough, 1995, p. 1). We might well exercise caution in assessing any claims we find: appeals to what is right tend to short-circuit genuine scientific inquiry.


    Ethnomethodology (see chapter 10) is the study of commonsense knowledge and practical reasoning. To convince yourself that you have such knowledge and do employ such reasoning, see what happens if you react ‘literally’ when someone next addresses you with such formulaic expressions as How do you do? or Have a nice day. For example, you can respond What do you mean, ‘How do I do?’ or How do you define ‘a nice day’? (Be careful!) You should find that commonsense knowledge tells you not to take every- thing you hear literally. So far as practical reasoning is concerned, collect examples of how people actually do reach conclusions, give directions, and relate actions to consequences or ‘causes’ to ‘effects.’ Do they do this in any ‘scientific’ manner?

    Bernstein, a British sociologist, has claimed that some children acquire a somewhat limited exposure to the full range of language use as a result of their upbringing, and may consequently be penalized in school. What kinds of evidence would you consider to be relevant to confirming (or disconfirming) such a claim?

    Labov (1970, p. 30) has described the sociology of language as follows: It deals with large-scale social factors, and their mutual interaction with lan- guages and dialects. There are many open questions, and many practical prob- lems associated with the decay and assimilation of minority languages, the development of stable bilingualism, the standardization of languages and the planning of language development in newly emerging nations. The linguistic input for such studies is primarily that a given person or group uses language X in a social context or domain Y. What are some of the ‘questions’ and ‘problems’ you see in your society, either broadly or narrowly defined, that fall within such a sociology of language? As a further instance of a topic that might be covered in the sociology of language, consider who speaks English in the world, where, and for what purposes? You might also contrast what you can find out about the uses of English with what you can find out about the uses of Latin, Swahili, French, Haitian Creole, Basque, and Esperanto.Studies of linguistic variation make use of the concept of the ‘linguistic variable.’ One simple linguistic variable in English is the pronunciation of the final sound in words like singing, running, fishing, and going (-ing or -in’) in contexts such as ‘He was singing in the rain,’ ‘Running is fun,’ ‘It’s a fishing boat,’ and ‘Are you going?’ and on various occasions (e.g., in casual conversation, in formal speech making, or in reading individual words out aloud). What do you find? How might you try to explain any differences you find?

    (Back to the Top)

    Methodogical Concerns

    Sociolinguistics should encompass everything from considering ‘who speaks (or writes) what language (or what language variety) to whom and when and to what end’ (Fishman, 1972b, p. 46), that is, the social distribution of linguistic items, to considering how a particular linguistic variable (see above) might relate to the formulation of a specific grammatical rule in a particular language or dialect, and even to the processes through which languages change. Whatever sociolinguistics is, it must be oriented toward both data and theory: that is, any conclusions we come to must be solidly based on evidence. Above all, our research must be motivated by questions that can be answered in an approved scientific way. Data collected for the sake of collecting data are of little interest, since without some kind of focus – that is, without some kind of non-trivial motive for collection – they can tell us little or nothing. A set of random observations about how a few people we happen to observe use language cannot lead us to any useful generalizations about behavior, either linguistic or social. We cannot be content with ‘butterfly collecting,’ no matter how beautiful the specimens are! We must collect data for a purpose and that purpose should be to find an answer, or answers, to an interesting question. Questions phrased in ways that do not allow for some kind of empirical testing have no more than a speculative interest.

    Those who seek to investigate the possible relationships between language and society must have a twofold concern: they must ask good questions, and they must find the right kinds of data that bear on those questions. We will discover how wide the variety of questions and data in sociolinguistics has been: correlational studies, which attempt to relate two or more variables (e.g., certain linguistic usages to social-class differences); implicational studies, which suggest that if X, then Y (e.g., if someone says tess for tests, does he or she also say bes’ for best?); microlinguistic studies, which typically focus on very specific linguistic items or individual differences and uses and seek possibly wide-ranging linguistic and/or social implications (e.g., the distribution of singing and singin’); macrolinguistic studies, which examine large amounts of language data to draw broad conclusions about group relationships (e.g., choices made in language planning); and still other studies, which try to arrive at generalizations about certain universal characteristics of human communication, e.g., studies of conversational structure.

    Since sociolinguistics is an empirical science, it must be founded on an adequate database. As we will see, that database is drawn from a wide variety of sources. These include censuses, documents, surveys, and interviews. Some data require the investigator to observe ‘naturally occurring’ linguistic events, e.g., conversations; others require the use of various elicitation techniques to gain access to the data we require and different varieties of experimental manipulation, e.g., the matched-guise experiments. Some kinds of data require various statistical procedures, particularly when we wish to make statements about the typical behavior of a group, e.g., a social class; other kinds seem best treated through such devices as graphing, scaling, and categorizing in non-statistical ways, as in dialect geography or the study of kinship systems.

    A bona fide empirical science sets stringent demands so far as data collecting and analysis are concerned, demands involving sampling techniques, error estimation, and the confidence level, or the level of significance with which certain statements can be made, particularly when arguments are based on numbers, e.g., averages, percentages, or proportions. As we will see, socio- linguists try to meet these statistical demands when they are required. However, many of the conclusions we can draw from sociolinguistic studies are of a non-statistical nature and leave no element of doubt. This is because much of language use is categorical (i.e., something is or is not) rather than statistical (i.e., some phenomenon occurs with this or that probability). A recurring con- cern, then, must be with considering the certainty with which we can draw our conclusions in sociolinguistics. What is the theoretical framework? What are the relevant data? What confidence can we have in the gathering of the data, and in the analysis? What do the results really show? How should they be interpreted in relation to such concepts as ‘identity,’ ‘power,’ ‘solidarity,’ ‘class,’ ‘gender,’ etc.? What do we mean by such concepts? How useful are they in trying to achieve an understanding of how people function in society? What kind of social theory do we subscribe to? In these respects sociolinguistics is like all other sciences, so we should expect no less than that these requirements be met.

    As part of an attempt to work out a set of principles, or axioms, which sociolinguistic investigations should follow, Bell (1976, pp. 187–91), drawing extensively on the work of Labov, has suggested eight as worthy of consideration:

    The cumulative principle. The more that we know about language, the more we can find out about it, and we should not be surprised if our search for new knowledge takes us into new areas of study and into areas in which scholars from other disciplines are already working.The uniformation principle. The linguistic processes which we observe to be taking place around us are the same as those which have operated in the past, so that there can be no clean break between synchronic (i.e., descriptive and contemporary) matters and diachronic, (i.e., historical) ones.
    The principle of convergence. The value of new data for confirming or interpreting old findings is directly proportional to the differences in the ways in which the new data are gathered; particularly useful are linguistic data gathered through procedures needed in other areas of scientific investigation.The principle of subordinate shift. When speakers of a non-standard (or subordinate) variety of language, e.g., a dialect, are asked direct questions about that variety, their responses will shift in an irregular way toward or away from the standard (or superordinate) variety, e.g., the standard language, so enabling investigators to collect valuable evidence concerning such matters as varieties, norms, and change. The principle of style-shifting. There are no ‘single-style’ speakers of a language, because each individual controls and uses a variety of linguistic styles and no one speaks in exactly the same way in all circumstances. The principle of attention. ‘Styles’ of speech can be ordered along a single dimension measured by the amount of attention speakers are giving to their speech, so that the more ‘aware’ they are of what they are saying, the more ‘formal’ the style will be. The vernacular principle. The style which is most regular in its structure and in its relation to the history of the language is the vernacular, that relaxed, spoken style in which the least conscious attention is being paid to speech. The principle of formality. Any systematic observation of speech defines a context in which some conscious attention will be paid to that speech, so that it will be difficult, without great ingenuity, to observe the genuine ‘vernacular.’

    The last principle accounts for what Labov has called the ‘observer’s paradox.’

    He points out (1972b, pp. 209–10) that the aim of linguistic research is to find out how people talk when they are not being systematically observed, but the data are available only through systematic observation. Somehow speakers must have their attention diverted away from the fact that they are being observed so that the vernacular can emerge. This can happen when speakers become emo- tional. Labov found that a question like ‘Have you been in a situation where you were in serious danger of being killed?’ nearly always produces a shift of style away from careful speech toward the vernacular, thus providing the linguist with the kinds of data being sought.


    The uniformation principle mentioned above proposes that there is a relationship between synchronic (i.e., descriptive) and diachronic (i.e., historical) statements made about a language. There has been a long advocacy in linguistics for separating the two (see Saussure, 1959, Bloomfield, 1933, and just about any introductory linguistics text written prior to the mid-1970s).

    To convince yourself that there are no ‘single-style’ speakers, try for an hour or two not to vary your speech style as circumstances change. For example, try to speak to your cat (or dog), your close friends, your teachers, and complete strangers with exactly the same degree of formality (or informality), principles of word choice, precision of articulation, and method of address (e.g., John, Mr Smith, Sir). Report what happened and how you felt about what you were doing as the setting and participants changed. How did others react? (Be careful: you might run into difficulties!)

    For Labov and other sociolinguists the vernacular is very important. What do you understand by this term? When do you use such a variety? How easy or difficult is self-observation of that variety?

    On the whole we will be concerned with the spoken varieties of languages rather than the written varieties. What are some of the essential differences between the two? What do linguists mean when they say that the spoken language is ‘primary’ and the written language is ‘secondary’? How do most people relate the spoken and written varieties?

    (Back to the Top)

    Language and Communities

    In language there are only differences. Strange the difference of men’s talk. Choice words, and measured phrase, above the reach Of ordinary men, a stately speech.


    Correct English is the slang of prigs who write history and essays. Language is by its very nature a communal thing; that is, it expresses never the exact thing but a compromise – that which is common to you, me and everybody.

    I include ‘pidgin-English’ . . . even though I am referred to in that splendid language as ‘Fella belong Mrs Queen

    (Back to the Top)