|Publisher:||Cambridge University Press|
|Sold by:||Barnes & Noble|
|File size:||6 MB|
About the Author
Read an Excerpt
Cambridge University Press
0521836069 - Sociolinguistics - by Florian Coulmas
Introduction: notions of language
Reality does not speak to us objectively, and no scientist can be free from constraints of psyche and society.
Stephen Jay Gould (2000)
Natural language and social language
As human beings we are able to change our behaviour. The idea that we act as free agents is fundamental to our self-conception. Every word we say reinforces this conviction, for whenever we speak we make choices. The ability to consider alternatives and opt for one is basic to intelligent life. It is restricted by our physical nature, the many things we cannot choose, such as the colour of our eyes, our IQ, or whether we are beautiful or ugly. All this may change soon, as the human species gets ready to do with itself what it has done with other species for a long time: interfere with nature's course, select, breed, grow and artificially manipulate their genetic makeup. The life sciences have made spectacular progress over the past several decades, constantly expanding the realm of culture – that which we control – at the expense of nature – that which controls us. No longer confined to science fiction novels, anthropotechnology has crossed the threshold into the real world and become a vital concern of legislation, the paradigm of deliberate regulation of behaviour. The prospects are tempting. Before long, we are told, we will be able to safeguard our offspring against congenital diseases, if not secure immortality for ourselves. At the same time, we are confronted with new challenges, which will be a lot more serious than how to retrain all those undertakers. We will have to decide whether to go down every pathway science opens up or to erect occasional warning signs, STOP HERE, at critical junctures. In short, at the present time, we are forced to rethink our place in the universe, the confines of nature and our own nature.
Language, the inborn
Language plays a peculiar role in this regard. People are born to speak, though they are not born speaking. It is no coincidence that the scientific study of language has been thoroughly impressed by, and, some would claim, has contributed to, the revolutionary changes in the life sciences. For language is seen as an evolutionary adaptation to communicate information. It is what most distinguishes us from other beasts, chatty chimps and brainy dolphins notwithstanding. The exploration of language, therefore, is indispensable if we want to understand our own nature. For language, as cognitive scientist Steven Pinker put it, 'is a distinct piece of the biological makeup of our brains'. 'It is not something that parents teach their children or something that must be elaborated in school' (Pinker grammar school because they think their children's language is in no need of elaboration, and not just because they have misgivings about the school's effectiveness in this regard.
Pinker could of course be right. He would not be the first scientist who gets the better of popular ideas. Language has been known for a long time to have a physical base in our brain, and of late the race is on among biologists to track down the language gene. Given the overwhelming importance of language to the survival of our species it is more than a remote possibility that it is genetically determined. Assuming that it is, we are or aren't equipped with it, and in the latter case no amount of schooling will make up for the deficit. All babies acquire language quickly because they have the ability to do so and because all societies use language. The ability to acquire language is universal and unrelated to intelligence. With the exception of some pathological cases, humans grow up to speak, the dumbest and the brightest. Evidence for that is all around us. What this suggests is that language is innate and common to the species. Those who are chasing the language gene may be on the right track, then. Language helps us survive. But does Italian, or Dutch, or Bengali? Such a proposition would be hard to defend, and no one – except for the authors of 'Survival Italian', etc. – really does.
It would be jumping to conclusions if we were to instruct would-be parents that they must not waste their time teaching their children language, and teachers that they need not bother to elaborate it because the kids are born with it anyway. Thanks to the astonishing nature of language, both sides are right, the researchers who tell us that teaching children language is unnecessary and the parents and teachers who spend so much time and effort doing just that. The disagreement between them is only apparent. This is so because to acquire language both are indispensable, our brain's physical equipment and our society – represented perhaps by a single caregiver – talking to us. Brain damage or genetic deformation and social deprivation will both make language acquisition impossible. Thus, language has two sides, the biological and the social, each of which must be studied in its own right.
Geneticists and other life scientists interested in language are concerned with language in the singular, invariant in space and time. Like-minded linguists and cognitive scientists are devoted to the quest for the ground plan of language that is hard-wired in the brain. They speak of 'natural language' and some of them, therefore, call their field 'biolinguistics'. Its main task is to elucidate the 'faculty of language' which Noam Chomsky, the most influential linguist of this school of thought, has defined as follows:
But compare the visual systems of the French and the Fulfulde and the Fukienese, and you will find that they are virtually identical. If, however, we compare the French, Fulfulde and Fukienese languages the differences are striking. Even French French and Quebec French differ in many ways. Biolinguists take notice of this diversity only in so far as it may help to clarify aspects of the abstract system of rules and principles underlying all languages. Their focus is on general immutable properties of language. Disregarding the still remote chances of genetic engineering to design a better language, no choice is possible here. Linguistic diversity must be dealt with, but, to the great embarrassment of linguistic scholarship, cannot be explained on linguistic grounds. If the faculty of language is part of our genetic heritage and an organ of the body, why does it come in so many vastly different guises? Why are languages so much more different than lungs and adrenal glands? The sobering fact is that there is no convincing answer to this question unless we open our eyes to the other side of language, the social one.
The faculty of language can reasonably be regarded as a 'language organ' in the sense in which scientists speak of the visual system, or immune system, or circulatory systems, as organs of the body.
Language, the historical
Language has been defined by Ferdinand de Saussure, another great linguist of the twentieth century, as a 'social fact' (Saussure social fact in that every language is a collective product, an artefact created by its speakers which, at the same time, enables higher forms of social planning and cooperation to evolve. Society is built on language. There is no human society that does not speak and use language as its central instrument of organization.
Social behaviour has instinctive components, too, but those that are learned predominate. Being socialized means learning the ways of one's society, including its language. Just by following their instinct no one will ever learn Swahili. Every language must be learnt, and it is the society that teaches its new members how to use it properly, how to conform with established conventions. Language, from a social point of view, is conventional, which is another way of saying that it could be different. Every language could be different from what it actually is. We know this because we know that today's languages were different in the past, that they have changed and will continue to change. For the conception of language as a social as opposed to a natural fact this is of utmost importance. Social facts are historical facts. They have many contingent features. Biolinguistics ignores the historicity of language because it is interested in invariance, but to sociolinguistics the historical dimension of language is central. William Labov, one of the leading figures in this field, has identified as his primary goal 'to determine what happened in the history of language or language family' because 'the fact of language change is difficult to reconcile with the notion of a system adapted to communication' (Labov dialects (see chapter 2). The existence of different languages is a historical fact, a result of language change.
The historical character of language and the fact that it must be learned are closely related. It is true that all people learn to speak, as pointed out above; but it is also true that the general ability to learn does not imply that we all learn the same, and equally well. There are good learners and not so good learners, and what they learn is never an exact replica of the model. For instance, the Germans learned from the French the word baguette, 'French bread'. They spell it like the French, and the pronunciation is very close, too. But they changed the gender. The French model is feminine, the German copy neuter. Why? Ignorance, perhaps. The Germans may have been unaware that a French baguette was feminine and simply given the new word the same gender as their own word for bread, das Brot, n. Perhaps more interesting structural reasons were involved, such as the asymmetry between the dual French gender system and the tripartite German one. Perhaps morphophonological rules make themselves felt here. There are many neuter nouns in German ending, like baguette, in [-et], such as Bett, Fett, Brett, Kabinett, Skelett and Sonett, but I couldn't find a single feminine one. However this may be, the gender change of baguette didn't happen naturally. Somebody performed the operation. What the example illustrates is that learning often implies change.
Since French and German are different languages it is not surprising that elements of one adapted to the other will undergo modification. But the same also happens within what presumably is one language. In England, sauce and source are usually homonyms, but in some parts of the United States they are distinct, source, true to the French original, but not sauce, having an audible [r]. Differences of this sort may or may not be indicative of ongoing change. The point here is the same as above, an explanation can be found. If both pronunciations coexist and continue to coexist for a long time it is hard to argue that one is systematically more essential or sound than the other. It is also hard to argue that these differences are superficial and unimportant, because it is sets of variations of this kind that, if they pile up, can lead to linguistic divergence, mutual unintelligibility, and hence the emergence of a new language. This is so because the distribution of source with and without [r] is not random. It distinguishes not individuals but groups of speakers.
Every language is transmitted from one generation to the next by learning and has its unique history. These two facts go a long way to explaining linguistic diversity. Diversity means two things: the multiplicity of human languages – 6,000 is a conventional count – and the enormous variety of coexisting forms in every language. This diversity is the result of many contingent factors working on human speech behaviour. Being open to contingencies, language is neither deterministic nor random. Without such openness, not allowing for adaptation and innovation, it would be rapidly outdated. Luckily, in the process of learning, we do not just repeat what our elders said, but recreate our languages anew adapting them to our purposes and hence bring about change.
Migration and diversity
If we want to appreciate the great diversity of human languages we need to consider another factor, migration. According to a famous dictum attributed to George Bernard Shaw, England and America are 'two countries separated by the same language'. Most speakers of English are aware of the hiatus between British and American speech, but find it quite unremarkable because the cause is so obvious: the Atlantic Ocean. English, French, Spanish, Portuguese and Dutch in the New World aren't quite what they are in the old. We take it for granted that over long periods of time geographic isolation brings about linguistic divergence. People living in different environments speak about different things; in the process they mispronounce words, create new ones, reinterpret morphological forms, borrow lexical items from others and put them together to form sentences in novel ways. This must have been so from the beginning (assuming that the beginning of humanity can ever be lifted out of the realm of speculation1). Where a substantial body of population moves out of one territory and into another, driven by demographic pressure, commerce, or the incursion of invaders, it will take its language with it, but after some time this language ceases to be the same as that spoken in its original territory. From a theoretical point of view, this is remarkable because it means that social factors are involved in language change. If language change were deterministic, thrust towards a goal and governed entirely by quasi-natural laws inherent in the language system, as in the past historical linguists have claimed,2 we should expect it to be unaffected by migration. In the event, English should continue to change along the same lines on both sides of the Atlantic. But as it turns out, once a group splits into two, language change is no longer synchronized. Since the two groups are stripped of the opportunity to adjust their speech to each other, the transmission and recreation of their language is propelled onto different trajectories.
Desires and norms
Migration usually induces language change, but a speech community's spatial contiguity and temporal continuity are no guarantee for maintaining linguistic homogeneity or a sufficient condition for bringing uniformity about. The obvious function of language as a marker of distinction dividing one speech community from another comes to bear within a single speech community as well. The argument that linguistic variation will decrease with intensity of communication has often been made, but there are good reasons to doubt a causal relationship in this connection. For variation serves important social functions.
As we will see in the course of this book, it is not at all rare that linguistic distinctions withstand ostensible forces of homogenization. Why linguistic distinctions are maintained in the face of both homogenization pressure and the opportunities offered by uniformity is one of the key questions that brought the discipline of sociolinguistics into existence. Efficiency of communication would seem to call for a reduction, if not elimination, of potentially disruptive distinctions in the speech of individuals and groups. Yet, such distinctions persist.
In highly stratified societies such as the caste societies of India, it is quite possible for people to be in constant and regular communication over long periods of time without adopting each other's speech patterns. It would seem that communication leads to uniformity only when there is both the possibility and the desire for social assimilation. Where social norms put a premium on social distinctness, linguistic symbols of such distinctness tend to be maintained.
(Gumperz 1967: 228)
Two important notions in the passage quoted above are 'the desire for social assimilation' and 'social norms [that] put a premium on social distinctness'. Both are invoked as causal factors in the process of language change. Whose desire he refers to Gumperz does not tell us, but it is clear that speech communities, social groups and their members are at issue. In what sense an assemblage of individual desires can be understood as a collective desire is a difficult question to which we will return later. For the present purposes the important thing to note is that mental dispositions such as the desire for assimilation (or division) influence language change. Desires and the willingness to adhere to, or breach, social norms make a difference, since it is by virtue of its members having desires and preferences that the speech community creates and perpetuates its language.
This is testimony to the intrinsically mental character of language. Speakers, rather than just being the bearers of abstract structures removed from conscious reflection which constrains their speech behaviour, are active, knowledgeable, purposeful agents who make choices whenever they use language. The ability to do so is at the heart of the nexus between language and society, and it is the vantage point of this book. Speakers make choices. The subsequent chapters will show that this holds for every level of language, structural and stylistic (chapter universal grammar, and every individual's language represents a choice of his or her collectivity's language. Social norms are restrictions on individual choices making deviations that imperil communication unacceptable, if not impossible.
Speakers cannot avoid making choices, for things can always be formulated differently. People high and low have strong feelings about the intentionality of their speech and they articulate what they believe they should, although they sometimes seem to belie their own words. 'I know what I believe. I will continue to articulate what I believe and what I believe – I believe what I believe is right.' So much for beliefs and articulation. US President George W. Bush articulated these words on a visit to Italy3 near the Forum Romanum where orators used to speak. Surely, they could have been chosen more adroitly, but chosen they were.
Choice is the pivotal notion of sociolinguistics, and I will have to discuss this notion in some detail to see what it means with regard to human action, in general, and to speech behaviour and language, in particular. Before doing so, let me summarize the main points discussed so far concerning the different conceptions of language as a natural fact and a social fact.
|Language as a natural fact||Language as a social fact|
|genetically fixed||culturally varied|
|governed by natural law||governed by convention|
In the language sciences it is sometimes thought that the two approaches dedicated, respectively, to the natural and the social side of language are irreconcilable. I prefer to think that they are complementary because neither of them can by itself fathom out the whole complexity of language. The language sciences all have their own notions of language allowing them to direct their attention to certain phenomena rather than to others, and there is no reason to believe in the superiority of one over another. It is necessary to emphasize this point because the sciences of historical complexity often occupy a lowly position in comparison with 'pure' and 'experimental' disciplines. A hierarchy that ranks biolinguistics or formal linguistics, as it used to be called, with the 'hard' sciences at the top and sociolinguistics with the 'soft' sciences at the bottom is useless and unjustified for two reasons. The late Stephen Gould, a professor of palaeontology, has formulated them clearly and elegantly. One is quoted at the beginning of this chapter: 'Reality does not speak to us objectively, and no scientist can be freed from constraints of psyche and society' (Gould Gould's conclusion is this:
Historical science is not worse, more restricted, or less capable of achieving firm conclusions because experiment, prediction, and subsumption under invariant laws of nature do not represent its usual working methods. The sciences of history use a different mode of explanation, rooted in the comparative and observational richness of our data.
Since human bodies consist of particles, their behaviour should be explained in terms of particles and the laws governing their movements. There is no room for a mind with a free will. At the same time, our everyday experience is that our reasonings and choices govern our behaviour, to a significant extent at least. This is the mind–body problem in a nutshell. No attempt will be made here to solve it, but we cannot altogether sidestep it, for, as we have seen, language has both a physical and a mental side, and these are not always easily kept apart. Sociolinguistics is the linguistics of choice, and, if only for that reason, we have to come to grips with the relationship of freedom of the will,4 human action and language, for choice is a notion which presupposes an agent rather than an automaton. The intricacy of the problem has been pinpointed by two scholars representing as it were the two sides, the neurologist John Eccles and the philosopher Karl Popper. Interestingly, they see the very origin of language indissolubly linked with choice. Here is what they say:
Choosing to speak before you know what to say, let alone know what language is, seems quite a feat, but, on reflection, it may be quite common. The important point is that making choices is a central part of the human condition. Interestingly, Eccles and Popper's notion of choice does not require full control and foresight. This is important, for, while I don't want to take a position here as to whether or not ontogeny repeats phylogeny, I want to argue that babies make choices, because every line we draw to show where intentionality begins is arbitrary. Our choices are subject to restrictions of various kinds from birth. The division of labour in the language sciences can be understood in terms of the restrictions on possible linguistic choices. Physical and cognitive restrictions are the field of biolinguistics and cognitive science; social restrictions on linguistic choices are for sociolinguists to investigate.
We could say that in choosing to speak, and to take interest in speech, man has chosen to evolve his brain and his mind; that language, once created, exerted the selection pressure under which emerged the human brain and the consciousness of self.
(Popper and Eccles 1977: 13)
For instance, our lifespan, or, less dramatically, the need to sleep, puts natural restrictions on the length of our sentences. The range of speech sounds is restricted by our auditory system which is designed to perceive and process sounds in a range between 12 and 20,000 cycles per second. Structural restrictions are the subject matter of grammar studies. Some are very general, forming the theme of universal grammar, whereas others are applicable to some languages or a single language only. Gender agreement between article, noun and adjective is a good example. French bon [bɔ ̃] and bonne [bɔn] are, respectively, the masculine and feminine forms of 'good'. It is bon mot, literally a 'good word', that is, a witty remark, but bonne action 'a good deed', because mot 'word' is masculine, whereas action 'deed' is feminine. Choice between [bɔ ̃] and [bɔn] is not up to the speaker's taste, but determined by agreement rules. Agreement rules are restrictions on choices. What they mean is that, if a language has a gender system, speakers are not free in their choices of gender forms of nouns and adjectives and articles. The requirement to match adjective and noun in terms of gender could be a peculiarity of French. Comparative studies reveal that it is much more common and typical of all languages that have a gender system.
The central theme of sociolinguistics is variety. To the observer, language presents itself as a seemingly infinite variety of forms, but this variety is patterned. That is, there are restrictions on choices between coexisting varieties. For instance, English words like fast have, in standard British pronunciation, a long vowel [a:]. If you want to sound a bit archaic, or Australian, you can pronounce it with a short [a], and in some American varieties it borders on [æ]. Australians living in London tend to lengthen their [a] along with other adjustments they make to blend in with their environment. Such fine-tuning has to do with preferences and social norms rather than structural rules, which is not to say that it is random. Quite the contrary, in the absence of patterning we would be unable to recognize speakers for what they are. Speech varieties are powerful markers of group membership. Outsiders, particularly children, can be observed making great efforts to use the right words and give their pronunciation the right tinge to conform with the group they are trying to join. It can be done. But as George Bernard Shaw's Eliza Doolittle under the able and loving guidance of Professor Higgins found out, discovering, in the great variety of available choices, the socially acceptable one is essential.
Professor Higgins, on a whim, bet his friend that Eliza, the flower girl with that ear-piercing drawl, could be transformed into a lady. The language was the key. With determination and linguistic genius she made him win his bet, substituting her native Cockney by the unmistakable accent of the upper class. Hers was one speaker's choice, remarkable and, though fictitious, of deep significance to the relationship of language, social class, gender and upward mobility, topics that will be dealt with in chapters 2 and 3 below. Eliza Doolittle is so noteworthy because her choice is readily recognizable as such, implying as it does a change in group affiliation. It is also remarkable because the nuances of pronunciation were raised to the level of conscious manipulation. More commonly we leave these to the automatic pilot and stay with our group, because we see no reason or opportunity to do otherwise. Strictly speaking, this is also a choice which corresponds to the above remark that everything can always be formulated differently. It implies that every speaker has the ability to change the way he or she speaks. The choices speakers make in this regard are not made in a vacuum but are constrained in many ways.
© Cambridge University Press
Table of Contents1. Introduction: notions of language; Part I. Micro-choices: 2. Standard and dialect: social stratification as a factor of linguistic choice; 3. Gendered speech: sex as a factor of linguistic choice; 4. Communicating across generations: age as a factor of linguistic choice; 5. Choice and change; 6. Politeness: cultural dimensions of linguistic choice; Part II. Macro-choices: 7. Code-switching: linguistic choices across language boundaries; 8. Diglossia and bilingualism: functional restrictions on language choice; 9. Language spread, shift and maintenance: how groups choose their language; 10. Language and identity: individual, social, national; 11. Language planning: communication demands, public choice, utility; 12. Select letters: a major divide; 13. The language of choice; 14. Research ethics.