Domains of Language Use: A Fundamental Concept for Framing Language Policy in South Africa ` – Chris Jeffery and Rajend Mesthrie On behalf of the English National Language Body of the Pan South African Language Board A valuable achievement of sociolinguistic theory has been to make explicit fundamental facts about language that everyone knows intuitively yet easily overlooks. Two such pertinent facts are: 1. Language is invariably, indeed necessarily, embedded in and given meaning by the situations in which it is used. 2.
All communities use language differently according to situation of use; and many communities go a step further and actually use different languages according to situation. Ultimately every situation and every individual user is unique, but for the purpose of analysis it is necessary to make abstractions by organising instances of this concrete reality into categories. Abstractions from individual habits, regarded as tokens, constitute types of linguistic behaviour known as registers or varieties; and, at a higher level, as languages.
As with language, so with situation: situation-types must be abstracted from tokens, that is the particular occasions, by asking who is using language to whom, how, why, when and where. The answers are arranged in sets of categories; and the basic, indispensable category is generally known as domain (of discourse). Linguistic variation typically correlates primarily with domain. The sociolinguistic notion of domain was formalised by Joshua Fishman (1972), who stressed that different settings characteristically call for the use of different languages in a multilingual society (or varieties of the same language in a monolingual society).
At one level a domain is a concrete setting like the home, the street, the classroom, a shop, university, a religious institution, the media etc. More plausibly, the determinant of using one language variety over another is not the physical setting alone, but the general activity (“event”) conventionally associated with the setting. So “church” stands for the range of activities associated with religion. In some societies this domain calls for a switch to another language 1 PDF created with pdfFactory Pro trial version www. pdffactory. com (Latin, Sanskrit, Hebrew, Ge’ez etc. . In all religious societies the domain calls for a formal, “high”, “sermonistic” and somewhat archaic variety of language. Other broad domains would commonly be home, school, work, sport and recreation; and each has distinct linguistic correlates. Boardroom and bedroom, dining-room and changing-room all differ in discourse. For the purpose of analysis narrower distinctions of domain may be required. For example, within the work domain, boardroom and shopfloor differ in discourse; likewise, a newspaper office and an IT company; and so on.
The association of particular languages or varieties of language with specific domains is a widespread convention in human societies, but it is not necessarily fixed: An influential leader may avoid an over-formal or ornate style in favour of language closer to more informal domains. The language of the home is often associated with customs and traditions of a community; however, where the community itself is in the process of change and/or modernisation, language practices may also change. Since speakers are at least intuitively aware that their practice is meaningful in itself, it is clear that meaning is associated with choice of form.
Furthermore, domain has a pervasive semantic effect. In order for the finite resources of language to encompass the potential infinity of meanings in the extra-linguistic world, polysemy is a necessary condition: words must have many meanings, which vary essentially with domain. So for example partners in the boardroom are not (normally) partners in the bedroom. Everyone knows that meaning depends on context; but not everyone appreciates how much more of that context is constituted by the situation than by the sentence. It was pre-eminently M. A. K. Halliday (especially 1978) who showed how meaning correlates with situation-type.
Rather than being inherently meaningful, discourse means what people expect it to mean, and their expectations are determined primarily by the domain that they are engaged in. In effect the domain is charged with meaning-potential, which is realised by uttering of words, and which reciprocally determines what those words can mean. A simple illustration of the principle is the shout of “Fire! ” in battle and in a crowded cinema: what can be meant in each domain is decidedly different; so it could be no excuse for alarming the patrons to explain that you were vociferously enjoying the shooting on screen.
The domain determines not just your linguistic choices but also, by virtue of the expectations which it primes, what those choices mean to other participants. (In accordance with this insight Halliday refined his concept of register from “variety of language” to “meaning2 PDF created with pdfFactory Pro trial version www. pdffactory. com potential”; but the term retains its primitive meaning for many users. ) Domain is the primary category determining situation-type (see above paragraph). Once that has been defined, analysis proceeds to ask other questions: What is the topic?
Is it being discussed in speech or writing? How many interlocutors are there? How do they stand in relation to each other? As these other parameters vary within domain, so do the linguistic choices, and their possible meanings. Thus, at work, interactions with superiors, peers and subordinates constitute significantly different situation-types, eliciting correspondingly different styles of discourse; likewise at home, with partner, children and guests; and so on. In each case the domain is defined, and then the relationships and attitudes, all of which determine choices, of both language and meaning.
Clear expression of relationships and attitudes can be found along the cline of formality. While the language associated with the religious domain stands at the most archaically formal end, at the other end the most informal types of language are associated with street and playground, domains in which young people typically experiment and play with language. The domain of education, by contrast with street and playground, utilises a more formal variety that is associated with the formal accumulation of knowledge.
By contrast with religion, on the other hand, it prefers progressive-sounding language to archaicsounding. In multilingual societies, which are probably commoner worldwide than unilingual ones, domain often determines the choice not of variety (or more precisely, register) within one language but between different languages. Thus, language A (such as Pidgin English) may be accepted as appropriate for business dealings, and language B (such as Latin or Sanskrit) for religion, while languages C, D, and E are preferred by various communities for interactions between families and friends.
It counts as socially inappropriate behaviour for anyone to contravene these norms, for example to use language A in a religious context instead of language B. This is the normal state of affairs in multilingual societies. A society in which every language enjoys equal status and usage in every domain is inconceivable. We have to deal with dynamic realities. South African government policy on multilingualism provides for parity of esteem for all our languages; and also for equitable usage.
Policy thus allows for the principle that not all languages are necessarily equally used in all domains, while still enjoying equal status. 3 PDF created with pdfFactory Pro trial version www. pdffactory. com The correlations between domain and choice of language in SA have still to be ascertained by research. Also desirable is a review of practice in other multilingual societies. Nevertheless, prior to detailed empirical studies, some sociolinguistic facts are immediately observable. It is clear that the widespread multilingualism in SA is mediated via the principle of domain specialisation.
A young person might speak Zulu in the home, use a formal variety of it in church, informal “street Zulu” with friends (or Tsotsitaal in mixed male peer groups), and Zulu and English at school. In most domains no intervention is necessary, as communities evolve their own norms by consensus over time. Thus, for example, English has come to predominate in the commercial domain in SA, as in much of the world, leaving small scope for equal deployment of the other ten languages in this one corner of the vastly complex global market. Education and other public domains (like the state media) are, however, different.
Here choices are made from above (at political or administrative level), rather than from below (by students or parents alone). The domain of education requires two language decisions: one concerns the earliest levels of schooling, where it is necessary to use a language that children can understand. The other concerns the choice of a language in which students will engage with knowledge at more advanced levels. In most countries this has been the language in which reporting of research as well as discipline-specific writing has been undertaken.
South Africa is not alone in the world in having to adapt to the use of an international language of learning at higher levels. Today many countries, including some which formerly had their own language featuring prominently in higher education, are having to concede to English in the scientific and technical fields. In these domains in the modern world English performs the function that during the Renaissance was performed by Latin, namely, a neutral and extremely useful means of communication between educated people across national boundaries and frequently within them.
These qualities neutral and useful may still need stressing in countries which remember that English used to be the language of empire. That is well in the past now, attitudes have changed, and it has turned out that erstwhile colonies are endowed with the global language as a readily accessible resource. Since that gives them a headstart over countries where English is utterly foreign, it would probably be wise for modern policy to aim to look ahead and capitalise on this resource. 4 PDF created with pdfFactory Pro trial version www. pdffactory. om In South Africa many parents seem to realise that English has the double advantage of being the vehicle both for wider contact with the world and for science and technology. More controversially, some parents see no need to teach languages that their children already speak. Here sociolinguists and applied linguists feel it their duty to advise parents that no language is safe unless it is maintained in the home and accrues some value in more public domains (such as administration and the lower echelons of the law on a regional basis).
Part of this profile would be the inculcation of African languages as school subjects for their literary and cultural value, and the appreciation and preservation of indigenous knowledge. In theory all languages have the potential to be used as the medium for university education. However, that potential has to be realised (as it once was for Afrikaans) by the will and the activities of writers, academics and leaders from the language communities. One difficulty for such persons is that educated professionals tend to communicate ideas across provincial, state and continental borders.
Creative writing and activity in all of South Africa’s languages should be supported. But for science and academic subjects, realistically, English should continue to be inculcated. This is not to the exclusion of other languages, and pupils could continue using their home languages to paraphrase concepts for better understanding where necessary. Code-switching has been shown to be a useful ally in the classroom. But the gains made by Black students over the years in the Sciences via English should be strengthened, not thrown away.
It would be more efficient for society, and beneficial for individuals, to equip every matriculant with the requisite English to function unfettered at university. In this country the regional distribution of languages is an important primary situational parameter beside domain, as witness the location of NLUs. As for the two without an identifiable home-territory, English broadly predominates in the cities, and Afrikaans elsewhere. One domain in which region might be discounted in principle is the interface between government and governed, because here, within reason, citizens should be free to use their own languages. PDF created with pdfFactory Pro trial version www. pdffactory. com Summary and Conclusions 1. Domain as a technical term defines activity rather than place: it means the type of activity in which language is being used; or, in other words, in which the discourse is embedded. 2. Domain is an extra-linguistic category. What is at issue is not categories of language but categories of discourse-situation: the situations in which language is used. 3. Domain is the primary category for defining types of situation. The term type is used by implicit contrast with token, because linguistic analysis works with abstractions rather than the actual instances underlying them. ) 4. Domain crucially determines linguistic meaning by setting meaning-potential, or register: what discourse can mean in a given situation-type. 5. Domain has an important bearing on SA’s multilingualism policy. While discussion of the policy hitherto has usually taken account of the regional distribution of our many languages, it has not accorded due significance to domain.
Policy should aim: O to empower all South Africans to use their own languages not necessarily in any and every domain but in appropriate domains; and furthermore: O to give all South Africans the opportunity to become sufficiently multilingual to gain confident entrance into those domains in which other languages happen to be appropriate. In order to implement the policy effectively, the task remains to ascertain the facts: firstly, about who uses what languages in which domains; and secondly, about the extent of the blurring of boundaries between languages, especially in informal domains.
So there is a clear need for a comprehensive programme of well-planned and thorough research. References Fishman, J. 1972. Domains and the relationship between micro- and macro-sociolinguistics. In J. J. Gumperz and D. Hymes (eds. ) Directions in Sociolinguistics: the Ethnography of Communication, 435-453. Oxford: Basil Blackwell. Halliday, M. A. K. 1978. Language as a Social Semiotic: the Social Interpretation of Language and Meaning. London: Edward Arnold. 6 PDF created with pdfFactory Pro trial version www. pdffactory. com