In April 2003, Charles Clarke, British education minister and notionally a Labourist, declared that public funds should no longer be devoted to ‘ornamental subjects’ like medieval history or classical literature. Such provocations, too deliberate and insistent not to intimate a policy, earned him epithets such as ‘philistine mobster’ and ‘intolerant yahoo’. In themselves, these attacks did not directly concern language: neither linguistics (which still enjoys a vague scientific aura) nor philosophy (which Mr. Clarke had the good grace to exclude from his attacks) are directly implicated. But to attack the history–culture nexus, the cultural past that is inscribed in the English language, out of which the English language is made, presupposes a conception of language as a tool and lingua franca, a simple instrument for the transmission of information and knowledge, without depth or past. Lecercle² has inveighed against the monolingual habitus of most English mother-tongue speakers and, correctly, puts this down to the fact that English is the language of globalisation.

In the context of a country that still has a sense of being in a distinctively transitional phase of its ‘development’, it is a matter of great interest to me as a sociologist of language to observe that, unlike most contemporary South African policy documents, some reference is made to the importance, indeed the centrality, of language in two reports that are being discussed currently – the Final Report on the Charter for Humanities and Social Sciences³(HSS) and the Consensus Study on the State of the Humanities in South Africa: Status, prospects and strategies⁴. It has always been strange to me that in spite of the linguistic diversity that characterises our social formation, most scholars, by ignoring the language question, have merely swum along with the tide of unthinking humanity, unable to see the rocks towards which their blind spot was propelling them. I suspect that we have to thank recent statements and actions on the part of the Minister of Higher Education and Training, Dr Nzimande, for their lighthouse effect in illuminating the space in which we can begin to rethink our own approaches and our specific research efforts in ways that deliberately take the language issue into account as one of the parameters that frame our efforts.

A few particularly significant remarks from both documents deserve to be highlighted here. In the Charter Report, the task team proposes that, as one of six ‘catalytic projects’ there be initiated

a national multidisciplinary project on how indigenous languages in South Africa could support the process of concept formation in the HSS, and furthermore, what know-hows in these languages could enrich social scientific thinking or pedagogy.^3(p.20)

In motivating this novel suggestion, the task team asks:

For example, does the distinction between ‘sebenza’ and ‘dala’ create new ways of understanding labour and work in sociological contexts? Does the fact that there are at least 23 ways of describing ‘poverty’ in the Nguni oral lexicon offer a way of understanding agencies better? Does the grammatical structure of local languages demand a different logic from the Boolean or the Aristotelian? A project, if carefully designed, would move the discussion of language to the heart of HSS.^3(p.38)

The task team return to this issue on pages 42 and 43, and, in spite of a definite vagueness in their formulations, there is a clear determination that research and other scholarly work in this domain will, if they have their way, receive not only more attention but also more appropriate funding and close monitoring. They also propose, as one of a series of ‘corrective interventions’, that the BA qualification be reviewed to take into account ‘mother tongue proficiencies’^3(p.22) and, in general, acknowledge the centrality of language to all science but especially to the humanities and social sciences. All of which is very good news to those of us who have for so many years pointed towards this very horizon against uninformed and unnecessary opposition.

The Consensus Study is less definite on the language question but equally committed to doing something about it. The statistics (p. 66) which the authors have collated on the decline in the numbers of students who take African languages as a major at universities and other post-secondary institutions, confirm the trend that has been signalled with an increasing sense of panic and desperation by linguists and other language professionals ever since 1996 at least. The fact that only 0.5% of all undergraduates are enrolled in African languages departments in South Africa (p. 131) is a damning indictment on social and especially educational policy during the past 15 years, and the study correctly characterises this as ‘a critical position’. In an African country, this situation is, to put it mildly, absurd, and it is clear that this domain has to become a research priority if we are to correct things and escape from the concomitant problems of mediocrity and perceived ‘failure’.

It is a very positive sign, therefore, that in their second recommendation, the authors of the Consensus Study call for a review of state funding for areas of study such as African languages and align themselves explicitly with the Charter Report in this regard, without expanding on their reasons for doing so. Recommendation 8 calls for ways of privileging languages throughout schooling up to Grade 12 so that students would be encouraged to consider further studies in these subject areas.

In the light of these positive signals, I want to use this opportunity to draw attention to some of the more fundamental insights of various schools of thought operating in the discipline known as ‘Sociology of Language’ with a view to providing one way of approaching the formulation of a theoretical basis for promoting the abovementioned ‘catalytic projects’ and ‘corrective interventions’. My immediate purpose is to alert especially the privileged elite layers of the new South Africa to the fact that we ignore the details of the language question at our peril. I should like, beyond that, to show that implementing a consistently democratic language policy is a critical component of the consolidation and expansion of the democratic society to which we are committed.

[Language] is built into the economic and social structure of society so deeply that its fundamental importance seems only natural. For this reason, language policies are often seen as expressions of natural, common-sense assumptions about language in society.

The purpose of his book is to rebut this all-pervasive notion and to demonstrate by way of many significant historical and contemporary examples that language policies are governmental strategies designed, mostly consciously, to promote the interests of specific classes and other social groups. Thus it is not true that languages simply develop ‘naturally’, as it were. They are formed and manipulated within definite limits to suit the interests of different groups of people. This intention is very clear in the case of so-called standard languages, as opposed to non-standard varieties (dialects and sociolects). ‘Standard languages’ are invariably the preferred varieties of the ruling class or ruling strata in any given society. They prevail as the norm because of the economic, political–military, or cultural–symbolic power of the rulers, not because they are ‘natural’ in any meaning of the term. The importance of this proposition derives from the fact that it validates the claim that languages, just like cities or families, can be planned. Indeed, it is a fact that in any modern state, whether or not it is explicitly acknowledged by governments, languages are always planned, in that legislation prescribes, often in great detail, where and how one or more languages are to be used. This practice is universal and has significant consequences in critical domains such as education. In regard to post-apartheid South Africa, it remains to be said that the principle – as well as the practice – of language planning is accepted. However, lack of implementation planning and, thus, of delivery, tends to negate the principle and to reduce it to mere lip service.

For reasons connected with the colonial history of southern Africa, the language of power in post-apartheid South Africa is undoubtedly English. Afrikaans continues to play an ancillary role in the processes of economic production in the formal sector of the economy even though there are determined attempts to reduce its significance in this domain as well as in other high-status domains. The question that we will have to consider presently is whether this fact in and of itself implies, as is often said and universally assumed, that ‘English is enough’ and what the implications of this belief are for democracy and development.

The other source of the power of language is its function as a transmission mechanism of ‘culture’ or, more popularly, its role in the formation of individual and social identities. I shall discuss this no further here, even though it is necessarily implicated in the general discussion of the broader topic of ‘language, class and power’. The reason for this is that consideration of the psychological and sociological issues involved in the question of identity would tend to blur our focus. Suffice it to say, therefore, that being able to use the language(s) one has the best command of in any situation is an empowering factor and, conversely, not being able to do so is necessarily disempowering. The self esteem, self-confidence, potential creativity and spontaneity that come with being able to use the language that has shaped one from early childhood (one’s mother tongue) are the foundation of all democratic polities and institutions. To be denied the use of this language is the very meaning of oppression. In the words of V.I. Lenin⁶:

… [a] democratic state is bound to grant complete freedom for the native languages and annul all privileges for any one language. A democratic state will not permit the oppression or the overriding of any one nationality by another, either in any particular region or in any branch of public affairs.

Forty years ago, Alexandre⁷ showed clearly that for post-colonial Africa, proficiency in the language of the former colonial power (English, French or Portuguese) constituted ‘cultural capital’ and was an index to the class location of the individual, as this ability almost automatically elevated the speaker into the ruling elite. The hierarchical relations between different varieties of a language, or between different languages, are a reflection of the historically evolved relations of domination and subjugation between the speakers of the relevant varieties or languages. In the South African case, Dutch, English and, later, Afrikaans, came to be the ‘legitimate languages’ in different periods of our history. This legitimacy was/is the result of colonial conquest in the first instance but, as the structural transformations that accompanied that cataclysmic series of events became routine, dominance was complemented and reinforced by hegemony. That is to say, consent of the victims of colonial subjugation became the major factor for the maintenance of English and Afrikaans as the legitimate languages (until 1994 approximately). In South Africa, unlike most other African countries in the British sphere of influence, the presence of a relatively large group of first-language speakers of English reduced the potential ‘profits of distinction’ that came with proficiency in the legitimate language, although the rate of profit remains relatively high. I make the point here, simply to stress the fact that there is a material reason for the maintenance of a particular language policy in any given period.

This insight, true as it is, does not tell us anything about the class consciousness or the class position of individual members of the elite. What has to be established in any given case, therefore, is the degree of consciousness of the ruling strata of the de facto policy of ‘elite closure’⁸ or exclusion of the masses by means of language policy.⁹ To do so is no easy task, because the levels of mystification and, more problematically, the veils of ignorance that delude policymakers and other power brokers into believing that their understandings are ‘scientific’, defy the logic of mere argument and historical experience. The recent hearings on and public discussion of the South African Languages Bill have reinforced this impression.

The relevant essential proposition is simple enough. It states that in a multilingual society, it is in everyone’s interest to learn the dominant language (of power), because doing so will help to provide equal opportunities in the labour market as well as in other markets. In post-colonial Africa, this approach has led to the almost complete marginalisation of the local languages of the people and the valorisation of English, French and Portuguese in the relevant African states. Indeed, in most other African states, the distinction between ‘official’ (European) and ‘national’ (African) languages ironically highlights in an unintended manner the social distance between the elite and the masses of the people. Because of the role model status of the middle class in most societies, the monolingual habitus^10,11 becomes generalised in such a manner that the vast majority of the people in so-called anglophone Africa come to believe that all that matters is knowledge of English. This belief means simply that most African people are willing to maintain their first languages in the primary contexts of family, community, primary school and religious practice but they do not believe that these languages have the capacity to develop into languages of power. Their consciousness reflects the reality of the linguistic market and they have become victims of a monolingual habitus,^10,11 in spite of the fact that most African people are proficient in two or more languages. This utterly disempowering disposition assumes the character of a social pathology which I have called the ‘Static Maintenance Syndrome’.¹²

To add insult to injury, Tollefson’s paradox⁵ notes that in modern societies, while vast resources are directed toward language teaching and bilingualism, especially involving English, more people than ever are unable to acquire the language skills they need in order to enter and succeed in school, obtain satisfactory employment, and participate politically and socially in community life. The great linguistic paradox of our time is that societies which dedicate enormous resources to language teaching and learning have been unable – or unwilling – to remove the powerful linguistic barriers to full participation in the major institutions of modern society. Inadequate competence is not mainly the result of poor books and other texts, inadequate pedagogy or lack of motivation and other similar suggested deficiencies. Instead

language competence remains a barrier to employment, education, and economic well being due to political forces of our own making. For while modern social and economic systems require certain kinds of language competence, they simultaneously create conditions which ensure that vast numbers of people will be unable to acquire that competence. A central mechanism by which this process occurs is language policy.^5(p.7)

Again, I can only point to the current debates and conflicts around the South African Languages Bill to confirm the accuracy of these propositions. While it is understandable, given the colonial and racist history of South Africa, that before 1973 the ruling class was fundamentally concerned with maintaining the limited markets in raw materials and semi-processed commodities which South Africa, because of its place in the international division of labour, had to provide to the transnational corporations and other imperialist entities, the implicit continuation of such policies in post-apartheid South Africa is something of an anomaly. For, whereas in apartheid South Africa, the rulers could afford to, and did, approach African languages as though they had no economic or cultural value, in the new South Africa, this attitude is clearly self-limiting and self-defeating, if not self-destructive. Unless we are prepared to grant that we are simply trotting along the footpaths pioneered by the neo-colonial states after 1960, where the indigenous languages of Africa were not seen as resources but as problems. In this connection, it is germane to our focus to point to the fact that Africa, including South Africa, is today subject to the intensified pressures of ‘globalisation’ and that the pressure to adopt English, which is incontestably the global language, as the only legitimate language is exceptionally strong in ‘anglophone’ territories.¹³ The focus on English, the ‘language of globalisation’, is also one of the corollary implications of the prioritisation of the science, technology, engineering and mathematics disciplines in South Africa, as it is in most other countries today. The HSS disciplines, conversely, have, as part of a counter-hegemonic strategy, to ensure that the local languages are not marginalised and endangered. Indeed, in a modern African country, these languages should be at the heart of all development, including in the formal economic sector.

In this respect, a caveat applies: we have to be exceptionally careful not to fall into the trap of prescribing cures that turn out to be worse than the sickness itself. A recent strand of analysis that attempts to view the political economy of language in terms of ‘functional multilingualism’ in economic life runs the risk, in my opinion, of promoting a kind of economic diglossia where the ‘minority’ languages are confined as instruments of communication in the processes of production, exchange and distribution to the so-called informal sector, against the nationally dominant languages (in Africa, these are almost without exception languages of European origin) that perform these powerful functions in the ‘formal’ economy. This approach to the issue derives from a dual-economy paradigm that has a long history but, even if it had been useful in earlier times, is particularly irrelevant and misleading in the era of globalisation. Ultimately, it may do no more than serve as an apologetic justification for the perpetuation of existing social stratification. In Africa specifically, the languages of the majority of the people have to become the dominant languages, in whatever combinations, in the respective economy, taken as a whole, of the individual countries. Only if this happens will the danger of a two-tier citizen-subject social model be countered in favour of a democratic system where all are citizens and all have similar life chances. Djité¹⁴ wrote a useful analytical essay on the subject from the point of view, not of minorities in Europe, North America and elsewhere, but of the ‘third world’, where the ‘informal economy’ is often the major contributor to the GDP or the main source of employment.

In the economically more-developed of these countries, this informal sector constitutes a set of niche markets in which, necessarily, local languages are essential for lubricating the economic processes.^14,15 These niche markets are often rapidly occupied by the products of multinational firms – one more indication that the notion of a ‘dual economy’ is no more than an abstraction.¹⁵ The economic benefits of multilingualism should be transferred to the central economies. In respect of Africa, communication facilitated in the local languages will remove the inefficiencies introduced by the selection and promotion of the official language, and policies that promote growth with equity are necessary to achieve socio-economic inclusion for all.¹⁴ English may be the language of global trading, but the ability to speak other languages nonetheless ensures a competitive edge.¹⁵ The multilingual populations of inner-circle countries are a valuable resource, which we overlook at our peril. Their contribution to international business is becoming increasingly evident in areas such as China and the Middle East. Initiatives that target minorities rely heavily on the knowledge and experience of minority-language speakers. Bilinguals are a marketable commodity – the ability to speak other languages opens up a far wider range of better-paid employment opportunities than might otherwise be the case. I need not, I am sure, belabour a point that is dramatically present on every street in the central business districts of just about every South African city today.¹⁶

The legacy of apartheid education in South Africa exacerbates the static maintenance syndrome referred to above, as most Black people continue to equate mother tongue based education with the ravages of Bantu education. I maintain without fear of contradiction that this tendency, even though there are currently the beginnings of some hesitant countervailing tendencies, will continue to undermine South Africa’s ability to expand and consolidate democracy and at the same time represents a built-in constraint on economic development, the magnitude of which remains to be established by means of carefully designed research in all branches of the economy. As I shall indicate below, it is one of three language-related areas of research that I believe we should prioritise in the next decade.

The following are a few examples of how we unnecessarily restrict the capabilities of our workforce and the efficiency of economic production besides the not unimportant factors of inadequate job satisfaction and a reduced work ethic. Very little detailed research has as yet been done in this area, so the numbers I quote probably do not reflect the real magnitude of the problem. African language speaking learners in the Western Cape tend to do badly in the matriculation examination largely because the medium of instruction and assessment is not the mother tongue, but a second or third language,¹⁷ illustrating all too clearly an avoidable continuity between apartheid and post-apartheid education. Significantly, the only ‘learning area’ in which all the matriculation candidates performed at comparable levels was the First Language (Higher Grade) subject (i.e. English, Afrikaans and isiXhosa). For the Xhosa first language speakers, this subject is the only one in which they were taught and assessed in their mother tongue.¹⁷ Simkins and Patterson¹⁸ confirm, in respect of the causal significance of the medium of instruction factor, that social and economic variables at the individual household level do not play an enormous role in determining performance, with the exception of the language variables. Pupils whose home language is an African language are at a considerable disadvantage in the language of instruction by the time they reach Grade 11 if the language of instruction is never spoken at home. This disadvantage can be offset somewhat if the language of instruction is spoken sometimes at home and it can be offset considerably if the language of instruction is spoken often at home.¹⁸

They also claim that competence in the language of instruction is crucial for performance in Mathematics:

Every extra per cent earned in the language test is associated with an addition of one-sixth of a per cent in the mathematics test in Grade 9 and one-third of a per cent in Grade 11.^18(p.34)

Their study, although limited and preliminary in many respects, has advanced the argument for mother tongue based education, whether single- or dual-medium is irrelevant in this context, by demonstrating a causal relationship between educational success and language medium. Other important variables such as a good meal once a day and a favourable home literacy environment are essential, of course, but for the first time in post-apartheid South Africa, the language medium has been demonstrated to be a central cause of success or failure. On the assumption that in a properly functioning educational system, a 90% pass rate would be reasonable, we have been wasting approximately R3 billion annually on paying the salaries of the teachers employed in Grades 10–12 who produce the average 50% failure rate we have experienced over the past 15 years or so. If these impressions do nothing else, they ought to demonstrate the need for in-depth educational research, in which the language issue, specifically the language-medium policy and practice, should feature centrally.

The academic authority of the Charter Report and of the Consensus Study, one can only hope, will serve as a counterweight to this lamentable shortsightedness. In my view, research on the viability and cost of radically transforming the educational system of post-apartheid South Africa by basing it in terms of language medium policy on the mother tongues (first languages, home languages) of the learners should be prioritised if we are to move away from the economic and social costs of the current English-based system. Such a research programme could take as a platform the research on mother tongue based bilingual education undertaken in recent years by organisations such as the Project for the Study of Alternative Education in South Africa, among others, as there is no need to reinvent the wheel.

Unless African languages are given market value, that is, unless their instrumentality for the processes of production, exchange and distribution is enhanced, no amount of policy change at school level can guarantee their use in high-status functions and, thus, eventual escape from the dominance and the hegemony of English. We have understood for many years already that the current language-medium practices caused cognitive impoverishment and, consequently, necessitated investment in compensatory on-the-job training by the private sector in order to enhance the ‘trainability’ of the just-from-school recruits. This wastefulness would have been completely avoidable if there had been a national development plan in which reform of education and economic development planning were integrated. Such a plan would mean that fundamental changes in the language-medium policy would be directly related to the increased use of the African mother tongues where relevant, in the public service and in the formal economic sectors. An articulated programme of job creation and employment on the basis of language proficiencies would, in the South African context, also serve as an organic affirmative action programme, one that would not have the unintended consequence of perpetuating and entrenching divisive racial identities inherited from the apartheid past.

In this connection, both the diagnostic report and the National Development Plan emanating from the Ministry in the Presidency responsible for planning are lamentably inadequate. The latter skirts this central issue with a bland statement to the effect that language policy needs to be informed by a greater appreciation of labour market imperatives; and that learners need to receive high-quality instruction in both their mother tongue and English from early in the foundation phase. For the first time since 1994, as far as I am aware, what was always implicit in previous language in education policy statements is openly stated: that the language of teaching after the foundation phase is assumed to be English. Most first-language Afrikaans-speaking pupils will undoubtedly not be touched by this uninformed policy choice, even though it lays the foundation for a new taalstryd.

At a more general level, it is my view that we have to move rapidly beyond mere posturing and gesturing in the direction of implementing a consistently democratic language policy in South Africa. We have to do so not only in order to improve and consolidate the democratic political culture that has been initiated here, but also in order to expand the potential of national economic development that will become possible because of a higher level of general education of the workforce and a deeper substratum of ordinary South Africans attuned to the needs and dynamics of modern science and technology that will have been mediated through local languages as well as English.

A second priority area for language-related research is, therefore, that of workplace multilingualism. As in the case of places of learning, we need detailed studies on the language effect at the workplace, that is, whether the use of appropriate combinations of languages in different economic sectors, branches and plants would enhance efficiency, productivity and job satisfaction. If we could establish this definitively, the language issue will cease to be seen as ‘a problem’ and our linguistic diversity will be seen to be an asset that can be planned and used like any material resource.

The third priority research area with respect to the language question is that of integrating systems of knowledge production, the ‘traditional’ and the ‘modern’, which are usually elegant variations for ‘African’ and ‘European’. Judging by some of the comments in the Charter Report, I suspect that this will become a controversial and prominent, if not a dominant, preoccupation of HSS disciplines in the coming period. In one sense, this preoccupation is as it should be in a country and in a time where the African renaissance concept is most likely to be promoted vigorously. However, we are deservedly known to be sensible about some of the most sensitive questions of social life, even if we are no good at dealing with them in practice. My advice in this regard is to explore this area with the utmost circumspection, for we should take as our point of departure the now generally accepted fact that for most of the past 70 000 years of the existence of modern humans, Africa has been the site of numerous modernities in comparison with which the 400-odd years of Europe`s dominance of the epoch of industrial and digital modernity pales into insignificance, in spite of the fact that its actuality imposes itself on our every thought. Indeed, given the subject of this commentary, it is worth reminding ourselves that language itself originated in Africa, possibly in some southern or eastern corner of the continent.¹⁹

Allow me to conclude with what many people consider to be a provocative and utopian challenge – a question first suggested by Amilcar Cabral: will South Africa’s middle class and its intellectuals find the courage and imagination to commit class suicide by moving away decisively from the current English-mainly and often English-only language policy, with all its negative consequences for a democratic polity? My answer is, that given the hegemonic status of neoliberal economic orthodoxy today, this will happen only if we can demonstrate the economic value of the African languages. Hence, my call for prioritising research in this domain.

The challenge, however, is not only to the political, business and cultural leadership of the country. It is a challenge also to applied language scholars, language practitioners and other social scientists of southern Africa. Above all, it is high time that the intelligentsia begin to move out of their comfort zones and accept that language policy, class and power are tightly interwoven and that unless we devise our own agendas in the interest of our people as a whole, we are willy-nilly carrying out the possibly nefarious agendas of others. By insisting on the centrality of language in social life and of language studies in the social sciences and humanities, we are beginning to orientate ourselves in the right direction.

This is an edited version of the inaugural lecture in the E. B. Van Wyk Honorary Lecture series at the University of Johannesburg given by the author on 22 February 2012.

FIGURE 1: The late Neville Alexander (centre) with young participants at the Vulindlela Reading Club (photo: Carole Bloch).

2. Lecercle J. A Marxist philosophy of language. Leiden and Boston: Brill; 2006.

3. Department of Higher Education and Training (DoHET). Final report: Charter for humanities and social sciences. Pretoria: DoHET; 2011. Available from: http://www.charterforhumanities.co.za/final-charter-report-released

4. Academy of Science of South Africa (ASSAf). Consensus study on the state of the humanities in South Africa: Status, prospects and strategies. Pretoria: ASSAf; 2011. Available from: http://www.assaf.org.za/wp-content/uploads/2011/08/25-July-Final.pdf

5. Tollefson J. Planning language, planning inequality. Language policy in the community. London and New York: Longman; 1991.

6. Lenin V. Lenin on language. Moscow: Raduga Publishers; 1983.

7. Alexandre P. An introduction to languages and language in Africa. London: Heinemann; 1972.

8. Myers-Scotton C. Elite closure as boundary maintenance. The case of Africa. In: Weinstein B, editor. Language policy and political development. Norwood, NJ: Ablex Publishing Corporation; 1990.

9. Bamgbose A. Language and exclusion. Hamburg: LIT-Verlag; 2000.

10. Bourdieu P. Language and symbolic power. Cambridge: Polity Press; 1991.

11. Gogolin I. Der monolinguale Habitus der multilingualen Schule [The monolingual habitus of the multilingual school]. Münster and New York: Waxmann; 1994. German.

12. Baker C. Fundamentals of bilingual education and bilingualism. Clevedon: Multilingual Matters; 1996.

13. Alexander N. The impact of the hegemony of English on access to and quality of education with special reference to South Africa. In: Harbert W, McConnell-Ginet S, Miller A, Whitman J, editors. Language and poverty. Clevedon: Multilingual Matters; 2008.

14. Djité P. Multilingualism and the economy. Unpublished mimeo; 2005.

15. Edwards V. Multilingualism in the English-speaking world. Oxford: Blackwell; 2005.

16. Djité P. The sociolinguistics of development in Africa. Clevedon: Multilingual Matters; 2008.

17. October M. Medium of instruction and its effect on matriculation examination results for 2000 in Western Cape secondary schools. Praesa Occasional Papers No. 11. Cape Town: Praesa; 2002.

18. Simkins C, Patterson A. Learner performance in South Africa. Social and economic determinants of success in language and mathematics. Cape Town: HSRC Press; 2005.

19. Alexander N. Linguistic rights, language planning and democracy in post-apartheid South Africa. In Baker S, editor. Language policy: Lessons from global models. Monterey, CA: Monterey Institute of International Studies; 2002.