Taaldebat - Language debateArgief
Tuis /
Briewe /
Kennisgewings /
Skakels /
Nuus /
Fiksie /
PoŽsie /
Taaldebat /
Language debate
Opiniestukke /
Boeke /
Film /
Teater /
Musiek /
Slypskole /
Opvoedkunde /
Artikels /
Visueel /
Expatliteratuur /
Expat literature
Gayliteratuur /
Gay literature
Nederlands /
Rubrieke /
Geestelike literatuur /
Religious literature
Hygliteratuur /
Erotic literature
Wie is ons? /
More on LitNet
LitNet is ’n onafhanklike joernaal op die Internet, en word as gesamentlike onderneming deur Ligitprops 3042 BK en Media24 bedryf.

Submission of the Taalsekretariaat[1] on the proposed national language policy for education as announced by the Minister of Education



  1. Principles for a national language policy for education
  2. Models for language policy in schools, technikons and universities
  1. The Laissez Faire Model
  2. The Ad Hoc Model
  3. The Centralised Designation Model
  4. The Numerical Formula Model
  5. The Incentive Model
  6. The Multilingualism Model
  1. Some important additional ingredients of a national language policy for education
  1. Language Learning
  2. Co-ordination
  1. Conclusion


The fact that Minister Kader Asmal is to announce a national language policy for education towards the end of October should be welcomed. Such a policy could put an end to the current conflict and uncertainty on this issue.

It is also high time that the language clauses in the Constitution was given legislative shape.

What are the options?

The rectors of the five bilingual (historically Afrikaans) universities recently requested the minister to leave decisions regarding language policy up to the educational institutions themselves. This laissez faire model has the advantage that it respects the autonomy of institutions of learning.

The disadvantage is that public institutions of learning are simply sacrificed to the destructive impact of market forces, with the inevitable consequence that the present tendency towards monolingualism continues unabated. No measures are taken to ensure that all South Africans enjoy equal access to mother-tongue education.

The approach currently followed by government, where unpredictable and one-sided language demands are imposed on particular institutions from time to time, is equally undesirable. It creates an atmosphere of uncertainty, thereby undermining proper planning and effective management. Moreover, experience has taught that such an ad hoc approach has a high conflict potential.

A proposal that enjoys wide support among activists for multilingualism is that of the Gerwel Committee, which was appointed by the minister to propose a policy on Afrikaans in higher education. The committee suggested that two universities be charged with the task of developing Afrikaans as a language of science and scholarship, while each of the African languages be assigned to at least one university for the same purpose.

This type of “designation model“ is certainly preferable to the present situation. However, it could lead to a never-ending struggle among language groups over the number of institutions that each is entitled to.

In its more extreme form, the designation model demands that certain institutions be handed over to independent community councils representing the different language groups. This would only aggravate the problem of how the education cake is to be distributed. It would also obstruct the essential process of social integration.

The basic weakness of this proposal is that it seeks to regulate everything to the smallest detail. Consequently different rules are made for different institutions - a procedure that is always vulnerable to (often legitimate) charges of discrimination.

A single rule that applies to all institutions and languages would solve this problem.

One such rule is already contained in the present education act, which determines that an educational facility has to provide tuition in a specific language if at least 30 learners insist on it: the “numerical formula model“. Minister Asmal has appealed to this law in support of his insistence that Afrikaans-medium schools introduce English-medium classes. Yet many questions arise. Can Afrikaans learners also insist on mother-tongue education in English-medium schools? How can an institution offer tuition in isiXhosa or xiTsonga if teachers do not understand these languages? Can one institution accommodate three or four languages, and change its language policy annually? Who will pay for it? That this regulation has so far remained a dead letter confirms that it is impracticable.

Another possibility hinted at by the minister is that of larger subsidies for multilingual institutions. Institutions already under pressure to introduce English in addition to Afrikaans (the opposite has yet to happen) would welcome such a step, as it would relieve their current financial burden. It is doubtful, however, whether this would increase multilingualism. The larger subsidy would not be an incentive for multilingualism, but would merely finance it. In the absence of any other incentive to introduce more languages, English-medium institutions would have no motivation to accommodate other languages, especially African languages. Thus the subsidy model, or “incentive model“, would simply strengthen existing trends.

Perhaps the time has come for more comprehensive and long-term thinking on language in education - a simple and uniform policy, which would retain English and Afrikaans as academic languages, while also improving the chances of the other languages to acquire that status. What would such a policy look like?

Suppose all educational institutions, from the primary to the tertiary level, were obliged by law to offer tuition in two official languages of their choice (the multilingualism model). That this is not impossible is proven by UNISA’s track record, and by the remarkable South African tradition of dual- and parallel-medium schools.

Such a policy would have to be announced in advance and implemented gradually. In the first year, it could be introduced only at grade 1 level, and thereafter in a further grade every year until, in the 12th year, the first matric group would be taught bilingually. Tertiary institutions would by then have had sufficient time to prepare for the introduction of a second language at that sophisticated level.

Learning materials should also be available in both languages, and it should be possible for learners to take any official language as a school subject at first-language level.

Probably most schools would adopt an African language as additional language, simply because it is the only language understood by most of their teachers and students. This would, in turn, put pressure on tertiary institutions to consider not only English and Afrikaans, but also African languages, as media of instruction. Yet the future of both English and Afrikaans would also be secured by such a policy, as long as the demand for instruction in those languages exists.

To prevent white youth from learning only English and Afrikaans, and neglecting the African languages, all learners could be expected to take three official languages as compulsory school subjects, one of them only at a practical, elementary level.

If there is one area in which the interests of all South Africans, black and white, coincide, it is in the area of language policy. A simple and uniform policy could make multilingualism in education a reality. It would secure the future of English and Afrikaans, while also making mother-tongue - ie better - education available to more South Africans.

I. Principles for a national language policy for education

Any national language policy for education in South Africa should comply with the following principles:

A. The policy should balance respect for the autonomy of institutions of learning with the obligation of the state to ensure that language policies do not infringe on the language, educational and other human rights of (prospective) learners.

B. The policy should be economically feasible and practical, and should not cause unnecessary disruption in the smooth running of institutions of learning.

C. The policy should not discriminate against any institution(s) or language(s): the same rights and duties should be accorded to all institutions of learning and all 11 official languages, ie the policy should be uniform and coherent.

D. While market forces should be taken into account, such forces should not be allowed to dictate the language policies and practices of institutions, especially in so far as these forces are destructive to multilingualism and blind to human rights.

E. The policy should result in more multilingualism in all institutions, and not only in some. Where full multilingualism is not yet achievable, present levels of multilingualism (eg bilingualism) should be retained. Under no circumstances should a policy lead to a regression to monolingualism.

F. The policy should comply with the Constitution of the Republic of South Africa in:

  1. according the same status to all the official languages,
  2. favouring no language over others,
  3. making mother-tongue education available to all South African learners,
  4. not discriminating against any person, community or institution on the grounds of language, and
  5. developing and empowering the previously marginalised languages.

II. Models for language policy in schools, technikons and universities

In the public debate about language in education, several possible models have been put forward. The purpose of this section is briefly to describe and evaluate some of the more prominent models, so as to assist the Department of Education in developing a model appropriate to the demands of the South African situation.

A. The Laissez Faire Model

According to this model, which is tacitly or openly supported by the governing bodies of most educational facilities in South Africa[2], and widely accepted by the country’s economic elite, language policy and practice should be left to the judgement of the institutions concerned.

The main arguments in favour of this model are:

  1. that it respects the autonomy of institutions of learning by proscribing any government interference in the language policies and practices of such institutions, and

  2. that it is believed to go some way towards realising the language rights of prospective students by allowing room for the use of different languages by different institutions, and through a “free market” process of supply and demand.

To this it may be objected that

  1. although the autonomy of (semi-) public institutions, like schools, technikons and universities, is protected by the Constitution as an indispensable ingredient of any truly democratic order, it is limited by, and should therefore be balanced against, other, equally indispensable rights enshrined in the Constitution, such as rights of access to public facilities, to (mother-tongue) education, to language equity, to protection against discrimination, and to affirmative action for previously disadvantaged sections of the population (and their languages);

  2. experience, both locally and internationally, has shown that a “free market” approach to language in education, like unrestrained “free market” approaches in other spheres of public life, does not lead to an equitable distribution of language and educational rights, but, on the contrary, tends to favour already privileged sections (language groups) for the simple reason that they are in a better position, socially and economically, to claim their rights and defend their interests.

In South Africa, as elsewhere, the influence of market forces (including capitalist globalisation, eg the increased international mobility of - especially white - South African professionals) has led, in recent years, to the increased dominance of one language (English) at primary, secondary and tertiary levels, with the corresponding marginalisation of Afrikaans as a language of science and scholarship, and the continued and complete neglect of the other official languages.

For example, two formerly Afrikaans universities (UWK and UPE) have already been wholly anglicised for all intents and purposes, while the remaining five are showing a marked tendency in the same direction, with an increasing number of courses offered only in English. Contrast this with the 17 exclusively English-medium universities already in existence. A similar pattern is observable in technikons and among primary and secondary schools. Despite the elevation to official status of the indigenous African languages, not a single tertiary institution has so far offered courses through the medium of any of those languages. The number of schools that offer tuition through the medium of an African language, even at the most elementary level (eg for the first four years), is negligible. This has happened, and continues to happen, despite the fact that the majority of South Africans of all races and languages would prefer mother-tongue education[3].

By definition this development has not improved access to English-medium (ie the vast majority of) educational facilities, and in historically Afrikaans institutions, it has benefited only that section of the population that is functionally proficient in English, ie all English speakers (mostly white), the majority of white Afrikaans speakers, a smaller number of black (including so-called “coloured”) Afrikaans speakers, and a minority (about 22%) of black African language speakers[4]. This small section of the population is already relatively privileged socially and economically, and is already more than adequately served, as far as language needs are concerned, by the large number of existing English-medium institutions.

The result has been that those sections of the population that are most in need of improved access to educational opportunities and the benefits of mother-tongue education[5], namely the previously disadvantaged black (including so-called “coloured”) community[6], have suffered, rather than benefited, from the “spontaneous” language shift that has been occurring. Given the language situation described above, it stands to reason that, to the extent that access has been improved for the 22% of black South Africans who are functionally proficient in English, such improvement must be due to factors other than language, eg the scrapping of apartheid laws, lowering of fees and other forms of financial assistance, bridging programmes, and the like[7]. Most importantly, due to the increasing dominance of English in the field of education, these improvements have not benefited the majority of previously disadvantaged South Africans, as has already been pointed out.

Our conclusion is that the laissez faire model is inadequate for the demands of the South African situation. The state has a duty to ensure that institutions of learning, whether at primary, secondary or tertiary level, implement language policies that respect the constitutional rights of all South Africans, whether or not such policies are “spontaneously” favoured by “market forces”. By withholding legislative pressure and financial support in this area, schools, technikons and universities are left completely at the mercy of forces that are blind to the needs and rights of the vast majority of South Africans, and that would lead inevitably to a situation of complete monolingualism, with all the attendant educational, economic and other human costs.

B. The Ad Hoc Model

In the absence of a clear and coherent language policy for education, this has become the default model adhered to in practice by the Department of Education. (The planned announcement of a national language policy for education suggests that the Department recognises the limitations of this approach.) In essence, it constitutes a variant of the laissez faire model, but with the proviso that the state may intervene in the language policies and practices of institutions of learning from time to time, so as to ensure that language is not used as a barrier to access.

Because this model recognises the duty of the state to ensure that language policies do not infringe on the constitutional rights of South Africans, it is in principle preferable to the laissez faire model in its most extreme form. However, the model has serious shortcomings:

  1. As is inevitable with any ad hoc approach to policy issues, interventions by the state have so far been one-sided. Such intervention - whether by national or provincial departments of education - has usually involved the insistence that particular institutions offer tuition in English. No institution has so far been ordered to use an African language for instruction and/or communication, and in no case has Afrikaans been protected against replacement or marginalisation by another language.

  2. Because occasional interventions are not part of a well-considered and coherent language policy, they tend to take place only as a response to emergency situations that arise in education. The result is that they do not engender any deep, structural changes in the existing language pattern. On the contrary, state intervention in the language policy of educational facilities has tended merely to strengthen the problematic sociolinguistic trends referred to in connection with the laissez faire model.

  3. Ad hoc interventions by the state are unpredictable. As such they contribute to an atmosphere of uncertainty in institutions of learning, thereby undermining efficient planning, especially as far as the internal language policies and practices of institutions and programmes for the improvement of access and diversity in such institutions are concerned.

  4. The ad hoc model has a very high conflict potential. Large institutions are, by their very nature, averse to outside interference. If the institutions concerned are educational facilities, and if the interference concerns language, wider communities of interest are also immediately involved. Every intervention by the state then evokes aggressive resistance from a variety of interest groups. Often the ad hoc interventions of the state are perceived by the affected institutions and communities to be motivated by a hidden agenda. In several cases this has led to costly and time-consuming legal disputes. In the past (eg Soweto 1976), similar ad hoc interventions have even led to violent conflict. Especially in a country where huge inequalities in education have to be addressed, and where language has been a controversial political issue for centuries, such unnecessary conflict must be avoided as far as possible.

We conclude that the ad hoc model, though in principle preferable to the laissez faire model, does not in practice make any difference to the unacceptable outcomes of the latter model. In addition, this model causes problems of its own. Therefore we welcome the decision to announce a national language policy for education, which, if well conceived, could bring clarity and calm to our education system.

C. The Centralised Designation Model

In this model the state assigns particular languages to specific institutions so as to ensure an equitable language dispensation. The designated institutions are then charged with the task of “developing” the language in question “as a language of science and scholarship”. This may involve using that language as the (primary or exclusive) medium of instruction. Thus the “educational cake” is divided up among the official languages of the country. Such division may be done according to the language demography of the population as a whole, according to the language preferences of students (or parents), or on the basis of regional (eg provincial) language demographies.

This model is implicit in the recommendations of the Gerwel Advisory Committee, appointed by the minister to investigate the role of Afrikaans in higher education[8]. It is also strongly supported by some cultural associations, like the Groep van 63, Aksie HoŽr Onderwys (AHO) and the Oorlegplatform. In its more extreme form this model can be developed in the direction of so-called “cultural autonomy”, where a cultural community demands that specific educational facilities be put under its control and reserved for use by its members. In this more extreme form the centralised designation model is strongly reminiscent of language policies under apartheid.

The main arguments in favour of this model are as follows:

  1. The model involves a serious attempt by the state to ensure, by means of legislation if necessary, that provision is made in the education system for the needs of speakers of all languages.

  2. Educational language policies along the lines of this model are being increasingly followed in multilingual states world-wide (eg Belgium and Canada).

  3. By securing the use of all official languages as languages of science and scholarship, the viability of the cultural communities whose languages they are, is ensured, since such communities depend for their continued long-term survival on the use of their language for “higher” functions, such as education.

  4. The model would not only protect Afrikaans against further marginalisation due to the increasing dominance of English, but would also enable the other official languages to take their place in the field of education.

However, the centralised designation model has distinct disadvantages:

  1. The idea that it is the state’s duty to ensure the continued long-term survival of cultural communities as communities can be questioned. Certainly, cultural communities, like religious communities for example, have the right to exist and flourish, and to be protected against unfair discrimination (unequal treatment). Yet, just as it is not the state’s duty to protect religions against degeneration, it is not up to the state to ensure that cultural communities use the level playing field of democracy to their own advantage. The right to freedom of association also implies the right of cultural, religious and other communities to disintegrate, assimilate, split or disappear. The language rights in the Constitution are not “group rights”, but individual rights possessed by every South African regardless of the community to which s/he belongs. This model is, therefore, not easily reconciled with the Constitution, and it comes as no surprise that its proponents often plead for constitutional changes as a precondition of language equity.

  2. While the centralised designation model is preferable to the laissez faire and ad hoc models, in that it might bring about structural changes to the existing language pattern by ensuring the use of several languages in education, it does so in way that is always vulnerable to charges of discrimination and inequitable division of resources. It would be very hard to come up with an appropriate formula according to which institutions of learning can be distributed among the different official languages. On the one hand, a distribution according to the language demography of the population as a whole, or of a particular province, would not be practicable at present, for the simple reason that the most widely spoken languages in South Africa, the Nguni and Sotho languages, have hardly ever been used in (especially higher) education before and therefore would have to be introduced gradually. On the other hand, a division according to the present language preferences of students would merely reflect and entrench existing patterns of privilege and exclusion. It can be expected that a policy of this kind would lead to endless wrangles between particular language communities and government, and among the language communities, over the number of institutions that they are entitled to. Thus the policy would have to be continually adapted.

  3. The element of discrimination or unequal treatment is also clearly discernible in the simple fact that, under a policy of this kind, different rights and duties are accorded to different institutions and languages. A uniform language policy, which makes the same rules for all institutions and languages, would be preferable.

  4. While the advantages of mother-tongue education are beyond dispute among experts[9], it remains true that a large number of South African parents and students of all races, especially among the economic elite, are not yet convinced of this. For some time to come, many prospective students will insist on being educated in their second or third language (usually English) because they believe such an education to be to their advantage. Institutions of learning charged with the exclusive use of indigenous African languages may, then, initially find themselves facing a shortage of students and/or loss of prestige. Moreover, while the state may and should inform people of the benefits of mother-tongue education, and should create opportunities and incentives for it, the state may not interfere in the personal choices of citizens by forcing them to study in specific languages other than their language(s) of preference. The fact that this happens in some modern states[10] constitutes no justification for such an infringement on personal liberty in our society.

  5. Because of past injustices there is a wide economic and social gap between white and black South Africans. Unfortunately this gap also corresponds roughly to language preferences and abilities. English is the mother-tongue of mostly white South Africans. A significant proportion of Afrikaans speakers (nearly half) are also white. All the other official languages are spoken nearly exclusively by black South Africans. In such a context, institutions designated as exclusively English or Afrikaans, with other institutions taking care of the African languages, will tend to become enclaves of privilege in a sea of poor and struggling African language institutions.

  6. Because of the factors outlined above, any distribution of, say, tertiary institutions among the languages in the foreseeable future would probably involve designating one or two universities as Afrikaans and all the rest English (with the proviso that some of the Afrikaans or English universities also try to “develop” the other languages). Such a policy would entrench the dominance of English and lessen the chances of the other languages ever really to take their place alongside English and Afrikaans. When the demand for instruction in the African languages arises (as it most certainly will), the national policy would first have to be changed in order to allow universities to respond to the demand. Moreover, the fact that nearly all opportunities for higher education are in English would add to the tendency among the public to opt for English as medium of instruction in primary and secondary schools, thereby making it harder for mother-tongue education to be implemented fully at that level.

In short, the centralised designation model exhibits weaknesses that are the exact opposite of those of the laissez faire model. It gives the state too strong a role in determining the language policy of educational facilities. Consequently, it has all the drawbacks associated with rigid proportionalism. It does not leave sufficient room for changes in language demography and language preference, and thus must either entrench existing patterns or be impracticable. It has a high conflict potential and could easily contribute to the linguistic, cultural and class fragmentation of South African society. It also rests on a problematic discourse of “group rights”, which is not easily reconciled with the principles underlying the South African Constitution, and which leads in practice to unequal treatment of languages and institutions.

D. The Numerical Formula Model

According to this model, the language policy of any educational facility is determined by public demand. The difference between this and the laissez faire model is that it is linked to a numerical formula: if a certain number or percentage of students insist on being served in a particular language, the institution in question is obliged to offer that language as an option. The minister of education, in interviews and public pronouncements, has appealed to existing legislation in support of this principle.

The numerical formula model has some major advantages over the other models that have been mentioned so far:

  1. Since the same formula can be prescribed for all institutions of learning, the model envisions a uniform language policy which applies to all languages and institutions equally. Thus it avoids the danger of different rights and duties being accorded to different institutions and languages.

  2. Since the model does not prescribe a specific distribution of languages, it can accommodate changes in language demography and preferences without having to be continually amended.

  3. The model takes seriously the constitutional right to mother-tongue education by empowering students and parents in claiming those rights, and by respecting their personal freedom.

However, the model also has significant weaknesses:

  1. Although it empowers South Africans to claim their language rights if they so wish, it offers no positive incentive for them to do so. Given the strong sociolinguistic forces promoting language convergence at present, a more pro-active policy may be required.

  2. Probably the greatest weakness of the model is the practical problems to which it could give rise. Many institutions serve learners from several language communities. However, it would be unreasonable to expect a single institution to accommodate three or four languages. The financial costs of such a procedure alone are already sufficient to make it impracticable. Moreover, offering tuition in a particular language depends on the extent to which teachers/lecturers are able to use the language in question. The language policy of a university must of necessity influence the appointments made and the language requirements set for appointees. If institutions have to reckon with annual changes in language preference, and perhaps even different preferences for different years and different courses, they will have great difficulty recruiting and equipping staff for the purpose.

  3. The model makes existing language preferences decisive for the language policy of institutions, which means that current language patterns - and the associated patterns of marginalisation and exclusion - are to some extent entrenched, albeit less so than in a centralised designation model based on existing language preferences. Thus a policy based on this model is not likely to increase multilingualism in the academy.

We conclude that the numerical formula model is in principle preferable to the laissez faire, ad hoc and centralised designation models. However, we recognise that this model may not provide strong enough incentives for choosing mother-tongue education, and that it may well be impracticable.

As far as the latter point is concerned, it should be noted that if this model is considered and the practical problems associated with it ironed out, a clear procedure should be created whereby students can notify institutions of their language preferences. Already when enrolling at an institution, they should have the opportunity to indicate their language(s) of preference, so that the institution in question can apply the numerical formula before the academic year commences.

Moreover, the government should finance the introduction of additional languages, including the costs associated with the recruitment and equipment of staff, the translation of learning materials etc. If the government refuses to do so, it would amount to discrimination against precisely those institutions that promote multilingualism in practice. This is part of the rationale behind the next model.

E. The Incentive Model

According to this model, which has also been raised as a possibility by the minister of education, institutions that use more than one language for instruction and communication should receive larger subsidies. These subsidies are intended to finance the use of an additional language and to encourage institutions to consider the multilingual option.

The arguments in favour of this model are as follows:

  1. The model takes seriously both the autonomy of institutions of learning and the duty of government to ensure that language policies do not discriminate or restrict access unfairly.

  2. It avoids the charge of discrimination against particular institutions and languages arising from the ad hoc and centralised designation models.

  3. Like the numerical formula model, it allows for shifts in language patterns, but without causing the administrative and financial havoc implied by that policy: language criteria can simply be included in the formula according to which institutions are currently subsidised by the state.

  4. It empowers learners to demand tuition in the language of their choice by removing the excuse of prohibitive costs often put forward by monolingual institutions in response to such demands.

However, some criticisms also need to be taken into account:

  1. The most important objection to this model is that it does not really offer an incentive for the promotion of multilingualism as it is intended to do. After all, the increased subsidies are meant to finance the costs associated with the introduction of additional languages, and would necessarily be used for that purpose, so that no additional financial benefits would flow from the implementation of a multilingual policy. Institutions that are compelled by market forces to introduce additional languages would find the increased subsidy a welcome relief, but other institutions would not be encouraged to introduce additional languages at all, for although such promotion of multilingualism would not cost them much, it would cost nothing at all to continue with a monolingual policy. In this sense, a policy based on the model in question would probably not turn the tide of language convergence, but would merely soften to some extent the devastating impact of that process on certain institutions.

  2. Interest groups opposed to this model often suspect that a policy of subsidising additional languages would be maintained only up to the point where the language shift (towards the increased dominance of English in all institutions) has become irreversible. At that point, it is argued, the additional subsidy would be withdrawn, and the less dominant languages would consequently disappear. Concern is also often expressed over the question whether the proposed subsidies would be sufficient. Concerns of this kind show the weaknesses of the incentive model. Although it is intended to promote multilingualism in education, it is perceived to depend too much on the continued commitment to multilingualism on the part of government, and to be too exposed to economic pressures.

We conclude that the incentive model, despite some serious shortcomings, is the best of all the models considered thus far. If the subsidies are sufficient to cover the true costs of introducing an additional language, and if the continued availability of such subsidies is secured through legislation, it is a model that should be seriously considered. However, since the model has significant shortcomings - especially the fact that it would probably not encourage multilingualism in presently monolingual institutions - the possibility of a more comprehensive and pro-active policy that retains the uniformity and simplicity of the incentive model should be seriously considered.

F. The Multilingualism Model

According to this model every institution of learning in the country - from the primary to the tertiary level - is required by law to use at least two official languages for instruction and communication across the board. The decision as to which two languages will be used is left to the governing body of each institution. As one of the languages, institutions (particularly at the tertiary level) should be allowed to choose a “language block” or “family of languages”, such as the Nguni languages, the Sotho languages or Afrikaans/Dutch as one of their languages.

It is required of all teachers/lecturers, as part of their employment contract, to be able to teach, answer questions, lead discussions and evaluate exams and projects in both languages[11]. Existing staff who do not meet this requirement are given a fixed time-frame within which to acquire the language(s) that they do not understand. They are also assisted with language acquisition courses.

All learning materials (except textbooks at the tertiary level), administrative documents and official correspondence are made available in both languages. Classes are conducted in both languages on either parallel-medium (separate classes) or dual-medium (both languages in the same classroom) basis, depending on student numbers and financial viability.

The system is introduced gradually on a year by year basis, starting with grade 1 in primary schools. This would give the Department of Education, in co-operation with institutions of learning, the private sector and civil society, the opportunity to ensure the availability of textbooks and other learning materials and to develop academic terminology at an increasing level of sophistication as the programme is progressively implemented. Naturally, syllabi and policy documents affecting the operation of learning facilities should be made available in all 11 official languages. By announcing the commencement of the programme at least a year in advance, and implementing it on a year by year basis, institutions are enabled to plan ahead. In the case of tertiary institutions, this would give the institutions 12 years to prepare for the introduction of an additional language.

This model has several advantages:

  1. It strikes a balance between respecting the autonomy of institutions of learning on the one hand and obliging such institutions to respect the language and other human rights of learners. It does this by making the use of at least two languages compulsory, while leaving the choice of languages up to each institution. Thus it avoids the pitfalls of both the laissez faire and centralised designation models.

  2. Because the system is introduced gradually, it allows for proper planning and budgeting (unlike the ad hoc model), and is not, like the numerical formula model, unnecessarily disruptive or impracticable. That full bilingualism at all levels and in all courses is possible in practice is proven by the long track record of the University of South Africa (UNISA) in this regard. At school level successful bilingual experiments were also quite common in the past. If language requirements are strictly enforced, and the necessary infrastructure created, bilingual institutions are most certainly economically and logistically feasible.

  3. The model leaves enough room for market forces (eg language preferences, affordability etc) to influence the language policies and practices of institutions, without taking those forces as normative or sacrificing institutions to their destructive, disempowering effects, as does the laissez faire model.

  4. A policy based on this model would not only make multilingualism in education possible or affordable, as would the numerical formula and incentive models respectively, but would ensure that multilingualism increasingly becomes a reality in all institutions, including the existing English-medium institutions.

  5. In all likelihood, this model would increase the range of languages available as media of instruction in the education sector. Especially at the level of primary and secondary education, the second language adopted (usually in addition to English) would be one of the indigenous African languages, for the simple reason that the language in question is already understood (usually better than English, which is used at present) by most or all of the teachers and learners in the institution. The gradual extension of mother-tongue education in the African languages, eventually up to the level of grade 12, would in turn exert pressure on the tertiary institutions to provide tuition in those languages. Thus, whereas tertiary institutions might initially opt in most cases for English and Afrikaans, because these languages have both long been used for academic purposes at all levels, the dynamics set off by the national language policy would make the option of at least one African language as one of the two languages increasingly attractive to tertiary institutions as well.

  6. This model does not discriminate against particular institutions or languages. It applies the same rule - and a very simple rule at that - to every educational facility and every language in the country. As such, it makes room for the important role of English as one of the international languages, but without putting Afrikaans as a language of science and scholarship in danger, and - most importantly - while increasing the chances of the indigenous African languages to acquire full public recognition within the foreseeable future.

  7. The model does not divide students into separate institutions based on language. Because every institution, from primary schools to technikons and universities, would eventually be fully bilingual, every South African learner would be exposed on a day to day basis to a much greater diversity of fellow-students than is currently the norm. This would also facilitate the learning of languages, and would enhance mutual understanding and respect for one another’s languages and cultures. As such, this model promotes integration and nation-building, while respecting and affirming diversity.

  8. Since every language, by its very nature, functions as a barrier to access for those who do not understand it, the use of an additional language would automatically make every institution of learning in South Africa accessible to a greater number of South Africans, and therefore more representative of the population as a whole. The increased use of African languages in education, eventually even up to the tertiary level, would dramatically increase access precisely for those young South Africans who are currently excluded from the benefits of a good education due to the lack of recognition of their languages.

  9. The implementation of a policy of this kind would affect all sectors of South African society. It would strengthen the public role of the African languages, not only in the academic field, but also in the workplace generally, the courts, the media, politics and economic life. Perhaps most importantly, it would contribute to a sense of dignity among the speakers of every language in South Africa, affirming to each of them that s/he is recognised and valued as a full citizen of this country, and therefore accommodated and made to feel welcome.

We conclude that the multilingualism model is the most suitable model for putting multilingualism into practice in the education sector, given the various considerations arising from the South African context, the constitutional framework within which the policy is to be integrated, and the needs and best interests of all South Africans.

III. Some important additional ingredients of a national language policy for education

A. Language Learning

Apart from promoting diversity as far as the languages used for instruction and (internal and external) communication are concerned, the policy should ensure that South Africans increasingly learn one another’s languages. Therefore we propose that every learner in South Africa should have at least three official languages as compulsory school subjects. Once again, the choice of languages offered at a particular institution should be left to the judgement of the governing body of that institution.

Such a policy would require the following:

  1. It should be possible to take any official language of South Africa, and not only English or Afrikaans, at first or second language level, ie the necessary syllabi and examinations should be made available for this purpose. The two languages chosen by a particular institution as media of instruction and communication should both be offered at first and second language level, so that learners can decide which language to take at which level.

  2. The third language taken as a compulsory subject should be offered at a third language or practical level, ie it should involve the acquisition of basic listening, reading, speaking and writing skills, but not the more advanced elements of literature, creative writing and the like. In order to make the acquisition of the language at this elementary level even more manageable, it should be included in the syllabus from grade 1 to grade 12. Nevertheless, the subject should be examined, and passing it should be a prerequisite for going on to the next level of schooling, obtaining matric or gaining entrance to university.

  3. The policy proposed some time ago, according to which schools would have to choose two languages to be offered as subjects - one “European” language (English or Afrikaans) and one African language (one of the other nine languages), should not be implemented. Such a policy would elevate English to a superior status and position above all the other languages, it would be the death-knell of Afrikaans as a language of science and scholarship, it would place the other official languages in a permanent position of inferiority, and it would entrench a culture of apartheid. Rather, the choice of languages should be left to the institutions concerned, and to ensure that formerly white schools do not simply choose English and Afrikaans, thereby maintaining the existing pattern where whites do not learn African languages, a third language should be made compulsory in all schools.

B. Co-ordination

The Department of Education should consult with other departments and public bodies in order to explore ways in which language policies in different spheres of public life can be co-ordinated so as to strengthen one another. While multilingualism is promoted in education and training, science and scholarship, other public institutions should ensure that multilingualism is also increasingly expanded as a permanent feature of public life. The public broadcaster has a special responsibility in this regard. The Pan South African Language Board should be consulted in connection with this and any other language policy to be adopted in future.

IV. Conclusion

We believe that the time has come for the Department of Education to take a significant stride forward in realising the vision of our Constitution as regards language in the sphere of education at all levels. This stride should be taken with an inspiring long-term vision in mind: the ideal of a truly multilingual society in which every citizen is respected, accommodated and treated fairly, regardless of the language that s/he speaks.

Language can no longer be allowed to bar access to full participation in public life to the majority of (especially poor) South Africans. Rather than expecting of the poor majority to learn the language of the wealthy elite, the latter should be obliged to use the languages of the former in order to improve access to opportunities for them.

“Multilingualism” should not mean forcing non-English institutions to use English, but rather requiring all public institutions to use as many languages as possible, thereby giving equal public recognition to all the official languages, as the Constitution wisely requires.

If the Department of Education were to follow this path, the seeds of conflict already sown by many participants in the public debate on language would wither and die, rather than sprout and smother our young democracy. The cultural riches of our people would contribute to our public life and our national identity. Our children would learn better. Our scholars and scientists would make more and better contributions to the common store of human knowledge. Our economy and technological infrastructure would develop more efficiently and be more in tune with the spirit of our people. Participatory democracy would be deepened and the opportunities of public life opened to everyone. We would be healed of our colonial and apartheid mindset, and come to experience in our day to day intercourse with one another that “South Africa belongs to all who live in it.”

* * *

Signed on behalf of the
Taalsekretariaat by:

Andre van der Walt
Koos du Toit
Gerrit Brand

Contact information:

15 Herold Street
Stellenbosch 7600
E-mail: taal@linguasek.co.za
Tel: 021-8872713 / 8872736
Fax: 021-8872710

[1] The Taalsekretariaat is an independent, non-politically aligned organisation with its own administrative structure, staff and funding, whose mission is to promote the indigenous languages of South Africa as adequate means of communication for both mother-tongue and non-mother-tongue speakers within and beyond the borders of the Republic of South Africa. (See last page for contact details.)

[2] See, eg, the proposals of the rectors of the bilingual (ie historically Afrikaans) universities recently made to the Minister of Education.

[3] This was the finding of a recent study by MarkData.

[4] These statistics concerning “functional proficiency” in English are taken from a study by Markinor.

[5] The Department of Education, in the persons of both the minister and deputy minister, has often declared in accordance with the incontrovertible findings of all studies in this area, that mother-tongue education is superior to second or third-language instruction, even as far as the acquisition of new languages (eg of English) is concerned. Since we agree wholeheartedly with this view, we shall not argue this point, but refer instead to the wealth of information made available by the Pan South African Language Board in this connection, some of which has been published on their website.

[6] The (mostly Afrikaans-speaking) “coloured” population presently shows the lowest level of entry to higher education, followed by “blacks”. Among the latter group it is especially among the poorest and most marginalised communities that proficiency in Afrikaans, rather than in English, is common. Of course, nearly all those belonging to this group have as their mother-tongue one of the indigenous African languages. The matric pass rate (in a context of nearly exclusive English-medium education) is also lowest among this group.

[7] The fact, often cited in support of the laissez faire approach, that students (and parents) from among this relatively privileged minority tend to prefer, and even insist on, English as the medium of instruction at the tertiary level, when contrasted with the language attitudes of South Africans generally (see footnote 4), merely confirms the extent to which English is functioning as a barrier to access to the majority of (black) South Africans, who struggle to pass matric, let alone enter higher education, and therefore never have the opportunity to have their demands heard. It is further confirmed by the fact that the state has found the medical schools of Spanish-medium universities in Cuba more accessible to previously disadvantaged South Africans than (both English and Afrikaans-medium) South African universities.

[8] The committee recommended that two institutions, one in the north and one in the south of the country, be charged with the development of Afrikaans as a language of science and scholarship, and that each indigenous African language be assigned to at least one university for the same purpose.

[9] See footnote 5.

[10] Eg Canada, where French-speaking parents in the French territory are prohibited by law from sending their children to English-medium schools.

[11] In this connection, the proposal currently being considered by the Education Department to reintroduce the requirement of a bilingualism certificate for teachers, but without limiting “bilingualism” to English and Afrikaans, deserves the support of all interested parties.


© Kopiereg in die ontwerp en inhoud van hierdie webruimte behoort aan LitNet, uitgesluit die kopiereg in bydraes wat berus by die outeurs wat sodanige bydraes verskaf. LitNet streef na die plasing van oorspronklike materiaal en na die oop en onbeperkte uitruil van idees en menings. Die menings van bydraers tot hierdie werftuiste is dus hul eie en weerspieŽl nie noodwendig die mening van die redaksie en bestuur van LitNet nie. LitNet kan ongelukkig ook nie waarborg dat hierdie diens ononderbroke of foutloos sal wees nie en gebruikers wat steun op inligting wat hier verskaf word, doen dit op hul eie risiko. Media24, M-Web, Ligitprops 3042 BK en die bestuur en redaksie van LitNet aanvaar derhalwe geen aanspreeklikheid vir enige regstreekse of onregstreekse verlies of skade wat uit sodanige bydraes of die verskaffing van hierdie diens spruit nie. LitNet is ín onafhanklike joernaal op die Internet, en word as gesamentlike onderneming deur Ligitprops 3042 BK en Media24 bedryf.