Gelyke Kanse vs Stellenbosch University: High Court judgment

IN THE HIGH COURT OF SOUTH AFRICA

(WESTERN CAPE DIVISION, CAPE TOWN)

CASE NO: 17501 /2016

In the matter between:

GELYKE KANSE -1st Applicant

DANIëL JOHANNES ROSSOUW - 2nd Applicant

THE PRESIDENT OF THE CONVOCATION OF THE STELLENBOSCH UNIVERSITY - 3rd Applicant

BERNARDUS LAMBERTUS PIETERS - 4th Applicant

MORTIMER BESTER - 5th Applicant

JAKOBUS PETRUS ROUX - 6th Applicant

FRANCOIS HENNING - 7th Applicant

ASHWIN MALOY - 8th Applicant

RODERICK EMILE LEONARD - 9th Applicant

And

THE CHAIRMAN OF THE SENATE OF THE STELLENBOSCH UNIVERSITY - First Respondent

THE CHAIRMAN OF THE COUNCIL OF THE STELLENBOSCH UNIVERSITY - Second Respondent

THE STELLENBOSCH UNIVERSITY - Third Respondent

PROF ESTIAN CALITZ (CHAIRMAN) a trustee for the time being of the Het Jan Marais Fund - Fourth Respondent

PROF WALTER CLAASSEN N.O., a trustee for the time being of the Het Jan Marais Fund - Fifth Respondent

GEORGE STEYN N.O., a trustee for the time being of the Het Jan Marais Fund - Sixth Respondent

DR JOHAN VAN DER MERWE N.O., a trustee for the time being of the Het Jan Marais Fund - Seventh Respondent

GYS STEYN N.O., a trustee for the time being of the Het Jan Marais Fund - Eighth Respondent

MARTHIE HEYL N.O., a trustee for the time being of the Het Jan Marais Fund - Ninth Respondent

Coram: Dlodlo J and Savage J

Date of Hearing: 14, 15 and 16 August 2017

Date of Judgment: 25 October 2017

JUDGMENT

DLODLO, J

INTRODUCTION

[1] The applicants seek orders reviewing and setting aside the decisions of the Senate and Council of Stellenbosch University (‘SU’), taken on 9 and 12 June 2016 respectively (the decisions), to adopt a new language policy for the SU (‘the Policy or the 2016 Policy’) in terms of section 27 (2) of the Higher Education Act 101 of 1997 (‘the Act’); as well as an order setting aside the Policy itself; and an order directing SU to implement its previous language policy approved by the Council on 22 November 2014 (‘the 2014 Policy’) until it is validly amended or replaced. Clearly final relief is sought by the applicants in this application. A mention must be made that since application was made in this case at the end of September 2016, the Supreme Court of Appeal handed down judgment in the matter of University of Free State v Afriforum 2017 (4) SA 283; 2017 (2) ALL SA 808 (SCA) on 28 March 2017. The latter judgment is binding on this Court in respect of issues which also arise in the current case by virtue of the doctrine of stare decisis.

Mr Heunis sought to distinguish the SCA decision from the present matter. This shall be dealt with later in this judgment.

[2] On behalf of the applicants it is contended that these proceedings were instituted ‘in an attempt to convince the court, for the reasons comprehensively set out in the affidavits filed, of the vital importance of the continuation of Afrikaans as a primary language of instruction at the SU and, to that end, to convince the Court that this can only be achieved by the Court reviewing and setting aside the SU’s newly adopted language policy which dispenses with Afrikaans as a primary language of instruction’. In oral argument the applicants conceded that the impugned decision did not involve administrative action as contemplated in the Promotion of Administrative Justice Act 3 of 2000 (‘PAJA’) but that the adoption of the new policy could be subject to a legality review on the ground that it was made in the exercise of a public power. However, in written argument the applicants contended that since the policy itself is attacked and has the legal consequences of adversely affecting the rights of people as well as a direct, external legal effect, that part of their case predicated upon the operation of PAJA remains alive. It is apparent therefore that the applicants approached this matter on the basis that they seek to have the policy set aside on constitutional as well as administrative law grounds. It remains common cause that the Higher Education Language Policy (‘the LPHE’) is implicated. Mr Heunis set out an overview of the statutory regime, including the provisions of the Act and the LPHE, as well as the Constitution and the Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000 (‘the Equality Act’). We are also referred to European law.

[3] The respondents were out of time with regard to the filing of the answering papers. Consequently there is an unopposed application for condonation in that regard which is accompanied by a tender for costs. The respondents also brought two striking out applications in terms of Rule 6 (15) of the Uniform Rules of Court. Furthermore, the applicants brought an application for the admission of a further affidavit. In this application the applicants also applied for the leading of oral evidence i.e. that the Court should subpoena an identified person to come and testify before Court. This application is strenuously resisted by the respondents. Prior to presentation of oral submissions, the parties informed the Court that they have reached an agreement to the effect that all these issues would be argued together with the main matter and this duly occurred.

THE APPLICANTS’ CASE AND SOME PASSING REMARKS THERETO

[4] Mr Heunis correctly contended that the language policy determined by the Council of the University has to be informed by the LPHE and all such policies (including the LPHE itself) have to comply with Sections 29 (1) (b) and 29 (2) of the Constitution. He referred us to some provisions of the LPHE which he described as directly pertinent and which echo provisions of the Constitution, namely: (a) The role of all South Africa's languages "working together" to build a common sense of nationhood is consistent with the constitutionally enshrined values of "democracy, social justice and fundamental rights". (See Section 6 (2) of the Constitution and para 3 of the LPHE: (b) Everyone has the right to use the language and to participate in the cultural life of his or her choice, provided that these rights may not be exercised inconsistently with any provision of the Bill of Rights. (See S30 of the Constitution and para 3.1 of the LPHE): (c) Everyone has the right to receive education in the official language or languages of his or her choice in public education institutions where such education is reasonably practicable. In order to ensure the effective access to, and implementation of this right, the state has to consider all reasonable educational alternatives, including single medium institutions, taking into account equity, practicability and the need to redress the results of past racially discriminatory laws and practices. (S 29 (2) of the Constitution and para 3.1.2 of the LPHE); (d) The role of language and access to language skills are critical to ensure the right of individuals to realise their full potential to participate in and contribute to the social, cultural, intellectual, economic and political life of the South African society. (para 4 of the LPHE); (e) The challenge facing higher education is to ensure the simultaneous development of a multilingual environment in which all South Africa's languages are developed as academic/scientific languages, while simultaneously ensuring that the existing languages of instruction do not serve as a barrier to access and success. This is what the policy framework, set out in the LPHE, seeks to address. (para 6 of the LPHE).

[5] Mr Heunis relied heavily on the fact that the Minister of Higher Education had invited Prof GJ Gerwel to convene an informal committee to provide him (the Minister) with advice specifically with regard to Afrikaans as a language of instruction. It is true that the committee referred to in this regard was tasked to advise on ways in which Afrikaans could be assured of continued long-term maintenance, growth and development as a language of science and scholarship in the higher education system without non-Afrikaans speakers being unfairly denied access within the system or the use and development of the language as a medium of instruction wittingly or unwittingly becoming the basis for racial, ethnic or cultural division and discrimination. Of course the reason for focusing on Afrikaans was that, with the exception of English, Afrikaans is the only other South African language which is employed as a medium of instruction and official communication in institutions of higher education. There is hardly a dispute that the framework for language in higher education also reflects the values and obligations of the Constitution, especially the need to promote multilingualism, and it commits (as it were) to an attempt to ensure that all the official languages are accorded parity of esteem. See para 12 of the LPHE.

[6] It remains a fact that in relation to languages of instruction the Ministry: (a) acknowledges the prevailing position of English and Afrikaans as the dominant languages of instruction in higher education and believes that it will be necessary to work within the confines of the status quo until such time as other South African languages have been developed to a level where they may be used in all higher education functions (para 15.1 of the LPHE); (b) acknowledges that Afrikaans as a language of scholarship and science is a national resource and, therefore, fully supports the retention of Afrikaans as a medium of academic expression and communication in higher education and is committed to ensuring that the capacity of Afrikaans to function as such a medium is not eroded (para 15.4 of the LPHE); (c) does not believe, however, that the sustainability of Afrikaans in higher education necessarily requires the designation of the University of Stellenbosch and the Potchefstroom University of Christian Higher Education (now the North West University ("NWU")) as "custodians" of the academic use of that language as proposed by the Committee (para 15.4.1 of the LPHE); (d) also agreed with the Rectors of the Historically Afrikaans Universities that the sustained development of Afrikaans should not be the responsibility of only some of the universities (para 15.4.2 of the LPHE); (e) is of the view that the sustainability of Afrikaans as a medium of academic expression and communication can be ensured through a range of strategies which include the adoption of parallel and dual language medium options which would, on the one hand, cater for the needs of Afrikaans language speakers and, on the other, ensure that the language of instruction is not a barrier to access and success, to which end the Ministry committed itself, in consultation with the historically Afrikaans medium institutions, to examine the feasibility of different strategies, including the use of Afrikaans as a primary but not a sole medium of instruction (para 15.4.4 of the LPHE).

It must be mentioned that the LPHE seeks to balance, on the one hand, the needs to transform higher education, and in particular to prevent institutions' languages of instruction from impeding access and success by people who are not fully proficient in English and Afrikaans on the other hand, the development of multilingualism in those institutions' day-to-day functioning and core activities, including the development of indigenous African and other languages as scientific and academic languages. It also seeks to assure the long-term maintenance and growth of Afrikaans as a language of science and scholarship in the higher education system.

[7] The LPHE (based on Prof Gerwel's Committee) acknowledges that Afrikaans, ‘as a language of scholarship and science is a national resource’ and it commits to ‘ensuring that the capacity of Afrikaans to function as such a medium is not eroded’. It is of significance that in the above regard the Supreme Court of Appeal in the University of the Free State v Afriforum and Another 2017 (4) SA 283; 2017 2 ALL SA 808 (SCA) made the following observation:

‘In November 2002 The Education Ministry outlined a framework for a Higher Education Language Policy (LPHE), which encouraged the promotion of multilingualism. It advocated "that attention and strengthening of Afrikaans as a language instruction" (sic), in historically Afrikaans Universities but it also acknowledged that this will practically create a tension with other constitutional imperatives, particularly considerations of equity, the need to redress past racially discriminatory laws and practices and practicability, identified in s 29(2) of the Constitution.’

In Mr Heunis’s contention in formulating its language policy, the SU had to have regard to, and comply with (or else justify departure) the following features of the LPHE: (a) The acknowledgement that Afrikaans as a language of scholarship and science is a national resource. (b) The Ministry's support for the retention of Afrikaans as a medium of academic expression and communication in higher education and its commitment to ensure that the capacity of Afrikaans to function as such, is not eroded. (c) The Ministry's position that the sustained development of Afrikaans is not the responsibility of only some of the historically Afrikaans universities. (d) The Ministry's view that the sustainability of Afrikaans as a medium of academic expression and communication can be secured through a range of strategies, including the adoption of parallel and dual language medium options, which would, on the one hand, cater for the needs of Afrikaans language speakers and, on the other, ensure that language of instruction is not a barrier to access and success.

[8] In Mr Heunis’s contention the SU failed to have regard to and comply with or else justify departure from the aforementioned features of the LPHE. Mr Heunis submitted as follows:

‘Electing to abandon a policy in terms of which Afrikaans was a primary language of instruction with equal status to English, particularly in circumstances where instruction in Afrikaans has been abandoned or significantly curtailed at other universities, is clearly inconsistent with the LPHE as evidenced in particular by the summary according to which the framework is designed to promote multilingualism and to enhance equity and access in higher education inter alia through the retention and strengthening of Afrikaans as a language of scholarship and science.’

In his submission as far as the adoption and implementation of their policies are concerned, the historical Afrikaans Universities, at which a significant number of Afrikaans-speaking students still enrol, have a primary obligation, which derives from the LPHE, in respect of the retention and strengthening of Afrikaans as a language of scholarship and science. Concluding reference and reliance on the LPHE, Mr Heunis submitted as follows:

‘What has been happening at the UJ, UFS, UP, and UNISA and now also at the SU, is self-evidently inconsistent with the LPHE. That inconsistency is the most dramatic in the instance of the SU which has taken a deliberate decision to end the status of Afrikaans as a primary language of instruction with full knowledge of what has happened at the other historical Afrikaans universities. It may be that this is the reason why the Second Respondent consistently denies, in the face of overwhelming evidence to the contrary, that the NLP would lead to a significant down scaling of the role of Afrikaans as a language of instruction at the SU.’

[9] Talking to the constitutional aspect of this case, Mr Heunis prefix his submissions by referring to the provisions of Section 239 of the Constitution. The latter Section provides as follows:

‘In the Constitution, unless the context indicates otherwise – … 'organ of state' means –

(a) any department of state or administration in the national, provincial or local sphere of government; or

(b) any other functionary or institution –

(i) exercising a power or performing a function in terms of the Constitution or a provincial constitution; or

(ii) exercising a public power or performing a public function in terms of any legislation, but does not include a court or a judicial officer.’

Mr Heunis pointed out that the concept ‘organ of state’ in Section 239 covers all instances in which a public power is exercised or a public function is performed in terms of legislation regardless of whether the person or institution exercising a power or performing the function is formally recognised as an organ of state or not. It is important to note that as far as possible at this stage an endeavour is made to document Mr Heunis’s contentions. I make passing remarks but would fully and comprehensively deal with these contentions later in this judgment. We have been referred to Western Cape Minister of Education v Governing Body of Mikro Primary School 2006 (1) SA 1; 2005 (10) BCLR 973 (SCA), a judgment which overturned a finding of this Court that the governing body of the School was not an organ of state and intended by the legislature to be independent of state or government control in the performance of its functions. The SCA concluded as follows:

"In terms of the definition in the Constitution, any institution exercising a public power or performing a public function in terms of any legislation is an organ of State. The second respondent, a public school, together with its governing body, the first respondent, is clearly an institution performing a public function in terms of the Act. It follows that it is an organ of State as contemplated in the Constitution."

Relying on the above authority Mr Heunis contended that by parity of reasoning, the SU is also an organ of state and, therefore, bound by the Bill of Rights by virtue of the provisions of Section 8 (1) of the Constitution. In this regard this Court was referred to the following authorities:

See Baloro and Others v University of Bophuthatswana 1995 (4) SA 97 (B) (university is an organ of state); Toerien en 'n Ander v De Villiers en 'n Ander 1995 (2) SA 879 (C), 885F (university is a public authority); National Union of Tertiary Employees of SA v Central University of Technology: Free State (2009) 30 ILJ 1620 (LC); Gardner and Others v Central University of Technology: Free State 2012 ZALAC 23 (25 July 2012) (respondent treated as higher education institution in terms of section 1 of the HEA, and as an organ of state in terms of section 239(b) of the Constitution).

It would of course appear that the SCA took this as axiomatic in the University of the Free State case supra.

[10] Mr Heunis invited the attention of this court to the known principles governing the interpretation of the Bill of Rights, namely: (a) A purposive interpretation is called for. This is not synonymous with the broadest or most generous meaning which can be given to a provision. The purpose of a right must be determined with reference to the language, history, larger character of the Bill of Rights, and, where applicable, to the meaning and purpose of related rights. See in this regard S v Zuma and Others 1995 (2) SA 642; 1995 (4) BCLR 401 (CC) para 15; S v Makwanyane 1995 (3) SA 391; 1995 (6) BCLR 665 (CC) para 9; (b) in Makwanyane supra the court talking to the interpretation principles stated thus: ‘In giving meaning to s 9, we must seek the purpose for which it was included in the Constitution. This purposive or teleological approach to the interpretation of rights may at times require a generous meaning to be given to provisions of chp 3 of the Constitution and at other times a narrower or specific meaning.’ (c) In Ferreira v Levine NO and Others; Vryenhoek and Others v Powell NO and Others 1996 (1) SA 984; 1996 (1) BCLR 1 (CC) para 172, Chaskalson P (as he then was) stated that the Court had adopted a purposive interpretation of the Constitution. One, in my view, may not conclude this aspect on interpretation without reference to Davis et al – Fundamental Rights in the Constitution, Commentary and Cases (1997) page 14 where the learned authors describe the difference between a generous and purposive interpretation as follows:

‘The generous interpretation of a charter right would require a court to interpret the language in the widest possible manner. By contrast the purposive interpretation is predicated upon the purpose of the right, with the result being that the widest possible interpretation will not inevitably be the one which will be supported.’

I unreservedly accept that in each case the language of the Constitution has to be the starting point and that (in casu), the purpose of Section 29 is informed by the content of, particularly, Section 6.

[11] Mr Heunis contended that the issue of language at the SU engages two constitutional principles which he mentioned as follows: (a) the constitution’s commitment to diversity, including linguistic diversity. In this regard he referred this Court to MEC for Education, KwaZulu-Natal and Others v Pillay 2008 (1) SA 474; 2008 (2) BCLR 99 (CC) para 65, where the Constitutional Court held that ‘our constitutional project … not only affirms diversity, but promotes and celebrates it’. The Court continued (per Langa CJ) by saying that ‘our Constitution does not tolerate diversity as a necessary evil, but affirms it as one of the primary treasures of our nation’. See para 92 of the judgment. According to Devenish- A Commentary on the South African Bill of Rights (1999) page 419, the protection of language rights secures more than merely the technical facility to communicate, it guarantees cultural viability and continuity. The writer continued and said that Section 92 (2) (c) is strengthened by Sections 30 and 31 of the Constitution. Mr Heunis relying on Sections 15, 30 and 31 of the Constitution, contended that, that commitment is manifest, for example, in the protection of religious and cultural rights. He submitted that more pertinently, it underlies (in part), the right to an education in the language of one's choice in Section 29 (2) and recognises Afrikaans as part of that diversity. In his view, it supports the need to sustain Afrikaans as a vibrant language and the need to increase Black (African) students at historically White public educational institutions such as the SU.

[12] According to Mr Heunis it may fairly be said that the underlying purpose of this right is to facilitate instruction, to promote educational progress and to foster linguistic diversity. He added that there are also important objectives of educational rights, including equitable access to quality education. Section 29 (2) requires authorities to take all these considerations into account in formulating balanced and fair language policies. In the second place is the Constitution's commitment to eradicating the legacy of South Africa's racial past including unequal access to public education, and it commits to eradicating it through positive measures to promote those who were previously disadvantaged. Section 9 (2) of the Constitution permits affirmative action measures that are ‘designed to protect or advance persons, or categories of persons disadvantaged by unfair discrimination’. See Minister of Finance and Another v Van Heerden 2004 (6) SA 121; 2004 (11) BCLR 1125 (CC) on the proper interpretation of Section 9 (2).

[13] Malherbe – The Constitutional Framework for Pursuing Equal Opportunities in Education, Perspectives in Education, Volume 22 (3), September 2004, talking to the commitment to diversity, including linguistic diversity states the following:

‘Section 29(2) guarantees the right of everyone to education in the official language or languages of their choice in public educational institutions where it is reasonably practicable, in other words, whenever it is reasonable to expect the State to provide such education. In the heated atmosphere generated by the ongoing disputes over single-medium institutions, the recognition of this basic right in the South African multilingual situation is being overlooked sometimes. Note that the right goes further than Article 2 of Protocol I of the European Convention on Human Rights and Fundamental Freedoms as it was interpreted and applied in the Belgian Linguistic cases, and that it indeed imposes a duty on the State to provide such education wherever reasonably practicable. The right is not confined to existing facilities either. The right applies to all education and is not restricted to basic education. Even institutions of higher education are therefore required to provide instruction in the languages preferred by students. However, the right extends only to the official languages and not to all languages used in South Africa. Although it would accommodate most South Africans, it therefore does not strictly speaking provide for a right to mother-tongue education. Of course the right includes mother-tongue education, which is significant not only for the protection of language rights, but also, as mentioned, because it has been proven over and over that the mother-tongue is the preferred medium of education, especially in the early phases, and is therefore a legitimate mechanism for creating equal educational opportunity.’

The aforementioned two constitutional principles raise a question for language at the SU, namely: How should it accommodate – (a) the rights of the Afrikaans-speaking students to their language and culture; (b) the promotion of multilingualism; and (c) the rights of primarily Black (African) people who are not conversant in Afrikaans to access a tertiary education at the SU? It does not appear that there is an ideal solution. In the absence of a language policy that assigns precisely equal weight to all eleven of the official languages in every subject of the curriculum, it will always be arguable that one or other language is subordinated relative to others. Importantly, the reality of limited resources entails that any recognition of a linguistic or cultural right may be to the detriment of a competing and arguably more deserving right.

[14] In Mr Heunis’s contention though, what has happened at the SU involves a move towards English domination, a development (he described) that would reinforce an already overwhelmingly dominant language – which is the mother-tongue of only a particularly privileged minority of White South Africans. He argued that this is clearly at odds with the State's duty under section 6 (2) of the Constitution ‘to take practical and positive measures to elevate the status and advance the use’ also of Afrikaans. In addition, so argued Mr Heunis, like all official languages, Afrikaans, under section 6 (4), enjoys parity of esteem and must be treated equally. His contention was that for one of the last universities that is not effectively English single medium to abandon Afrikaans as a primary language of instruction is clearly inconsistent with these provisions of the Constitution. I mentioned earlier that I shall fully deal with these contentions. But I cannot resist the temptation to point out that in my understanding of the 2016 Policy which is under attack nothing can be construed as amounting to the abandonment of Afrikaans as a primary language of instruction at SU.

[15] The Court was referred to Professor Haysom in Cheadle et al – South African Constitutional Law (Issue 1, 2005) 25-3 where the learned Professor notes as follows in the context of language rights:

‘The increasing tendency to recognise and protect language rights in more recent national constitutions reflects a concern with resisting the 'globalisation' of dominant languages, and, much more specifically, with the recognition of the value of diversity as opposed to uniformity – an appreciation that equality does not imply sameness.’

The above sentiments were echoed by Justice Kriegler with respect to Afrikaans in particular in Ex Parte Gauteng Provincial Legislature: In re Dispute Concerning the constitutionality of Certain Provisions of the Gauteng School Education Bill of 1995 1996 (3) SA 165; 1996 (4) BCLR 537 (CC) para 39 (SALR editors’ translation):

‘Language – and particularly the preservation of Afrikaans – evokes deep-rooted emotions. For that reason it is vitally necessary that sober and considered attention be given to the implications of this matter … (I)t (section 32(c) of the interim Constitution, the equivalent of section 29(3) of the Constitution) … is and remains a bulwark against the swamping of any minority's common culture, language or religion. For as long as a minority actually guards its common heritage, for so long will it be its inalienable right to establish educational institutions for the preservation of its culture, language or religion.’

In the same case Justice Sachs held that although ‘(a)t present, the imperatives of equalising access to education are strong’, those imperatives should not override constitutionally protected rights in relation to language and culture. See para 52 of the latter case. It is true that the Constitution recognises all the interests at stake and requires organs of state to find a reasonable balance between them. Indeed the Constitutional Court in the Pillay case supra employed the phrase ‘reasonable accommodation’ in a similar context describing the principle as follows:

‘At its core is the notion that sometimes the community, whether it is the State, an employer or a school, must take positive measures and possibly incur additional hardship or expense in order to allow all people to participate and enjoy all their rights equally. It ensures that we do not relegate people to the margins of society because they do not or cannot conform to certain social norms.’

Mr Heunis placed reliance on the aforementioned formulation and contended that in the context of SU policy it means that the University may need to take additional steps to accommodate students who are unable to succeed with what may appear to be a neutral language policy, but what it cannot do is to take away Afrikaans’ status as a primary language of institution and, by so doing depriving Afrikaans –speaking students of an existing right.

[16] Indeed the key to the present challenge is Section 29 of the Constitution. It is apposite that the provisions of Section 29 of the Constitution be set out in order to facilitate this discussion. It provides as follows:

‘(1) Everyone has the right–

(a) to a basic education, including adult basic education; and

(b) to further education, which the State, through reasonable measures, must make progressively available and accessible.

(2) Everyone has the right to receive education in the official language or languages of their choice in public educational institutions where that education is reasonably practicable. In order to ensure the effective access to, and implementation of, this right, the State must consider all reasonable educational alternatives, including single medium institutions, taking into account–

(a) equity;

(b) practicability; and

(c) the need to redress the results of past racially discriminatory laws and practices.’

Our Courts of Appeal i.e., the Constitutional and the Supreme Court of Appeal have had occasion to consider and analyse the import of section 29 of the Constitution. For instance in Gauteng School Education Bill supra, though in the context of the predecessor to section 29 of the Constitution (Section 32 of the Interim Constitution), the Constitutional Court observed: (a) That the Constitution provides that every person shall have the right to establish educational institutions. Linguistically and grammatically, it provides a defensive right to a person who seeks to establish such educational institutions and it protects that right from invasion by the State, without conferring on the State an obligation to establish such educational institutions. (b) The object of subparagraph (c) is to make it clear that while every person has a right to basic education through instruction in the language of his or her choice, those persons who want more than that and wish to have educational institutions based on a special culture, language or religion which is common, have the freedom to set up such institutions based on that commonality, unless it is not practicable. Thus interpreted, section 32 (c) is neither superfluous nor tautologous; it preserves an important freedom. (c) The interpretation of section 32 (c) as a defensive right, based on its grammatical and linguistic structure, seems to be supported by its context within section 32 itself. Section 32 (a) creates a positive right that basic education be provided for every person and not merely a negative right that such a person should not be obstructed in pursuing his or her basic education. (d) Section 32 (b), recognising the diversity of languages in our country, again creates a positive right for every person to instruction in the language of his or her choice, where this is reasonably practicable, not merely a negative right to prevent any obstruction if such person seeks instruction in the language of his or her choice. Section 32 (c), by contrast, guarantees a freedom - a freedom to establish educational institutions based on a common culture, language or religion. It is that freedom which is protected by section 32 (c). A person can invoke the protection of the Court where that freedom is threatened, but the language of section 32 (c) does not support a claim that such educational institutions, based on a commonality of culture, language or religion, must be established by the State, or a claim that any person is entitled to demand such establishment, notwithstanding the fact that his or her right to basic education and to instruction in the language of his or her choice is, where practicable, otherwise being satisfied by the State. In the same judgment, Justice Kriegler observed as follows:

‘[41] Secondly, it should be clearly understood what the debate is really about in this case. Subsections (a) and (b) of s 32 of the Constitution record and confirm the right of everyone to a basic education, equal access to educational institutions and, where reasonably practicable, instruction in the language of the pupil's choice. The government is constitutionally obligated to that. The standard of reasonable practicability is elastic - as it necessarily has to be in order to leave room for a wide range of circumstances. It is, however, objectively justiciable, which means that arbitrary governmental action can be restrained by the Courts. Accordingly, meaningful numbers of language-speakers have an enforceable right against the government to instruction in the language of their community as long as it is reasonable practicable.

[42] Section 32(c) enlarges on this. As my Colleague Mahomed DP indicates - and I emphasise - the Constitution keeps the door open for those for whom the State's educational institutions are considered inadequate as far as common culture, language or religion is (sic) concerned. They are at liberty harmoniously to preserve the heritage of their fathers for their children. But there is a price, namely that such a population group will have to dig into its own pocket therefor. In a sense, the present dispute is not about a people's heritage but about money.’

[17] In the Minister of Education, Western Cape v Governing Body, Mikro Primary School supra the SCA held: (a) The right of everyone to receive education in the official language or languages of their choice in public educational institutions where that education is reasonably practicable is a right against the State. (b) The Constitution recognises that there may be various reasonable educational alternatives available to the State to give effect to this right and has left it to the State to decide how best to do so. In order to ensure the effective access to, and implementation of, this right, the State must in terms of the provision consider all reasonable educational alternatives, including single medium institutions.(c) Section 29 (2), therefore, empowers the State to ensure the effective implementation of the right by providing single-medium educational institutions. This is a clear indication that, in terms of section 29 (2), everyone has a right to be educated in an official language of his or her choice at a public educational institution to be provided by the State if reasonably practicable, but not the right to be so instructed at each and every public educational institution subject only to it being reasonably practicable to do so.

[18] In Head of Department, Mpumalanga Department of Education and Another v Hoërskool Ermelo and Another 2010 (2) SA 415 (CC), the Constitutional Court: (a) Observed that unequal access to opportunity, including private and public education were among the many scars left by apartheid. (b) Observed that the Constitution ardently demands that this social unevenness be addressed by a radical transformation of society as a whole and of public education in particular. This, according to the then Deputy Chief Justice, the Constitution does in a cluster of warranties which include: section 1(a) which entrenches respect for human dignity, achievement of equality and freedom; section 6(1) read with section 6(2) which warrants and widens the span of our official languages from a partisan pair to include nine indigenous languages which for long have jostled for space and equal worth; sections 9(1) and (2) which entitle everyone to formal and substantive equality; section 9(3) which precludes and inhibits unfair discrimination on the grounds of, amongst others, race and language or social origin; section 31(1) which promises a collective right to enjoy and use one's language and culture; and section 29(1) which entrenches the right to basic education and a right to further education which, through reasonable measures, the State must make progressively accessible and available to everyone. (c) Recognised and embraced the tribute Sachs J paid to minority language rights in general and to Afrikaans in particular in the Gauteng School Education Bill supra:

‘The fourth assumption is that the Afrikaans language is one of the cultural treasures of South African national life, widely spoken and deeply implanted, the vehicle of outstanding literature, the bearer of a rich scientific and legal vocabulary and possibly the most creole or "rainbow" of all South African tongues. Its protection and development is therefore the concern not only of its speakers but of the whole South African nation. In approaching the question of the future of the Afrikaans language, then, the issue should not be regarded as simply one of satisfying the self-centred wishes, legitimate or otherwise, of a particular group, but as a question of promoting the rich development of an integral part of the variegated South African national character contemplated by the Constitution. Stripped of its association with race and political dominance, cultural diversity becomes an enriching force which merits constitutional protection, thereby enabling the specific contribution of each to become part of the patrimony of the whole.

At the same time, these assumptions have to be located in the context of three important considerations highlighted by the Constitution.’

(d) Analysed section 29 of the Constitution on the following basis:

(i) That the provision is made up of ‘two distinct but mutually reinforcing parts’. The first part places an obvious premium on receiving education in a public school in a language of choice. (ii) That right, however, according to the Constitutional Court, is internally modified because the choice is available only when it is ‘reasonably practicable’. (iii) When it is reasonably practicable to receive tuition in a language of one's choice will depend on all the relevant circumstances of each particular case. These would include the availability of and accessibility to public schools, their enrolment levels, the medium of instruction of the school that its governing body has adopted, the language choices that learners and their parents make, and the curriculum options offered.

[19] In short, the reasonableness standard built into section 29 (2) (a) imposes a context-sensitive understanding of each claim for education in a language of choice. (iv) An important consideration will always be whether the State has taken reasonable and positive measures to make the right to basic education increasingly available and accessible to everyone in a language of choice.(v) It must follow that when a learner already enjoys the benefit of being taught in an official language of choice the State bears the negative duty not to take away or diminish the right without appropriate justification.(vi) The second part of section 29(2) of the Constitution points to the manner in which the State must ensure effective access to and implementation of the right to be taught in the language of one's choice. It is an injunction on the State to consider all reasonable educational alternatives which are not limited to, but include, single-medium institutions. (vii) In resorting to an option, such as a single or parallel or dual medium of instruction, the State must take into account what is fair, feasible and satisfies the need to remedy the results of past racially discriminatory laws and practices.

[20] In the University of the Free State case supra, the SCA held inter alia: (a) The legal standard of reasonableness, of necessity, involves a consideration of constitutional norms, including equity, redress, desegregation and non-racialism. (b) The factual criterion is practicability, which is concerned with resource constraints and the feasibility of adopting a particular language policy. (c) Even if a language policy is practical because there are no resource constraints to its implementation, it may not be reasonable to implement because it offends constitutional norms. The policy would therefore not meet the reasonably practicable standard. (d) Once the standard is met and the right to a language of choice exists, the State bears a negative duty not to take it away or diminish the right without justification. But this does not mean that once the right exists it continues, regardless of whether the context and the circumstances have changed. A change in circumstances may materially bear on the question whether it is reasonably practicable to continue with a policy. What is required of a decision-maker, when there is a change in circumstances, is to demonstrate that it has good reason to change the policy. In other words, it must act rationally and not arbitrarily. (e) That the dispute raises potentially difficult constitutional questions, including whether the new policy’s pursuit of racial integration and equality has the effect of unfairly discriminating against linguistic and cultural minorities; impermissibly promoting majoritarian hegemony at the expense of linguistic and cultural diversity, or undermining the fundamental language scheme of our constitutional order, which requires the State to take practical and positive measures to elevate the status and advance the use of all official languages, instead of diminishing their importance. (f) That such questions may only be confronted through a substantive constitutional challenge to the State’s language policy, and not somewhat diffidently or obliquely through judicial review, as the respondents have done in that case. (g) The Court concluded that the respondents sought an order reviewing and setting aside the decision of UFS to adopt a single-medium English language policy. That decision, according to the Court was not reviewable under PAJA; and the respondents had failed to make out a proper case for review under the principle of legality. In light thereof, the SCA concluded that the UFS was entitled to adopt a new policy because it was no longer reasonably practicable to continue with the 2003 policy, which had the effect of segregating the student community along racial lines. The UFS was under no legal obligation to apply the LPHE and was free to depart from it for good reason. It had done so.

[21] Mr Heunis in an endeavour to distinguish the Free State case from the present case argued that in the Free State case the Afriforum case did not deal with a substantive constitutional challenge to the language policy; instead (so he argued) the matter was approached on the basis of judicial review. He contended that in contradistinction, the present challenge is also underpinned by a frontal substantive constitutional challenge. He submitted that the constitutional values underpinning the right to dignity and the right to have one’s dignity respected and protected, also comes into play when language rights are derogated from. Relying on Governing Body of the Juma Musjid Primary School and Others v Essay N.O. and Others (Centre for Child Law and Another as amici curiae) 2011 (8) BCLR 761 para 37, Mr Heunis contended that the right to basic education in Section 29 (1) (a) which is ‘immediately realisable’, the right to further education is progressively realisable and subject to reasonable measures. Importantly, unlike other socio-economic rights such as the right to housing, healthcare, food, water and security, the right to further education) is not expressly made subject to the availability of resources. See in this regard Sections 26 and 27 of the Constitution and Woolman & Bishop at cit page 37: Constitutional Law of South Africa. Mr Heunis correctly conceded on this aspect stating that although the absence of this internal limitations clause have some meaning, he submitted that it does not render resource constraints irrelevant since that would undermine the limitations of reasonableness and progressive realisation.

[22] Even though the internal limitations of reasonableness and progressive realisation have not been explored in the context of Section 29 (1) (b), Courts have dealt with this and have provided guidance which is to be found from the meaning assigned to them when the other socio-economic rights were interpreted. Courts have spoken thus: (a) ‘Reasonable measures’ generally requires the state to have a program ‘capable of facilitating the realisation of a right’. This obligation rests on all the levels of government involved but, in the context of higher education, it rests primarily on the national government and higher education institutions since tertiary education is a functional area of national legislative and executive competence. (See Government of the Republic of South Africa and Others v Grootboom 2001 (1) SA 46; 2000 (11) BCLR 1169 (CC) para 41; Schedule 4 of the Constitution). (b) Progressive realisation calls for the progressive facilitation of accessibility calling for the examination and, where possible, lowering over time legal, administrative, operational and financial hurdles. See Grootboom case para 45. (c) Like other socio-economic rights, section 29(1)(b) includes both positive and negative aspects requiring the state to take positive measures to improve the availability of further education and to gradually improve the quality of that education while, on the other hand, providing protection against unreasonable exclusion from existing access to higher education. (d) The negative dimension arises from the general non-retrogression principle that applies to all socio-economic rights in South African law, and internationally. See International Covenant on Economic, Social and Cultural Rights, Geneva Comment 3: The Nature of State Parties, Obligations (1990) para 9. (e) For example, a measure which allows a person to be deprived of existing access to housing will violate the negative dimension of the right to housing. Similarly, a blanket denial of access to higher education to asylum seekers limits the right. See Minister of Home Affairs v Watchenuka and Another 2004 (4) SA 326; 2004 (2) BCLR 120 (SCA). The negative part of the right is obviously not subject to the limitations of progressive realisation and reasonableness. Mr Heunis’ submission in above regard was as follows:

‘It stands to reason that no single university can be solely responsible for fulfilling the positive element of the right to further education and that universities must do so together with one another and the national government. This implies that when they are called upon to determine their language policies, they should do so with regard to the national picture in terms of offer and demand. What is clear, however, is that they cannot, when they determine language policy, summarily dispense with existing rights.’

[23] Noticeably, the phrase ‘reasonable measures’ is used in several of the socio- economic rights in the Bill of Rights. It does require the State to have a policy or program that must be capable of facilitating the realisation of the right in question. We accept that the State does have a wide latitude in choosing the policy or program. A court considering the reasonableness of measures adopted by an organ of state may not enquire whether other more desirable or favourable measures could have been adopted, or whether public money could have been better spent. ‘Progressive available and accessible’ as a phrase means and must mean that the hurdles to the realisation of the right to further education - legal, administrative, operational and financial – must be identified by the State and, where possible, lowered over time. Given the significant number of mainly Black (African) students and prospective students who have enrolled or will in future enrol at SU and who are not conversant or sufficiently conversant in Afrikaans, but who are conversant or sufficiently conversant in English, what this means is that SU must provide and implement measures aimed at an adequate English offering. Mr Heunis is, however, fully aware of the aforegoing statement of fact. He contended that SU is able to achieve this without dispensing with Afrikaans as a primary language of instruction precisely because of its achievement in this regard consistently with the language policy which stands to be replaced by the New Language Policy (the 2016 Policy).

[24] There is no denying that the right in section 29 (2) serves two purposes, namely: (a) to improve access to education by ensuring it is available in a language that is understood. Talking to this purpose, Mr Heunis opined that it is relevant to those students and prospective students who are not conversant in Afrikaans, but who are conversant in English. (b) to promote linguistic communities, including Afrikaans-speakers, and to protect linguistic populations against assimilation and the erosion of their culture associated with the language. Mr Heunis contended that this is particularly important given the increasing prevalence of English in South Africa, including in public higher education where most institutions lecture exclusively or mainly in English. Mr Heunis relied heavily on Gauteng Provincial Legislature, supra where the Constitutional Court recognised that Afrikaans (like all languages) is not simply a means of communication and instruction. The Constitutional Court found that Afrikaans is Central to the cohesion and identification of the Afrikaans-speaking community. In the words of Sachs J ‘the Afrikaans language, like all languages, is not simply a means of communication and instruction, but a central element of community cohesion and identification for a distinct community in South Africa. We are accordingly dealing not merely with practical issues of pedagogy, but with tangible factors, that as was said in Brown v Board of Education of Topeka [347 US 483 (1954)], form an important part of the educational endeavour. In addition, what goes on in schools can have direct implications for the cultural personality and development of groups spreading far beyond the boundary fences of the schools themselves.’

[25] In the above regard this Court was also referred to Solski (Tutor of) v The Quebec (Attorney-General) 2005 1 SCR 201 para 3, a Canadian Supreme Court matter which quoted Doucet-Boudereau v Nova Scotia (Minister of Education) 2003 3 SCR 3 para 26 as follows:

‘(e)ducation rights play a fundamental role in promoting and preserving minority language communities. Indeed, '[m]inority language education rights are the means by which the goals of linguistic cultural preservation are achieved'.

Mr Heunis contended that this concern is not only compelling at the level of primary and high schools where the failure to provide education would quickly result in the death of a language, but is also important at the level of higher education since universities train the professionals and academics of tomorrow and if that training does not occur in a particular language, the language will suffer over time, both as a social language and as a language of the professions, business, science and so on.

[26] In Mikro, case supra, Streicher JA held that "(t)he right of everyone to receive education in the official language or languages of their choice in public educational institutions where that education is reasonable practicable is a right against the State". See para 31 of judgment. In Ermelo, Moseneke DCJ, speaking for the Court, stated as follows:

‘The right to receive education in the official language of one's choice in a public educational institution where it is reasonably practicable is located in s 29(2) of the Constitution. In order to give effect to this right, the same provision imposes a duty on the State to consider all reasonable educational alternatives, including single-medium institutions, taking into account what is equitable, practicable and addresses the results of past racially discriminatory laws and practices.

Turning to examine section 29(2), he said the following:

‘The provision is made up of two distinct but mutually reinforcing parts. The first part places an obvious premium on receiving education in a public school in a language of choice. That right, however, is internally modified because the choice is available only when it is 'reasonably practicable'. When it is reasonably practicable to receive tuition in a language of one's choice will depend on all the relevant circumstances of each particular case. These would include the availability of and accessibility to public schools, their enrolment levels, the medium of instruction of the school that its governing body has adopted, the language choices that learners and their parents make, and the curriculum options offered. In short, the reasonableness standard build into s 29(2)(a) imposes a context-sensitive understanding of each claim for education in a language of choice.’

Malherbe – the Constitutional Dimension of the Best Interests of the Child as applied in Education (2008 TSAR 267), points out, "learner numbers, costs, availability of facilities and educators, the distance to the nearest similar institution that is able to provide education in the chosen language, and the chosen medium of instruction in the case of universities, can be relevant factors that may determine whether, in a particular case, it is reasonably practicable to provide such education.’

The applicants place reliance also on Woolman & Bishop, supra op cit page 59 where the learned authors say that language of choice instruction is reasonably practicable ‘where sufficient numbers of learners request instruction in their preferred language…and no adequate alternative school exist to provide such instruction’. The authors opine that at that point, the educational institution ‘is under an obligation – with assistance from the State – to provide instruction in the language of Choice’.

[27] Speaking to the Section 29 (2) requirement, Malherbe supra made the following observation:

‘Although this provision does not provide for a right to single-medium institutions, it imposes a particular duty on the State and on any applicable organ of State. In choosing the appropriate institution in general or in a particular case, the State must consider all reasonable alternatives in a bona fide way, taking into account what is educationally appropriate, as well as the listed factors of equity, practicability and the need for redress. The factors carry equal weight and must be balanced. What may be equitable to everybody concerned may not be practicable or educationally reasonable or appropriate, and what may be practicable may not serve to redress of historical inequalities. This duty applies in the case of existing institutions as well.’

Of course the Supreme Court of Appeal pointed out in Mikro case supra at paragraph 30 of the judgment, that Section 29 (2) does not mean that:

"Everyone has the right to receive education in the official language of his or her choice at each and every public education institution where this was reasonably practicable. If this were the correct interpretation of s 29(2), it would mean that a group of Afrikaans learners would be entitled to claim to be taught in Afrikaans at an English medium school immediately adjacent to an Afrikaans medium school which has vacant capacity provided they can prove that it would be reasonably practicable to provide education in Afrikaans at that school. So interpreted, since the right in question extends to 'everyone', this would entail that boys have a constitutional right to be educated at a school for girls if reasonably practicable."

The fact of the matter is simply that once it is shown that education in the language of choice is reasonably practicable, it becomes necessary to consider the second part of Section 29 (2), i.e. the means to fulfil the right. In truth, at that point, as the Constitutional Court said in Ermelo case, the second sentence of Section 29 (2) places ‘an injunction on the State to consider all reasonable educational alternatives’ to achieve the right. In the determination of what alternatives to employ, ‘the State must take into account what is fair, feasible and satisfies the need to remedy the results of past racially discriminatory laws and practices’.

Woolman & Bishop op cit page 61 pointed out that the combination of these factors means that ‘the State cannot simply invoke an overriding commitment to “equality” or “transformation” in order to dismantle single medium institutions’.

See also in this regard Laërskool Middleburg en ‘n Ander v Departmentshoof Mpumalanga Departement van Onderwys, en Andere 2003 (4) SA 160 (T). Mr Heunis contended as follows:

‘One of the crucial flaws in the decisions which led to the adoption of the new language policy is precisely that the Council and the Senate of the SU did not consider what was "reasonably practicable" at the University and has clearly overlooked that, as an organ of state, it is co-responsible for taking the desired measures, and not to abolish measures that were in place and were consistent with the LPHE and the Constitution, particularly in the face of the Ermelo decision's affirmation of the principle of non-retrogression. This doctrine stands squarely in the way of a decision that has the effect of curtailing vested rights that claim the protection of the Constitution.’

[28] In his submission although the SU is a historically Afrikaans University, it has established English as a language of learning and teaching to a considerable extent and the issue is not whether it should offer learning in English at all (that was the point of contention in the case concerning the Afrikaans-medium in Mikro Primary School). The issue is what the nature and extent of the SU's English and Afrikaans offering should be. Section 29 (2) truly obliges the State to consider all reasonable educational alternatives in order to achieve the right. Needless to mention that in Ermelo case supra, the Constitutional Court emphasised that when determining what alternative to employ (as already mentioned), the State is obligated to take into account what is fair, feasible and what satisfies the need to remedy the results of past racially discriminatory laws and practices. It is true that in the Ermelo matter the Constitutional court also held that when a person already enjoys the benefit of being taught in an official language of choice, the State bears the negative duty not to take away or diminish the right without appropriate justification. See para 53 of the Ermelo judgment.

[29] Mr Heunis drew the attention of the Court to Section 23 of the Canadian Charter of Rights and Freedoms which grants parents who are minority language speakers in a province – whether French or English – the right ‘to have their children receive primary secondary school instruction in that language in that province’. The latter Section (S 23 of the Canadian Charter of Rights and Freedoms) is, according to Mr Heunis closely analogous to our Section 29 (2) under discussion. In Association des parents De L’ecole Rose – Des – Vents v British Colombia (Education) 2015 2 SCR 139 para 27, the Canadian Supreme Court held:

‘The gradual loss of the mother tongue is inevitable without some institution to give formal instruction in the language and to enhance its prestige by according it some social recognition.’

In Canada the right is granted to parents, not to children and is only incidentally concerned with the quality of the education. In Mahe v Alberta 1990 1 SCR 342, the Canadian Supreme Court interpreted the limitations imposed in Section 23 (3) to create a sliding scale of the right to minority-language education. At the upper end of the scale is what is promised in Section 23 (3) (b) namely ‘minority language educational facilities provided out of public funds". Indeed at the lower end is the simple ‘instruction’ mentioned in Section 23 (3) (a). Therefore ‘S 23 guarantees whatever type and level of rights and services is appropriate in order to provide minority language instruction for the particular number of students involved’. See page 366 of the Mahe judgment. When the number of learners warrants the provision of education at the upper end of the scale, then the Courts in Canada apply a test of ‘equivalency’. The minority-language facilities must be equivalent to the majority-language facilities. The Canadian Supreme Court has held that ‘the educational experience of the children (must) be of meaningfully similar quality to the educational experience of majority language students’. See Association des parents case supra para 33. The question is whether parents would be ‘deterred from sending their children to a minority language school because it is meaningfully inferior to an available majority language school’. See Association des parents case para 34.

[30] A three-pronged argument presented by Mr Heunis on the application of the Section 29 (2) criteria maintaining that it has to take into account of: (a) the importance of the retention and protection of Afrikaans' status as an academic and science language and as a national asset which can only be secured if it is used as a language of instruction at the tertiary level of education; (b) that it could never have been the intention of section 29(2) that the right of Afrikaans-speaking students to choose Afrikaans as a language of instruction at public institutions of tertiary education would be systematically phased out bearing in mind also that the other indigenous languages have to be developed in the interests of a multilingual society and that the exclusion of Afrikaans through its replacement by English cannot be conducive to the multilingual ideal; (c) that the erosion of Afrikaans at the tertiary level of education will inevitably put pressure on schools to also treat English as the dominant language and will no doubt have a domino effect which will result in the right of mother-tongue education, which derives from section 29(2), being eroded also at schools with obvious detrimental implications for Afrikaans as also the quality of education and the promotion of multilingualism.

[31] In Mr Heunis’ contention regarding what is fair and what will remedy the results of past racially discriminatory laws and practices, the language policy (the 2014 Policy) which is being replaced by the New Language Policy (the 2016 Policy) clearly passes muster. In this regard the contention advanced on behalf of the applicants is as follows:

‘The decimation of the Afrikaans lecture offering is not justifiable, particularly since there is only one other option for Afrikaans learning in South Africa and none in the Western Cape Province. The people of this Province have a legitimate expectation that the SU will cater primarily (although not exclusively) to students from the Province. There is also a legitimate concern that abrogating the Afrikaans lecture offering will adversely and irreversible affect the role of Afrikaans at the university and, as a result, in higher education in South Africa generally.’

I point out without expanding on this issue that the fact of the matter is that students at all universities (including SU) come from various parts of the Provinces of South Africa. They also come from the other countries, not only of the African continent but also European countries. They choose to study at a particular university for various reasons. Some universities have acquired international reputation, even students born in the Western Cape enrol (by choice) to other universities in the Country. Nothing binds them to only enrol at SU.

[32] It is absolutely true that when weighing the competing interests of Afrikaans- speakers and Blacks (Africans) with no or inadequate Afrikaans, it is also important to bear in mind that the Constitution aims at achieving an equality of opportunities not inequality or equivalence of burdens. The Constitutional Court put it rather eloquently in Minister of Home Affairs and Another v Fourie and Another; Lesbian and Gay Equality Project and Others v Minister of Home Affairs and Others 2006 (1) SA 524; 2006 (3) BCLR 355 (CC) by stating that when section 9 (2) of the Constitution speaks of the State taking measures to promote the achievement of equality, it ‘calls for equality of the vineyard not the graveyard’.

Mr Heunis, talking to the feasibility, told this Court that he knows that the SU has the financial, infrastructure and personnel resources for equality teaching because that was required in terms of the language policy which is now being replaced (the 2014 Policy). He referred this Court to Gauteng Provincial Legislature case supra particularly where Kriegler J stated the following:

‘The standard of reasonable practicability is elastic – as it necessarily has to be in order to leave room for a wide range of circumstances. It is, however, objectively justiciable, which means that arbitrary governmental action can be restrained by the Courts. Accordingly, meaningful numbers of language-speakers have an enforceable right against the government to instruction in the language of their community as long as it is reasonably practicable.’

Reliance is also placed on the concurring judgment of Sachs J in the same matter, particularly the following portions:

‘[46] The first assumption is that the 'never again' principle, which I feel should be one of our guides to interpretation, applies not only to bitter experiences of former State enforced segregation, but also to those of past compulsory assimilation. This was a major theme at the National Convention held to draft the document which became the Constitution of the Union of South Africa in 1910.

[47] The second assumption is that the Afrikaans language, like all languages, is not simply a means of communication and instruction, but a central element of community cohesion and identification for a distinct community in South Africa. We are accordingly dealing not merely with practical issues of pedagogy, but with intangible factors that, as was said in Brown v Board of Education of Topeka, form an important part of the educational endeavour. In addition, what goes on in schools can have direct implications for the cultural personality and development of groups spreading far beyond the boundary fences of the schools themselves.

[48] The third assumption is that there exists amongst a considerable number of people in this country a genuinely-held, subjective fear that democratic transformation will lead to the down-grading, suppression and ultimate destruction of the Afrikaans language and the marginalisation and ultimate disintegration of the Afrikaans-speaking community as a vital group in South African society.

[49] The fourth assumption is that the Afrikaans language is one of the cultural treasures of South African national life, widely spoken and deeply implanted, the vehicle of outstanding literature, the bearer of a rich scientific and legal vocabulary and possibly the most creole or 'rainbow' of all South African tongues. Its protection and development is therefore the concern not only of its speakers but of the whole South African nation. In approaching the question of the future of the Afrikaans language, then, the issue should not be regarded as simply one of satisfying the self-centred wishes, legitimate or otherwise, of a particular group, but as a question of promoting the rich development of an integral part of the variegated South African national character contemplated by the Constitution. Stripped of its association with race and political dominance, cultural diversity becomes an enriching force which merits constitutional protection, thereby enabling the specific contribution of each to become part of the patrimony of the whole.’

But Justice Sachs added the following important observation:

‘[49] Of course, vital parts of the 'patrimony of the whole' are indigenous languages which, but for the provisions of s 6 of the Constitution, languished in obscurity and underdevelopment with the result that at high-school level, none of these languages have acquired their legitimate roles as effective media of instruction and vehicles for expressing cultural identity.

[50] And that perhaps is the collateral irony of this case. Learners whose mother tongue is not English, but rather one of our indigenous languages, together with their parents, have made a choice to be taught in a language other than their mother tongue. This occurs even though it is now settled that, especially in the early years of formal teaching, mother-tongue instruction is the foremost and the most effective medium of imparting education.’

[33] Woolman & Bishop op cit page 59 supra, point out that the right to receive education in the official language of one's choice in public educational institutions is not an unqualified right but is subject to a standard of reasonable practicability presupposing sufficient numbers of learners requiring instruction in a preferred language and that a failure to demonstrate that request for instruction is reasonably practicable ends the enquiry. The latter was also a finding by the Supreme Court of Appeal in Mikro case supra. The second sentence in Section 29 (2) requires that all reasonable educational alternatives that would make mother-tongue or preferred language instruction possible, ought to be considered. For instance, for a single medium institution to be preferred to another reasonable practicable institutional arrangement, such as dual medium instruction or parallel medium instruction, it has to be demonstrated that it is more likely to advance or satisfy the three listed criteria of equity, practicability and historical redress. See in this regard Woolman & Bishop supra op cit p. 57-60.

[34] There are three factors which require consideration in the interpretation and implementation of Section 29 (2) of the Constitution. It is not necessary to further explore these factors. It suffices to mention, though that in Mr Heunis’ contention, the third factor (which talks to redress) weighs strongly in favour of ensuring that the language is not a barrier to access for Black (African), Coloured and Indian students. In his contention though, this consideration (for reasons advanced by him infra) does not favour the new policy over the old. He proffered these reasons: (a) The old policy favoured multilingualism and sustaining the use of Afrikaans. (b) While Afrikaans is a barrier to Black (African) students, English is a barrier to many coloured students who were also victims of the past discrimination and a move that decreases the Afrikaans offering would negatively affect them, particularly when regard is had to the diminishing other options for Afrikaans-language higher education. (c) It will not benefit Black (African) students since the previous policy was not a barrier to access for them because in the prevailing parallel medium environment there is a 100% English offering. An important contention put forth by Mr Heunis is that the facts regarding language demography in the feeder areas of universities, as also statistics regarding the language offer, on one hand, and the demand of Afrikaans-speaking matriculants, on the other, have to be important considerations when decisions are made as to whether or not Afrikaans as a language of choice is reasonably practicable. The concern expressed by the applicants is that not one university remains as a single medium Afrikaans University. This, according to Mr Heunis, is a fact testifying to compliance with (particularly) the third criterion in Section 29 (2). He brought it to the attention of the Court that as a consequence of the developments the NWU and the SU were (until the latter decided to adopt the impugned 2016 Policy), the only universities at which Afrikaans-speaking students had the benefit of a 100% Afrikaans offering. In the latter regard, the submission made on behalf of the applicants is:

‘In our submission the fact that English has been introduced at all historical Afrikaans universities as a language of instruction, especially to comply with the redress criterion in section 29(2), does not mean that Afrikaans must inevitably be replaced by English as the dominant language of instruction since that would clearly fall foul of the fairness criterion without any commensurate benefit viewed from the perspective of the demand which derives from the redress criterion.’

[35] The applicants also postulate that the Constitution's recognition of community rights, associational rights, religious rights, cultural rights and linguistic rights, creates a set of background conditions against which the claim of continued parallel medium instruction at the SU has to be considered. The view taken on behalf of the applicants is that ‘an overriding commitment to “equality” or “transformation” cannot simply be invoked to dispense with Afrikaans as a medium of instruction. See Woolman & Bishop supra op cit p. 60. Lastly on this aspect, the applicants place reliance on the following observation made by Professor Malherbe supra:

‘A balance must also be struck between the constitutional values of dignity, equality and freedom. Aspects of current education policies fail to appreciate this, especially when it comes to reflecting language and religious diversity in education. Policies that deny this diversity, and impose uniformity (including language uniformity) in the name of equality, will fail in the long run, because a unified nation cannot be built by rejecting the bricks one has to use. As such policies marginalise people, and deny their self-respect and self-worth; they affect their human dignity. A clearer understanding is needed of what nation building is about, and in pursuing everyone's equal worth, it must be appreciated that equality will remain an illusive dream if people's uniqueness is ignored, and if we fail to pursue equality within the context of their diversity. In the final analysis it is a quest for human dignity rather than equality. That is what Brown v the Board of Education is about. That is what democracy in South Africa should be about.’

JURISPRUDENCE THE APPLICANTS RELIED ON – DOMESTIC AND FOREIGN

(re: FREEDOM AND DIVERSITY)

[36] The Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000 (‘the Equality Act’) recognises in clause 2 of the Schedule thereto that the failure to reasonably and practicably accommodate diversity in education is an example of an illustrative unfair practice in the educational sector. Thus in MEC for Education, KwaZulu-Natal v Pillay 2008 (1) SA 474 (CC), the Constitutional Court recognised the significance of freedom and diversity to the constitutional agenda. What the Constitutional Court observed in this regard is the following: (a) The centrality of freedom as one of the underlying values in the Bill of Rights and the injunction on the Courts to interpret all rights to promote the underlying values of human dignity, equality and freedom. (b) A necessary element of freedom and of dignity of any individual is an entitlement to respect for the unique set of ends that the individual pursues. (c) That our constitutional project not only affirms diversity, but promotes and celebrates it. The acknowledgment and acceptance of difference is particularly important in our country given its history. The Constitution thus acknowledges the variability of human beings (genetic and socio-cultural), affirms the right to be different, and celebrates the diversity of the nation.

The Court was also referred to Prince v President of the Republic of South Africa and Others 2002 (2) SA 794 (CC) which dealt with freedom of religion and the protection of the associational nature of cultural, religious and language rights. I accept that international law does have an important role to play in the interpretation of the Bill of Rights in our Constitution. Thus Section 39 (1) of the Constitution provides:

‘(1) When interpreting the Bill of Rights, a court, tribunal or forum-

(a) must promote the values that underlie an open and democratic society based on human dignity, equality and freedom;

(b) must consider international law; and

[37] The relevance of international law to the South African constitutional framework was explained by the Constitutional Court as follows in Glenister v President of the Republic of South Africa and Others 2011 (3) SA 347 (CC) at para 97:

“[95] To summarise, in our constitutional system, the making of international agreements falls within the province of the executive, whereas the ratification and the incorporation of the international agreement into our domestic law fall within the province of Parliament. The approval of an international agreement by the resolution of Parliament does not amount to its incorporation into our domestic law. Under our Constitution, therefore, the actions of the executive in negotiating and signing an international agreement do not result in a binding agreement. Legislative action is required before an international agreement can bind the Republic.

[96] This is not to suggest that the ratification of an international agreement by a resolution of Parliament is to be dismissed 'as a merely platitudinous or ineffectual act'. The ratification of an international agreement by Parliament is a positive statement by Parliament to the signatories of that agreement that Parliament, subject to the provisions of the Constitution, will act in accordance with the ratified agreement. International agreements, both those that are binding and those that are not, have an important place in our law. While they do not create rights and obligations in the domestic legal space, international agreements, particularly those dealing with human rights, may be used as interpretive tools to evaluate and understand our Bill of Rights.

[97] Our Constitution reveals a clear determination to ensure that the Constitution and South African law are interpreted to comply with international law, in particular international human-rights law. Firstly, s 233 requires legislation to be interpreted in compliance with international law; secondly, s 39(1)(b) requires courts, when interpreting the Bill of Rights, to consider international law; finally, s 37(4)(b)(i) requires legislation that derogates from the Bill of Rights to be 'consistent with the Republic's obligations under international law applicable to states of emergency'. These provisions of our Constitution demonstrate that international law has a special place in our law which is carefully defined by the Constitution.

[98] But treating international conventions as interpretive aids does not entail giving them the status of domestic law in the Republic. To treat them as creating domestic rights and obligations is tantamount to 'incorporat[ing] the provisions of the unincorporated convention into our municipal law by the back door'.”

The fact is that our Country (South African) ratified the International Covenant on Economic, Social and Cultural Rights (‘the International Covenant’). Several General Comments have of course been issued under the International Covenant which have provided guidance to the Constitutional Court in its interpretation of certain rights in the Bill of Rights. An example would be Motswagae v Rustenberg Local Municipality 2013 (2) SA 613 (CC) at footnote 6; Residents of Joe Slovo Community, WC v Thubelisha Homes (Centre on Housing Rights & Evictions, Amici Curiae) 2010 (3) SA 454 (CC) para 237 and Government of the RSA v Grootboom 2001 (1) SA 46 (CC) paras 30 and 31.

[38] In support of the relief sought in these proceedings Mr Heunis referred to what he called the threshold of justification. He elucidated that the position at SU (a) was initially one of single medium Afrikaans instruction; (b) thereafter of dual and parallel medium English and Afrikaans instruction; and (c) currently of predominantly English medium instruction to the virtual exclusion of Afrikaans. He maintained that the consequence is that until the adoption of the current language policy, Afrikaans-speaking students at the university had the right and option of being taught in Afrikaans. In his contention, the current policy (the 2016 Policy) deprives them of this right and it thus implicates the negative elements of the right protected by Section 29 of the Constitution. In Mazibuko v City of Johannesburg 2010 (4) SA 1 (CC) at para 47, the Constitutional Court reaffirmed that traditionally, constitutional rights (especially civil and political rights) are understood as imposing an obligation upon the State to refrain from interfering with the exercise of the right by citizens (the so-called negative obligation or the duty to respect). According to the Constitutional Court social and economic rights are no different in that the State bears a duty to refrain from interfering with social and economic rights just as it does with civil and political rights. For example in Government of the Republic of South Africa and Others v Grootboom supra at para 34, the Constitutional Court held that a negative obligation placed on the State and all other entities and persons to desist from preventing or impairing the right of access to adequate housing. See too Ex parte Chairperson of the Constitutional Assembly: In re Certification of the Constitution of the Republic of South Africa, 1996 (4) SA 744 (CC) at 79 and Minister of Health and Others v Treatment Action Campaign and Others No2 2002 (5) SA 721 (CC) para 46.

[39] It is true that the Supreme Court of Appeal reaffirmed in Afriforum matter that a negative duty on the State exists not to take away or diminish the right to being taught in Afrikaans without justification. Key to the assessment of the justification, according to the SCA, is whether the context and the circumstances have changed, and if so, whether good reason has been proffered for the change of policy. Mr Heunis argued that the current language policy constitutes a retrogressive measure in relation to the Afrikaans speaking students’ rights to education whereas Section 29 (1) (b) of the Constitution requires that the State make rights to further education progressively available and accessible. The Court is privy to the fact that, drawing from International experience in this regard (in the context of the International Covenant), the Committee on Economic, Social and Cultural Rights stated in General Comment No. 3 that the duty to progressively realise rights imposes an obligation to move as expeditiously and effectively as possible towards the goal of realising the right. The committee commented that ‘any deliberately retrogressive measures in that regard would require the most careful consideration and would need to be fully justified by reference to the totality of the rights provided for in the Covenant and in the context of the full use of the maximum available resources’. See General Comment No. 3 para 9.

[40] In Mr Heunis’ contention the evidence tendered by the University does not meet the above threshold test because Afrikaans speaking students, although they are no longer more than 50% of the total student population, remain the largest group. He conceded that there is a significant number of students who require to be educated in English. He hastened to add that there would be no justification for not lecturing the Afrikaans speaking students in Afrikaans. A mention must be made that according to the Constitutional Court, determining when it is reasonably practicable to receive tuition in a language of one’s choice will depend on all the relevant circumstances of each particular case. I agree that the Constitutional Court has not considered the issue in the context of university education. It has indeed provided a non- exhaustive list of factors in the school context and to this end has emphasised the context-specific approach that must be employed. As far as the latter approach is concerned, the applicants contend that the following evidence is of relevance: (a) Historically, there were 7 universities that catered for Afrikaans medium of instruction. This has changed since the inception of democracy. Currently, there is only one university that offers Afrikaans as a medium of instruction. (b) In the Western Cape, the evidence demonstrates that despite the Afrikaans speaking population accounting for almost 50%, there is currently not a single university offering Afrikaans as a medium of instruction. (c) In the Western Cape, Afrikaans is the first language of a large majority of persons of colour whose interests are affected. The submission on behalf of the applicants is that when considering the threshold of reasonable practicability, this Court must have regard to other rights in the Constitution which emphasise the importance of language to the Constitutional Court agenda. A reference to Section 6, 31 and 9 was made.

[41] The Court was also referred to Hartson v Lane N.O. 1998 (1) SA 300 (CC) where the Court set out the stages for an equality enquiry. This was done with reference to the Interim Constitution. In Prinsloo v Van der Linde 1997 (3) SA 1012 (CC) reliance is placed by Mr Heunis on the following:

‘It must be accepted that, in order to govern a modern country efficiently and to harmonise the interests of all its people for the common good, it is essential to regulate the affairs of its inhabitants extensively. It is impossible to do so without differentiation and without classifications which treat people differently and which impact on people differently. It is unnecessary to give examples which abound in everyday life in all democracies based on equality and freedom. Differentiation which falls into this category very rarely constitutes unfair discrimination in respect of persons subject to such regulation, without the addition of a further element. What this further element is will be considered later.’

A point is made on behalf of the applicants that a clear differentiation exists in the present case in that Afrikaans is being ousted as a language of instruction on an exclusive basis but SU has not demonstrated that the differentiation meets a legitimate government objective. The contention is that since differentiation is on the ground of language (listed in terms of section 9 (3) of the Constitution) it is presumed to be unfair. The impugned language policy is also attacked on the basis that its impact is that it impedes the constitutional objective of diversity as opposed to enhancing it. It is not only the Afrikaans speaking students, but it is also preferred by students of colour from this Province whose mother-tongue is Afrikaans. The 2016 Policy adopted by SU is described as having failed to foster diversity in language. According to Mr Heunis, the impugned policy instead, imposes a singular language option of English notwithstanding the provisions of Sections 6 and 31 of the Constitution. I undertake to consider and analyse the 2016 Policy.

[42] Mr Heunis is of the view that the Afrikaans speaking persons who are unable to communicate adequately or at all in the language of instruction at the SU (English), they will, in all likelihood forego the opportunity to study at SU, and if they are unable to access another university that has Afrikaans as a medium of instruction, they may forego the opportunity for tertiary study completely. Talking to unfair discrimination a reference was made to Equality Court. In the latter instance the claimant has to show that there is discrimination. Direct discrimination occurs when a law or policy expressly singles out a group for inferior treatment. On the other hand, indirect discrimination (also argued Mr Heunis) happens when a law or policy appears to be neutral, but has a disproportionate adverse impact on the protected class of persons. In Mr Heunis’ contention, since the vast majority of Afrikaans-speaking students are White and Coloured, the downscaling of Afrikaans as a language of instruction amounts to discrimination against them. One must observe that the argument that the policy discriminates on the grounds of language postulates that it withholds benefits from Afrikaans-speaking students that are enjoyed by English-speaking students since preference is given to English and there is no longer a clear commitment to equality of the two languages. The determination of fairness is not at all an easy task to be involved in. In anticipation of an argument to be presented on behalf of the respondents, Mr Heunis contended as follows:

‘Anticipating an argument that because the use of English imposes a burden on (primarily) Black (African) students whose home language is not English, it would be fair to expect Afrikaans students (including both White and Coloured students) to endure a similar burden by being taught in a language other than their home language, namely English, we submit that this type of formal equality is not what the Constitution (or the Equality Act) envisages. It is an argument for "the equality of the graveyard where all people must be equally badly off’.

[43] In Fourie supra, the Constitutional Court responding to an argument that equality could be obtained by the State refusing to issue marriage licences to either heterosexual or homosexual couples made the following observation:

‘Levelling down so as to deny access to civil marriage to all would not promote the achievement of the enjoyment of equality. Such parity of exclusion rather than of inclusion would distribute resentment evenly, instead of dissipating it equally for all. The law concerned with family formation and marriage requires equal celebration, not equal marginalisation; it calls for equality of the vineyard and not equality of the graveyard.’

One bears in mind that since the decision to adopt the new policy is apparently said to be subject to review in that it was made in the exercise of a public power, the question calling for consideration is whether, viewed objectively, the decision was rationally connected to the purpose for which the power was given. See University of the Free State supra and Pharmaceutical Manufacturers Association of South Africa and Others 2000 (2) SA 674 (CC) paras 85-86. Of course the above remains a factual enquiry and if a decision maker acts within its powers, and considers the relevant material in arriving at a decision so that there is a rational link between the power given, the material before it and the end sought to be achieved, the rationality threshold would be met. Of course if the decision maker misconstrues its power, it ordinarily will offend the principle of legality thereby rendering the decision made reviewable. See Masetlha v President of the Republic of South Africa and Another 2008 (1) SA 566 (CC) para 81. Concluding on this aspect Mr Heunis accused the Working Group, the Senate and Council as follows:

‘One of the most important shortcomings of the decision-making process which led to the adoption of the NLP is that neither the Working Group nor the Senate and the Council considered what would be reasonably practicable and overlooked the fact that the SU, as an organ of State, is co-responsible for taking steps to implement the right which derives from section 29(2) of the Constitution. The NLP falls foul of the LPHE, other provisions of the Constitutions and the requirement that existing rights may not be compromised without justification.’

In the process the SU is said to have abandoned its negative duty not to abrogate an existing right without proper justification.

INTERNATIONAL AND FOREIGN LAW

[44] Admittedly, the jurisprudence in this area is not well developed. The court was nevertheless referred to some international instruments and case law. Indeed there is some international and foreign authority suggesting that there may be an obligation upon States (in certain circumstances) to recognise and progressively realise such a right; to deploy available resources to support the exercise thereof; and not to withdraw such a right once it has vested, save where retrogression in the implementation of the right can be shown to be justified. This court is obligated to interpret the Constitution and the law such that the interpretation complies with the relevant international law to the extent the latter is not inconsistent with our law. This was best elucidated by the Constitutional Court in Glenister v President of South Africa and Others supra at para 97 where the Court observed as follows:

‘Our Constitution reveals a clear determination to ensure that the Constitution and South African law are interpreted to comply with international law, in particular international human-rights law. Firstly, s 233 requires legislation to be interpreted in compliance with international law; secondly, s 39(1)(b) requires courts, when interpreting the Bill of Rights, to consider international law; finally, s 37(4)(b)(i) requires legislation that derogates from the Bill of Rights to be 'consistent with the Republic's obligations under international law applicable to states of emergency'. These provisions of our Constitution demonstrate that international law has a special place in our law which is carefully defined by the Constitution.’

Article 26 of the Universal Declaration of Human Rights (‘UDHR’) provides that ‘Everyone has the right to education’. The UN Committee on Economic, Social and Cultural Rights has stated that measures entailing the withdrawal of a vested right:

‘Would require the most careful consideration and would need to be fully justified by reference to the totality of the rights provided for in the covenant and in the context of the full use of the maximum available resources.’

See UNDOC HR1/GEN 1 Rev 5.20; Liebenberg, Socio-Economic Rights (2009) p.189.

[45] Mr Heunis referred also to the International Covenant on Civil and Political Rights (‘ICCPR’) which affords protection to the right to education. Article 27 of ICCPR provides:

"In those States in which ethnic, religious or linguistic minorities exist, persons belonging to such minorities shall not be denied the right, in community with the other members of their group, to enjoy their own culture, to profess and practise their own religion, or to use their own language."

See GA res 2200A (XXI), 21 UN GAOR supp. (No 16) at 52, UNDOC A/6316 (1966), 999 UNTS 171, which came into force on 23 March 1976. Although the above is expressed negatively, the provision is accepted by the UN Human Rights Committee as conferring a positive right. It places an obligation upon the State to protect a minority’s identity ‘and the right of its members to enjoy and develop their culture or language’. See Human Rights Committee, General Comment No. 23 (50): The Rights of Minorities (Art 27) 08/09/04 CCPR/C/Rev. 1/Add. 5 (1994), paras 6.1 and 6.2. A reference was also made to the United Nations Convention on the Rights of the Child (‘CRC’). Perhaps a mention must be made that while the CRC protects the right of a child from a minority group to ‘enjoy his or her own culture’ and ‘to use his or her own language’, it does not guarantee a right to be taught in one’s mother tongue or freedom from the assimilary effects of schooling, particularly in the State sector. See H Cullen –Education Rights or Minority Rights? (1993) 7 International Journal of Law and the Family p. 143. Article 2 (1) of the UN General Assembly Declaration on the Rights of Persons Belonging to National or Ethnic, Religious or Linguistic Minorities defines the right of persons belonging to minorities as regards culture, language and religion in positive terms as follows:

‘Persons belonging to national or ethnic, religious and linguistic minorities … have the right to enjoy their own culture, to profess and practice their own religion, and to use their own language, in private and in public, freely and without interference or any form of discrimination.’

The Court was also referred to the European Convention on Human Rights (‘ECHR’) particularly article 2 of the First Protocol providing thus:

‘No-one shall be denied the right to education. In the exercise of any functions which it assumes in relation to education and to teaching, the State must respect the right of parents to ensure such education and teaching in conformity with their own religious and philosophical convictions.’

The judgment of the ECHR in Belgian Linguistics (No.2) (1979-80) 1 EHRR 252, is regarded the most influential authority on the interpretation of A2PI. The question was whether the failure to make French-language education available in the Flemish region, and the withholding of grants from schools which did not give instruction in Flemish, violated A2P1 read with article 14, by discriminating on the ground of language. It was held that Article 2 of Protocol 1 (A2P1) to the ECHR ‘does not specify the language in which education must be conducted in order that the right to education should be respected", the right "would be meaningless if it did not imply … the right to be educated in the national language or one of the national languages, as the case may be’.

SUBMISSIONS IMPLICATING PAJA SPECIFICALLY

[46] Mr Heunis contended that the impugned decisions and the New Language Policy (NLP) are subject to being reviewed and set aside under PAJA in that the SU is an organ of State. He referred to Altech Autopage Cellular (Pty) Ltd v The Chairperson of the Council of the Independent Communications Authority of South Africa (Case No 2002/08 (TPD). The latter is an unreported case where the Court reviewed and set aside ministerial ‘policy’ directives, rejecting arguments that the directions were not susceptible to review, because they constituted executive rather than administrative action. Davis J held that even if not so, the directives could not escape review since they would then be subject to the constitutional principle of legality. The latter judgment only has persuasive value that does not bind this Court. In Mr Heunis’ contention the NLP and decisions which led thereto are invalid, reviewable and fall to be set aside by reasons of the provisions of Section 6 (2) (a) (iii), 6 (2) (c), 6 (2) (d) and further provisions of PAJA. He attacked the procedure followed contending that the process was procedurally unfair. The following are some of the accusations labelled against the respondents: (a) the decisions were influenced by errors of law; (b) they were taken for ulterior purposes or motives; (c) decisions were infected with bias; (d) were so unreasonable that no reasonable person could have supported such. Importantly, the submission made is that ignoring the comments of interested parties in respect of the draft language policy overwhelmingly supportive of the retention of Afrikaans as a primary language of instruction testifies to bias and the inappropriate attachment to a pre-determined outcome. In Mr Heunis’ contention, the whole process was designed to create the pretence of a consultation without there having been any real consultation. The respondents are accused of having not honoured undertakings made to Afriforum and the convocation regarding consultation. On the applicants behalf neither the Working Group nor the Senate and the Council considered the Constitution and the LPHE. Having documented the submissions made on behalf of the applicants, it is now time opportune to respond thereto. In the nature of this matter, the best response must be governed by subtopics. The discussion will be best facilitated if I first briefly describe the main elements of the 2014 Policy and the impugned 2016 Policy. I record briefly the reasons advanced by SU as to why it replaced the 2014 Policy with the 2016 Policy.

THE ELEMENTS OF 2014 AND 2016 POLICIES

[47] The SU adopted its first official language policy and an accompanying language plan on 12 December 2001 following the publication of the national language policy for higher education (‘the LPHE’) under Section 3 of the Act in November 2002. A mention must be made that under the 2002 Policy, Afrikaans was the default language of undergraduate learning and instruction, with the use of English being allowed only after the reasons had been thoroughly considered. Of course Afrikaans was also the default institutional language with English being used alongside Afrikaans as a language of internal communication as circumstances may require. Both Afrikaans and English were used in postgraduate learning and instruction. Afrikaans, English and, where possible, isiXhosa were the languages of external communication. IsiXhosa would (reportedly) be promoted as a developing academic language.

[48] On 22 November 2014, the SU Council adopted the 2014 Policy and it made consequential amendments, to the language plan. Under the 2014 Policy (a) Afrikaans and English were SU’s languages of learning and teaching – it was committed to purposefully extending the academic application of both; (b) Afrikaans and English would be employed in various usage configurations, which were spelled out in more detail in the Plan; (c) Parallel-medium teaching and real-time educational interpreting were the preferred options where practically feasible and affordable; (d) Postgraduate learning would happen in both languages, with significant utilisation of English; (e) Documentation of prime importance had to be available in Afrikaans and English; (e) Afrikaans and/or English and where feasible, isiXhosa had to be used by SU for external communication.

[49] The Plan, as amended by the Council on 22 November 2014, gave substance to the 2014 Policy. In particular, it created the following language – specifications for undergraduate teaching in the following order of preference: (a) for the first two years of undergraduate studies, normally: (i) parallel-medium teaching in separate groups for modules with 250 students or more (A+E); or (ii) real-time interpreting from Afrikaans to English (A+i) or from English to Afrikaans (E+i)), depending on the language the lecturer was more comfortable with. (b) For the third year of undergraduate studies and onwards: (i) preferred options: A+E; A+I or E+i, depending on the language the lecturer was more comfortable with; or (ii) provided the relevant faculty can show: (a) the preferred options are not feasible; and (b) the support offered for students who are not sufficiently academically literate in Afrikaans or English: (i) dual-medium teaching, i.e. the balanced use of Afrikaans and English to one class group, with the Afrikaans offering at least 50% (T-specification). The Plan states this exploits the proven advantages of bi-or multilingual teaching and is particularly suited in the senior years of study, when students’ proficiency in the two languages should be more strongly developed. However, the Plan acknowledges the T-specification supposes a certain minimum proficiency in both these languages. Consequently, the Plan that students who do not understand one of the languages at all will miss content in a context where they are not well supported. (ii) English only (E), if the lecturer is not proficient enough in Afrikaans for the T-specification; (iii) Afrikaans only (A) where the resources for multilingual presentation are not yet available. (c) The Plan did however, allow for the use of the T, E or A-specifications in the first two years of undergraduate study, provided it was indicated how students who lacked sufficient Afrikaans or English language skills would be supported to benefit from the full content of lectures.

WHY CHANGE THE 2014 POLICY?

[50] The 2014 Policy and Plan were intended to make it easier for English-speaking students to obtain an education at SU. We are told that during 2015 and the first half of 2016 it became clear that the 2014 Policy and Plan (although it was not their purpose) excluded students who were proficient in English but not proficient in Afrikaans. The majority of those students excluded were Black (African) students. As a result of their poor Afrikaans, the majority of Black (African) students (i) could not fully understand the lectures presented in the A or T specifications; (ii) they felt stigmatised by the real-time interpretation, which was almost solely used for translating the lectures they could not understand; and (iii) they felt excluded from other aspects of campus life, like residence meetings and official SU events which took place in Afrikaans, without interpretation.

[51] It is not disputed that by contrast, nearly all Afrikaans-speaking students were sufficiently proficient in English to understand SU’s academic content presented in English. Thus to require them to take certain lectures in English would not impose a comparable burden on them. Importantly for many years SU has prescribed text books in English, the result being that its students have at least to be able to read and understand English. We gather from the answering papers that SU undertook a study of the cost of an immediate change to a full parallel medium system and this indicated that it would be an amount of about R640 million in infrastructure and about R78 million per year for additional personnel. This reportedly, translated to an approximately 20% increase in fees (up by R8100 from about R40 000 per year).

[52] Consequently SU decided to adopt a new language policy (the 2016 Policy) which would result in a 100% English offering, but would not similarly increase the Afrikaans offering. According to the answering papers, instead, it (SU) would manage the sum total of the Afrikaans offering so as to maintain access for students who choose to study in Afrikaans and to further develop Afrikaans as a language of instruction where reasonably practicable. As hinted in the introduction the 2016 Policy was adopted by Senate on 9 June 2016 and by the Council on 22 June 2016.

THE 2016 POLICY

[53] The applicants in their contentions assume that the 2016 Policy will cause the ‘virtual exclusion’ of Afrikaans. The respondents, however, are of the view that this assumption is totally mistaken. Looking squarely at the 2016 Policy, one gathers that it does not reduce the Afrikaans offering at SU. In fact the expressly stated goal of the Policy (at its para 7.4.1.2) is to maintain and if possible increase the Afrikaans offering subject to demand and resources. Of course it does at the same time adopt a preference for English in certain circumstances in order to advance SU’s multiple goals, namely, equal access, multilingualism, integration, and preserving Afrikaans, all within available resources. The purpose and aims of the Policy are: (a) The Policy expressly states that its purpose is to ‘give effect to Section 29 (2) (language in education) and 29 (1) (b) (access to higher education) read with Section 9 (equality and the prohibition against direct and indirect unfair discrimination) of the Constitution’. (b) It records that ‘[a]pplying and enhancing the academic potential of Afrikaans is a means to empowering a large and diverse community in South Africa’. See Para 2 of the Policy. It explains in detail how SU will ‘advance the academic potential of Afrikaans’ in Para 7.5.3 of the Policy. (c) It repeatedly notes SU’s commitment to multilingualism ‘as a differentiating characteristics of SU’, and devotes an entire Section to how SU will promote multilingualism and particularly the use of Afrikaans and isiXhosa. (d) One of its principles is that ‘[l]anguage should promote access… and should not constitute a barrier to students or staff’, particularly in the light of past racial discrimination. See Policy para 6.1. (e) It emphasises that the Policy ‘and its implementation are informed by what is reasonably praticable’. See Policy para 6.8. It then goes on to explain that reasonably practicability included an assessment of:

‘The number of students who will benefit from a particular mode of implementation, the language proficiency of the students involved, the availability and language proficiency of staff members, timetable and venue constraints, as well as SU’s available resources and the competing demands on those resources.’

[54] The operational parts of the Policy must be interpreted in light of the above stated goals, purposes and principles. The provisions regulating the use of Afrikaans as set out in the answering papers are the following: As to learning and teaching, the Policy provides: (a) Afrikaans and English are SU’s two languages of learning and teaching; (b) undergraduate modules will be taught mainly in parallel medium (separate lectures in Afrikaans and English) or dual medium (during each lecture all information is conveyed at least in English and summaries or emphasis on content are also given in Afrikaans), or, in a limited range of circumstances, in either Afrikaans or English (Namely, where the nature of the subject matter of the module justifies doing so, where the assigned lecturer is proficient to teach only in Afrikaans or English or where all the students in the class group have been invited to vote by means of a secret ballot and these students who have voted, agree unanimously to the module being presented in Afrikaans only or English only]; (c) in postgraduate learning and teaching, including final year modules at NQF level 8, any language may be used provided all the relevant students are sufficiently proficient in that language.

[55] In addition to the general policy provisions governing and teaching set out above, the following further provisions govern the use of Afrikaans at SU: (a) In dual-medium module lectures questions in Afrikaans are answered in Afrikaans. (b) In dual-medium module lectures and single-medium module lectures in English, during the first year of study SU makes simultaneous interpreting available in Afrikaans; and during the second and subsequent years of study, simultaneous interpreting is made available upon request by a faculty, if the needs of the students warrant the service and SU has the resources to provide it. (c) For all undergraduate modules, all SU module frameworks and study guides are available in Afrikaans compulsory reading material (excluding published material) is also provided in Afrikaans where reasonable practicable and students are supported in Afrikaans during a combination of appropriate, facilitated learning opportunities (e.g. consultations during office hours, or scheduled tutorials and practicals). (d) Question papers for tests, examinations and other summative assessments in undergraduate modules are available in Afrikaans and students may answer all assessments and submit all written work in Afrikaans. (e) A variety of information and communication technology (ICT) enhanced learning strategies, including podcasts and vodcasts of lectures, are made available to students in Afrikaans for the further reinforcement of concepts and for revision purposes. (f) Afrikaans (together with English) is used for internal institutional communication, including in all documentation of primary importance. (g) Afrikaans (together with English) is used for external communication.

[56] Generally, SU advances the academic potential of Afrikaans by means of, for example, teaching, conducting research, holding symposia, presenting short courses, supporting language teachers and hosting guest lecturers in Afrikaans; presenting Afrikaans language acquisition courses; developing academic and professional literacy in Afrikaans; supporting Afrikaans reading and writing development; providing language services that include translation into Afrikaans, and editing of and document design for Afrikaans texts; developing multilingual glossaries with Afrikaans as one of the languages; and promoting Afrikaans through popular-science publications in the general media. Additionally the following further policy provisions govern the use of English at SU: (a) In dual-medium module lectures questions in English are answered in English. (b) In dual-medium module lectures, during the first year of study SU makes simultaneous interpreting available in English; and during the second and subsequent years of study, simultaneous interpreting is made available upon request by a faculty, if the needs of the students warrant the service and SU has the resources to provide it. (c) In single-medium module lectures in Afrikaans, SU makes simultaneous interpreting available in English. (d) For all undergraduate modules, all SU module frameworks and study guides are available in English, all compulsory reading material is provided in English except where the module is about the language itself and students are supported in English during a combination of appropriate, facilitated learning opportunities (e.g. consultations during office hours, or scheduled tutorials and practicals). (e) Question papers for tests, examinations and other summative assessments in undergraduate modules are available in English and students may answer all assessments and submit all written work in English. (f) A variety of ICT-enhanced learning strategies, including podcasts and vodcasts of lectures, are made available to students in English for the further reinforcement of concepts and for revision purposes. (g) English (together with Afrikaans) is used for internal institutional communication, including in all documentation of primary importance. (h) English (together with Afrikaans) is used for external communication.

[57] In summary, the Policy creates three language specifications, namely, parallel medium, dual medium and single medium. The parallel medium is employed where it is reasonably practicable and pedagogically sound. Where parallel classes are not possible or appropriate, classes are taught in dual medium meaning that: all material is conveyed in English; (b) summaries or emphasis of content is provided in Afrikaans; and (c) questions are answered in the language in which they are asked. Additionally, (i) All first year dual medium classes are supported by simultaneous translation; and (ii) Lectures in later years will be translated if there is a request by the faculty, the needs of students warrant it, and SU has the resources to provide it.

[58] Single medium classes are offered in only three limited circumstances: (a) where the subject matter justifies it; (b) where the lecturer is only proficient in one language; or (c) where the students unanimously vote by secret ballot to be taught in a single language. Where the lecture is single medium because of the lecturer’s proficiency:

(a) SU will always provide simultaneous translation from Afrikaans to English; and (b) It will provide simultaneous translation from English to Afrikaans; (i) for all first year modules; and (ii) in second and third year modules if there is a request by the faculty, the needs of students warrant it, and SU has the resources to provide it.

[59] The details below testify to the assertion that the Policy is designed to grant the greatest possible tuition in English and Afrikaans, within SU’s available resources. Indeed there are only three ways in which the Policy treats English differently from Afrikaans and these are (a) in dual-medium module lectures all information is conveyed at least in English, whereas summaries or emphasis of content is also given in Afrikaans. However, simultaneous translation is made available in all first year dual medium modules, and in later years on request, considering student needs and available resources. (b) for undergraduate modules where the assigned lecturer is proficient to teach only in Afrikaans, SU will make simultaneous interpreting available in English during all years of undergraduate study. It is only during the second and subsequent years of study that there is a distinction. In those, English, simultaneous interpreting will only be made available upon request by a faculty, if the needs of the students warrant the service, and SU has the resources to provide it. (c) whereas all compulsory reading material is provided in English (the exception being where the module is about another language), there are two limitations on the provision of compulsory material in Afrikaans: (i) Material which is not published in Afrikaans need not be made available in Afrikaans; and (ii) Non-published compulsory material is made available in Afrikaans where reasonably practicable.

[60] In all other ways, it would appear, English and Afrikaans are treated identically. While English enjoys preference, it can safely be mentioned that the impact on Afrikaans speakers is extremely limited. The aforegoing is so because: (a) in the first year of study there is no difference at all. All lectures are given simultaneous translation, students will have equal access; (b) the limitations are all linked directly to what is reasonably practicable. Whether SU will offer a module in parallel medium, and whether it will offer simultaneous translation in dual-medium or English lectures in later years of study is expressly made subject to what is ‘reasonably practicable’, or to the needs of students and SU’s resources; (c) The

slight preference only applies to lectures and, to a limited degree, materials. For pedagogical reasons, SU intends – like other universities across the world – to move away from the lecture being the sole focus of learning and teaching. Other facilitated learning opportunities will become increasingly central to the learning process. Those will be equally available in English and Afrikaans and increasingly in IsiXhosa; (d) the Policy creates an accountability mechanism to ensure that Afrikaans teaching is not reduced significantly from pre-2016 Policy level and in increased where this is possible. Paragraph 7.4.1.2 of the Policy reads: ‘The Afrikaans offering is managed so as to sustain access to SU for students who prefer to study in Afrikaans and to further develop Afrikaans as a language of tuition where reasonably practicable’. The Senate is obligated in terms of paragraph 7.4.3 to approve all language plans and so can send a plan back to the faculty for reconsideration if it fails to meet this requirement. The import of this provision is that: (i) the Afrikaans offering cannot be reduced materially as that would not ‘sustain access’ for Afrikaans students; and the Afrikaans offering should be increased to the extent that is logistically and financially practicable.

[61] It is doubtful that there will be any reduction in the Afrikaans offering (to the level suggested on behalf of the applicants) compared to what was offered under the 2014 Policy. Obviously, that will depend on how faculties implement the policy. Arguably, it may be that the 2016 Policy under discussion will lead to more parallel medium classes and more simultaneous interpretation which will increase the total amount of Afrikaans tuition. Even if the reduction becomes a reality, that cannot be described as the intent of the Policy and will certainly not be an inevitable consequence of implementing the Policy. It clearly will be a direct consequence of the nature of student demand and the limits of SU’s resources. I find it difficult to accept that the Policy intends to reduce Afrikaans. In my understanding, the Policy is crafted and/or designed to retain the extent of Afrikaans tuition under the 2014 Policy and to offer as much Afrikaans tuition as SU is reasonably able to do so, considering what is reasonably practicable (particularly the needs of students and SU’s resources).

A CHALLENGE TO SU 2016 POLICY - BUT THE STATE POLICY NOT CHALLENGED

[62] It is abundantly clear from the aforegoing discussion that the applicants have sought to review and set aside SU’s 2016 Policy and the decisions of the Senate and the Council adopting it. It has been sufficiently demonstrated that the right to receive education in the official language of one’s choice in Section 29 (2) of the Constitution is at the Centre of the applicants’ attack. The applicants have not sought to challenge the State’s language policy- the LPHE referred to earlier. It is trite that the LPHE has provisions that: (a) reject the idea that SU and the (then) Potchefstroom University for Christian Higher Education should be designed as ‘custodians’ of Afrikaans as an academic language, because doing so could concentrate Afrikaans speaking students in those institutions and thereby set back ‘the transformation agendas of [the other] institutions that have embraced parallel or dual medium approaches as a means of promoting diversity]; (b) reject the idea of Afrikaans universities, as district from universities which accept institutional responsibility for promoting Afrikaans as an academic medium, because Afrikaans universities would be contrary to the end goal of a transformed higher education system which, as indicated in National Plan for Higher Education, envisages ‘the creation of higher education institutions whose identity and cultural orientation is neither black nor white, English or Afrikaans-speaking, but unabashedly South African’; and (c) State that to achieve the goal of sustaining Afrikaans as medium of academic expression and communication, there must be ‘a range of strategies’ including ‘the adoption of parallel and dual language medium options, which would on the one hand cater for the needs of Afrikaans language speakers and, on the other, ensure that language of instruction is not a barrier to access and success’.

[63] Indeed the fact that there is no challenge to the State’s language policy is of importance in the light of the SCA’s holdings in UFS v Afriforum supra. Needless to mention that the challenge there was limited to a review of the decision to adopt UFS’S language policy. The main constitutional ground of attack in the UFS case was that Section 29 (2) required the UFS to continue with its existing parallel medium policy because there were no resource constraints stopping it from doing so (and even though in practice it led to segregation along racial lines with mainly white students attending the Afrikaans lecture). It must be pointed out that, like the challenge in the present matter, the challenge in UFS v Afriforum supra did not extend to the LPHE.

[64] In view of the confined target of the challenge in UFS v Afriforum, the SCA held that difficult underlying questions about whether UFS’s policy unfairly discriminated against linguistic and cultural minorities, or promoted ‘majoritarian hegemony at the expense of linguistic and cultural diversity’, or undermined ‘the fundamental language scheme of our constitutional order, which requires the state to take practical and positive measures to elevate the status and advance the use of all official languages, instead of diminishing their importance’, did not arise for decision. As Cachalia JA explained, ‘such questions may only be confronted through a substantive constitutional challenge to the State’s language policy, and not somewhat diffidently or obliquely through judicial review, as the respondents have done in this case’.

[65] Regard being had to the aforegoing one may go so far as to say that SU is not at all responsible for the fate of Afrikaans throughout South Africa. Its responsibility in this regard stretches to the boundaries of the University itself. The deeper issues about ‘majoritarian hegemony’ must be dealt with through an attack on the State’s policy, as expressed in the LPHE. SU’s Policy complies with the LPHE which allows each university to take reasonable decisions on their own language policy. The Applicants’ real complaint appears to be the cumulative effect of those decisions by multiple universities that negatively impact Afrikaans-speakers. As the SCA held, the target then is the State’s language policy, not SU’s Policy. The respondents contend that the 2016 Policy is constitutionally compliant. The applicants have, in my view, not persuaded this Court that the SU 2016 Policy is in any way unconstitutional.

DO THE IMPUGNED DECISIONS CONSTITUTE EXECUTIVE ACTION OR ADMINISTRATIVE ACTION?

[66] Notably both the applicants and the respondents have pleaded this case on the basis that the impugned decisions constitute administrative action as defined in PAJA. Indeed that was the position the parties and the High Court adopted in Afriforum v University of Free State (A701 [2016] ZAFSHC 130 (21 July 2016). Seemingly, the same approach was adopted by the Full Court in Afriforum and Another v Chairperson of the Council of the University of Pretoria and Others [2017] 1 ALL SA 832 (GP) (even though in the latter case, there is no clear finding on the aspect). However, in UFS v Afriforum supra, the Supreme Court of Appeal held that the decision to adopt a language policy taken by the Council of the UFS was executive in nature and ‘does not constitute administrative action as contemplated by PAJA.’ See para 18 of the judgment. Thus the challenges to the decisions of the Senate and Council could not be adjudicated under PAJA.

[67] As highlighted earlier in this judgment, the applicants contend that, because they (unlike UFS) have challenged the Policy, PAJA does apply. The point is, however, although the policy itself was not challenged in the University of the Free State case, Cachalia JA made it clear that the Policy itself does not amount to administrative action. He held: ‘the policy itself does not adversely affect the rights of any person or have the capacity to do so. Neither does it have direct, external legal effect’. See UFS v Afriforum supra para 18. It seems, it would only be decisions taken in the implementation of the policy that would be subject to administrative review. Therefore on the authority of the Supreme Court of Appeal PAJA remains inapplicable to the present application. Of course the decisions and the Policy are subject to review under the principle of legality. That would essentially mean that the grounds of review that apply are more circumscribed and that the intensity of review is reduced. The Supreme Court of Appeal observed guidingly as follows in this regard in UFS v Afriforum supra:

‘The question to be considered in this context is whether, objectively viewed, the decision was rationally connected to the purpose for which the power was given. This is a factual enquiry and courts must be careful not to interfere with the exercise of a power simply because they disagree with the decision or consider that the power was exercised inappropriately. If, therefore, the decision-maker acts within its powers, and considers the relevant material in arriving at a decision so that there is a rational link between the power given, the material before it and the end sought to be achieved, this would meet the rationality threshold. The weight to be given to the material lies in the discretion of the decision-maker; so too does the determination of the appropriate means to be employed towards this end. But if a decision-maker misconstrues its power, this will offend the principle of legality and render the decision reviewable.’

[68] Therefore, the only grounds of review that the applicants can rely on are: (a) That the decision was substantively irrational. See Pharmaceutical Manufacturers Association of South Africa and Another: In re Ex Parte President of the Republic of South Africa and Others 2000 (2) SA 674 (CC). This of course is a far lower standard than reasonableness required under PAJA. See Democratic Alliance v President of South Africa and Others 2013 (1) SA 248 (CC) at paras 29-32. The important message to bear in mind is that Chaskalson CJ explained that, it is not for the courts to decide that there were better ways for the executive to achieve its goal. He observed as follows:

‘The fact that there may be more than one rational way of dealing with a particular problem does not make the choice of one rather than the others an irrational decision. The making of such choices is within the domain of the executive. Courts cannot interfere with rational decisions of the executive that have been made lawfully, on the grounds that they consider that a different decision would have been preferable.’

See Bel Porto School Governing Body v Premier, Western Cape 2002 (3) SA 265 (CC) at paras 41-5. (b) That the decision was procedurally irrational. This of course does not require full procedural fairness but merely a rational connection between the procedure adopted and the purpose of the decision. See Albutt v Centre for the Study of Violence and Reconciliation 2010 (3) SA 293 (CC) at paras 49, 72 and 74; Democratic Alliance v President of South Africa and Others supra at paras 27 and 34; National Treasury and Another v Kubukeli 2016 (2) SA 507 (SCA) at paras 16-18. (c) That the decision was unlawful. This includes that: (i) the decision and the policy are inconsistent with the Constitution; (ii) The decision-makers were biased or improperly influenced; (iii) The decision was taken for an ulterior purpose; and (iv) The decision was ultra vires.

Importantly, the applicants cannot attack the decisions or the Policy on the basis that they were unreasonable or procedurally unfair, nor on the ground that information was not considered, unless the failure to do so tainted the rationality of the process as a whole. The difficulty faced by the applicants is that even if PAJA is nevertheless applicable, the Policy would still have been lawfully adopted and will thus survive substantive administrative law review.

THE IMPLEMENTATION OF THE 2016 POLICY IS NOT BEING ATTACKED

[69] The application was launched in September 2016 far before the Policy was implemented (it was implemented on 1 January 2017). The applicants clearly believed that the 2016 Policy and the process followed to adopt same are irredeemably flawed. In the evaluation of the substantive attack on the constitutionality of the Policy, this Court is duty bound to evaluate it as written accepting implementation as a reality. Several constitutional court cases have spoken to this. In S v Jordan and Others (Sex Workers Education and Advocacy Task Force and Others as Amici Curiae) 2002 (6) SA 642 (CC), the applicants argued that a law criminalising sex work was unfairly discriminatory because (in practice) it was only enforced against the sex workers, who were almost all female. The Court rejected the argument for the following simple reason: ‘What happens in practice may therefore point to a flaw in the application of the law but it does not establish a constitutional defect in it.’

[70] In Minister of Public Works and Others v Kyalami Ridge Environmental Association and Others (Mukhwevho Intervening) 2001 (3) SA 1151 (CC), the Government established a transit camp to aid flood victims. An organisation challenged the decision to establish the camp on the basis that it was unlawful. The High Court upheld the challenge because the government could not implement its decision without obtaining the necessary consents under various pieces of legislation (which had not yet been obtained). The Constitutional Court concluded that the High Court had been wrong to take that approach because it ‘failed to distinguish between the taking of the decision and its implementation.’ As Chaskalson P (as he then was) explained, it was possible that the decision could ‘be lawfully implemented if the necessary consents are obtained. While the absence of the consents was a basis to interdict implementation, ‘it is not a ground on which the decision can be set aside’, so the Constitutional Court reasoned. South African Police Service v Solidarity obo Barnard 2014 (6) SA 123 (CC) concerned the reverse situation – a challenge to implementation instead of the underlying policy itself. The Constitutional Court again stressed that the two types of challenges are distinct. As the policy itself – in that case an employment equity plan – had not been challenged, the Court had to accept that it was valid. The reverse is also true – where implementation has not been challenged, it must be accepted that it will occur according to the policy.

[71] Mr Muller took a point that the third applicant (Adv. JC Heunis SC) participates in these proceedings as a party – (as distinct from his role as lead counsel for the applicants) in his official capacity as the President of the SU Convocation; and that as the President of the Convocation he does not have legal capacity to bring an application of this kind against SU, since the Convocation itself also does not have the requisite legal capacity and both the Convocation and its President are organs of SU and cannot adopt a position in litigation adversarial to the University. In this regard a reference was made to Registrar of Pension Funds v Howie NO and Others [2016] 1 ALL SA 694 (SCA). In the latter case it was held that the Financial Services Board does not have locus standi to review a decision of the Board of Appeal established by the Financial Services Board Act 97 of 1990 because it cannot adopt a position adversarial to the Board of Appeal. The above point was, however, not persisted with in view of the fact that there are a number of applicants involved in this matter. The point is thus not dispositive of the issues in this application.

THE CHALLENGES TO THE CONTENT OF THE 2016 POLICY

[72] It has been demonstrated above that on authority of the Supreme Court of Appeal in the University of the Free State matter, the decisions and the Policy are executive action and subject only to limited review on grounds of rationality and legality. We are mindful that the primary case presented by the applicants in the light of the findings in UFS v Afriforum supra is that: (a) the Policy is inconsistent with Section 29 (2) of the Constitution; (b) the Policy constitutes unfair discrimination against Afrikaans speakers and White and Coloured students; and (c) the Policy is contrary to the right of access to higher education in Section 29 (1) (b) of the Constitution. It is by now common cause that the aforegoing arguments rest on the standard of reasonableness and fairness. The contention by SU is that the Policy will not result in any reduction in the Afrikaans offering. In any event the differential treatment of English and Afrikaans is justified by the necessity to ensure that Black (African) students are not excluded from SU, to promote multilingualism, and to ensure integration. Another point made is that it is consistent with what SU is reasonably able to provide given its resources.

THE RIGHT TO EDUCATION IN THE LANGUAGE OF CHOICE (S 29 (2) OF THE CONSTITUTION)

[73] The legal position in the above regard has now been definitively set out by the Supreme Court of Appeal in UFS v Afriforum supra. The position is the following: (a) What is ‘reasonably practicable’ is an assessment of equity and historical redress; (b) Courts should be extremely hesitant to interfere with a university’s determination of what is reasonably practicable; and (c) it is rational for a university to conclude that it is not reasonably practicable to teach in Afrikaans because it will result in an unconstitutional situation on its campus, such as segregated classrooms. A mention must be made that the Supreme Court of Appeal upheld the UFS’s language policy which ended Afrikaans tuition almost entirely, solely to ensure that the campus was racially integrated. There was no suggestion that UFS lacked the resources to continue to provide Afrikaans tuition, or that any students were unfairly discriminated against as a result of the UFS’s prior policy. I am of the view that the fact that SU acted on the basis of a more onerous understanding of the legal limits on its power to determine its own policy cannot be allowed to count against it.

INTERPRETATION OF S 29 (2) ITS HISTORICAL CONTEXT, PURPOSE AND STRUCTURE

[74] The above task has been embarked upon and concluded by the Supreme Court of Appeal in the recent UFS v Afriforum judgment. The right to own language education protected in Section 29 (2) has indeed a pedigree in international human rights law. See, for example, UNESCO Convention Against Discrimination in Education, art 5; The Document of the Copenhagen Meeting of the Conference on the Human Dimension of the CSE, para 34; UN Declaration of the Rights of Persons Belonging National or Ethnic, Religious and Linguistic Minorities, art 4.3 (‘States should take appropriate measures so that, wherever possible, persons belonging to minorities may have adequate opportunities to learn their mother tongue or to have instruction in their mother tongue’); and the Framework Convention for the Protection of National Minorities, art 14. None of the instruments that South Africa has ratified includes this right. None of the following conventions contain an express right to own-language education: The African Charter, the ICESCR, the ICCPR, the Convention on the Rights of the Child, and the African Charter on the Rights and Welfare of the Child. They include only general prohibitions against discrimination on the basis of language, general protections of minorities’ rights to use their language, and rights to basic education (not higher education).

[75] As the applicants note, some commentators have argued that a limited right to own-language education can be gleaned from these provisions. SU accepts that. But it does not alter the analysis of s 29(2) because the international law does not impose a higher standard on SU than the Constitution does. The right is normally claimed by vulnerable minorities who have been disadvantaged by past or current oppression by a majority. It is vital for those communities to maintain their language and their community. One of the key ways in which they achieve those goals is through education in their own language.

[76] Nobody can deny that the South African context is more complicated. Afrikaans- speakers are a linguistic racial and ethnic minority in this country. White Afrikaans-speakers, as a group are also (together with White English-speakers) the beneficiaries of Apartheid and are as such undoubtedly economically and educationally an advantaged group. Given the historical advantage given to Afrikaans it enjoys far better reach educationally than any other official language, save for English. In this regard Moseneke DCJ has observed in Head of Department of Education and Another v Hoërskool Ermelo and Another 2010 (2) SA 415 (CC) (‘Ermelo’) at para 46:

‘It is so that white public schools were hugely better resourced than black schools. They were lavishly treated by the apartheid government. It is also true that they served and were shored up by relatively affluent white communities. On the other hand, formerly black public schools have been and by and large remain scantily resourced. They were deliberately funded stingily by the apartheid government. Also, they served in the main and were supported by relatively deprived black communities. That is why perhaps the most abiding and debilitating legacy of our past is an unequal distribution of skills and competencies acquired through education.’

This was of course in the context of schooling. See also MEC for Education in Gauteng Province and Other v Governing Body of Rivonia Primary School and Others 2013 (6) SA 582 (CC) at paras 1-2.

[77] The Constitution addressed this history by allowing and mandating the State to take measures to address past discrimination. See Section 9 (2) of the Constitution which reads as follows:

‘Equality includes the full and equal enjoyment of all rights and freedoms. To promote the achievement of equality, legislative and other measures designed to protect or advance persons, or categories of persons, disadvantaged by unfair discrimination may be taken.’

Section 9 (2) needs to be read together with Section 7 (2) which imposes a positive obligation on all organs of State to ‘respect, protect, promote and fulfil the rights in the Bill of Rights’. Afrikaans speakers remain the bearers of the rights under Section 29 (2). The Constitutional Court has recognised that Afrikaans ‘is one of the cultural treasures of South African national life’. See In Re Dispute concerning the Constitutionality of Certain Provisions of the Gauteng School Educational Bill of 1995 supra at para 49, quoted with approval in Ermelo case supra para 48. I do not understand the case presented by SU to be denying that Afrikaans speakers are fully entitled to the rights in Section 29 (2) of the Constitution. But the truth is that the nature of the right must be analysed with full acknowledgement of the historical context of State support for Afrikaans, disregard for other indigenous languages, and racial exclusion from education. In City of Tshwane Metropolitan Municipality v Afriforum and Another [2016] ZACC 19; 2016 (6) SA 279 (CC), Mogoeng CJ observed as follows:

‘we have made a solemn undertaking to embark on an all-inclusive constitutional project, geared at achieving national unity and reconciliation. The injustices of the past are not to be pampered or approached with great care or understanding or sympathy. And the immeasurable damage racism or cultural monopoly has caused requires that stringent measures be taken to undo it.’

Clearly the commitment to equality, to redress and to equal access to further education (motivating the LPHE as well) are and remain fundamental parts of the Constitution’s mission.

[78] Talking to the purpose of Section 29 (2) Mr Muller submitted that it serves two purposes, namely (a) ‘To improve the quality of education, as people learn better in their mother tongue. It is therefore related to section 29(1) of the Constitution and, in the context of the University, to section 29(1)(b) which guarantees the right to “further education, which the state, through reasonable measures, must make progressively available and accessible’ and (b) to promote and maintain cultural communities. In Gauteng School Education Bill supra Sachs J explaining the language observed that ‘is not simply a means of communication and instruction, but a central element of community cohesion and identification for a distinct community in South Africa.’ It is linked to culture. See Section 31 of the Constitution. The structure of Section 29 (2) is such that it achieves its purposes in ‘two distinct but mutually reinforcing part’. See Ermelo case supra para 52. It is so that the first part reads ‘Everyone has the right to receive education in the official language or languages of their choice in public educational institutions where that education is reasonably practicable’. This clearly affords a right to own-language tuition only where that is ‘reasonably practicable’ calls for a ‘context-sensitive understanding’. Indeed in the context of basic education, the inquiry has been held to demand a consideration of ‘all the relevant circumstances of each particular case’ including: ‘the availability of and accessibility to public schools, their enrolment levels, the medium of instruction of the school that its governing body has adopted, the language choices that learners and their parents make, and the curriculum options offered.’

See Ermelo supra para 52. See also Afriforum v UFS supra para 15 (‘factors such as learner numbers, costs, availability of facilities and educators, the distance to the nearest similar institution that is able to provide education in the chosen language, and the chosen medium of instruction in the case of universities, can be relevant factors that may determine whether, in a particular case, it is reasonably practicable to provide such education’). The second part explains how the state should provide the right if it is triggered by the first part – if own-language education is reasonably practicable. Section 29 (2) places ‘an injunction on the State to consider all reasonable educational alternatives’ to achieve the right. See Ermelo supra at para 53. In order to determine what alternatives are to be employed, ‘the State must take into account what is fair, feasible and satisfies the need to remedy the results of past racially discriminatory laws and practices.” See Ermelo supra.

REASONABLY PRACTICABLE AND REASONABLE EDUCATIONAL ALTERNATIVES

[79] Indeed the most important finding in the Supreme Court of Appeal’s judgment in UFS v Afriforum supra remains the recognition that what is reasonable includes an assessment of the factors in the second part of Section 29 (2). It is needless perhaps to mention that the SCA was confronted with two competing interpretations of the term ‘reasonably practicable’. Afriforum argued that it was limited to logistical factors. UFS, on the other hand, contended that it included an assessment of other, substantive factors, including the constitutional need to promote integration. The SCA sided squarely with UFS and held as follows:

‘The legal standard is reasonableness, which of necessity involves a consideration of constitutional norms, including equity, redress, desegregation and non-racialism. The factual criterion is practicability, which is concerned with resource constraints and the feasibility of adopting a particular language policy.

It follows, in my view, that even if a language policy is practical because there are no resource constraints to its implementation, it may not be reasonable to implement because it offends constitutional norms. The policy would therefore not meet the reasonably practicable standard.’

See UFS v Afriforum supra paras 26-7.

[80] It is therefore settled law that the assessment of what is reasonably practicable requires a consideration both of resource constraints and logistics (the factual criterion), and what is reasonable which clearly includes considerations of equity, redress, and non-racialism (the constitutional criterion). In UFS v Afriforum supra, the SCA concluded that UFS’s decision to almost completely end tuition in Afrikaans was a rational executive decision because Afrikaans tuition was no longer reasonably practicable. Therefore, the UFS had considered, inter alia, the following: (a) The ‘ever-increasing numbers of black students opting for English-medium language instruction, and correspondingly fewer numbers of white Afrikaans students seeking Afrikaans- medium instruction’; (b) The resulting racial segregation; and (c) The perception that Afrikaans (White) students received closer supervision than English (Black) students because they were in smaller classes.

[81] The above consideration ‘led UFS to conclude that the continuation of the 2003 policy is not only reasonably practicable, but absolutely impossible’. Because that conclusion was reached with ‘the support of the overwhelming majority of the University community’ and ‘after proper research, debate and deliberation’ a court ‘should be slow to interfere with [it] on review.’ See UFS v Afriforum judgment supra, para 29. Additionally, the SCA held, UFS’s policy was ‘carefully calibrated’. It allowed currently-registered students to complete their studies in Afrikaans, would be piloted in only three faculties, kept Afrikaans for tutorials (particularly for first year students), and kept Afrikaans in specific faculties. UFS, therefore, had accordingly correctly construed Section 29 (2), and had ‘been exemplary in the manner it approached the decision to reconsider the 2003 policy and adopt a new policy’. See para 30 of UFS v Afriforum supra. The same analysis was adopted in Afriforum v University of Pretoria supra. There as well, both the Senate and the Council had considered a range of factors, but supported the new policy on the basis that integration was the most important factor. Kollapen J (writing for the full court) who seemed to be applying the ordinary test for reasonableness rather than rationality – held that:

‘both Senate and Council applied their minds to a number of relevant and often competing considerations and properly considered what was before them. The weight that they afforded to the different considerations that were before them is not a matter for the Court to prescribe. In any event it hardly appears that the considerations that occupied them were neither cogent nor relevant to the determination of what the law required of them.’

He concluded that UP had been responsive to the constitutional rights of Afrikaans students, but that ‘[b]eing responsive can hardly equate to having to positively respond to the request made’. The UP was held to have considered its language policy at ‘a high level of engagement, thoroughness and transparency and the ultimate conclusion that it would not be reasonably practicable was reached after a proper consideration of all the necessary and relevant factors in a context-sensitive understanding within which the claim was located.’ See para 30 of University of Pretoria judgment supra. Mr Muller pointed out that the prohibitive cost of full parallel-medium lectures, and the incorporation of the ‘reasonably practicable’ standard into the determination of when to offer parallel-medium or simultaneous translation, means that the Policy must stand even on this more restrictive approach. The latter is an aspect still to be considered infra. Having had regard to what the applicants contend, it perhaps suffices merely to state that the UFS v Afriforum supra is and remains the governing precedent and that its approach to what is reasonably practicable makes it clear that what is logistically possible, but constitutionally offensive is not reasonably practicable.

[82] As soon as the right has been established to exist, a university need only consider the reasonable education alternatives only if education in the language of choice is ‘reasonably practicable’ as properly understood. I mention, however, that neither the SCA in UFS v Afriforum supra, nor the High Court in Afriforum v University of Pretoria supra, reached this question because both courts held that continued parallel-medium tuition was not reasonably practicable. Mr Muller pointed out that it was rational for SU to conclude that Afrikaans tuition beyond what is offered in the Policy is not reasonably practicable. This, is an aspect which I consider later in this judgment. The High Court in UFS v Afriforum supra held that a single medium institution could only be preferred to another reasonably practicable option that would provide mother tongue education ‘if it is more likely to advance or satisfy the three listed criteria of equity, practicability and historical redress’. See para 28 of the University of the Free State v Afriforum and Another (SCA) supra: (a) From the perspective of equity, it held that doing away with Afrikaans would not affect Black (African) students because the vast majority are neither English nor Afrikaans speaking. It would violate Afrikaans-speakers’ Section 29 (2) negative right. (b) There being no suggestion that the previous, parallel medium position was impracticable, UFS could not claim its new policy was justified on that ground. (c) On the question of redress, the Court held that this factor ‘weighs strongly in favour of ensuring that language is not a barrier to access for Black (African), Coloured and Indian students’. The High Court’s general approach to Section 29 (2) was overruled by the SCA. It is of cardinal importance to mention that given the clear inter-connectedness of the two parts of the test, it is highly likely that the SCA may have adopted an approach that leaned more in favour of access and integration, and in the process, gave universities more leeway to determine their own language policies. Concluding on this aspect, Mr Muller contended that even adopting the High Court’s approach to the matter, when these factors are applied to SU, it is plain that the 2016 Policy is a ‘reasonable educational alternative’ that fully complies with the obligations of Section 29 (2) of the Constitution. In the answering papers it is demonstrated that SU accepts that it was functionally possible to continue to provide tuition in terms of the 2014 Policy. It had the necessary infrastructure, staff and monetary resources to do so. But that was the case for both the UFS and UP when their policies were challenged in court. However, doing so would not have been consistent with the constitutional criterion.

[83] The 2014 Policy referred to above was but a brave attempt to move away from SU’s past and to recognise the equal status of English and Afrikaans. That much is demonstrated in the answering papers and remain undisputed. The answering papers explain that the 2014 Policy was adopted because of the changing demographics of SU’s student body and the increasing demand for English. However, despite the intent to make SU equally accessible to all, that did not apparently eventuate during the implementation of the 2014 Policy. The unintended consequence became that the 2014 Policy served to exclude Black (African) students from full and equitable access to SU. That resulted from a combination of the linguistic and racial demographics of SU’s student body, and the manner in which that policy was implemented. Statistically 63% of the 539 first years without Grade 12 Afrikaans were Black (African); 61% of all Black (African) first years did not have Grade 12 Afrikaans. Additionally 82.7% of the Afrikaans-speaking students were White; and only 17% of the Afrikaans-speaking students were Coloured, while 62% of such Coloured students were English-speaking. The statistics appear from the figures in the Breitenbach and Bishop opinion and were incorporated by reference into the answering affidavit.

[84] As we gather from the answering papers (not disputed in reply), the majority of African students could not learn in Afrikaans. The 2014 Policy as shown earlier, adopted various language specifications or options. How it functioned and how it became necessary to replace it with the impugned 2016 Policy has been dealt with supra. In passing one may mention that there were significant complaints by both the Student Representative Council and Open Stellenbosch about implementation and how simultaneous translation was provided, lecturers were said to be unable or unwilling to lecture in both languages and it being alleged that sometimes they ended up teaching almost entirely in one language. The T-option in 2014 Policy is described to have been more burdensome for the English-speaking students who could not understand Afrikaans, than for Afrikaans students who were sufficiently proficient in Afrikaans. Students complained labelling the interpretation as often of poor quality. The simultaneous translation was primarily used to translate from Afrikaans to English. Thus in 2015 and the first half of 2016 it became clear to SU that the 2014 Policy discriminated directly against English speakers, and indirectly against Black (African) students. As explained in the answering papers, it was easier for White students to understand lectures than Black (African) students. Of course this created a serious burden for Black (African) students to access further education. This burden was not experienced by their white counterparts.

RETROGRESSIVE MEASURES

[85] One must acknowledge that the general rule against retrogressive measures in socio-economic rights in both our law and the international law as contended by Mr Heunis, also applies in the context of Section 29 (2) of the Constitution. The truth is that since the judgment in Ex Parte Chairperson of the Constitutional Assembly: In re Certification of the Constitution of the Republic of South Africa 1996, 1996 (4) 744 (CC) (‘the Certification Judgment’), the Constitutional Court has repeatedly recognised that socio-economic rights include a duty not to take away or diminish existing access. The meaning of the negative right in Section 29 (2) was (as alluded to earlier) pertinently been addressed by the Constitutional Court in Ermelo supra where the Court held, inter alia: ‘[W]hen a learner already enjoys the benefit of being taught in an official language of choice the State bears the negative duty not to take away or diminish the right without appropriate justification.’ Mr Heunis has of course referred to this authority. However, the decisions in UFS v Afriforum and Afriforum v University of Pretoria make it abundantly clear that the negative element of the right does not substantially alter the inquiry. In UFS v Afriforum, supra, Cachalia JA indicated that he was aware of the above statement from Ermelo case, but held as follows:

‘But this does not mean that once the right exists it continues, regardless of whether the context and the circumstances have changed. A change in circumstances may materially bear on the question whether it is reasonably practicable to continue with a policy. What is required of a decision-maker, when there is a change in circumstances, is to demonstrate that it has good reason to change the policy. In other words, it must act rationally and not arbitrarily.’

The same point was made by the full bench in Afriforum v University of Pretoria supra as follows:

‘I do not understand that there are two tests of application here, one located in section 29(2) in so far as it relates to a request for education in the language of choice, and the other that applies to instances where the right is already enjoyed. While it is and must be so when the State seeks to take the right away or diminish it, as is the case with the introduction of the 2016 language policy, there has to be sufficient justification. Such justification in this instance is to be found in the successful activation of the test of reasonable practicability found in Section 29(2). To suggest a different or a more onerous justification would have the effect of impermissibly entrenching language rights.’

See para 54 of Pretoria judgment supra.

[86] An important point to make is that neither the SCA nor the North Gauteng full bench considered a separate test of ‘appropriate justification’ in finding that the universities in those cases had complied with their obligations under Section 29 (2). The point though is that if SU can show that retaining the 2014 Policy was not reasonably practicable, or that the impugned 2016 Policy is a reasonable educational alternative, then it has acted constitutionally. One need not shy away from mentioning that the fact that Afrikaans speakers previously enjoyed greater rights does not entitle them to retain those rights where it is not reasonable for them to do so. In my view, the applicants are wrong in stating that Section 29 (2) means that SU ‘may not take away or diminish the rights of Afrikaans-speakers to receive education in Afrikaans in order to increase the English offering.’ Plainly SU may do so provided that increasing the English offering is otherwise consistent with Section 29 (2) because, for example, it is necessary to ensure all students can have equitable access to SU. Mr Muller is correct in contending that whether or not there is any reduction in the Afrikaans offering is a question of implementation and is not a necessary consequence of the impugned Policy. The truth is that the implementation of the Policy is not before us. If one assesses the 2016 Policy holistically, one finds or comes to the realisation that it is proportional to the goals it seeks to achieve. I can think of no better and carefully crafted policy. I am of the view that SU has indeed advanced an ‘appropriate justification’ for any possible reduction in Afrikaans tuition that flows inevitably from the Policy.

THE NATURE OF THE OBLIGATION IMPOSED BY SECTION 29 (2) AND THE RELIANCE ON FOREIGN AND COMPARATIVE LAW

[87] The Supreme Court of Appeal in Mikro supra removed any conceivable doubt as to whose primary obligations are contemplated in Section 29 (2) of the Constitution. The primary obligations imposed by Section 29 (2) rest solely on the State and not on individual universities. The Supreme Court of Appeal speaking to this aspect in Mikro supra stated it categorically as follows:

‘everyone has a right to be educated in an official language of his or her choice at a public educational institution to be provided by the State if reasonably practicable, but not the right to be so instructed at each and every public educational institution subject only to it being reasonably practicable to do so.’

Almost similarly, the full bench in the University of Pretoria supra put it thus:

‘Clearly the claim to the right must be located and adjudicated upon within the context of the education system as a whole and the resources and other means that exists within it, as opposed to the confines of any single public educational institution at any given time where such a claim may arise’.

[88] That is essentially the reason why the SCA in UFS v Afriforum supra held that the more difficult questions had to be addressed by a constitutional challenge to ‘the State’s language policy’, not a review of a single university’s language policy. See UFS v Afriforum supra para 31. This certainly must be correct. Students cannot go to any university and demand tuition in the language of their choice where they can access education in the same language at another university nearby. The aforegoing must not, however, be understood to mean that SU has no obligations at all arising from Section 29 (2) and can simply defer the problem to the national government. SU is and remains an organ of state and as such is a vehicle through which ‘the State’ provides higher education. SU is required to comply with Section 29 (2) when it adopts a language policy. But it is important to emphasise that it is not SU’s to ensure that Afrikaans students across the country have access to Afrikaans tuition. SU’s obligation is limited to providing Afrikaans education where reasonably practicable and through reasonable educational alternatives. Its obligation in this regard must necessarily be based on its own existing and predicted student base, its own financial resources, and its own commitments to equity and redress. It clearly cannot be the case that the lawful and constitutional decisions by UFS, UP and other universities to stop teaching in Afrikaans impose a greater obligation on SU to continue teaching in Afrikaans. The point is simply that each individual university must be assessed individually based on its own peculiar and particular facts and circumstances.

[89] It has been indicated above that the applicants’ place reliance on Canadian law and international law. Section 39 (1) (c) of the Constitution is authority for the proposition that this court may consider foreign law and that it must consider international law. However, I am still to be persuaded that the Canadian and international authorities are particularly helpful in interpreting Section 29 (2) of our Constitution. Our courts have already conclusively interpreted Section 29 (2) in this precise context. I have reservation about the necessity and appropriateness for this Court to turn to foreign or comparative law to seek to second-guess the interpretations of the Constitutional Court, Supreme Court of Appeal and the full bench in Pretoria. In any event it is to be noted that the Section of Canadian law was also referred to in the opinion of Breitenbach SC and Bishop. One must perhaps paraphrase Advocates Breitenbach and Bishop’s opinion: (a) The Constitutional Court has warned repeatedly about the danger of borrowing uncritically from comparative jurisprudence. See for example Bernstein and Others v Bester and Others NNO 1996 (2) SA 751 (CC) at para 133; Alexkor Ltd and Another v Richtersveld Community and Others 2004 (5) 460 (CC) at para 33. More specifically, it has been held:

‘The special provisions of Section 23 of the Charter makes it a unique set of constitutional provisions quite peculiar to Canada’.

See Gauteng Provincial Legislature In re: Gauteng School Education Bill 1996 (3) SA 165 (CC) at para 15, quoting Attorney-General of Quebec v Quebec Association of Protestant School Board et al [1984] 10 DLR 321 (S.C.C.) at 331. That statement itself must be read with some caution as the Court was dealing with Section 32 (c) of the Interim Constitution, the equivalent of Section 29 (3). It noted that Section 23 was indeed analogous to Section 23 (b) of the Interim Constitution, the equivalent of Section 29 (2). (b) There are a number of important differences between the Canadian protection of minority language rights, and the right to education in a language of choice in Section 29 (2). The Canadian language rights are part of Canada’s unique constitutional history and settlement. Hence the focus on provinces in which French or English speakers are in the minority. A mention must also be made that the nature of discrimination and historical disadvantage against linguistic minorities while somewhat present in Canada, is most certainly vastly different from South Africa’s Apartheid past which imposed Afrikaans on Black learners. (c) The Canadian right belongs to the parents, not the children. See Association des parents de l’ècole Rose-des-vents v British Columbia (Education) [2015] 2 SCR 139 at paras 34-5. That alone highlights its importance as a right aimed at the preservation of linguistic and cultural communities, rather than as a right directed to facilitating access to education. Our Section 29 (2) serves both purposes, neither of which predominates. (d) Section 23 of the Charter (unlike Section 29 (2) of the Constitution) is expressly limited to primary and secondary education. As shown in the answering papers, the concerns of assimilation that animate the decisions of the Canadian courts seem less weighty in the context of mother tongue education. The same hold true for the issues of practicability and access. The fact is there will always be fewer universities and colleges than schools.

NOT REASONABLY PRACTICABLE AND ABSENCE OF REASONABLE EDUCATIONAL ALTERNATIVE

[90] On the approach adopted in UFS v Afriforum and University of Pretoria supra, continuing with that position was not reasonably practicable because it was inconsistent with the demands in Section 29 (2) for equity and redress. In the current case, it could not be equitable for the majority of Black (African) students to be denied equal access to SU. It was plainly inconsistent with SU’s obligation to provide redress to continue to exclude Black (African) students in that way. Undoubtedly, it was only reasonably practicable for SU to offer Afrikaans tuition to the extent it could do so without excluding English-speakers. One may therefore competently argue that Afrikaans-speakers had no Section 29 (2) right to demand Afrikaans tuition beyond what SU was reasonably able to provide without excluding English speakers. What the 2016 Policy does (in practical terms) is that it provides as much Afrikaans tuition as is reasonably practicable without excluding the majority of Black (African) students. It grants some measure of preference to English in limited ways connected to the available resources: (a) Parallel-medium teaching is the first option. According to the impugned 2016 Policy the latter option must be used when it is ‘reasonably practicable and pedagogically sound’. This will depend on the size of classes and the availability of lecturers and classrooms. Other learning opportunities are provided jointly in order to avoid the difficulties associated to lack of integration. (b) Dual medium teaching is used when it is either not reasonably practicable or not pedagogically sound. In this category while not all information will be provided in Afrikaans by the lecturer, the important information is, and translation is provided in all first year modules, and where resources allow, in other years. See para 7.1.4 of 2016 Policy. (c) Single medium English teaching is only permitted where the subject matter requires it, or where no lecturer is available to teach in Afrikaans. Even in this category SU provides translation for all first-year modules, and for other modules if SU has the resources to do so. It remains very difficult to conceive how SU could have crafted a language policy that offered greater Afrikaans tuition, without excluding Black (African) students.

[91] The apparent difficulty one encounters is that the applicants in the founding papers do not provide any reasonable basis as to how the Policy should be amended to prevent exclusion while providing for more Afrikaans tuition. All they suggest is that SU should have adopted a fully parallel-medium solution, or offer increased interpretation services. As is apparent from the answering papers, these are not realistic solutions to the challenges SU faced. The answering papers reveal that SU cannot afford full parallel medium, demonstrated by the study that was conducted. We are told it would cost SU R640 million to create the additional infrastructure, R78 million per year to employ the additional staff. That would translate to a 20% increase, or at current fee rates, R8 100 per year increase in average fees. SU states that this would not be financially feasible. This is an aspect criticised at some length by the applicants. But the fact of the matter is that the relevance of the study is merely to show that moving to full parallel medium tuition was not reasonably practicable for the same reason advanced in UFS v Afriforum supra - it would result in a segregated campus. In fact, for the reasons pertinently identified in UFS v Afriforum and University of Pretoria, it may well be constitutionally undesirable to move to a full parallel-medium policy. It would result in two separate streams of students who would have limited interaction. Although the 2016 Policy continues to use some parallel-medium lectures, this is necessitated by the fact that SU’s student body remains almost 50% Afrikaans and 50% English.

[92] The answering papers explain that one of the reasons SU wishes also to continue using both languages in a single lecture is to prevent segregation and promote multilingualism. The provision of interpretation services for all lectures is not necessarily a viable option in that: (a) it would be too expensive and SU may not have the resources to do so; (b) it is not a long term solution because interpretation will never be as good as a lecturer in the original language. Although the applicants maintain that no students were denied access under the 2014 Policy, it si apparent that the 2014 Policy led to significant discrimination against Black (African) students who were unable to understand Afrikaans. One must bear in mind that Afriforum repeatedly sought to prevent SU from increasing the English offering during the very 2016 when SU sought to address the discrimination which had arisen in this regard. In conclusion there is sufficient evidence testifying to the fact that it was not reasonably practicable for SU to continue with the 2014 Policy. It clearly would not be reasonably practicable to offer more Afrikaans tuition than what is currently provided for in the 2016 Policy.

[93] The applicants’s case on the constitutional question referred to supra boils down to a series of complaints about the 2016 Policy, and primarily the implementation thereof. By way of a summary, the applicants essentially contend that: (a) SU failed to consider the broader context of its decision; (b) SU has exploited Afrikaans-speaking students’ bilingualism; and (c) for various reasons, the implementation of the Policy will result in a substantial reduction in Afrikaans offering. The most appropriate way of evaluating these complaints, is to consider them as subtopics (some of which are combined).

THE BROADER CONTEXT OF THE DECISION AND THE EXPLOITATION OF AFRIKAANS STUDENTS

[94] The applicants argue that SU’s constitutional obligations to Afrikaans-speakering students have been more powerful due to two factors, namely: (a) several other universities cancelling or diminishing the extent of Afrikaans tuition at the same time; and (b) If SU decreased its Afrikaans offering, that would reduce demand for Afrikaans education at primary and secondary level which would have severe effects on Afrikaans culture. The aforegoing factors are clearly beyond SU’s direct control. Perhaps it is necessary to state categorically that the aforegoing factors do not (in the least) affect the legality of the 2016 Policy or the Decisions. SU as an institution cannot be said to be responsible for the provision of Afrikaans tuition throughout the Country. I have demonstrated earlier in this judgment that the obligation rests on the State as a whole and not individual universities. A point that must be made also is that SU’s obligations under Section 29 (2) cannot be determined by the decisions taken by other universities. I emphasise that SU is of course obligated to merely act rationally and consistently with Section 29 (2) based on its own particular circumstances. Perhaps the applicants must take the issue of what is perceived to be a countrywide pattern (that Afrikaans is being systematically phased out of higher education), up with the National government. The point is that SU continues to teach in Afrikaans and the impugned 2016 Policy requires it to maintain or increase Afrikaans offering. Countrywide occurrences concerning Afrikaans usage do not result from SU’s actions.

[95] The SCA in UFS v Afriforum supra, held that the Senate and the Council must merely have acted rationally. This Court cannot second-guess the way the SCA weighed various factors. The 2016 Policy is clearly rationally related to SU’s goals of ensuring equitable access, promoting multilingualism, and providing as much Afrikaans tuition as possible, all within its available resources. SU, which continues to provide Afrikaans tuition cannot conceivably, in my view, be found to have acted unconstitutionally because other universities have lawfully abolished Afrikaans tuition altogether. It would be strange to single out a university that continues to provide the most support for Afrikaans. It adopted the 2016 Policy in order to ensure that students had full and equal access to higher education.

[96] The applicants accuse SU of cynically ‘exploiting’ or ‘capitalising’ on the bilingualism of Afrikaans students to reduce Afrikaans tuition. It is true that SU justifies the 2016 Policy on the basis that Afrikaans students are universally bilingual whereas the majority of Black (African) students are not. The reality is that Afrikaans students will suffer less harm being required to learn in English than English students will incur if they have to learn in Afrikaans. If SU did not take account of the aforegoing, it would have acted irrationally. The complaint about capitalising on bilingualism is inconsistent with the argument that SU is seeking equal disadvantage for all students. A similar argument surfaced before the Constitutional Court in the context of same-sex marriage in Minister of Home Affairs v Fourie supra.:

The submission by the applicants is that because English is already the second language for many Black (African) students, SU seeks to justify its policy on the basis that Afrikaans students should suffer equal disadvantage by being required to learn in their second language. In my understanding of the answering papers, SU had to choose between: (a) requiring the majority of Black (African) students to attend lectures in a language they could not understand at all; or (b) requiring Afrikaans students to attend lectures in a language they can understand, but is not their mother tongue. That certainly is substantive equality that takes account of South Africa’s history and the need to promote equitable access for all students. In the University of Pretoria case supra, the full bench made the following observation:

‘It is true that instruction in English only may be to the advantage of a small number of mainly white English speakers. All students will, to some extent, be disadvantaged ‘equally’. This is sometimes referred to as the ‘graveyard’ option, often illustrated by the analogy of a swimming pool being closed because black people are denied access to it, resulting in the area being left without a swimming pool. Whether an unlawfully discriminating swimming pool is better than no pool is a choice to be made.’

See para 51 of the Pretoria matter. Similarly, if the choice is between some equal disadvantage for all, and disadvantage based on race, then SU must be held to have been justified in choosing the former.

THE IMPLEMENTATION OF THE 2016 POLICY AND THE ASSOCIATED COMPLAINTS

[97] There are various complaints about how the 2016 Policy will be implemented. The applicants claim, inter alia, that the implementation result in a significant decrease in the extent of the Afrikaans offering. However, the Decisions and the 2016 Policy are either valid or invalid based on the manner in which they were taken, and the content, and not how they are implemented. What is before Court is not a challenge about improper implementation. That is not a case the respondents are called upon to meet. I accept that concerns about implementation could only possibly be relevant if the evidence established that it was intended or inevitable that the Policy would have certain results, notwithstanding the internal accountability mechanisms designed to ensure proper implementation. The applicants associated arguments advanced are discussed hereunder.

‘The applicants complain that the Faculties and Management have too much power’

[98] The nature of the above complaint is that the Policy under attack delegates too much power to faculties and management to determine what language option will be used in each module. Para 7.4.3 of the Policy does require faculties to determine their language specification for each module. That is subject to the approval by Senate. Additionally, the Policy makes provision for: (a) annual agreements between the Vice-Rector: Learning and Teaching and the deans of the faculties on mechanisms to ensure accountability for the implementation of the policy, including the faculties’ annual reports on the realisation of the implementation of their language implementation plans and the compliance reports which the faculties must submit after each semester; and (b) an annual report by the Vice-Rector: Learning and Teaching to the Council, via the Rector’s Management Team (‘RMT’) and Senate, on the accountability mechanisms agreed with the deans for the next ensuing year. See Policy para 7.4.1 read with paras 8.1 and 8.3.

[99] A reference to SU’s institutional statute reveals the following: (a) ‘The day-to-day management of the University is the responsibility of the Rector’s Management Team collectively or individually’. (b) Faculty boards are responsible for submitting recommendations to Senate on academic programmes, which must include the language of instruction. (c) Council, by contrast, ‘exercises a general supervisory responsibility in respect of academic and operational matters and institutional policy and strategy’. (d) Senate has very specific obligations that do not include determining language specifications. See generally Statute of Stellenbosch University serving as Annexure “GMS4” in the answering papers. In passing I mention that the respondents contend that the structure of decision-making adopted in the Policy is consistent with SU’s institutional statute. In any event, common sense and logic dictate that Faculties know what resources they have and how best to use them. They know which lecturers are competent and available to teach which courses. The decisions they take are then inevitably subject to appropriate oversight by Senate and Council to prevent any abuses that are inconsistent with the purpose of the Policy. That is how universities function.

[100] I am not persuaded that (as to the delegation linkage) it has been shown that Management and the Faculties are biased against offering Afrikaans tuition. The applicants complain that, in their contention, the Policy does not have adequately measurable goals. It is true that the 2016 Policy sets no strict requirements of, for example, the percentage of modules that must be offered in parallel medium, dual medium, or the percentage of second and third years classes that must have simultaneous translation. Mr Muller pointed that, this was a deliberate policy choice. He conceded that the advantage of strict measures is that they provide clarity. But, he elucidated, the disadvantage is that they may unintentionally inhibit the achievement of the policy’s real goals, be unrealistic and unachievable, or be unresponsive to changing circumstances. This is fairly clear. I understand why SU opted for a different form of accountability that rests ultimate authority with the Senate.

[101] The different form of accountability opted for operates in the following manner as gathered from the Policy itself: (a) Each faculty must annually prepare a Language Implementation Plan (LIP) and submit it to Senate. The LIP describes how the faculty will implement the Policy, including which modules will be offered in parallel, dual and single medium. Senate has the power to either accept the LIP, or refer it back to the faculty (para 7.4.3). (i) If changes are made to the LIP outside of the annual review (for example because a lecturer becomes unavailable) those must be reported to Senate and to the faculty board (para 7.4.4). (ii) Each faculty is required to report to the Vice-Rector: Learning and Teaching after each semester on its compliance with the Senate-approved LIP. The faculty must identify and describe each instance of non-compliance with the LIP, “the reasons for it fully and the steps the faculty is or will be taking to avoid future deviations from the [LIP]” (para 8.3). (iii) Each faculty reports annually to the Vice-Rector about any difficulties in implementing the Policy, any mechanisms to improve implementation, and any suggestions for amendments to the Policy (para 8.1). (iv) The Vice-Rector must agree further accountability mechanisms with the deans of the faculties and must report on those mechanisms to the Council, the Senate and the Rectors Management Team. Those accountability mechanisms must have due regard to the principles in paragraph 6, and the following two principles (para 7.4.1):

‘The English offering is revised upwards so as to achieve full accessibility to SU for academically deserving prospective and current students who prefer to study in English.

The Afrikaans offering is managed so as to sustain access to SU for students who prefer to study in Afrikaans and to further develop Afrikaans as a language of tuition where reasonably practicable.’

I agree that these accountability mechanisms must necessarily provide SU with the necessary flexibility to manage the implementation of the Policy in the light of changing circumstances, but within defined principled boundaries. In my finding, it was a reasonable policy choice for SU to make to achieve accountability in this way, rather than through numerical targets.

[102] Another complaint is that the possibility for lectures to be offered in Afrikaans alone is illusory because there are only 8 lecturers who are proficient only in Afrikaans, whereas there are 202 lecturers who are only able to lecture in English. The answering papers accept the aforementioned factual basis as accurate. There are, however, respects that water-down the applicants’ argument in this regard. These are set out as follows both in the answering papers and the contentions advanced on behalf of the respondents:

‘The result of the disparity in lecturers is that there will be more modules offered in parallel-medium or in English than in Afrikaans alone. The Applicants can have no complaint about an increase in parallel-medium lectures. Nor can there be a meaningful complaint about English lectures where it is not possible for SU to offer the module in dual medium (because a lecturer is not available) or parallel medium (because the size does not justify the additional expense). That was the position even under the 2014 Policy.

The real complaint is that, in second and further years, Afrikaans lectures will automatically be translated, whereas English lectures will only be translated “upon request by a faculty, if the needs of the students warrant the service and SU has the resources to provide it”. (para 7.1.5.2(a)). That minor relative burden is justified by the reality of Afrikaans bilingualism, the limited ability of a majority of Black students to learn in Afrikaans, SU’s commitment to equitable access, and the limited available resources.

The extent of the burden will depend on how often requests for translation are granted. There is no evidence on this score and so this court must accept that they will be granted whenever it is reasonable to do so. On that basis, it is impossible to declare that the Policy is unlawful merely because there are more English-only lecturers.’

I cannot fault the aforegoing. In addition to a claim based on s 29(2), the Applicants argue that the Policy constitutes direct unfair discrimination against Afrikaans-speaking students; and indirect unfair discrimination against White and Coloured students, belongs to a different forum and not this Court. The Constitutional Court has made it clear that complaints that the policies or conduct of organs of state violate the right not to be unfairly discriminated against must be brought under the Equality Act, not under the Constitution. See MEC for Education: Kwazulu-Natal and Others v Pillay 2008 (1) SA 474 (CC) at para 40. See also My Vote Counts NPC v Speaker of the National Assembly and Others 2016 (1) SA 132 (CC) at para 57; De Lange v Presiding Bishop of the Methodist Church of Southern Africa for the time being and Another 2016 (2) SA 1 (CC) at para 53. This is of course a result of the operation of the principle of subsidiary. The same point was correctly made in the opinion of Breitenbach SC and Bishop.

[103] Litigants must bear in mind that where an applicant has multiple claims, one of which is an equality claim, the correct procedure is to launch the case in both the High Court and the Equality Court and then seek to have the matter heard by the same judge, sitting as a judge of both the Equality Court and the High Court. See De Lange (n 201) at para 46 above, citing Manong & Associates (Pty) Ltd v Department of Roads and Transport Eastern Cape and Others (No 2) 2009 (6) SA 589 (SCA) at paras 54 and 57; Manong & Associates (Pty) Ltd v Department of Roads and Transport, Eastern Cape, and Another (No 1) 2009 (6) SA 574 (SCA) at paras 30-1; and Minister of Environmental Affairs and Tourism v George and Others 2007 (3) SA 62 (SCA) at paras 12-3.

Simply put, absent an application in the Equality Court, a High Court judge has no power to hear complaints in terms of the Equality Act. That is why in De Lange supra, the Constitutional Court refused to hear an unfair discrimination claim that had been raised only in the High Court. See De Lange judgment at para 59. The Applicants have launched this application exclusively in the High Court. This Court lacks jurisdiction to hear the complaint. I demonstrate infra that even if this court had the necessary jurisdiction to entertain this complaint, the applicants would still be faced with difficulties. I say so because even if the limited preference granted to English under the 2016 Policy constitutes discrimination on the basis of language, it does not constitute indirect discrimination on the basis of race. In order to establish that claim the applicants would have to demonstrate, at least prima facie that the Policy has a disproportionate impact on White or Coloured students. That is, for example, is said to have occurred under the 2014 Policy which imposed a disproportionate burden on Black (African) students because the majority of them could not speak Afrikaans at all. The majority of all students who could not speak Afrikaans were Black (African). A consideration of the relevant demographic information demonstrate that the same cannot be said to be true of the 2016 Policy.

[104] The test for fairness is set out in Section 14 (2) of the Equality Act. The latter Section requires a Court to consider the context and the following factors:

‘(3) The factors referred to in subsection (2) (b) include the following:

(a) Whether the discrimination impairs or is likely to impair human dignity;

(b) the impact or likely impact of the discrimination on the complainant;

(c) the position of the complainant in society and whether he or she suffers from patterns of disadvantage or belongs to a group that suffers from such patterns of disadvantage;

(d) the nature and extent of the discrimination;

(e) whether the discrimination is systemic in nature;

(f) whether the discrimination has a legitimate purpose;

(g) whether and to what extent the discrimination achieves its purpose;

(h) whether there are less restrictive and less disadvantageous means to achieve the purpose;

(i) whether and to what extent the respondent has taken such steps as being reasonable in the circumstances to-

(i) address the disadvantage which arises from or is related to one or more of the prohibited grounds; or

(ii) accommodate diversity.’

[105] If one considers the aforementioned factors properly, one is compelled to conclude that the 2016 Policy is fair in that: (a) SU accepts that access to tuition in the language of one’s choice has a connection to human dignity and that (assuming there is discrimination at all) it will have some negative impact on Afrikaans speakers. At the risk of being repetitive, SU still offers tuition in Afrikaans and accordingly any conceivable impairment of dignity is minimal. Importantly, the majority of Afrikaans speakers are able to learn in English, particularly with the additional assistance offered by SU in the first year of study. (b) Those who are disadvantaged are primarily White Afrikaans speakers. The truth is that they generally occupy an historic and current position of privilege in society. Certainly, that weighs in favour of fairness. (c) The discrimination is thus limited. It is systemic in the sense that it flows from a policy, but there is systematic discrimination against White Afrikaans people generally. (d) The 2016 Policy serves the legitimate purpose of ensuring equitable access to SU, and ensuring that Black (African) students are not prevented from learning. By ensuring that all information will be available in English, it achieves that purpose. (e) There (as explained earlier) are no less restrictive means to achieve that purpose within SU’s available resources. The reason is because the Policy requires the maximum possible Afrikaans offering within SU’s resources. (f) The Policy promotes (as it were) diversity in that it makes SU an attractive and accommodating space for all students, regardless of race. Accordingly, any conceivable discrimination is most certainly fair.

THE RIGHT OF ACCESS TO FURTHER EDUCATION (S 29 (1)( b) OF THE CONSTITUTION), INDIGENOUS LANGUAGE (S 6 (2) OF THE CONSTITUTION AND THE LPHE

[106] Some of the above topics have already been touched on. But their importance and the pivotal role they have been employed to play by the applicants have earned them some passing observations. Section 29 (1) (b) of the Constitution grants everyone the ‘right to further education, which the State, through reasonable measures, must make progressively available and accessible.’ Unlike the right to basic education in Section 29 (1) (a) which is ‘immediately realisable’, the right to further education is progressively realisable and subject to reasonable measures. See in this regard Governing Body of the Juma Musjid Primary School & Others v Essay N.O. and Others 2011 (8) BCLR 761 (CC) at para 37. Maybe one needs to encupulate the applicant’s argument in this regard. They argue that the 2016 Policy violates Section 29 (1) (b) because ‘it will exclude, alternatively impede access to the SU by Afrikaans speaking learners’. The argument continues, and it says that, this is because Afrikaans speakers who cannot obtain Afrikaans tuition at SU will not study there, and if they cannot study in Afrikaans elsewhere, will forego tertiary education altogether. I fail to comprehend the above argument. The point is in terms of the 2016 Policy, first year students can learn entirely in Afrikaans. In later years, Afrikaans is used as much as students’ needs demand, within SU’s resources. Unlike Black (African) students who cannot learn in Afrikaans, Afrikaans-speaking students can learn in English. Certainly, any barrier to access is rather minimal, considering the measures taken to teach in Afrikaans in the first year. There is no evidence (non preferred by the applicants) that potential Afrikaans-speaking students will rather not attend university than attend SU if they must attend under the 2016 Policy rather than the 2014 Policy. I hold that there is no violation of Section 29 (1) (b) of the Constitution.

[107] The Applicants appear to launch a separate constitutional attack based on Section 6 (2) of the Constitution. Section 6 (1) of the Constitution identifies the 11 official languages of South Africa. Section 6 (2) provides: ‘Recognising the historically diminished use and status of the indigenous languages of our people, the State must take practical and positive measures to elevate the status and advance the use of these languages.’ The applicants complain that SU failed to appreciate that Afrikaans was an indigenous language contemplated in Section 6 (2) and that SU was obliged to take practical and positive measures to elevate the status and advance the use of Afrikaans. In their contention, the 2016 Policy is inconsistent with that obligation. Nobody can deny the fact that Afrikaans is certainly a South African and African language. I accept that it could very well be described as ‘indigenous’ in general terms. But let the truth be told, Afrikaans is not an indigenous language as contemplated in Section 6 (2) of the Constitution. This is abundantly clear from the introductory phrase that refers to the ‘historically diminished use and status of the indigenous languages’ that the Constitution is not at all referring to Afrikaans. Nobody can dispute that Afrikaans received massive State support in order to develop it as a language of scholarship and science which it is today. The massive State support was intended to compel Afrikaans use in schools and to promote its use in government and business. Of course, the languages that qualify for corrective measures under Section 6 (2) are the other African languages that received no such similar investment. See in this regard I Currie ‘Official Languages and Language Rights’ in S Woolman & M Bishop (eds) Constitutional Law of South Africa (2 ed, 2005) ch65-p15 (‘Is Afrikaans an 'indigenous language of our people'? … While its origins might qualify it for the label 'indigenous', it is unlikely that Afrikaans qualifies for the corrective measures required by FC S 6 (2). The language was the beneficiary of decades of active promotion by the National Party government and can hardly be considered 'historically diminished' in use and status’).

[108] The Constitutional Court itself seems to have recognised the above in Ermelo supra. Referring to Sachs J’s statement that reducing Afrikaans reduced the ‘patrimony of the whole’, Moseneke DCJ wrote:

‘Of course, vital parts of the “patrimony of the whole” are indigenous languages which, but for the provisions of section 6 of the Constitution, languished in obscurity and underdevelopment with the result that at high school level, none of these languages have acquired their legitimate roles as effective media of instruction and vehicles for expressing cultural identity.’

See Ermelo at para 49. Indeed the clear implication of the above is that Afrikaans is not an indigenous language as meant in Section 6 because it most certainly did not ‘languish in obscurity and underdevelopment’.

[109] The fact of the matter is that even if Afrikaans qualifies to be referred to as an indigenous language under Section 6 (2) (it does not so qualify), it is a general obligation that rests on ‘the State’, not a specific obligation on SU. Undoubtedly, the type of obligation envisaged by Section 6 (2) is for the State as a whole to take a variety of measures (over time), to promote the indigenous languages. It hardly provides a legal basis to declare a particular policy of a particular university invalid. In the latter regard Mr Muller correctly contended as follows: ‘even if the obligation applies to SU in this context, it has complied with it. SU has – over the years and in its 2016 Policy – taken “practical and positive measures to elevate the status and advance the use” of Afrikaans. It continues to invest heavily in Afrikaans teaching, scholarship and research. It continues to employ Afrikaans as one of two official languages in its events and communications.’

[110] The complaint regarding the LPHE is that SU acted contrary to the LPHE. This aspect features later in this judgment. For present purposes one needs to point out that in the first place even though the LPHE is so important, it is not binding on SU. Section 27 (2) of the Higher Education Act 101 of 1997 allows universities to determine their language policies ‘subject to the policy determined by the Minister’. The phrase ‘subject to’ is usually used to mean that the subordinate document ‘may not be inconsistent with’ the superior one. See UFS v Afriforum supra at paras 34-5. That it is not the case in this matter is beyond question. In UFS v Afriforum the SCA held that ‘the words ‘subject to’ in Section 27 (2) [of the HEA], contextually understood, do not impose a legal obligation on any university to adopt the LPHE. The LPHE goes no further than to provide a policy guideline for the universities from which they are free to depart.’ See UFS v Afriforum para 39. Clearly, the only obligation on SU was to justify any departure from the LPHE. Importantly, in UFS v Afriforum supra, the SCA held that UFS had adequately justified its departure. I have indicated that this shall be revisited later in this judgment. It suffices at this stage to merely observe that the Policy under attack is fully consistent with the LPHE. The LPHE at para 15.4 recognises the importance of Afrikaans as a language of instruction. What the LPHE, however, does not do is that it does not specify how that tuition should be provided, nor does it impose any specific duty on SU not to alter its existing Afrikaans offering.

[111] To the contrary, the LPHE’s focus is on ensuring equitable access. It stresses that the ‘continued long term maintenance, growth and development’ of Afrikaans, must be done “without non-Afrikaans speakers being unfairly denied access with the system, or the use and development of the language as a medium of instruction wittingly or unwittingly becoming the basis for racial, ethnic or cultural division or discrimination’.

[112] The LPHE requires universities to adopt ‘a range of strategies’ to both preserve and grow Afrikaans, while not excluding non-Afrikaans speakers. As stated by the High Court in Afriforum supra at para 39.

‘The [LPHE] seeks to balance, on the one hand, the need to transform higher education, and in particular to prevent institutions’ languages of instruction from impeding access and success by people who are not fully proficient in English and Afrikaans, with, on the other hand, the development of multilingualism in those institutions’ day-to-day functioning and core activities, including the development of indigenous African and other languages as scientific and academic languages. It also seeks to assure the long-term maintenance and growth of Afrikaans as a language of science and scholarship in the higher education system.’

The latter is of course consistent with what the SCA held in UFS v Afriforum supra. See also University of Pretoria at para 15. I agree with Mr Muller that the 2016 Policy is carefully calibrated to achieve the balanced approach endorsed in the LPHE.

THE ADMINISTRATIVE-LAW-BASED CHALLENGE

[113] It has been established earlier in this judgment that on authority of the Supreme Court of Appeal in UFS v Afriforum supra that neither the Decisions nor the Policy are reviewable under PAJA. As shown above they are only reviewable under the less stringent principle of legality. A point must, however, be made that even if they are reviewable under PAJA (they are not), they would withstand the scrutiny. The applicants contend that as regards the process which led to the adoption of the 2016 Policy: (a) SU Management determined the process, notwithstanding the fact that it must have been obvious to the Senate and the Council that SU Management could not adopt an objective and independent position regarding the matter; (b) a radical minority prescribed to the SU Management what the language policy should be and the SU Management did the same to the SU Council; and (c) the process was designed to create the pretence of a consultation, whereas there was no real consultation. The applicants allege that the real reasons SU adopted the Policy were the improper purposes of regularising SU’s deviations from the 2014 Policy and Plan and averting political pressure and the prospect or threats of violence.

[114] Importantly the allegation is that SU failed to take into consideration a wide range of relevant considerations, namely: (a) the Constitution and the LPHE; (b) the various legal opinions from counsel; (c) the status of Afrikaans as an indigenous language which is being marginalised by English; SU’s capacity to offer tertiary education in Afrikaans; (d) the difficulty Afrikaans-speaking students will experience when transitioning from Afrikaans-language schooling to English-language teaching at university; (e) the knock-on effect of moving from Afrikaans to English teaching at universities on Afrikaans-language schooling – school learners will switch to English to avoid being handicapped at university; (f) SU’s own survey which showed strong support for the retention of Afrikaans as a primary medium of instruction; (g) the growing demand for tertiary education in Afrikaans in the Western Cape Province and South Africa as a whole; (h) the demands of higher education at the national level, especially the fact that only one of the other 16 universities in South Africa (the North West University) still has a language policy that provides for Afrikaans as a primary medium of instruction; (i) the demographics of the Western Cape and Northern Cape Provinces; and (j) the fact that the “Coloured” community in the Western Cape Province is predominantly Afrikaans speaking and the most under-represented population group at tertiary institutions.

[115] The applicant allege that SU was biased and obstinately adhered to a pre-determined outcome, as evidenced by its reckless disregard of relevant considerations, its slavish adherence to irrelevant political considerations, its abject pandering to the threats of a militant minority, its attempts to suppress the invocation of the Constitution and its dismissal or failure to take into consideration the comments of interested parties and the Convocation about the draft language policy which supported the retention of Afrikaans as a primary language of instruction. These categories of challenges shall be dealt with infra. Having said so, I need to mention that the adjudication must be based on the assumption that the Policy is (as already found above) substantively lawful and consistent with SU’s constitutional obligations. I address the challenges documented above.

THE PROCESS FOLLOWED BY SU SUMMARISED

[116] It has been emphasised above that the only permissible procedural challenge to the Decisions is that they were procedurally irrational, i.e. there is no rational connection between the procedure adopted and their purpose. Having found that the Decisions are not administrative action, there is no requirement that they be procedurally fair. The process SU followed to adopt the Policy is comprehensively set out in the answering papers. I document hereunder a mere summary of the process SU followed.

[117] On 9 February 2016, at its first weekly meeting in the 2016 academic year, the RMT decided that a formal process for the review of the 2014 Policy and Plan should start and requested the Vice-Rector to make proposals for the possible extent of the review and the composition of a task team. On 12 February 2016, the SU Management issued a formal communication which included that SU had initiated a review of its language policy and plan with a view to submitting a new draft before the end of the first semester and that SU would ensure that a participatory process would be followed and interest groups would be given a fair opportunity to give inputs. On 16 February 2016, the Vice-Rector proposed a timeline for the review of the language policy to the RMT. This timeline was provided to members of the Council prior to their meeting of 20 February 2016.

[118] On 16 February 2016, at the RMT’s second weekly meeting, the Vice-Rector submitted proposals for the review of the 2014 Policy and Plan together with a timeline. At this meeting the RMT decided to establish a Language Policy Review Working Group (the Working Group). The RMT further decided that there should be wide ranging consultations, and consequently sufficient time for the Working Group to prepare a first draft, for public consultation, to consider proposals from interested parties for inclusion in the second and final drafts, to receive feedback from external and internal interested parties, and to consult with the Council and receive its feedback. The 18 members of the Working Group included academics from seven faculties and members of SU’s administration and support services with knowledge and experience of language planning and the implementation of language arrangements in teaching and administrative environments, as well as two student representatives nominated by student bodies.

[119] During the review process the members of the Working Group utilised a wide range of documents, including the Constitution, the LPHE, various internal SU documents and legal opinions (including the legal opinions by Advocates Breitenbach SC and Bishop), relevant demographic information about South Africa generally and the Western Cape in particular, the trends over the past few years regarding SU students’ home languages and language preferences for study, the results of a survey about the home languages and language preferences of SU students and the feedback from external and internal interested parties received during the public and internal processes of consultation described below. On 20 February 2016 there was an extraordinary meeting of the SU Council about the problems SU was experiencing with deviations in certain faculties from the 2014 Policy and Plan in the 2016 academic year. This had led, amongst other problems, to urgent litigation in this Court against SU by Afriforum and others instituted on 5 February 2016. The February 2016 Afriforum litigation, which was aimed at compelling SU to adhere to the 2014 Policy and Plan and the resulting language specifications for modules in the 2016 Yearbook, was settled on 12 February 2016 when SU gave an undertaking that it would take steps to ensure that all faculties implemented the 2014 Policy and Plan and those language specifications.

[120] When giving the undertakings, SU’s attorneys recorded that a process aimed at reviewing the 2014 Policy and Plan had commenced and SU would ensure all stakeholders, including Afriforum, were afforded a reasonable opportunity to make a contribution to the plan and the review process and further that the review would be completed within a reasonable time. Prior to the SU Council meeting on 20 February 2016, the members of the Council were provided with copies of the Vice-Rector’s language policy review proposal, the instructions to the Working Group, and the timeline for the review process. Between 3 March and 6 June 2016, before sending the final draft of a new policy to the Senate for consideration, the Working Group met six times. Each of those meetings and the work done by members of the Working Group between the meetings are described in detail in the Answering Affidavit. At its first meeting on 3 March 2016 the Working Group decided to appoint a three-member sub-group (the Sub-Group) and to task it with preparing a first draft policy for the Working Group to consider at its next meeting and, thereafter (i.e. between meetings of the Working Group), with considering and summarising comments on the draft, doing further work on the draft and if necessary preparing discussion documents for the Working Group.

[121] In an unrelated development, on 7 March 2016 Afriforum and several others brought a second urgent application in this Court to compel SU to adhere to the 2014 Policy and Plan and the language specifications in its 2016 Yearbook with immediate effect. SU abided the decision of the Court in this second urgent application, on condition the order would take effect at the start of the second term on 29 March 2016. SU did so because its investigations had revealed deviations from the language specifications when the 2016 lectures began on 22 February 2016, but it would take several weeks to ensure that all deviating modules changed back to the language specifications and to put measures in place to reduce the adverse impact of the change for students who were not proficient in Afrikaans. On 11 March 2016 this Court issued a rule nisi operating as an interim interdict directing SU to implement the language specifications in the 2016 Yearbook with effect from 29 March 2016. SU duly complied with this order. On 19 May 2016, by agreement between Afriforum and SU this Court confirmed the rule nisi and Afriforum withdrew the further relief it was seeking against SU, thereby bringing the March to May Afriforum litigation to a close.

[122] Returning to the language policy revision process, at the second meeting of the Working Group on 17 March 2016 it considered a first draft policy prepared by the Sub-Group and finalised it (the First Draft Policy). This draft was then posted on SU’s website in both English and Afrikaans on 22 March 2016 for public comment. On 22 and 23 March 2016 SU sent messages to all students and other stakeholders, including Afriforum, inviting them to comment on the First Draft Policy. Comments were due one month later, by 22 April 2016. In addition to soliciting comments, during April 2016 SU conducted an online survey of all 19 648 undergraduate students about their language preferences. The responses received from 5 196 students showed that 59% of the respondents preferred to be taught at SU in English, 26% in Afrikaans and 15% in both English and Afrikaans. In addition, only 65.8% of respondents who did Grade 12 in Afrikaans preferred to be taught at SU in Afrikaans. In all, SU reportedly received 514 responses to its invitation for comments on the first draft language policy, including one from Afriforum. All the comments were saved in a Dropbox folder that was accessible to all members of the Working Group. In addition, on 13 May 2016, all the comments, as well as summaries of the comments prepared by the Sub-Group, were made available to members of the SU Council through the Dropbox folder.

[123] At the third meeting of the Working Group on 15 April 2016, it discussed the Sub-Group’s summaries of all the comments that had been received up to 8 April 2016. During this meeting, the Working Group made changes to the First Draft Policy (which was projected on a screen at the meeting) with reference to the comments. After the meeting, using all the comments received between 22 March and 22 April 2016 the Sub-Group continued working on the draft policy, a revised version of which was circulated to all members of the Working Group on 28 April 2016. On 3 May 2016, the Working Group held its fourth meeting. By that stage, in addition to the revised version of the First Draft Policy, all the comments received between 22 March and 22 April 2016 had been made available to the members of the Working Group (this was done on 27 April 2016). At this meeting the Working Group considered the further comments received (i.e. after 8 April 2016) and worked on preparing the second draft of the policy (the Second Draft Policy).

[124] The factors which influenced the content of the Second Draft Policy included the feedback received during the public consultation process. The Second Draft Policy was completed on 5 May 2016 and sent to the faculties for consideration at their faculty council meetings between 9 and 13 May 2016 and to the Council for consideration at its meeting on 9 May 2016. At the Council meeting of 9 May 2016 (which was an ordinary, scheduled meeting), the Vice-Rector briefed the Council on the review process. The Council supported the continuation of the review process and unanimously adopted a motion that a special meeting of the Council be held where it could consider the language policy and give its feedback to the Working Group. (The First Applicant would later welcome this decision.) The special meeting of the Council took place over a full day on 21 May 2016. The meeting began with a presentation by Advocate Breitenbach SC about the constitutional principles that governed the Council’s decision. The presentation included a summary of the 2014 Policy and Plan; a summary of the three legal opinions he and Adv Bishop had provided to the SU (the third of which was that legal challenges to the 2014 Policy and Plan based on section 29(2) of the Constitution and the Equality Act would probably succeed); a summary of the important facts and data with which he had been briefed (including the relevant demographic statistics and the costs of moving to full parallel-medium instruction); and a discussion of what he considered to be the main sources of the legal principles relevant to the language policy, namely sections 3 and 27(2) of the Higher Education Act, the LPHE, sections 6, 9 and 29(1)(b) and (2) of the Constitution and sections 6, 7 and 14 of the Equality Act.

[125] During the course of the ensuing discussion, (which as stated lasted a day) the Council adopted a series of resolutions to be submitted to the Working Group for its consideration, including: (i) That the Council accept Advocate Breitenbach’s presentation as a guiding document and that it be given to the Working Group; (ii) That “[t]he Policy must be enabling and not shackling. There should not be micro management”; (iii) That “[t]he Policy must enable equitable access to SU for all academically deserving students on the basis of language, i.e. language should not be a barrier to access”; and (iv) ‘That the English offering be extended to the extent that no admitted student is excluded, that it is ensured that the Afrikaans offering is not reduced, and that Afrikaans as language of tuition is further developed, and also that the commitment to the development and promotion of isiXhosa as academic language is honoured’.

[126] The fifth meeting of the Working Group took place on 23 May 2016 over eight hours. It considered the principles adopted by the Council, the feedback from the faculties and the presentation of Advocate Breitenbach SC (who attended the meeting), and in the process revised the draft policy. After a long and intense debate, the Working Group adopted a revised draft of the policy to be submitted to the Institutional Forum and the Senate for consideration (the Third Draft Policy). Over the next week, the Institutional Forum, the Executive Committee of the Council and the Executive Committee of the Senate all met (on 25 May, 26 May and 1 June 2016 respectively) and gave feedback on the Third Draft Policy. The Council’s comments included that the Third Draft Policy did not fully encompass the policy principles on which the Council had resolved at its meeting on 21 May 2016, including that no provision was made for the expansion or even the maintenance of the Afrikaans offering.

[127] The Senate met on 3 June 2016 to consider the Third Draft Policy. After a long and intense discussion it approved the document, subject to specific reservations. It required the Working Group (augmented by Professors Quinot and Liebenberg of the Law Faculty) to reconsider parts of the policy and circulate a revised policy to the members of the Senate for a special Senate meeting on 9 June 2016. The Working Group held its sixth and last meeting on 6 June 2016. It included Professors Quinot and Liebenberg. Again, there was intense debate. The result was the approval by consensus of an amended version of the Policy (the Final Draft Policy). The Final Draft Policy, in the form in which it was then circulated to the Senate, showed (using different colours) the changes to the Third Draft Policy suggested by the Executive Committee of the Senate on 1 June 2016, the Working Group on 6 June 2016 and the Institutional Forum on 7 June 2016. At its meeting of 9 June 2016, the Senate approved the Final Draft Policy, more specifically the Third Draft Policy with the changes suggested by the Executive Committee of the Senate on 1 June 2016 and the Working Group on 6 June 2016, and recommended to the Council that it be adopted. On 17 June 2016 Advocate Breitenbach SC provided a memorandum on the constitutionality of the Final Draft Policy. He advised that, in his view, the policy was constitutionally sound. A copy of the memorandum was provided to the Council. At its ordinary meeting on 22 June 2016 the Council – after a presentation by the Vice Rector and two hours of debate – adopted the Final Draft Policy by a majority of 16 votes in favour and 9 votes against. (On 28 November 2016 the Council approved an Afrikaans translation of the Policy as the official translation.)

[128] Unavoidably, a comparison needs to be made between the process followed by SU (summarised above) and process followed by both UFS and UP in their respective matters. Because of the concern about integration, in 2015 the UFS established a Language Committee to formulate a new language policy. That committee’s work ‘spanned several months and involved thorough investigation, vigorous debate and full deliberation.’ See University of Free State judgment at para 8. The Committee reported to the UFS’s Council and Senate together with faculty comments. Afriforum participated throughout the process. A similar approach was taken by the UP. In both cases, the process followed was never questioned as being inadequate. It must be remarked that like both UFS and the UP, SU assigned the responsibility of developing a draft policy to a task team (the Working Group) the members of which had relevant expertise and experience. It prepared a draft policy, called for and considered inputs from the faculties, the Senate and the Council on revised drafts of the policy and received and considered legal advice. It did not act procedurally irrationally or even unfairly. The process documented above has not been disputed by applicants in reply.

THE RATIONAL CONNECTION; THE SU MANAGEMENT DETERMINED THE PROCESS (WAS THERE UNLAWFUL DICTATION?)

[129] The submission made by Muller is that the purpose of the making of a language policy by the Senate and Council of a higher education institution is to guide the use of language at the institution in a manner consistent with Section 29 (2) of the Constitution, the other applicable legal principles and the particular facts and circumstances of the institution. This is indeed a sound submission and I fully endorse its correctness. The point was made earlier in this judgment and perhaps it is worth repeating. The point is that the principle of legality requires that when making a language policy a higher education institution must choose and implement a means (i.e. a process) which, viewed as a whole, is rationally related to (i.e. capable of achieving) that purpose. See Albutt supra at paras 49, 72 and 74; Democratic Alliance supra at paras 27, 34, 36 and 37. The persons preparing the policy for consideration by the Senate and Council (i.e. the members of the Working Group) were suitably qualified and experienced. It is an undeniable fact that they considered the most relevant documents and information (including the applicable legal principles, the LPHE, relevant demographic information, a well-supported contemporaneous survey of student preferences and the cost of moving to full parallel-medium education). They afforded students and other stakeholders an opportunity to give input, which they considered when preparing a revised draft of the policy. They considered and responded (by preparing amended drafts) to input received from the faculties, the Council, the Executive Committee of the Senate, the Executive of the Council, the

Institutional Forum and the Senate. After long and intense debates in the Working Group, the Senate and the Council, the resulting (fourth) draft of the policy was formally approved by the Senate and the Council. The requirement was met in this case.

[130] It is correct that the process and timetable for the review of the language policy was determined by the Vice-Rector and approved by the RMT. In principle there was nothing wrong with that. The point is that there is nothing in the Higher Education Act, the Statute of SU or the 2014 Policy or Plan which precluded SU Management from initiating and determining the process of review. We gather from the answering papers that in view of the respective roles and capacities of the Council and SU Management, that it is normal that the process and timetable are prepared for acceptance or amendment by the Council. It was appropriate that SU Management initiated the process of reviewing the language policy because it (especially the Vice-Rector) had the capacity to assess the changing circumstances and the impacts of the 2014 Policy and Plan. In any event, the Vice Rector was the ‘owner’ of the 2014 Policy and the RMT is responsible for the day-to-day management of the university. The process that was adopted assigned the responsibility for the review to a specially appointed body (the Working Group) which comprised a majority of persons from outside the administration of SU. Importantly, the Council did not question the design of the process because it was clearly a considered and transparent one.

[131] The Applicants allege that a radical minority prescribed to the SU Management what the language policy should be and the SU Management did the same to the SU Council. The latter allegation is strangely missing in the founding papers. It surfaced for the first time in the heads of argument. What appears in the founding papers is the allegation that Open Stellenbosch (the members of which are the alleged radical minority) were given an unreasonable opportunity to make representations early in the process and in so doing to exercise greater influence than other organisations like Afriforum. The response by the respondents as documented in the answering papers is briefly the following: (a) In May 2015 Open Stellenbosch handed a memorandum to SU Management in which it raised concerns about the content and implementation of the 2014 Policy and Plan. (b) The Vice-Rector thereupon appointed a task team to investigate and make suggestions about Open Stellenbosch’s concerns, as well as concerns raised in a separate memorandum from the Students Representative Council. (c) In addition, SU Management engaged in discussions with various student groups, including Open Stellenbosch, about the 2014 Policy and Plan and its implementation.

[132] According to SU Management, during these discussions Open Stellenbosch made a significant contribution which was aimed at ensuring that the use of language at SU was not an obstacle to access to SU or the successful completion of studies there. The RMT openly acknowledged Open Stellenbosch’s positive contribution in its media release of 11 November 2015 in which it made concrete proposals for changes to the use of language at SU. (d) On 9 September 2015 the task team delivered its report, following which the SU did not give Open Stellenbosch any opportunity to give input regarding the language issue or the review process until the public notice-and-comment process in March 2016 (when it could participate along with all other interested parties like Afriforum). Although the 2016 Policy eventually adopted by the Senate and the Council emphasised the principle that language should facilitate access to and success at SU, the details of the policy differed significantly from the concrete proposals in the RMT’s media statement of 11 November 2015. There is no evidence that to the extent Open Stellenbosch influenced SU, it did so otherwise than through the strength of its arguments.

[133] It would appear, regard being had to the aforegoing that the applicants’ allegation in this regard is without substance. The allegation too that the SU Management prescribed to the SU Council what the language policy should be, makes no logical sense. I say so because SU Management is subordinate to the Council. Both Section 17 (1) of the Higher Education Act and article 11 (1) the SU Statute make it clear that the Council is the ultimate decision-making body of SU. I have been unable to find any evidence to support the allegation that the SU Management forced the Council to act contrary to its preference. Clearly this must mean that it did not happen.

[134] The answering papers deal exhaustively with the applicants’ allegation that the process was designed to create the pretence of a consultation whereas there was no real consultation. In brief the response is that: (a) At no stage during the language policy review process was it a foregone conclusion that the 2014 Policy and Plan would be replaced with a new plan, let alone that that the First Draft Policy published for public comment (on 22 March 2016) would be approved by the Senate and the Council. (b) If the 22 March 2016 First Draft Policy is compared with the 6 June 2016 Final Draft Policy that was approved by the Senate and the Council, it is clear that the former was materially amended. (c) In the period between 22 March 2016 and the decision of the Council to adopt the Policy on 22 June 2016, there was no consensus within SU about the proposed new policy. (d) There is no evidence of a conspiracy between SU Management and majorities of the Working Group, the Senate and the Council to manipulate the process. I am not persuaded that there are merits in the allegation discussed under this paragraph. Moreover, the chairperson of the Council, the Rector and the Vice-Rector testified that they did not have control over the precise contents of the policy and had no idea what the outcome of the relevant meetings of the Senate and the Council would be. Notably, the answering affidavit states unequivocally that all comments received were in fact considered. The applicants have not put up any evidence that comments were excluded from consideration. Perhaps in passing I need to mention that the fact that the comments of those in favour of, an increase in the Afrikaans offering or equal offerings for Afrikaans and English, were not accepted, does not mean they were not considered.

IMPROPER PURPOSES & FAILURE TO CONSIDER RELEVANT INFORMATION

(AND THE APPLICABLE STANDARD)

[135] Indeed the second category of procedural attack is the applicants’ allegation that the real reasons SU adopted the Policy were the improper purposes of regularising SU’s deviation from the 2014 Policy and Plan and averting political pressure and the prospect or threat of violence. In my view the aforegoing attack is but unfair and is devoid of merits. One must immediately point out that SU’s deviations from the 2014 Policy and Plan are well documented and fully explained in answering papers. We are told (and have no reason not to accept) that the Council and SU Management took a range of steps aimed at bringing the deviations to an end by the start of the second term of the 2016 academic year on 29 March 2016 and that they succeeded in doing so. As a result, the first problem of the deviations had been resolved long before the Senate and the Council adopted the 2016 Policy. In any event, the point regarding the alleged political pressure as being the real reason the Senate and the Council adopted the new Policy is not contained in the founding papers. As to alleged improper influence to amend the 2014 Policy and Plan by threats (particularly from Open Stellenbosch), it suffices to mention that the answering papers make it abundantly clear that the risk of violence did not influence the content of the 2016 Policy.

[136] The latter Policy was determined instead by the educational and constitutional considerations set out and discussed above. This appears from the evidence of the numerous meetings, wide-ranging deliberations and detailed inputs from linguistic and legal experts. The answering papers also show that the only relevance for the language policy issue of the widespread disruptions on campuses countrywide during 2015 and 2016, was that it emphasised the need for a decision about the issue sooner rather than later. I suppose the risk of violence or disruption merely and plainly created urgency to resolve the issue. In my view if SU did not consider that risk, it could competently be found to have been irrational. SU took firm measures designed to prevent occurrence of violence and disruptions on campus during 2016. Maybe to conclude I must quote the answering affidavit deposed to by Mr Steyn where he states unequivocally that threats of violence did not influence the content of the Policy: ‘Die risiko van geweld en ontwrigting het nie die inhoud van die 2016 Taalbeleid beïnvloed nie. Ek kan did onomwonde stel dat die inhoud van die 2016 Taalbeleid bepaal is deur opvoedkundige en grondwetlike oorwegings’. It is of importance that I mention (without elaborating) that because the Applicants are seeking final relief on motion and have not sought to refer this or any other issue to oral evidence, and Mr Steyn’s evidence is not inherently implausible, it must therefore be accepted as correct for purposes of the adjudication of the present matter. See Fakie NO v CCII Systems (Pty) Ltd 2006 (4) SA 326 (SCA) paras 55-56; Wightman t/a JW Construction v Headfour (Pty) Ltd and Another 2008 (3) SA 371 (SCA) para 12. Failure to consider various relevant factors or documents alleged by the applicants is, in effect, a procedural rationality challenge. SU contends that all the relevant information was in fact considered. I do not for a moment doubt this assertion. Moreover, even if this Court were to find that (I do not so find) a particular piece of relevant information was not considered, that alone would not taint the rationality of the process followed.

[137] The law does not require that each and every member of Senate and Council must read every relevant document. In Minister of Environmental Affairs and Tourism and Another v Scenematic Fourteen (Pty) Ltd 2005 (6) SA 182 (SCA), the SCA explained guidingly what is required of a decision-maker in situations where he is required to consider a large amount of information. The case concerned the allocation of fishing rights, a decision that ultimately had to be taken by the Deputy Director-General. He did not consider every application himself, but relied on detailed advice provided by his subordinates. The SCA held that there was nothing improper about this process:

‘[I]t does not follow that a functionary such as the DDG in the present case would have to read every word of every application and may not rely on the assistance of others. Indeed, given the circumstances, Parliament could hardly have intended otherwise. What the functionary may not do, of course, is adopt the role of a rubber stamp and so rely on the advice of others that it cannot be said that it was he who exercised the power. If in making a decision he were simply to rely on the advice of another without knowing the grounds on which that advice was given the decision would clearly not be his. But, by the same token, merely because he was not acquainted with every fact on which the advice was based would not mean that he would have failed properly to exercise his discretion. This would be particularly so if that advice was merely one of the factors on which he relied to arrive at his ultimate decision.’

See para 20 of the judgment, quoted with approval in Walele v City of Cape Town and Others 2008 (6) SA 129 (CC) at para 114 (the dissenting judgment). It is even needless to mention that the issue is and must be complicated when one is called upon to evaluate decisions taken by multi-member bodies. It is certainly not possible to know exactly what documents each member of Senate or Council read or did not read, or what information they regarded as relevant or irrelevant. The question is whether the relevant information was presented to them and the relevant documents were available for them to consider.

THE INFORMATION CONSIDERED

AND THE EXTENT TO WHICH CONSIDERED

[138] The applicants argue as demonstrated above that SU failed to consider the Constitution and the LPHE. Even though this has, to some extent been dealt with earlier, its importance necessitate a few remarks. SU is accused of having attempted to ‘suppress invocation of the Constitution’. One simple response is that the Policy itself states its essence as including aiming ‘to give effect to section 29(2) of the Constitution in relation to language usage in [SU’s] academic, administrative, professional and social contexts. The Policy aims to increase equitable access to SU for all students and staff and to facilitate pedagogically sound teaching and learning. Since our campuses are situated in the Western Cape, we commit ourselves to multilingualism by using the province’s three official languages, namely Afrikaans, English and isiXhosa’.

[139] It cannot be denied that SU received detailed input and legal advice during the policy-making process, including Advocate Breitenbach SC’s presentation regarding the applicable constitutional principles on 21 May 2016 and his memorandum regarding the constitutionality of the Final Draft Policy adopted by the Senate on 9 June 2016. As stated earlier, the LPHE was considered in detail throughout the process. Even if SU’s understanding of the LPHE and the Constitution markedly differed, that would not mean that the LPHE and the Constitution were not taken into account. Indeed the contents of the Policy does show that SU was well aware of the fact that speakers of the various South African languages use English to communicate with one another. SU was also aware of the significant academic, business and international value of English and of the need to advance the academic potential of Afrikaans. SU proceeded from the assumption that it was reasonably practical to deliver Afrikaans tuition. It must be emphasised that although SU knew that almost all Afrikaans –speaking students were sufficiently proficient in English to study in English and this was a key factor which allowed it to adopt the 2016 Policy, it did consider the need for a smooth transition to a university environment in which there would be more teaching in English than in Afrikaans.

[140] The above is evidenced by the fact that SU tries as far as possible to offer all first-year and second-year lectures in parallel medium. Of course this is further evidenced by the fact that the Policy provides that during the first year of undergraduate study SU will provide real-time interpreting from English to Afrikaans in all dual-medium lectures and all lectures in English; and that in the second and later years it will do so, subject to available resources, upon request of the faculties in cases where Afrikaans-speaking students need the service. Additionally, in all undergraduate modules all SU module frameworks and study guides will be available in Afrikaans. Finally, in this regard, all question papers for tests, examinations and other summative assessments in undergraduate modules will be available in Afrikaans and students may answer all assessments and submit all written work in Afrikaans. As to the knock-on effect on moving from Afrikaans to English teaching at universities on Afrikaans-language schooling, the answering papers reveal that SU was aware that there may be a link between the availability of Afrikaans tertiary education and the demand for Afrikaans primary and secondary education. This, however, did not result in SU altering the balance it sought to strike between Afrikaans-language teaching (access would be at least sustained) and English-language teaching which would be increased to a 100% offering so as to facilitate access for students especially Black (African) students who are not sufficiently proficient in Afrikaans.

[141] SU considered the results of its own survey which showed strong support for the retention of Afrikaans as a primary medium of instruction. However, while the results showed support for Afrikaans, they also showed significantly more support for dual-medium teaching and for English-only teaching. Mr Muller contended that SU did not consider in detail the demand for Afrikaans tertiary education in the Western Cape or nationally because its point of departure was that there is currently a significant and continuing demand for Afrikaans education in the Western Cape Province and nationally. Of course this appears from the statistics set out in the 27 November 2015 opinion of Advocates Breitenbach SC and Bishop. What SU also considered is the fact that other universities had reduced or ended their Afrikaans offering. It is said to also have been aware of, and took into account, the changes taking place at other universities during 2016, especially the changes to the language policies of the UFS and the UP. I am told that while SU considered this information, it did not let itself be led by what other universities had decided. The submission made on behalf of Stellenbosch in this regard is that the choices of those universities do not determine the legality of SU’s decision to amend its language policy. One must, perhaps conclude this aspect by categorically stating that when assessing this challenge it is important to do so in the light of what the 2016 Policy actually does. Although the Policy does not increase the Afrikaans offering to the same level (100%) as the increase English offering, the Policy nevertheless retains Afrikaans as a language of learning and teaching.

[142] According to the answering papers, SU considered the demographics of the Western Cape Province in particular. It reportedly focussed on the 15 to 24 years age category in the Western Cape Province because that is the primary source of its student body. It reportedly analysed the language and racial demographics of this category in detail and compared them with the national demographics and the demographics of SU’s 2015 first-year intake. It is of significance that these figures are set out in the first of the opinions by Advocates Breitenbach SC and Bishop dated 27 November 2015 and referred to again in Advocate Breitenbach SC’s presentation to the SU Council on 21 May 2016. As far as the Northern Cape is concerned, Mr Muller submitted that SU did not separately consider the demographics relating thereto. He hastened to add that in his submission SU was not obliged to do so as only 3.4% of its students came from that province in 2015 and only 1.92% of its students did so in 2016. The complaint that SU did not consider the fact that the ‘Coloured’ community in the Western Cape is predominantly Afrikaans speaking is correctly grounded. The answering papers make it clear that SU was well aware of this. In fact, the first of the opinions by Advocates Breitenbach SC and Bishop dated 27 November 2015 identified the fact that Coloured students were also victims of past discrimination, that ‘37.6% of SU’s first year coloured students use Afrikaans as their home language’ and that SU offered the only possibility for Afrikaans tertiary education in the Western Cape. According to the answering papers, SU was also well aware that the majority of its ‘Coloured’ students, came from the Western Cape.

[143] According to the submissions made on behalf of the respondents SU considered all the relevant information. In any event if this Court concludes that some piece of relevant information was not considered or was not (as it were) adequately considered, the question then becomes whether that must render the entire policy invalid? It cannot. It was pointed out earlier in this judgment that not every procedural failure renders a decision invalid. It is only when that failure ‘had an impact on the rationality of the entire process’ that it will render the decision invalid. See Democratic Alliance v President of South Africa and Others 2013 (1) SA 248 (CC) at para 39. According to the latter authority, there is a three-step process to determine when the failure to consider information taints an entire process:

‘The first is whether the factors ignored are relevant; the second requires us to consider whether the failure to consider the material concerned (the means) is rationally related to the purpose for which the power was conferred; and the third, which arises only if the answer to the second stage of the enquiry is negative, is whether ignoring relevant facts is of a kind that colours the entire process with irrationality and thus renders the final decision irrational.’

It is therefore our law that even if this Court finds that any particular item of information was not considered, that alone will be insufficient to render the entire process irrational. It will be irrational only if ignoring that particular item of information colours the entire process with irrationality.

[144] If the applicants’ claim must be understood to be saying that the factors were considered but that SU did not accord appropriate weight to those factors, that would mean it is a claim that the decision was not reasonable contrary to Section 6 (2) (h) of PAJA. But that is, of course, not a basis to attack an executive act such as the Decisions and the Policy. It was precisely in this context that the SCA held in Bel Porto School Governing Body v Premier, Western Cape 2002 (3) SA 265 (CC) at para 45:

‘The weight to be given to the material lies in the discretion of the decision-maker. And the fact that there are other means of achieving the same purpose is not something the court can consider: “Courts cannot interfere with rational decisions of the executive that have been made lawfully, on the grounds that they consider that a different decision would have been preferable.’

In truth even on the PAJA standard of reasonableness, the applicants’ claim clearly must fail for the general reasons outlined in Bato Star Fishing (Pty) Ltd v Minister of Environmental Affairs and Tourism and Others 2004 (4) SA 490 (CC). It is not for a court to second-guess how much weight to afford each factor in a polycentric administrative decision. Navsa JA and Swain AJA in MEC For Environmental Affairs and Development Planning v Clairison’s CC 2013 (6) SA 235 (SCA) put it as follows:

‘It has always been the law, and we see no reason to think that PAJA has altered the position that the weight or lack of it to be attached to the various considerations that go to making up a decision, is that given by the decision-maker. As it was stated by Baxter:

‘The court will merely require the decision-maker to take the relevant considerations into account; it will not prescribe the weight that must be accorded to each consideration, for to do so could constitute a usurpation of the decision-maker's discretion.’

That above dictum was cited in University of Pretoria case supra as follows:

Accordingly, in the instant matter, once it is shown that SU considered the relevant factors, the weight it afforded them is not at all an issue that this Court needs to address. The SCA in UFS v Afriforum supra held that UFS had considered other factors, but concluded that they were outweighed by the need to promote integration. That was held to have been a rational decision. I conclude by stating that SU appears to have decided that its multiple purposes of preventing exclusion, promoting multilingualism, ensuring integration, and fostering Afrikaans are best served by the 2016 Policy it adopted. It clearly considered multiple factors and weighed them all. This Court is commanded by the law not to second-guess that extremely difficult process, unless the outcome is obviously irrational.

BIAS AND NATURE THEREOF

[145] Hoexter-Administrative Law in South Africa (2ed, 2012) at 452, points out that claims of bias are generally reserved for cases involving judicial or quasi-judicial decisions where there is a clear contest between different arguments. Yes, similar claims against administrators are generally ‘couched in the language of abuse of discretion’. I do not accept an argument advanced to the effect that SU acted for an ulterior purpose. The decision-makers were the Senate and Council and not SU Management. The fact that Management had a view about what language policy it wished the Council to adopt simply does not found a case for bias. Management was entitled to form and publicly state its view – which it did in its controversial media statement of 11 November 2015. It attempted to persuade the Senate and Council to accept its view. Both the Council and the Senate are multi-member bodies and together they represent a range of different interested parties. In conclusion, the applicants do not appear to rely on extrinsic evidence of bias. Their argument is that bias is the only possible explanation for the decision. There is no basis on the papers to draw a conclusion of bias.

APPLICANTS’ APPLICATION FOR LEAVE TO ADMIT A FURTHER AFFIDAVIT

AND TO LEAD ORAL EVIDENCE OF THERON

[146] On 3 August 2017 the applicants delivered an application for leave to admit a further affidavit by Mr Daniel Rossouw (‘Rossouw’), the applicants’ attorneys, and for Mr Johan Theron (‘Theron’) to be subpoenaed and to testify to the content of his statement of defence, Annexure ‘DJR5’ to Rossouw’s affidavit. The application is opposed by the first to third respondents (‘the respondents’). It was contended that this court should not in the exercise of its discretion permit the adducing of further evidence at this late stage. Accordingly the submission was that both the application for leave to adduce Rossouw’s further affidavit, and the application for leave to subpoena and adduce the evidence of Theron, should be dismissed with costs. The general rule is that only three sets of affidavits are permitted in motion proceedings. See Uniform Rule 6 (5) (e). It is in the interests of the administration of justice that the number and sequence of affidavits should ordinarily be observed. See James Brown & Hamer (Pty) Ltd (previously named Gilbert Hamer & Co Ltd) v Simmons NO 1963 (4) SA 656 (A) 660E; Hano Trading CC v JR 209 Investments (Pty) Ltd 2013 (1) SA 161 (SCA) para [12]. However, that rule is not always rigidly applied and the Court enjoys a discretion to permit the filing of further affidavits. The Rule provides that ‘(e) Within 10 days of the service upon him of the affidavit and documents referred to in sub-paragraph (ii) of paragraph (d) of subrule (5) [the answering affidavit] the applicant may deliver a replying affidavit. The Court may in its discretion permit the filing of further affidavits.’

[147] Perhaps a mention must be made that while the Courts may permit the filing of further affidavits in exceptional circumstances, the Court (a) will not exercise its discretion in the absence of an explanation of why it is necessary to file the affidavit concerned; and (b) will always act subject to considerations of fairness and justice and the absence of prejudice to other parties. See Transvaal Racing Club v Jockey Club of SA 1958 (3) SA 599 (W) at 604A-E; Gold Fields Ltd v Motley Rice LLC 2015 (4) SA 299 (GJ) at para [123]; Standard Bank of SA Ltd v Sewpersadh & Another 2005 (4) SA 148 (C) at 154D. James Brown & Hamer supra 660 D-H.

[148] Importantly, other considerations will include the degree of materiality of the evidence, the stage which the litigation has reached, and the general need for finality in judicial proceedings. See Erasmus, Superior Court Practice, Vol. 2 page D1-68. Of course this follows from the trite rule of practice that an applicant must, generally speaking, stand and fall by his founding papers. The most important consideration of all is that adhering to the principles ensures that disputes between the litigants are resolved in terms of a procedure which is just, orderly and well recognised. See Gold Fields Ltd v Motley Rice LLC, supra, par 125; Union Finance Holdings Ltd v IS Mirk Office Machines II (Pty) Ltd and Another 2001 (4) SA 842 (W) at 847J-848E. Needless to mention that this is not at all pedantry. It is and remains an integral part of the principle of legal certainty. It is therefore the rule of law, because the other party is entitled to know precisely the case it has to meet. See Pilane & Another v Pilane & Another 2014 (4) BCLR 431 (CC) at para [49] ft nt 40; Standard Bank of SA Ltd v Sewpersadh & Another, supra, at 154A, relying in James Brown & Hamer, supra at 660.

[149] The point is where (as in the present case), affidavits are tendered both late and out of the their ordinary sequence, the party tendering has no right to do so, but seeks an indulgence from the Court. Such party must both advance an explanation of why the affidavits are out of time, and satisfy the Court that, although the affidavits are late, they should, having regard to all the circumstances of the case, nevertheless be received. See James Brown & Hamer supra at 660F; Sewpersadh & Another supra at 154C-E. The courts have correctly refrained from delineating precisely all the considerations that will be taken into account in exercising the discretion to admit or reject a late tendered affidavit, but what is invariably accepted as being one pre-requisite is ‘ the adequacy or otherwise of the explanation for the late tendering of the affidavit.’ See James, Brown & Hamer supra at 660H; Sewpersadh & Another supra, at 154E. There must indeed be a proper and satisfactory explanation as to why the information contained in the affidavit was not put up earlier. Quite apart from this being an indulgence sought from the Court, the applicant for such filing must show that he has not been mala fide or culpably remiss in not having put up the information in the further affidavit earlier. The Court must be satisfied that no prejudice will be occasioned to the other party if the new affidavit is admitted. It is axiomatic that to permit the filing of further affidavits severely prejudices the party who now has to meet a case based on those submissions. See Hano Trading CC v JR209 Investments (Pty) Ltd 2013 (1) SA 161 (SCA) at para [12]. An unfortunate development is developing in this regard. Parties would be well advised that courts will stamp out the practice that tends to show laxity and non-adherence to the rules regarding the three essential affidavits and the contents of each of them.

[150] In Mr Muller’s submission the further evidence which the applicants seeks to adduce by way of Rossouw’s further affidavit and the testimony of Theron is irrelevant to the determination of the relief sought in the main application. In the alternative, Mr Muller contended that it is insufficiently relevant to permit further evidence to be adduced in this way and at this late stage, because it concerns an hypothesis and allegations in support of it which lack any plausible factual basis. It is trite that in paragraph 3 of Rossouw’s affidavit (read in context) he purports to set forth the ostensible purpose of that affidavit and the intended oral evidence of Theron. In essence the ostensible purpose is to place information before this Court: (a) which has not yet been put before Court and could not have been adduced earlier; (b) which supports the Applicants’ version and legal conclusions in support of the relief sought in the main application, as set out in their founding and replying papers in the main application; (c) which was not placed before Judge Howie for purpose of his investigation in 2016 into the deviations from SU’s 2014 Language Policy and Plan (‘the 2014 Policy’) in the first term of 2016; (d) which indicates that Prof Wim De Villiers, the Rector of SU (‘De Villiers’), misled this management team (‘RMT’), and the RMT was never prepared to honour the undertaking given to Afriforum on 12 February 2016; (e) which indicates that the new Language Policy was accepted in June 2016 simply to clothe with legality unlawful departures from the previous language policy in the first term of 2016; and (f) which indicates that Mr George Steyn (‘Steyn’), the chair of the SU Council, was initially against the actions of the RMT, but later supported its actions as the way of least resistance and in fact became a pawn of the RMT.

[151] In paragraph 4 of Rossouw’s affidavit he asserts that the above information explains why the outcome of the review of the 2014 Policy during the first half of 2016 was (as he contends) a foregone conclusion and simply an attempt to regularise the contraventions of the 2014 Policy. The respondents have, however, given exhaustive reasons in their answering affidavits and in their subsequent submissions. Essentially, according to the respondents (a) there is no connection between the deviations from the 2014 Policy in the first term of 2016 and the new Language Policy which was approved by Senate and Council in June 2016 (the deviations from the 2014 Policy were rectified and ceased with effect from the start of the second term of 2016 on 29 March 2016); (b) the reasons for the deviations from the 2014 Policy were thoroughly investigated by Judge Howie in March and April 2016 and his report and conclusions were extensively debated at, and accepted by an overwhelming majority of, the SU Council at its meeting on 9 May 2016; and (c) the litigation between SU and Afriforum about the deviations from the 2014 policy was settled on 19 May 2016. It is thus clear that the deviations from the 2014 Policy and the controversy and litigation concerning them had been finally resolved more than a month before the SU Council adopted the new Policy on 22 June 2016.

[152] In my view, the applicants’ argument that the new Policy was adopted to regularise the 2014 Policy deviations lacks any factual foundation. One must say so because deviations had long ceased and had also been addressed by an investigation undertaken by a respected retired Judge and settled in hotly contested litigation between the SU and the Afriforum and other applicants. This court cannot conceivably validly exercise its discretion in favour of permitting the applicants to file a fourth set of papers and grant leave for a third party to be subpoenaed, in circumstances where the contents of the papers and the intended oral evidence are aimed at ‘proving’, or ‘substantiating, an hypothesis and allegations in support of it made by the applicants which clearly lack any plausible factual foundation and is consequently not relevant, or is insufficiently relevant to the relief sought.

[153] Of importance in this application is Judge Howie’ report which I briefly summarise infra: (a) During the course of his investigation Judge Howie interviewed sixteen individuals. These included Theron, Steyn, Professor PW van der Walt (‘Van Der Walt’), Proff. CS Human and A Schoonwinkel. De Villiers was the last person interviewed. Theron had the opportunity to say to Judge Howie whatever he wanted to concerning, for example, the deviations from the 2014 Policy, his discussions and meetings with De Villiers, Steyn, Van Der Walt, Human and Schoonwinkel and generally the issue of language at SU (all matters dealt with in Theron’s statement of defence). (b) All the individuals mentioned in the preceding paragraph, plus several others, including Prof JH Knoetze (whose correspondence is relied on by Theron in his statement of defence), were interviewed by Judge Howie and their versions taken into account for purposes of his report. (c) The Howie report was considered and debated at the meeting of the SU Council held on 9 May 2016. Those present at the meeting included Theron himself, as well as Professor Carstens and Advocate Jan Heunis SC. By closed ballot Council they voted overwhelmingly to approve the report and its findings, by a majority of 20 votes to 1, with 2 abstentions.

[154] As far as the adoption by SU in June 2016 of the new Policy the following must be mentioned: (a) On 9 June 2016 the Senate approved the new Policy by a vote of 113 in favour to 10 against. (b) On 22 June 2016 the Council adopted the new Policy by a majority of 17 votes in favour to 9 against. (c) The lengthy process which preceded and culminated in these decisions by the Senate and Council respectively has been dealt with by all parties fully in the papers. I accept, in any event, that at no stage during the language policy review process was it a forgone conclusion that the 2014 Policy would be replaced with a new plan, let alone that the First Draft Policy which was published for public comment on 22 March 2016 would be approved by the Senate and the Council – not least because during the period 22 March to 22 June 2016 there was no consensus whatsoever within SU about the proposed new policy. Given these facts, I am not persuaded that there is a scope for the hypothesis that the process which culminated in the approval of the new Language Policy by the Senate and Council in June 2016 was a foregone conclusion or was not properly considered by the Senate and Council. The important thing is that the applicants’ contentions in this regard are already before the Court. In my finding, Rossouw’s further affidavit and the proposed evidence of Theron take these matters no further. What they do is that they merely serve to repeat allegations already made and seek to demonise individuals such as Steyn, De Villiers and Schoonwinkel even further.

[155] I bear in mind that were the applicants permitted to adduce the further affidavit evidence of Rossouw and the oral evidence of Theron (quite apart from the lack of relevance discussed above), this would substantially prejudice the respondents. It shall be remembered that the application to adduce further evidence was delivered: (a) by e-mail on the afternoon of Thursday, 3 August 2017 (only 5 Court days prior to the commencement of the hearing of this matter on Monday, 14 August 2017; (b) after both parties had already filed their heads of argument. If this Court were to allow the further evidence of Rossouw and Theron to be adduced, that will entitle the respondents to deal with numerous issues mentioned in Rossouw’s further affidavit. I do not deem it necessary to document such issues in this judgment. They are contained in Rossouw’s further affidavit.

[156] Theron’s statement of defence is annexed as ‘LVH1’. Were this Court to permit the applicants’ to subpoena and call Theron to testify, it would not have been practicable for the respondents’ legal representatives to obtain instructions necessary properly to cross-examine Theron on the many aspects of his statement of defence during the hearing scheduled for 14 August 2017. Importantly, in any event, were such an unusual procedure in a matter of this kind and magnitude to be countenanced at this very late stage, the respondents might be required to also make application to adduce oral evidence of their own in response to such evidence that Theron might give. If that eventuality becomes a reality, that would have the unavoidable and inevitable effect of delaying the hearing of this application and be severely prejudicial to both the applicants and the respondents. The truth is that, this being an attack on SU’s new language policy, it is in the interests of the SU community that the matter be heard and finalised on dates agreed to by the parties. The respondents have a procedural right that the matter proceed and be concluded speedily, it is quite obvious that SU community and the respondents will be prejudiced by the conceivable postponement which will be necessitated if the further affidavit of Rossouw is admitted and/or the oral evidence of Theron is permitted.

[157] According to the applicants, the only motivation for permitting the further affidavit of Rossouw and to subpoena and adduce oral evidence of Theron, is the following:

‘Aangesien Theron se verweerskrif feite openbaar wat Gelyke Kanse se aansoek teen die US in belangrike opsigte staaf, en belangrike inligting wat daarin vervat is nie tot die Applikante se beskikking was ten tye van die opstel van die antwoordende (sic) en repliserende beëdigde verklaring ingedien word.’

Rossouw does not even attempt to identify which part(s) of Theron’s statement of defence contains information which was not available to the applicants when their affidavits were prepared. Most (if not all) of it was or would have been. The contents of the statement can be described as a repetition of allegations which already appear in the applicants’ papers. There is no explanation why such information as is in Theron’s statement of defence and which might not have been known to the applicants, was not known to them and could not have been included in the applicants’ papers from the outset. It remains incumbent on an applicant seeking to adduce additional evidence out of its ordinary sequence, to provide a full explanation for why the information sought to be adduced could not have been obtained earlier. It is now extremely late and parties even filed heads of argument. Rossouw does not assert that it was not possible to confer with Theron earlier or that Theron had previously declined to provide any relevant information.

[158] The founding and answering papers make it clear that Theron was part of the contingent of Council members opposed to any revision of the 2014 language policy and bitterly resisted adoption of the 2016 Policy. He is referred to and relied on several times by the applicants in the founding papers. He spoke up frequently at Council meetings and he left the Council meeting in protest at the adoption by the Council on 22 June 2016 of the 2016 language Policy (this appears from the minutes of that meeting) and is a fact also mentioned in the founding papers. After this meeting Theron gave an interview to the Afrikaans press expressing his dissatisfaction with the 2016 Policy and the adoption process. Perhaps Mr Muller puts it rather accurately by stating the following:

‘throughout the process Theron has been a leading and outspoken critic of the 2016 Policy, its adoption by Senate and Council, and the process which preceded such adoption. The Applicants knew this.’

[159] Regard being had to the aforegoing, when preparing their founding and replying papers the applicants knew or must have known that Theron was a witness with information which might be of relevance, but they evidently failed to consult with him. Had the applicants done so they could have adduced his evidence in the founding papers or, had he then taken up the attitude (which he now reportedly has done) that as a sitting member of the SU Council he cannot give evidence for the applicants without an order of this Court ordering him to do so, the applicants could have applied for the order requiring that he testify orally far sooner than the eve of the hearing. In conclusion I emphasise that in the absence of any explanation as to whether the applicants interviewed Theron, or why they did not interview him when preparing the founding or at least replying papers, I am obliged to find that there is simply no adequate explanation furnished for seeking to adduce Rossouw’s further affidavit or to compel the oral testimony of Theron at this extremely late stage by way of the unusual procedure in motion proceedings of seeking an order that he be subpoenaed to testify. The application to admit the further affidavit and to subpoena Theron to testify stands to be dismissed forthwith.

THE STRIKING OUT APPLICATIONS

[160] SU has launched two striking out applications in terms of Uniform Rule 6(15) of the Uniform Rules of Court. The first application is dated 24 February 2017. This application is aimed mainly at the substantial quantity of hearsay and what the respondents describe as irrelevant allegations contained in the applicants’ founding papers. This application is resisted by the applicants. The second application is dated 27 July 2017 and it pertains to the applicants’ replying papers, as papers that are replete with new and irrelevant matters. Although the Rule under discussion deals expressly with the striking out of scandalous, vexatious or irrelevant matter, those grounds are not at all exhaustive or intended to be exhaustive as to the grounds on which a court will strike out allegations; the court enjoys an inherent jurisdiction to grant relief where appropriate. See Titty’s Bar and Bottle Store (Pty) Ltd v ABC Garage (Pty) Ltd and Others 1974 (4) SA 326 (T) at 368F-H. It is trite that new matter raised for the first time in reply falls to be struck out. Therefore, for an application to strike out to succeed, (a) the matter to be struck out must be of an offending kind as indicated above; and (b) the court must be satisfied that if the matter is not struck out the party seeking to have the matter struck out would be prejudiced. See Bredenkamp v Standard Bank of SA Ltd 2009 (5) SA 304 (GSJ) at para 75-77. The requirement of prejudice does not, however, (I point out), mean that if the offending allegations remain the innocent party’s chances of success will be reduced. On the contrary, it is substantially less than that. See Vaatz v Law Society of Namibia 1991 (3) SA 563 (NM) at 566J-567B.

[161] Perhaps before proceeding further, it becomes necessary to point out that the nature of the applicants’ case and the relief sought form important background to the determination of these interlocutory applications. The respondents have been at pains to emphasise that the applicants attack is on the 2016 Policy itself which they seek to have reviewed and set aside together with the decisions of the Senate and Council approving it. Obviously, matters pertaining to the implementation of the Policy are irrelevant to the determination of that relief. Mr Muller correctly contended that an attack on the implementation of the Policy would be an entirely different case, one which is not made out in the founding papers and one which the SU has consequently not been called upon to answer.

[162] The allegations sought to be struck out from the founding papers in the first striking out application are specifically set out in the schedule to the application. These amount to approximately 250 pages and need not be set out in this judgment. They are found at pages 3676-3678 of the record of proceedings. The overwhelming quantity is uncorroborated hearsay and irrelevant matter. There are a few allegations which are described as vexatious. A reading of the schedule reveals that much of the uncorroborated hearsay evidence is also irrelevant. It is clear that it is advanced on account of the applicants’ approach, which is to make SU and its language policy responsible for the fate of Afrikaans throughout South Africa. That this in an incorrect and erroneous approach is beyond question. This amounts to substantial allegations and documentation dealing inter alia with – (a) the role of mother-tongue instruction in education; (b) press coverage and/or commentary of language-related developments at SU; (c) commentary on the hegemony of English; and (d) commentary on the ‘death’ of Afrikaans language and Afrikaans culture. The above is completely irrelevant to the relief sought.

[163] The general rule is that hearsay evidence is not permitted in affidavits. The aforegoing rule is subject to the Law of Evidence Amendment Act. I accept that such evidence could, in terms of Section 3, be admitted in the present circumstances only on the basis that this Court forms the view that it is in the interests of justice to do so, having regard to: (a) the nature of the proceedings; (b) the nature of the evidence; (c) the purpose for which the evidence is tendered; (d) the probative value of the evidence; (e) the reason why the evidence is not given by the person upon whose credibility the probative value of the evidence depends; (f) any prejudice to the SU that the admission of the evidence might entail; or (g) any other factor that should in the opinion of the court be taken into account. As shown above invariably the hearsay evidence tendered by the applicants is also irrelevant to the determination of the relief. The point is if the hearsay evidence is found to be irrelevant (as I hereby find), it must follow necessarily that no purpose would be served in further considering whether it can be admitted in terms of Section 3 (1) of the Act.

[164] In my finding, those hearsay allegations which may survive the relevance inquiry utterly fail to satisfy the ‘interests of justice’ requirement. The Applicants have not (let alone attempted) to satisfy the requirements of Section 3 of the Act in relation to any hearsay allegations tendered. Why the author of a newspaper article could not provide an affidavit for purposes of this litigation, remains a mystery to me. No steps have been taken in order to comply with Section 3 of the Act. SU contends that it is prejudiced by this. This cannot be gainsaid. SU is obliged to deal with this which they correctly describe as unnecessary and irrelevant matter. Importantly SU shall incur costs in that regard. The first application to strike out stands to succeed.

[165] The second application to strike out is indeed a composite application. It is directed at the striking out of new, irrelevant and hearsay matter raised in reply by the applicants. Of importance is the fact that if the impermissible new matter in reply is not struck out, then the Court shall be obliged to allow SU leave to adduce a further set of affidavits in order to address that matter. The offending allegations complained of pertain to the following: (a) New allegations concerning the implementation of the Policy in various faculties of SU (by far the bulk of the offending allegations, including numerous affidavits from various students at SU); (b) New allegations concerning an alleged meeting in March/April 2015 between Professors Koopman and Schoonwinkel and students Burger and Pieters; (c) The new allegation that SU never tried to engage in discussion with the national Higher Education Ministry about the feasibility of different strategies aimed at sustaining Afrikaans as a medium of academic expression and communication; (d) The new allegation that the implementation of the Policy in 2017 has resulted in a notable reduction in the Afrikaans offering; (e) Whether the legal opinions of advocates Breitenbach SC and Bishop dated 27 November 2015 and 30 March 2016 were made available to the members of the Senate and Council; (f) The omission from the table in SU’s main answering affidavit of the fact that all the responses received between 22 March and 22 April 2016 were made available to the Working Group on 27 and 28 April 2016; (g) Whether at the Working Group meeting of 3 May 2016 a screen-projected version of the draft language policy was used to make changes; (h) Whether the report on the costs of a fully parallel-medium language offering served before the Institutional Forum, the Senate and the Council; and the criticisms of the contents of that report; and (i) Whether SU ever tried to engage in discussion with the national Higher Education Ministry about the feasibility of different strategies aimed at sustaining Afrikaans as a medium of academic expression and communication.

[166] It is of importance to mention that the applicants seek to justify the inclusion of this new matter on the ostensible basis that:

‘Aangesien die US nie die implementering van die nuwe taalbeleid opgeskort het hangende die beregting van hierdie aansoek nie, is die Applikante nou by magte om nuwe feite wat voortspruit uit die implementering van die taalbeleid en wat die Applikante se voorspellings en gevolgtrekkings staaf aan die Agbare Hof voor te hou.’

Indeed the aforementioned purported justification is manifestly without merit. That these allegations complained of constitute new matter raised in reply is beyond question. In fact the applicants effectively concede by (at the outset and before the SU had challenged it or launched the striking out application) seeking to justify its inclusion. The law is clear on this aspect. It is not open to the applicants to introduce this new matter in reply. The implementation of the Policy was not challenged in the founding papers. The fact is that the implementation of the Policy is a heavily fact-based issue which was not canvassed in the founding papers. It has not been properly ventilated in the papers before court. The applicants’ allegations are vague and unparticularised, inadmissible hearsay, and/or simply false. The allegations put forth in reply are uncorroborated hearsay and/or palpably false. It is true that there are what one may call speculative allegations concerning proposed implementation in the founding papers. But the law is clear in this regard. An applicant is not in our law permitted to set up a ‘skeleton’ case in its founding papers and to flesh it out in reply. See Bergkelder v Delheim Wines (Pty) Ltd 1980 (3) SA 1171 (C) at 1176H; Johannesburg City Council v Bruma 32 (Pty) Ltd 1984 (4) SA 87 (T) at 91F-92F; Herbstein & Van Winsen The Civil Practice of the High Courts of South Africa (5^th ed) p 441. Only existing facts may be alleged. It is improper to allege facts in anticipation of an event.

[167] The following authorities demonstrate that the general principle against making out a new case in reply has indeed hardened in recent times. In Pilane and Another 2013 (4) BCLR 431 (CC), the majority of the Constitutional Court held, without any qualification, that it is impermissible for an applicant in motion proceedings to make out a new case in reply. An applicant must stand or fall by what is contained in his founding affidavit. The majority consequently disregarded the issues raised in the relevant parties’ replying papers in the court a quo when deciding the matter on appeal. In this regard, the Court said the following in a lengthy footnote:

‘In Director of Hospital Services v Mistry 1979 (1) SA 626 (AD) at 635H–636B, the Appellate Division held:

“When ... proceedings are launched by way of notice of motion, it is to the founding affidavit which a Judge will look to determine what the complaint is. As was pointed out by Krause J in Pountas’ Trustees v Lahanas 1924 WLD 67 at 68 and as has been said in many other cases: ‘... an applicant must stand or fall by his petition and the facts alleged therein and that, although sometimes it is permissible to supplement the allegations contained in the petition, still the main foundation of the application is the allegation of facts stated therein, because those are the facts which the respondent is called upon either to affirm or deny’.

Since it is clear that the applicant stands or falls by his petition and the facts therein alleged, ‘it is not permissible to make out new grounds for the application in the replying affidavit’ (per Van Winsen J in SA Railways Recreation Club and Another v Gordonia Liquor Licensing Board 1953 (3) SA 256 (C) at 260)."

[168] In South African Transport and Allied Workers Union and another v Garvas and others [2012] ZACC 13, 2013 (1) SA 83 (CC), 2012 (8) BCLR 840 (CC) (Garvas) at para [114], this Court held as follows:

“Holding parties to pleadings is not pedantry. It is an integral part of the principle of legal certainty which is an element of the rule of law, one of the values on which our Constitution is founded. Every party contemplating a constitutional challenge should know the requirements it needs to satisfy and every other party likely to be affected by the relief sought must know precisely the case it is expected to meet.”’

The rule against allowing new matter in reply was held in Shephard v Tuckers Land and Development Corporation (Pty) Ltd 1978 (1) SA 173 (W) 178A; and in Poseidon Ships Agency (Pty) Ltd v African Coaling and Exporting Co (Durban) (Pty) Ltd and Another 1980 (1) SA 313 (D) at 315F-316A, to be capable of being departed from only in special or exceptional circumstances. The latter decisions are, however, old and they predate the stance adopted by the Constitutional Court. Even if one were to rely on Shephard and Poisedon Ships Agency supra, one will be unable to find that such exceptional circumstances have been shown to exist in this matter.

[169] Quite apart from the new matter in reply, the hearsay evidence which is challenged comprises mainly – (a) uncorroborated press clippings; and (b) references to the representations, comments etc. of third parties without identifying them or providing confirmatory affidavits from them and the further replying affidavit of Mr Rossouw (dated 4 July 2017) consisting of references to social media pages and comments of and belonging to various third parties. The obvious needs to be mentioned. The prejudice to SU is clear inasmuch as the admission of such matter unnecessarily broadens the issues for determination and requires time and resources to answer. Of course SU does not enjoy the right to answer. In any event, I accept that, to do so would be wasteful in circumstances where it is unnecessary for purposes of determining the relief sought. The second striking out application is well grounded and stands to be granted.

Having regard to the issue of costs in respect of the main application, I can find no reason as to why costs, including in respect of the use of two counsel, should not follow the result. It was not in any event contended differently by counsel for either party.

ORDER

[170] In the result, the following orders are made in this matter:

(a) The application for condonation launched by the respondents seeking condonation for the late filing of the answering papers is hereby granted; the respondents are liable to pay the applicants’ costs in this regard as tendered.

(b) The application by the applicants for the admission of further affidavit deposed to by Mr Daniel Rossouw (‘Rossouw’) and for Mr Johan Theron to be subpoenaed and to testify to the content of his statement of defence, Annexure “DJR5” to Rossouw’s affidavit, is hereby dismissed with costs.

(c) The two striking out applications launched by the respondents are hereby granted; the applicants shall pay costs in this regard.

(d) The main application (in which orders are sought reviewing and setting aside the decisions of the Senate and Council of Stellenbosch University taken on 9 and 22 June 2016 respectively to adopt a new language policy for the Stellenbosch University in terms of Section 27 (2) of the Higher Education Act 101 of 1997 as well as the setting aside the 2016 Policy itself) is hereby dismissed with costs.

(e) The costs mentioned in (b), (c) and (d) above shall include the costs occasioned by the employment of two counsel and shall be paid by the applicants jointly and severally.

____________________________

D V DLODLO

Judge of the High Court

I agree.

_________________________

K M SAVAGE

Judge of the High Court

APPEARANCES:

For the Applicants: Adv. J Heunis (SC)

Instructed by DJ Rossouw of West & Rossouw

For the First to Third Respondents: Adv. J Muller (SC)

Adv. N De Jager

Instructed by L Van Niekerk of Cluver Markotter Inc.

Source: www.saflii.org

DOCUMENTS

The medical sector is sickening

ANC-EFF-ActionSA mayoral committee bad news - Cilliers Brink

Investigation claims by GoolamMV and News24 are false - Paul Mashatile

Allegations of multiple sex relations against Queenstown Girls High teacher shocking – Committee

Mbeki vs the race-baiters

The Donald or Kamala?

Going out with a bang

Wilgenhof: A reply to Adriaan Basson

“The most dangerous racist organisation”

Yahya Sinwar: The butcher of Khan Younis

🔒 The Economist: What to watch for on US election night, and beyond

NewsWrap: US election update; SAPS; Gauteng water crisis; Maimane on GNU; De Beer on Mashatile; Summers on Pick ‘n Pay

Why I came back to South Africa: brains gained explained

🔒 FT: Elon Musk has parked his Tesla on the White House lawn

SA politics at a crossroads: Maja and Maimane speak out on ANC’s decline and failing municipalities

iSERVICE FEED

"SAPS “Mafia”: The “biggest gang” in the country – charges laid against top cops…" - BizNews

"RW Johnson: The Suzman Foundation and Imtiaz Sooliman" - BizNews

"Integrity at stake: Wilgenhof; Justice Cameron; and Advocate de Jager – Martin Luitingh" - BizNews

"Rob Hersov: Why capitalism works and socialism does not" - BizNews

This content can be edited under Website Copy - Metered Paywall

DOCUMENTS