South Africa: Constitutional CourtYou are here: SAFLII >> Databases >> South Africa: Constitutional Court >> 1996 >>  ZACC 10 | Noteup | LawCite
Du Plessis and Others v De Klerk and Another (CCT8/95)  ZACC 10; 1996 (3) SA 850; 1996 (5) BCLR 658 (15 May 1996)
Download original files
Links to summary
Bookmark/share this page
CONSTITUTIONAL COURT OF SOUTH AFRICA
CASE NO CCT 8/95
In the matter of:
DU PLESSIS, D First Appellant (Defendant)
THE PRETORIA NEWS (PTY) LTD Second Appellant (Defendant)
LAUTENBACH, D Third Appellant (Defendant)
ALLIED PUBLISHING (PTY) LTD Fourth Appellant (Defendant)
DE KLERK, G F J First Respondent (Plaintiff)
WONDER AIR (PTY) LTD Second Respondent (Plaintiff)
Heard on: 7 November 1995
Delivered on: 15 May 1996
 KENTRIDGE AJ: The Pretoria News is a daily newspaper published in Pretoria. The first appellant is the editor of the Pretoria News, the second appellant is the owner and publisher of the newspaper, the third appellant is a journalist employed on the newspaper and the fourth appellant is its distributor. During February and March, 1993, the newspaper published a series of six articles dealing with the supply by air of arms and other material to the Angolan rebel movement, UNITA. The tenor of the articles was that South African citizens were engaged in these operations, that the operations were covert, and that they entailed the evasion of South African air control regulations. The flights were described in the articles as “illegal” and as “pirate flights.” The articles suggested that those responsible for the flights were “fuelling the war in Angola”, and were doing so for motives of personal gain, notwithstanding the disastrous effect of the Angolan civil war on the inhabitants of that country. The articles were published under the by-line of Dale Lautenbach, the third appellant.
 The last two in the series of articles, published on 9th and 11th March, 1993, mentioned by name Mr Gert de Klerk, the first respondent herein and his company Wonder Air (Pty) Ltd, the second respondent. The article published on 9th March, stated that the Department of Foreign Affairs had been calling in a number of private air operators “following suspicions that individual companies might be fuelling the war in Angola with supplies.” The first respondent was named as one of those summoned. The article published on 11th March, again in the context of illegal flights to supply the UNITA rebels, referred to “the mystery airstrip” owned and operated by the respondents. In consequence of these publications the respondents issued a combined summons out of the Transvaal Provincial Division of the Supreme Court claiming damages for defamation against the appellants jointly and severally. The first respondent claimed damages of R750 000.00 for injury to his reputation and his feelings; the second respondent claimed R5 million for loss of business and damage to its commercial reputation. I shall hereafter refer to the respondents as “the Plaintiffs” and to the appellants as “the Defendants.”
 On 25th May, 1993, the Defendants filed a joint plea. The Defendants admitted publishing the articles, but denied that they meant that the Plaintiffs were involved in illegal activities, or that the articles were defamatory of the Plaintiffs. In the alternative the Defendants alleged that the general subject matter of the articles was a matter of public interest. On this basis they pleaded a “rolled-up” defence of fair comment - namely that in so far as the references to the Plaintiffs were expressions of opinion, those opinions constituted fair comment made in good faith on matters of public interest, and were based on facts truly stated in the articles themselves; and that in so far as the articles contained allegations of fact those allegations were true and were matters of public interest. There was a further allegation by way of defence that the Defendants had published the articles in good faith in pursuance of a duty to its readers and to the public in general to keep them informed of “facts, opinions and allegations” concerning the civil war in Angola, that its readers had a corresponding right to be so informed and that in the premises the publication of the articles “was not unlawful.” All allegations of damage were denied.
 I have given only a brief and simplified summary of the Defendants’ plea because it is not in issue in the proceedings in this Court. What has brought the Defendants, as appellants, to this Court is the fate of an application to amend their plea by adding a further defence. Notice of intention to amend the plea was given by the Defendants on 7th October, 1994. The significance of this date is that it was subsequent to the coming into force of the interim Constitution on 27th April, 1994, in terms of section 251(1) of the Constitution of the Republic of South Africa Act 200 of 1993. The Plaintiffs objected to the proposed amendment, and it is necessary to set out in full both the proposed amendment and the grounds on which the Plaintiffs objected to it.
 The notice of intention to amend read as follows -
“KINDLY TAKE NOTICE that the defendants intend to amend their plea in the following way -
By the insertion after paragraph 12.14 of the following:
“12.15 In addition to the aforegoing, the publication of the article was not unlawful by reason of the protection afforded to the defendants by section 15 of the Constitution of the Republic of South Africa (Act 200 of 1993) which provides:
` (15) (1) Every person shall have the right to freedom of speech and expression, which shall include freedom of the press and other media, and the freedom of artistic creativity and scientific research.’
12.15.1 The articles in question were published against the background and in the circumstances described in paragraphs 12.1 - 12.9 hereof in good faith and without the intention of defaming the plaintiffs.
12.15.2 The articles concern matters of public interest and were published pursuant to a duty to keep members of the public informed of facts, opinions and allegations concerning the on-going civil war in Angola and a corresponding right or legitimate interest on the part of readers of the Pretoria News to be informed of such facts, opinions and allegations.
12.16 By virtue of the facts and contentions set out in paragraphs 12.15, the publication of the said articles were not unlawful and such publication is protected by section 15 of the Constitution.”
The grounds of objection were the following -
“The Plaintiffs object to the proposed amendment on the following grounds:
1. That the proposed amendment would render the Defendants’ plea excipiable;
2. The Constitution of the Republic of South Africa, Act 200 of 1993, was at no relevant stage in force when the Defendants published the defamatory material of and concerning the Plaintiffs;
3. The damage caused to the Plaintiffs consequent upon and as a result of the publication of the defamatory material was caused prior to the promulgation introduction of Act 200 of 1993;
4. The South African Constitution is not retroactive;
5. In the alternative, the Constitution has no application horizontally, alternatively does not apply to disputes of the present nature;
6. Further alternatively, Section 15 of the Constitution does not grant any of the Defendants leave and licence to publish defamatory material, either as alleged or at all;
7. In particular, Chapter 3 of the Constitution protects the Plaintiffs’ right to their physical and emotional integrity, reputation, unrestricted participation in public and commercial affairs and their right to an untarnished reputation;
8. These rights, inasmuch as they may come into conflict with the Defendants’ right to publish defamatory material (the existence of which right is denied), takes precedence over any right claimed by the Defendants; alternatively
9. The Defendants’ right to publish defamatory material (which is denied) is limited in terms of Section 33 of the Constitution and the common law by the Plaintiffs rights as aforesaid;
10. Consequently, the proposed amendment of the Defendants’ plea does not disclose a defence and should not be granted.”
(I have not corrected the grammatical errors in the two documents.)
 The opposed application to amend the plea was heard by Van Dijkhorst J in the Transvaal Provincial Division. On 10th November, 1994, he gave judgment refusing the application for amendment. The learned judge’s approach to the application was that an amendment which would render a pleading excipiable should not be allowed, and he held that the proposed amendment would be excipiable on two separate grounds. The first ground was that the proceedings before the court were “proceedings which immediately before the commencement of the Constitution were pending before any court of law ... exercising jurisdiction in accordance with the law then in force ...”, in terms of section 241(8) of the Constitution, and therefore had to “be dealt with as if this Constitution had not been passed.” This meant, according to the learned judge, that the provisions of the Constitution could not be invoked by any party to the pending proceedings. He followed his own judgment in Kalla v The Master and Others, in which he had given extensive reasons for the conclusion “that section 241(8) precludes retrospective operation of the Constitution.”
 The second ground on which Van Dijkhorst J held the proposed amendment to be excipiable was that set out in paragraph 5 of the Plaintiffs’ notice of objection, viz.-
“In the alternative the Constitution has no application horizontally, alternatively does not apply to disputes of the present nature.”
 The question whether Chapter 3 of the Constitution (Fundamental Rights) has only a “vertical” application or has in addition a “horizontal” application has been the subject of considerable debate by commentators on the Constitution. There have been similar debates, both academic and judicial, in other countries with constitutional Bills of Rights. The term “vertical application” is used to indicate that the rights conferred on persons by a Bill of Rights are intended only as a protection against the legislative and executive power of the state in its various manifestations. The term “horizontal application” on the other hand indicates that those rights also govern the relationships between individuals, and may be invoked by them in their private law disputes. Although the terms “vertical” and “horizontal” are convenient they do not do full justice to the nuances of the jurisprudential debate on the scope of Chapter 3. Does Chapter 3 entitle a party to private litigation to contend that a statute relied on by his opponent is invalid as being inconsistent with the Constitution? To what extent does Chapter 3 have an impact on the common law in either the criminal or the civil field? Does the vertical application of the Constitution cover private law disputes between a citizen and the state? These and no doubt other related questions are open questions in this Court at least. At this point in the present judgment it is sufficient to record that Van Dijkhorst J, upon an analysis of the relevant constitutional provisions, held that Chapter 3 had only vertical and not horizontal application, and that in consequence a defendant could not invoke section 15 as a defence to a civil action for damages for defamation.
 In due course the Defendants applied to Van Dijkhorst J for leave to appeal to this Court. The learned judge held that in view of conflicting decisions at first instance it was imperative that the constitutional issues which had been decided against the Defendants be resolved by the Constitutional Court. On 1st March, 1995, he accordingly referred those issues to this Court under section 102(2) of the Constitution, alternatively under section 102(8). Further in terms of Rule 18(e) of the Rules of the Constitutional Court he certified -
(1) These two Constitutional issues are of substance and a ruling thereon by the Constitutional Court is desirable.
(2) They can be disposed of on the pleadings and no evidence is necessary.
(3) In view of conflicting decisions in the Supreme Court on both issues there is a reasonable prospect that another court may reach a different conclusion should permission be granted to bring the appeal.
 On 9th June this Court granted leave to appeal against the whole of the judgment and order of Van Dijkhorst J of 10th November, 1994. As the issue of the correct interpretation of section 241(8) of the Constitution had in the interim been resolved by this Court in its judgment in S v Mhlungu and Others, this Court formulated the first issue on which it required argument in the appeal as follows -
“(a) Are the Defendants entitled to invoke the provisions of the Constitution notwithstanding that -
(i) publication of the offending material had already occurred; and/or
(ii) action was instituted; and/or
(iii) all relevant facts had occurred
before the Constitution came into operation?”
It also reformulated the judge’s question, “whether the Constitution has horizontal application.” The parties were asked to address this question -
“(b) Are the provisions of Chapter 3 of the Constitution - and more particularly section 15 - capable of application to any relationship other than that between persons and legislative or executive organs of state at all levels of government?”
 Thereafter, on 20th October, 1995, the parties were requested by the President of the Court to address the Court on the following additional matters:
“i. In view of the finding by the judge in the Court a quo that the proposed amendment does not raise the issue whether the common law of defamation should be developed to make it consistent with the Constitution, is it competent to raise this as an issue in the appeal?; and if so
ii. Is the development of the common law within the jurisdiction of the Appellate Division or the Constitutional Court or both Courts?; and if the latter
iii. Should the appeal on this issue have been noted to the Appellate Division and dealt with by it in terms of Section 102 (4), (5) and (6) of the Constitution?”
At the hearing before us on 7th November counsel addressed us on all the above issues.
 In their written argument the Defendants contended that the amendment which they had sought ought to have been granted. At an early stage of the oral argument, however, Mr Gilbert Marcus, who appeared for the appellant Defendants, was faced with a difficulty which proved to be insuperable. The Constitution, in terms of section 251(1), came into operation on 27th April 1994, and on that day a new legal order came into existence in the country. In S v Mhlungu and Others, supra n6, a case much relied on in the Defendants’ written argument, this Court held that from that day onward any person in South Africa was entitled to, and could invoke, the rights conferred by Chapter 3 of the Constitution. Cases such as Kalla v The Master and Others, supra n4, which had held, in reliance on section 241(8), that those rights were not available in proceedings which were pending immediately before the commencement of the Constitution, were overruled. The purpose of section 241(8), was held to be essentially to preserve the authority of pre-Constitution courts to continue to adjudicate in pending cases. On and after 27th April the Constitutional guarantees were available to accused persons in pending cases as they were to all other persons. Accordingly, Mhlungu and other persons accused in cases pending on 27th April, 1994, were entitled to invoke their constitutional rights so as to preclude the use against them of the presumption contained in section 217(1)(b)(ii) of the Criminal Procedure Act 51 of 1977, a presumption which this Court had held in S v Zuma and Others to be unconstitutional and hence invalid.
 It was in that limited sense, if at all, that S v Mhlungu and Others, supra n6, held that Chapter 3 had “retrospective” operation. It most certainly did not decide that the Constitution operated retroactively in the meaning which I endeavoured to explain in my dissenting judgment in that case. A statute is said to be retroactive if it enacts that “as at a past date the law shall be taken to have been that which it was not,” so as to invalidate what was previously valid, or vice versa. The Constitution does not operate retroactively in that sense. I do not believe that this proposition is in any way inconsistent with the majority judgments in Mhlungu’s case. Thus Kriegler J said, in paragraph 99 -
“In the true sense of the words it [i.e. the Constitution] is not retroactive nor retrospective. What it does mean, though, is that the moment when the judicial officer has to deal with a claim under Chapter 3 he or she has to ask whether such right exists.”
Mahomed J, in paragraphs 39 and 41 also made it clear that the Constitution did not affect acts performed before its commencement. See also per Sachs J paragraphs 132 and 144.
 Consequently, the difficulty facing the Defendants in this Court was their inability to point to anything in the Constitution which suggests that conduct unlawful before the Constitution came into force is now to be deemed to be lawful by reason of Chapter 3. Indeed, all indications in the text are to the contrary. First, there is section 251(1) itself, which fixes the date of commencement. Then there is section 7(2), which provides that Chapter 3 should apply “to all law in force and all administrative decisions taken and acts performed during the period of operation of this Constitution.” (My emphasis). In this sub-section “acts” may mean only administrative acts. Nonetheless if the provisions of Chapter 3 do not apply to administrative acts performed before the Constitution came into operation there is no reason to suppose that it was intended to apply to any other act performed before that date. Again section 98(6) provides -
“(6) Unless the Constitutional Court in the interests of justice and good government orders otherwise, and save to the extent that it so orders, the declaration of invalidity of a law or a provision thereof -
(a) existing at the commencement of this Constitution, shall not invalidate anything done or permitted in terms thereof before the coming into effect of such declaration of invalidity; or
(b) passed after such commencement, shall invalidate everything done or permitted in terms thereof.”
That sub-section enables this Court, where the interests of justice and good government require it, to ante-date the operation of a declaration of invalidity. Although there is no express limit on the power to ante-date a declaration of invalidity, it could hardly be suggested that any such declaration could refer to a date earlier than the date of the commencement of the Constitution. See the orders made by this Court in S v Zuma and Others, supra n9, and S v Mhlungu and Others, supra n6.
 It follows, as Mr Marcus was constrained to accept, that a pleading alleging that articles published in 1993 were, by reason of section 15 of the Constitution, “not unlawful” and were protected by that section, must be bad in law. The appeal against the order of Van Dijkhorst J must therefore be dismissed. That, however, does not conclude the proceedings before this Court. There is the judge’s reference of the issue of horizontality’ to this Court under section 102 of the Constitution to be considered. Further Mr Marcus on behalf of the Defendants has it in mind to apply in due course for an amendment to the plea so as to invoke section 15 of the Constitution on a different basis possibly by reference to section 35(3) of the Constitution. Whether he can invoke section 15 on any basis depends on the answer to the first issue on which this Court required argument.
 The Defendants argued that even if the Constitution does not make lawful what was previously unlawful, the protections of Chapter 3, including section 15, are available to relieve them from the consequences of a previously unlawful act. They rely by way of analogy on the right of persons convicted and sentenced before the commencement of the Constitution to invoke their constitutional right not to undergo cruel and inhuman punishment. The previous lawfulness of the sentence did not preclude their relying on their Chapter 3 rights to avoid its consequences. Similarly, they say, they are now entitled to rely on section 15 to relieve them from the obligation of paying damages for their earlier unlawful act.
 With all respect to the arguments of counsel, the analogy is false. This Court has held that the death penalty and the whipping of juveniles were in themselves unconstitutional and therefore unlawful by reason inter alia of section 11(2), which provides that no person shall be subject to cruel, inhuman or degrading punishment. Although the sentences were lawful when imposed, their execution became unconstitutional once the Constitution came into operation. The obligation to pay damages is obviously not in such a category. Another fundamental difference is that the commission of the delict and the liability to pay damages cannot be separated. The right to damages accrues at the moment the defamation is published. No-one could sensibly assert that the state has an accrued right to inflict a punishment. It cannot be disputed that since 27th April, 1994, the Defendants have been entitled to exercise their right of freedom of expression and freedom of the press under section 15. If their case on the interpretation of section 15 and on its horizontal application is correct, it may allow them to repeat their allegedly defamatory publications with impunity. But it is not in that sense that the Defendants wish to invoke their right of free speech.
 The Defendants also argue that it would be absurd and unjust to allow the
“arbitrary selection of one category of persons who would become entitled to enjoy the human rights guarantees of the Constitution and the arbitrary exclusion of another group of persons from such entitlement.”
The arbitrariness to which Mahomed J was referring related to the suggested exclusion of litigants in cases pending on 27th April, 1994, from the right to invoke constitutional guarantees after that date. As appears from section 7(2) of the Constitution, referred to above, there can be nothing arbitrary, absurd or unjust in the distinction between acts done (including delicts committed) before the Constitution commenced and those done thereafter.
 The Defendants also submit that the articles which are the subject-matter of the civil action could have led to a prosecution for the common law offence of criminal defamation. On the hypothesis that the existence of that common law offence is inconsistent with the right of freedom of speech under section 15, that section could be properly invoked, they say, as a defence to a prosecution notwithstanding the fact that the offence was committed before the Constitution came into force. The same principle might apply, it is suggested, to a prosecution for the common law crime of blasphemy. I shall assume for the purpose only of the present argument that that submission is correct. The attempt to extend it to civil law delictual claims is, however, unsound. At common law the statutory abolition of a criminal offence did not ordinarily affect a prosecution for an offence committed before the abolition. Under the Constitution different policy considerations may apply. The state may possibly be precluded from prosecuting for an offence which has by reason of the Constitution ceased to exist. The state cannot be said to have vested rights which will be affected, nor is any other person adversely affected and it may be said that to punish a person for an offence which has ceased to exist is an infringement of one or other of his protected fundamental rights. It is unnecessary and would be undesirable to express any view on these arguments. What is obvious is that very different considerations must apply to a civil claim for damages for defamation. There is another party whose rights would indeed be affected by depriving him of a claim for damages which had vested in him before the commencement of the Constitution. A right of action is a form of incorporeal property. Whether it is property entitled to protection under section 28 of the Constitution need not be decided. What is clear is that there is no warrant in the Constitution for depriving a person of property which he lawfully held before the Constitution came into force by invoking against him a right which did not exist at the time when the right of property vested in him. The Defendants’ citation of the well-known authorities on the need for a generous rather than a legalistic interpretation of a Constitution hardly supports an argument directed to depriving an individual of an existing right.
 I have dealt with the question of the retrospective or retroactive operation of Chapter 3 of the Constitution in general terms. As stated in paragraphs 13 and 14 above, the Constitution does not turn conduct which was unlawful before it came into force into lawful conduct. It does not enact that as at a date prior to its coming into force “the law shall be taken to have been that which it was not”. The consequences of that general principle are, however, not necessarily invariable. In the present case we are dealing with the right to damages for a defamation committed before the Constitution came into operation, and we hold that nothing in the Constitution impairs that right. But we leave open the possibility that there may be cases where the enforcement of previously acquired rights would in the light of our present constitutional values be so grossly unjust and abhorrent that it could not be countenanced, whether as being contrary to public policy or on some other basis. It is not necessary to spell out examples. It is sufficient to say that cases such as the one before us obviously do not fall into that category.
 I would therefore hold that the Defendants are not entitled to invoke section 15 as a defence to an action for damages for a defamation published before the Constitution came into operation. I have reached this conclusion without reference to foreign authority, but at this stage it may be appropriate to refer to some decisions on another constitutional instrument which has given rise to problems of retrospectivity in one sense or another, namely the Canadian Charter of Rights and Freedoms.
 The Canadian approach is summarised as follows by Professor PW Hogg -
“Section 58 of the Constitution Act, 1982 provides that the Act is to come into force on a day to be fixed by proclamation. That proclamation was issued by the Queen, who came to Canada for the purpose, at a ceremony in Ottawa on April 17, 1982; and the proclamation fixed April 17, 1982 as the day upon which the Constitution Act, 1982 was to come into force. The Charter of Rights accordingly came into force on that day, and operates only prospectively from that day.
A statute (or regulation or by-law or other legislative instrument) which was enacted before April 17, 1982, and which is inconsistent with the Charter, will be rendered “of no force or effect” by the supremacy clause of the Constitution, but only as from April 17, 1982. Action of an executive or administrative kind, such as search, seizure, arrest or detention, which was taken before April 17, 1982, cannot be a violation of the Charter, because the Charter was not in force at the time of the action.”
 In R. v. Longtin Blair JA, in the Ontario Court of Appeal, held that the Charter did not operate retrospectively. In the same Court, some years later, in R. v. James; R. v. Dzagic, Tarnopolsky JA remarked that that assertion of Blair JA had not been questioned, but added that the issue had rather been whether, in any particular case, giving effect to a Charter provision did or did not amount to a retrospective application. In that case the Ontario Court of Appeal held that section 8 of the Charter, which protects against unreasonable search and seizure had no application where the seizure took place before the Charter came into force, and that the material seized could accordingly be used in post-Charter proceedings. The Court held that the law to be applied was that in force at the time when the act complained of occurred. Our own Constitutional provisions and our own weighing of the competing public interests in South Africa may or may not produce a different approach to the admissibility in evidence of material wrongfully seized. That is not the issue here. What I would take from the case and respectfully endorse are some general remarks by Tarnopolsky JA at the end of his judgment -
“It is not an effective way to promote respect for Charter rights to apply new effects to actions taken before the Charter came into effect... it is important that actions be determined by the law, including the Constitution, in effect at the time of the action.”
An appeal from this judgment was dismissed, without written reasons by the Supreme Court of Canada. In another case in the Ontario Court of Appeal, R. v. Lucas; R. v. Neely, there were two prosecutions for the statutory offence of having sexual intercourse with a female under the age of fourteen. Although the offences occurred before the Charter came into force, a lower court had acquitted the accused on the ground that the statute was invalid, being inconsistent with the equal rights provision of the Charter. On appeal by the Crown to the Court of Appeal the submission that new substantive law should not be applied to past events was upheld.
 The generous approach of the Canadian courts to the interpretation of the Charter is well known. Perhaps, therefore, the Canadian cases put into perspective the Defendants’ contention that failure to uphold their submissions would result in absurdity and injustice.
 What remains to be considered, as far as the Defendants are concerned, is whether they can nonetheless derive any assistance from section 35(3) of the Constitution, a point related to the questions put to the parties by the President on 20th October, 1995. Before attempting to deal with those issues it is, however, necessary to revert to the second question referred to this court by Van Dijkhorst J, namely “whether the Constitution has horizontal application.”
 That reference was made under section 102(8) of the Constitution which provides-
“(8) If any division of the Supreme Court disposes of a matter in which a constitutional issue has been raised and such court is of the opinion that the constitutional issue is of such public importance that a ruling should be given thereon, it may, notwithstanding the fact that the matter has been disposed of, refer such issue to the Constitutional Court for a decision.”
In previous cases this Court has left open the precise connotation of the expression “disposes of a matter.” Whatever the precise scope of the expression, I have no doubt that in this case the learned judge had disposed of the matter before him. That matter was the application to amend the plea so as to introduce a new defence. His judgment refusing the amendment on the ground that the new plea would be bad in law, effectively eliminated that defence from the case.
 I find a useful analogy in the decisions of the Supreme Court on the appealability of judgments dismissing or upholding exceptions. The test applied is whether the order made has a final and definitive effect. Generally, the dismissal of an exception is not regarded as final, whereas the upholding of an exception to a pleading on the ground that it is bad in law is regarded as final and appealable. The reasons given for this distinction are instructive. In Trakman NO v Livshitz and Others a procedural application had been made in the court below and had been dismissed. The Appellate Division held that the order dismissing the application was appealable because it -
“ ... was final and not susceptible of alteration by the court a quo; it was definitive of the parties’ rights in respect of the application for review; and it disposed of all the relief claimed in such application”. (My emphasis)
In Liquidators, Myburgh, Krone & Co. Ltd v Standard Bank of South Africa Ltd and Another, in explaining why an order upholding an exception was final and therefore appealable, Innes CJ said -
“Where an order, though made during the progress of a litigation is not reparable at the final stage; or to put it another way, where the final word has been spoken on the point dealt with, then that order is final and not interlocutory”.
The order of Van Dijkhorst J did dispose finally of all the relief claimed in the application for amendment. He spoke the final word on that application. The conclusion that the judge had disposed of the case before him is reinforced by the consideration that the trial of the action need not be heard by the same judge.
 This conclusion is not affected by the possibility that an appeal may lie against the decision of Van Dijkhorst J. Section 102(8) refers to “any division of the Supreme Court”, which indicates that the power conferred is not limited to a court of final appeal.
 As to policy and convenience, I cannot see why the framers of the Constitution should have wished to exclude from the operation of sub-section (8) a case such as this one, where there has been a claim for specific relief and that claim has been finally disposed of. I see no reason of policy why, before a referral, the whole of any relevant proceedings must be completed, proceedings which may be protracted and which have no bearing on the constitutional issue. The whole basis for a referral under sub-section (8) is that a constitutional issue of great public importance has been raised. As far as the proceedings before the Supreme Court are concerned the issue may be moot. The losing party may not wish to appeal, or the parties may have reached a settlement. Nonetheless, provided there is a compelling public interest, the constitutional issue may properly be referred. I would add that a referral such as this does not disturb the “logic” of the appeal routes provided in the Constitution. Theoretically, no doubt, the learned judge might have granted leave to appeal to a full bench of the Transvaal Provincial Division but, given that the constitutional issue is of such public importance as to call for a referral to this Court, that possibility can be disregarded. In practical terms this is the only Court competent to review the judgment of the learned judge on the constitutional issues. Before sending the case to this Court he had dealt fully with those issues. I can discern no ground on which his referral can be faulted.
 Accordingly, although the appeal has been dismissed without the necessity of dealing with the “horizontality” issue the referral on that issue remains to be dealt with by this Court. Whether in any circumstances this Court has a discretion to refrain from deciding an issue validly referred to it I need not now decide. Even if such a discretion exists I would not exercise it, notwithstanding the dismissal of the appeal. The issue is plainly of public importance, especially in the light of the conflicting decisions in the Supreme Court referred to by the learned judge, and has been the subject of written and oral argument before us. (I add, in parenthesis, that the alternative referral under section 102(2) was not appropriate. See our decisions in S v Mhlungu and Others and S v Vermaas; S v Du Plessis.)
 The “horizontality” issue has arisen in other countries with entrenched Bills of Rights and the parties have supplied us with a wealth of comparative material both judicial and extra-judicial, for which we are grateful.
 In the court below the learned judge, having endorsed the purposive approach to constitutional interpretation, analysed the purpose of the Chapter on Fundamental Rights as follows -
“When interpreting the Constitution and more particularly the Bill of Rights it has to be done against the backdrop of our chequered and repressive history in the human rights field. The State by legislative and administrative means curtailed the common law human rights of most of its citizens in many fields while the Courts looked on powerless. Parliament and the executive reigned supreme.
It is this malpractice which the Bill of Rights seeks to combat. It does so by laying down the ground rules for State action which may interfere with the lives of its citizens. There is now a threshold which the State may not cross. The Courts guard the door.”
Having considered the interpretation of entrenched Bills of Rights in the Constitutions of other countries, he concluded that in general, fundamental rights are protected against state action only. “Horizontal protection,” he said,
“sometimes occurs to a limited extent but when it is intended over the broad field of human rights, it is expressly so stated”
Horizontal application of Chapter 3 would in his view create an undesirable uncertainty in private legal relationships which could not have been intended by the framers of our Constitution. After an analysis of certain provisions of the Constitution he held that the fundamental rights set out in Chapter 3 were of vertical application only, and that the contrary conclusion of Van Schalkwyk J in Mandela v Falati was clearly wrong. It should be noted that in Motala and Another v University of Natal Hurt J refused to follow the opinion of Van Dijkhorst J and held that at least sections 8 (equality) and 32 (education) had horizontal application. In Potgieter en n Ander v Kilian the Natal Provincial Division disagreed with these two judgments and endorsed the opinion of Van Dijkhorst J.
 There can be no doubt that the resolution of the issue must ultimately depend on an analysis of the specific provisions of the Constitution. It is nonetheless illuminating to examine the solutions arrived at by the courts of other countries. The Court was referred to judgments of the courts of the United States, Canada, Germany and Ireland. I would not presume to attempt a detailed description, or even a summary, of the relevant law of those countries, but in each case some broad features are apparent to the outside observer. A comparative examination shows at once that there is no universal answer to the problem of vertical or horizontal application of a Bill of Rights. Further, it shows that the simple vertical/horizontal dichotomy can be misleading. Thus under the Constitution of the United States the First to Tenth Amendments (the “Bill of Rights”) and the Fourteenth Amendment, insofar as they confer rights on individuals, would at first sight appear to be vertical, in the sense of being directed only against state power. Yet the courts of that country have in some cases at least reached what is effectively a horizontal application of constitutional rights by holding that the judicial power is a state power against which constitutional protections may invoked.
 So, in Shelley v. Kraemer an African-American couple had bought property which was subject to a restrictive covenant under which the seller had undertaken to sell only to whites. Owners of restricted property in the same neighbourhood sued to prevent the couple from taking possession of the property. The United States Supreme Court reiterated earlier holdings that the Fourteenth Amendment did not reach private conduct, however discriminatory, but held that official actions by state courts and judicial officials were subject to the Fourteenth Amendment, with the result that the discriminatory covenant could not be enforced by the courts. Vinson CJ said -
“... state action in violation of the Amendment’s provisions is equally repugnant to the constitutional commands whether directed by state statute or taken by a judicial official in the absence of statute.”
It was on this principle that the United States Supreme Court was able to hold in New York Times Co. v. Sullivan, an action between private litigants, that the law of defamation of the State of Alabama was an unconstitutional impairment of the right of freedom of speech. A complex case law suggests that the rule in Shelley v. Kraemer, supra n47, is not invariably available in private law disputes. The reasoning behind the decision has also been cogently criticised. It may nonetheless be accepted that by identifying some state involvement in private transactions (sometimes with great ingenuity) United States’ courts have found a way of enforcing fundamental constitutional rights in disputes between private litigants.
 Irish cases indicate that in some instances at least, constitutional rights have been directly applied in private disputes so as to override a rule of common law. An example is C.M. v T.M. in which Barr J held that the common law doctrine that a wife’s domicile was dependent on that of her husband was inconsistent with the principles of equality before the law and equality between husband and wife embodied in Articles 40 and 41 of the Irish Constitution.
 Very different models of constitutional adjudication are to be found elsewhere. There is a valuable comparative overview of the application of constitutional rights in the private law of a number of countries in Constitutional Human Rights and Private Law, a work by Justice A. Barak, of the Supreme Court of Israel, from which it appears that there are several jurisdictions which reject the horizontal application, or at least the direct horizontal application of constitutional rights. I propose to confine my further consideration of the comparative material to the Canadian and German position, particularly as argument on these two systems was specifically addressed to us.
 The leading Canadian case is Retail, Wholesale & Department Store Union, Local 580 et al. v. Dolphin Delivery Ltd. a judgment of the Supreme Court (to which I shall refer hereafter as Dolphin Delivery). That case arose from a labour dispute, in which the defendant trade union threatened to picket the plaintiff’s premises unless it ceased to do business with another company with which the union was in dispute. A trial judge found that the defendant’s conduct constituted the tort of inducing a breach of contract and granted an injunction restraining the threatened picketing. The union appealed on the ground that the injunction infringed its Charter right of freedom of expression. In dismissing the appeal the court held (among other grounds) that while the Charter applied to common law as well as statute law, it did not apply in litigation between private parties in the absence of any reliance on legislation or governmental action. McIntyre J, who gave the leading judgment, based his judgment on the terms of section 32 of the Charter which expressly provide that the Charter applies to “the Parliament and government of Canada” and to “the legislature and government of each province.” By “government,” he held, was meant the executive and administrative branch of government. An order of court was not to be equated with governmental action.
 The essence of the court’s conclusion is to be found in the following passage from the judgment of McIntyre J -
“It is my view that s. 32 of the Charter specifies the actors to whom the Charter will apply. They are the legislative, executive and administrative branches of government. It will apply to those branches of government whether or not their action is invoked in public or private litigation. It would seem that legislation is the only way in which a legislature may infringe a guaranteed right or freedom. Action by the executive or administrative branches of government will generally depend upon legislation, that is, statutory authority. Such action may also depend, however, on the common law, as in the case of the prerogative. To the extent that it relies on statutory authority which constitutes or results in an infringement of a guaranteed right or freedom, the Charter will apply and it will be unconstitutional. The action will also be unconstitutional to the extent that it relies for authority or justification on a rule of the common law which constitutes or creates an infringement of a Charter right or freedom. In this way the Charter will apply to the common law, whether in public or private litigation. It will apply to the common law, however, only in so far as the common law is the basis of some governmental action which, it is alleged, infringes a guaranteed right or freedom.”
What follows from this is - (a) if a party to private litigation founds a claim or defence on some piece of legislation (whether an act of Parliament, a by-law or regulation) or on some executive act, (such as the issue of a licence) its constitutionality under the Charter is an issue which may properly be raised; (b) in litigation between private parties no inconsistency between the common law and the Charter may be relied on; but (c) the Charter applies to the common law in a dispute between government and a private litigant - for example where the government relies on a common law prerogative. (In a subsequent case the Canadian Supreme Court has held that the Charter applies to the state even in respect of activities which are contractual or commercial in nature). The Defendants in the present case point to differences in wording between the Charter and our own Constitution, and deny that Dolphin Delivery provides any assistance in interpreting the latter. They have also referred us to the academic criticisms of Dolphin Delivery noted by Friedman JP in Baloro and Others v University of Bophuthatswana and Others. I shall return to Dolphin Delivery later in this judgment.
 The German jurisprudence on this subject is not by any means easy to summarise, especially for one who does not read German. There are, however useful, accounts of the German approach in some of the South African literature, as also in the work of Justice Barak, which I have mentioned above. I have also had the benefit of reading an extensive article entitled “Free Speech and Private Law in German Constitutional Theory” by Professor Peter E. Quint, to which I am much indebted.
 The German model may be described as the indirect application model. The rights of individuals entrenched in the Basic Law are directly available as protection against state (including legislative) action, but do not directly apply to private law disputes. The values embodied in the Basic Law do, however, permeate the rules of private law which regulate legal relations between individuals. A constitutional right may override a rule of public law, but it is said to “influence” rather than to override the rules of private law. Private law is therefore to be developed and interpreted in the light of any applicable constitutional norm, and continues to govern disputes between private litigants. Private law rules are not completely superseded. This approach was authoritatively laid down by the German Constitutional Court in the leading case of Lhth, a case concerning the right of free expression under Article 5 of the Basic Law. Later cases, such as the Mephisto case in 1971, and the Deutschland-Magazin case in 1976, established that it was for the ordinary courts to apply the constitutional norms to private law. This was likely to involve a balancing of constitutionally protected interests against one another (for example the right of free expression against the right of human dignity under Article 1) or against established private law rights such as confidentiality or privacy. The facts of the particular case are also to be taken into account in the balancing process. The German Constitutional Court will exercise, if necessary, a power of review, but it will do so with restraint - usually only when it is satisfied that the ordinary courts have proceeded on a seriously wrong interpretation of the basic constitutional rights under Basic Law. Quint makes two comments of particular interest. One is that the deference of the Constitutional Court to the ordinary courts on questions of private law stems from the fact that, unlike the United States Supreme Court, its basic function is to decide constitutional questions only. This consideration may prove in due course to have some relevance to the practical application of section 35(3) of our own Constitution. The second is that in some cases the impact of the German Basic Law upon private law under the “indirect”doctrine may be stronger than that of the United States Constitution on American common law under the “state action” doctrine, precisely because the ordinary German courts are entitled and obliged to take the Basic Law into account without searching for an element of state action.
 The doctrine of the application of the norms of the Basic Law in the field of private law (“Drittwirkung”) is subtle and is the subject of considerable debate in Germany itself. The analyses of Justice Barak and Professor Quint might not command universal acceptance, still less my own brief interpretation of the doctrine. It is not, however, my purpose to provide a definitive statement of German law, even if I were competent to do so. The purpose of this perhaps overlong account of constitutional adjudication elsewhere is to see what guidance it might provide in the interpretation of the South African Constitution. In my opinion there is at least one positive lesson to be learnt from the Canadian and German approaches to the problem before us. Both Canada and Germany have developed a strong culture of individual human rights, which finds expression in the decisions of their courts. Yet, after long debate, both judicial and academic, in those countries, the highest courts have rejected the doctrine of direct horizontal application of their Bills of Rights. On this issue, as on the retrospectivity issue, the example of these countries seriously undermines the Defendants’ contention that anything other that a direct horizontal application of Chapter 3 must result in absurdity and injustice.
 As I have already indicated the issue of horizontal or vertical application of Chapter 3 has been hotly debated in the South African legal literature. Arguments of substance have been deployed on both sides of the debate. I have read much of this literature, I hope with advantage. It is not out of any disrespect to the authors that I refrain from listing all those to be found on each side of the controversy, or from analysing their respective arguments. I propose instead to turn without further delay to consider what I take to be the relevant provisions of the Constitution.
 In relation to the application of Chapter 3 of the Constitution there are, as Professor Cockrell has explained, two inter-related but nonetheless different questions to be considered. The first is to what law the Chapter applies - does it apply to the common law, or only to statute law? The second question is what persons are bound by the Chapter - do the rights give protection only against governmental action or can they also be invoked against private individuals? There are, of course, subsidiary questions, such as what bodies can be considered to be organs of government, and whether executive action in the private law sphere is “governmental.”
 The plain answer to the first question emerges from section 7(2) of the Constitution, which states -
“This Chapter shall apply to all law in force and all administrative decisions taken and acts performed during the period of operation of this Constitution.”
The words “all law in force” may have some ambiguity, in that they are capable of being read as being limited to statute law. However, any ambiguity is removed by the Afrikaans version, where the equivalent words are “alle reg wat van krag is.” The word “reg” (as distinct from “wet”) unambiguously embraces common law as well as statute law. Although the Afrikaans version of Act 200 of 1993 was the original signed version, by virtue of section 15 of Act 2 of 1994 the English version is deemed to be the signed version. The latter version would therefore prevail in case of a conflict between the two versions. But where there is no conflict between them there is another well-established rule of interpretation: if one text is ambiguous, and if the ambiguity can be resolved by the reference to unambiguous words in the other text, the latter unambiguous meaning should be adopted. There is no reason why this common-sense rule should not be applied to the interpretation of the Constitution. Both texts must be taken to represent the intention of Parliament. Moreover, Afrikaans remains an official language with undiminished status in terms of section 3 of the Constitution. The term “reg” is used in other parts of Chapter 3 as the equivalent of “law,” for example in section 8 (“equality before the law”) and section 33(1) (“law of general application”). Express references to the common law in such sections as 33(2) and 35(3) reinforce the conclusion that the law referred to in section 7(2) includes the common law, and that Chapter 3 accordingly affects or may affect the common law. Nor can I find any warrant in the language alone for distinguishing between the common law of delict, contract, or any other branch of private law, on the one hand, and public common law, such as the general principles of administrative law, the law relating to acts of state or to state privilege, on the other. By contrast, many provisions of the Constitution use the word “wet” as the equivalent to “law”, in contexts which may assist in finding the answers to the second question.
 The second question too seems to have a plain answer. Section 7(1) states -
“This Chapter shall bind all legislative and executive organs of state at all levels of government.”
Entrenched Bills of Rights are ordinarily intended to protect the subject against legislative and executive action, and the emphatic statement in section 7(1) must mean that Chapter 3 is intended to be binding only on the legislative and executive organs of state. Had the intention been to give it a more extended application that could have been readily expressed. One model which would have been available is Article 5 of the Namibian Constitution, which provides -
“The fundamental rights and freedoms enshrined in this Chapter shall be respected and upheld by the Executive, Legislature and Judiciary and all organs of the Government and its agencies and, where applicable to them, by all natural and legal persons in Namibia, and shall be enforceable by the Courts in the manner hereinafter prescribed.”
It would be surprising if as important a matter as direct horizontal application were to be
left to be implied.
 Another strong indication that a general horizontal application was not intended is section 33(4) -
“This Chapter shall not preclude measures designed to prohibit unfair discrimination by bodies and persons other than those bound in terms of section 7(1).” (My emphasis)
If Chapter 3 has a general horizontal application, who can the bodies and persons be who are not bound? Then there is section 35(3) -
“In the interpretation of any law and the application and development of the common law and customary law, a court shall have due regard to the spirit, purport and objects of this Chapter.” (My emphasis)
Again, one asks why such a provision would be needed if the Chapter could be directly applied to common law disputes between private litigants.
 Nor do I believe that the absence of reference to the judiciary in section 7(1) is an oversight. One of its effects is to exclude the equation of a judgment of a court with state action and thus prevent the importation of the American doctrine developed in Shelley v. Kraemer, supra n47. This Court, like the provincial and local divisions of the Supreme Court, is bound to apply the law, which in a proper case includes Chapter 3 but that does not permit the courts to ignore the limitation contained in section 7(1). It has, I believe, sometimes been suggested that section 7(2) somehow overrides or extends section 7(1). This reading, unpersuasive in itself, results from a failure to keep in mind the two different questions which I earlier identified. Section 7(1) answers one of them, section 7(2) the other. It may be asked why then in private litigation a litigant may contend that a statute relied on by the other party is invalid as being unconstitutional. That such a contention is open to a litigant is hardly disputable. There are two reasons why it must be so. First, as Chapter 3 expressly binds the legislature, every person is protected against the operation of unconstitutional legislation. Second, section 4 of the Constitution (which is outside Chapter 3) provides -
“(1) This Constitution shall be the supreme law of the Republic and any law or act inconsistent with its provisions shall, unless otherwise provided expressly or by necessary implication in this Constitution, be of no force and effect to the extent of the inconsistency.
(2) This Constitution shall bind all legislative, executive and judicial organs of state at all levels of government.”
In this section the Afrikaans equivalent of “law” (in the phrase any law or act’) is not “reg” but “wet”, which unambiguously connotes a statute. This means that any statute inconsistent with the Constitution is of no force and effect. Any litigant must therefore be able to rely on this section in any litigation. To adopt the language of the reformulation of the referral issue (b) set out in paragraph 10 above, any litigant contesting the constitutionality of a statute is applying Chapter 3 to the relationship between himself and the legislature, not to his relationship to the opposing (private) litigant.
 Having referred to section 4 of the Constitution I should deal briefly with an argument which has been raised in the literature, namely that section 4, by nullifying any law inconsistent with the provisions of the Constitution, implies that the common law governing relations between individuals also falls to be tested directly against the provisions of Chapter 3. Quite apart from the consideration that “law” in section 4 apparently means statute law, the argument overlooks the proviso -
“unless otherwise provided expressly or by necessary implication in this Constitution”.
If on a proper construction of Chapter 3 its operation is intended to be vertical only, the argument based on section 4 loses any force which it may have had.
 To recapitulate, by reason of the sections to which I have referred -
a) Constitutional rights under Chapter 3 may be invoked against an organ of government but not by one private litigant against another.
b) In private litigation any litigant may nonetheless contend that a statute (or executive act) relied on by the other party is invalid as being inconsistent with the limitations placed on legislature and executive under Chapter 3.
c) As Chapter 3 applies to common law, governmental acts or omissions in reliance on the common law may be attacked by a private litigant as being inconsistent with Chapter 3 in any dispute with an organ of government.
In sub-paragraph (c) I refer to “governmental acts or omissions”. For the purposes of this judgment it is unnecessary to attempt to define that concept. In particular, I leave open the question whether (as in Canada) it would include state activities in the commercial or contractual sphere.
 In argument before us it was urged that this result was anomalous. It is fortuitous in modern times whether a rule of private law remains a common law rule or is embodied in a statute. Examples were given of some rules of common law which may be inconsistent with the rights of the individual set out in Chapter 3. It is also pointed out that some statutes embody the common law, and that various statutes have altered the common law in some parts of South Africa but not others. Thus the statute abolishing the marital power does not apply in the territories of the former Transkei, Bophuthatswana or Venda. Other examples mentioned were common law crimes such as blasphemy or criminal defamation which, it was said, may be inconsistent with Chapter 3 rights; if so, they must be susceptible to attack although no statute is involved.
 Pausing to remark that difficulties and anomalies arise on the vertical as well as horizontal approaches, I believe, with all respect to the submissions of counsel and of those writers who support them, that the supposed irrationalities of the vertical interpretation are exaggerated. Such as there may be flow from the structure and wording of the Constitution. This requires further analysis. I have already pointed out that in some parts of the Constitution “law” (“wet” in the Afrikaans) means statute law, at whatever level. One instance is section 4, which nullifies statutes inconsistent with the Constitution, but not common law rules. This distinction between common law and statute becomes of primary importance in relation to section 98, the section which confers jurisdiction on this Court. Under section 98(2) this Court has jurisdiction, “as the Court of final instance over all matters relating to the interpretation, protection and enforcement of the provisions of this Constitution.” Then certain specific matters are set out, including -
“(c) any inquiry into the Constitutionality of any law, including an Act of Parliament, irrespective of whether such law was passed or made before or after the commencement of this Constitution.”
Here too the Afrikaans text has “wet”, and the reference to passing or making a law is obviously inappropriate to a rule of common law. There is no similar reference to the constitutionality of any rule of common law. Sub-sections 98(5) and 98(6) also relate to statute law only. They provide that if the Constitutional Court finds a law (“wet”) inconsistent with the Constitution it shall declare such law invalid to the extent of its inconsistency. The Court may require Parliament or any other competent authority to correct the defect in the law (again “wet”) during which time that law remains in force.
 The operation of a declaration of invalidity of a law (“wet”) is dealt with in sub-section (6), but section 98 nowhere provides for a declaration that a rule of common law is invalid. Such a declaration would be highly unusual, and would give rise to much difficulty. If a statute, including one embodying a private law rule, is struck down, the previous common law (or earlier statute law) is presumably restored. But what would result from holding a rule of common law to be unconstitutional? What would follow is that the relevant common law would require to be reformulated. But reformulation of the common law is the task of the Supreme Court. In Shabalala, supra n80, we held that the state’s claim of docket privilege was inconsistent with the Constitution. The extent of the inconsistency was defined in the order in that case, but as was stated in paragraph 58 of the judgment of Mahomed DP, “the details as to how the Court should exercise its discretion in all these matters must be developed by the Supreme Court from case to case, but always subject to the right of an accused person to contend that the decision made by the court is not consistent with the Constitution”. This Court’s jurisdiction derives only from section 98. Unlike the Supreme Court of the United States, the Australian High Court or the Supreme Court of Namibia, it has no inherent or general jurisdiction. It cannot re-write the common law governing private relations. If this is borne in mind most if not all the suggested irrationality of the vertical doctrine disappears.
 In many cases a holding of unconstitutionality would leave a gap in the law. Take the rule of the common law referred to in the Defendants’ Heads of Argument that the widow of a customary union has no action for loss of support. If that rule were held to be unconstitutional what specific rights are to be accorded the widow, having regard to other rules of customary law regarding widowhood? To take another of the examples put before us, assume that, in the absence of a statute, the marital power at common law were to be “struck down” as unconstitutional, how would existing marriages in community of property be dealt with? Section 11(3) of the Matrimonial Property Act 88 of 1984, as amended by section 30 of Act 132 of 1993, which statutorily abolished the marital power, provided a detailed regime for the governance of marriages in community of property. This Court would have had no power to fill the gap. Defendants point out that if this is so, striking down a statute may leave an even worse common law regime in place. The lesson is to be circumspect in attacking statutes. The radical amelioration of the common law has hitherto been a function of Parliament; there is no reason to believe that Parliament will not continue to exercise that function.
 Where the state in its executive or administrative capacity is concerned there is no difficulty in the vertical application of Chapter 3 against it in the field of common law. If the common law offences of blasphemy and defamation are incompatible with the provisions of Chapter 3, the executive action of the state in prosecuting and inflicting punishment for those offences could be called into question. This is provided for in paragraph (b) of section 98(2) which gives this Court jurisdiction in any dispute over the constitutionality of any executive or administrative act or conduct. In this regard section 98(7) provides -
“In the event of the Constitutional Court declaring an executive or administrative act or conduct or threatened executive or administrative act or conduct of an organ of state to be unconstitutional, it may order the relevant organ of state to refrain from such act or conduct, or, subject to such conditions and within such time as may be specified by it, to correct such act or conduct in accordance with this Constitution.”
It was by reason of these provisions that we were able in Shabalala to declare that certain practices hitherto adopted by prosecuting authorities were unconstitutional and to state what was required of them so as to ensure that an accused’s right to a fair trial was not infringed. It will not have been overlooked that there is no provision similar to section 98(7) in relation to private persons - a strange hiatus if horizontality were intended.
 Another pointer in the same direction is section 33(1) which provides that rights entrenched in Chapter 3 may be limited by law of general application. That “law” may be common law, but the problem of applying section 33(1) to private relationships governed by the common law seems almost insurmountable. The common law addresses problems of conflicting rights and interests through a system of balancing. Many of these rights and interests are now recorded in the Constitution and on any view that means that as a result of the terms of the Constitution the balancing process previously undertaken may have to be reconsidered. A claim for defamation, for instance, raises a tension between the right to freedom of expression and the right to dignity. The common law compromise has been to limit both rights to a certain extent, allowing damages to be recovered for what is regarded as “unlawful expression” but allowing “dignity” to be infringed in circumstances considered to be privileged. Section 33(1) could hardly be applied to such a situation.
 I have arrived at the conclusions set out above without any reference to the drafting history of Chapter 3, and in particular of section 7. We heard no argument on that history, but it is referred to frequently in the literature which I have cited. It is perhaps sufficient to say that there is nothing in the legislative history referred to in that literature which requires the adoption of the horizontal interpretation. Nor have I so far referred in any detail to the considerations of policy which point to the vertical solution as the correct one. One consideration is adverted to by McIntyre J in Dolphin Delivery, supra at 196 -
“While in political science terms it is probably acceptable to treat the courts as one of the three fundamental branches of government, that is, legislative, executive, and judicial, I cannot equate for the purposes of Charter application the order of a court with an element of governmental action. This is not to say that the courts are not bound by the Charter. The courts are, of course, bound by the Charter as they are bound by all law. It is their duty to apply the law, but in doing so they act as neutral arbiters, not as contending parties involved in a dispute. To regard a court order as an element of governmental intervention necessary to invoke the Charter would, it seems to me, widen the scope of Charter application to virtually all private litigation. All cases must end, if carried to completion, with an enforcement order and if the Charter precludes the making of the order, where a Charter right would be infringed, it would seem that all private litigation would be subject to the Charter.”
Those remarks seem to me to be fully applicable to Chapter 3 of our own Constitution.
 The limitation of the jurisdiction of this Court to constitutional matters, and the preservation of the role of the Appellate Division as the final court of appeal in other matters also appear to me to lead inexorably to the conclusion that Chapter 3 is not intended to be applied directly to common law issues between private litigants. Section 101(5) of the Constitution states-
“The Appellate Division shall have no jurisdiction to adjudicate any matter within the jurisdiction of the Constitutional Court”.
Numerous provisions of Chapter 3 could and would be invoked in private litigation if direct horizontal application of the Chapter were permitted. For example, cases of injuria including defamation, invasion of privacy and breach of confidentiality would call for the application of sections 10, 13 and 15 of the Constitution. Section 15 would also be relevant to civil contempts of court. In employment cases sections 8 and 27 would apply. Section 26 would be applied to contracts in restraint of trade and section 29 in actions for nuisance. Section 30 would be applied in custody and maintenance cases. The consequence would be that appeals in all such cases would lie to the Constitutional Court, and the Appellate Division would be deprived of a substantial part of what has hitherto been its regular civil jurisdiction. At the very least, appeals to the Appellate Division would routinely result in referrals of common law cases to the Constitutional Court. I do not believe that such a state of affairs could ever have been intended by the framers of the Constitution.
 Our jurisdiction under section 98 is not suited to the exposition of principles of private law. I have made this point in relation to the Matrimonial Property Act 1984. The common law of defamation illustrates this point even more clearly. We are asked to find that the law currently applied by the courts is inconsistent with section 15 of the Constitution. Let that be so. What regime is to replace the existing law? In the development of the common law of defamation a multitude of choices is available. The Defendants, it would seem from their written arguments, are attracted by the far-reaching revision of the common law adopted by the United States Supreme Court in New York Times Co. v. Sullivan, supra n49, in terms of which a “public
person”, however grossly defamed in relation to his or her public conduct, can only succeed in an action for defamation by proving that the defamatory statement was false and, what is more, by proving with “convincing clarity” that it was made by the defendant with knowledge of its falsity or with reckless disregard whether it was false or not. I would suggest that before adopting this rule as part of our law, a court would have to consider among other things the sharp criticisms of that rule both academic and judicial, within the United States, and its rejection by the Supreme Court of Canada in Hill v. Church of Scientology of Toronto. Presumably a court would also wish to consider the rule adopted by the High Court of Australia in the interests of freedom of speech, namely that in an action for defamation by a person engaged in politics or government it is a defence for the defendant to prove that he honestly and reasonably believed in the truth of what he published. The Australian rule introduces the concept of a duty to exercise care into the law of defamation. A South African court would have to consider the appropriateness of introducing such an element into a delict of intent (injuria) in which hitherto culpa has not been an element. It would also doubtless consider whether the Australian rule was not right in placing the burden of proof on the defendant rather than the plaintiff - in that respect among others refusing to follow New York Times Co. v. Sullivan, supra n49. At least equally important would be the consideration of the development of the South African law of defamation. Unlike some of the other rights embodied in Chapter 3, freedom of speech and of the press is not a newly created right. When not suppressed or restricted by statute it was emphatically endorsed and vindicated in many judgments of South African courts. Any law of defamation is a restriction on freedom of speech in the interest of other rights thought worthy of protection. More particularly, in cases of defamation, courts have tried to strike a balance between the protection of reputation and the right of free expression. Presumably, too, a court would wish to take account of the fact that our Constitution, like that of Germany but unlike that of the United States, expressly recognises the right to dignity and to personal privacy, and might find guidance in the German cases to which I have referred as well as in the American cases. On the other hand a court might also wish to consider the desirability of cutting down the concept of a defamatory statement in the interests of freer political criticism. It may similarly consider whether the rule that the press and the broadcasting media, unlike other litigants, cannot avail themselves of the defence of absence of animus injuriandi ought to be varied in the light of the values embodied in section 15 of the Constitution. Those values might also require the development of a broader concept of the public interest, entailing a reconsideration of the Neethling case, supra n2. For present purposes the point is that these are not choices which this Court can or ought to make. They are choices which require consideration perhaps on a case by case basis by the common law courts. The common law, it is often said, is developed on incremental lines. Certainly it has not been developed by the process of “striking down”.
 The consequences which I have outlined in paragraph 57 above are well illustrated by the judgment of Cameron J in Holomisa v Argus Newspapers Ltd to which I referred in paragraph 21 above. The learned judge had regard to section 15 of the Constitution and to much South African and foreign case law, and considered various possible forms which a law of defamation might take. In the context of the case before him he fashioned a principle of the law of defamation which is completely novel in this country. Whether his reformulation of the law is a desirable one is a question quite outside the purview of this judgment. He reached his conclusion by attempting to apply the precepts of section 35(3) - a provision to which I shall advert in a subsequent paragraph - and not by a direct application of section 15. If, however, section 15 had a direct horizontal application the task of formulating an appropriate law of defamation would fall to this Court on appeal. But that could not be reconciled with our limited jurisdiction under section 98(2). What is in my view certain is that section 15 of the Constitution does not mandate any particular rule of common law. Our jurisdiction, which is to interpret, protect and enforce the provisions of the Constitution, cannot empower us to choose one among a number of possible rules of common law all of which may be consistent with the Constitution. It would be equally impossible, for reasons which I have already explained, for this Court simply to declare that a particular rule of the law of defamation is invalid, leaving a lacuna in the law.
 Fortunately, the Constitution allows for the development of the common law and customary law by the Supreme Court in accordance with the objects of Chapter 3. This is provided for in section 35(3) -
“In the interpretation of any law and the application and development of the common law and customary law, a court shall have due regard to the spirit, purport and objects of this Chapter.”
I have no doubt that this sub-section introduces the indirect application of the fundamental rights provisions to private law. I draw attention to the words “have due regard to” in section 35(3). That choice of language is significant. The lawgiver did not say that courts should invalidate rules of common law inconsistent with Chapter 3 or declare them unconstitutional. The fact that courts are to do no more than have regard to the spirit, purport and objects of the Chapter indicates that the requisite development of the common law and customary law is not to be pursued through the exercise of the powers of this Court under section 98 of the Constitution. The presence of this sub-section ensures that the values embodied in Chapter 3 will permeate the common law in all its aspects, including private litigation. I incline to agree with the view of Cameron J in the judgment already referred to, that section 35(3) makes much of the vertical/horizontal debate irrelevant. The model of indirect application or, if you will indirect horizontality, seems peculiarly appropriate to a judicial system which, as in Germany, separates constitutional jurisdiction from ordinary jurisdiction. This does not mean that the principles evolved by the German Constitutional Court must be slavishly followed. They do however afford an example of how the process of influencing the common law may work in practice. The German Basic Law, Article 1(3) provides -
“The following basic rights shall bind the legislature, the executive and the judiciary as directly enforceable law.”
It has no equivalent to section 35(3). Yet, as I pointed out earlier in this judgment, the German courts nonetheless apply a model of indirect and not direct application of the basic rights provisions in private litigation.
 There is also some practical guidance to be found in the Canadian authorities, always bearing in mind that there is no separation of constitutional and ordinary jurisdiction in Canadian courts. In R. v. Salituro Iacobucci J said -
“Judges can and should adapt the common law to reflect the changing social, moral and economic fabric of the country. Judges should not be quick to perpetuate rules whose social foundation has long since disappeared. Nonetheless there are significant constraints on the power of the judiciary to change the law. ... in a constitutional democracy such as ours it is the Legislature and not the courts which has the major responsibility for law reform ... . The judiciary should confine itself to those incremental changes which are necessary to keep the common law in step with the dynamic and evolving fabric of our society.”
In Bank of British Columbia v. Canadian Broadcasting Corp., the court engaged in the process of weighing up the claims of the Charter and the common law rules of defamation, in circumstances where a bank had sued a public broadcaster for libel following reports that it was in imminent danger of failing. The court extracted what it perceived to be the core value underlying the freedom of expression, press and media provision in section 2(b) of the Charter, namely the right to gather and disseminate truthful information free from government interference. “The ultimate purpose of the Charter protection is so that truth may be uncovered and made known ... . The Charter speaks to our highest values. Truth is one of them.”  What section 2(b) did not do, it was held, was to provide a special privilege to the press in the context of private litigation. In Hill’s case, supra n90, plaintiff instituted action for libel following the bringing of contempt proceedings against him by the Defendants. The court ruled out direct application of the Charter but emphasised that “the common law must be interpreted in a manner which is consistent with Charter principles. This obligation is simply a manifestation of the inherent jurisdiction of the courts to modify or extend the common law in order to comply with prevailing social conditions and values.”  Having weighed up the Charter requirements and the common law rules, the court concluded that the common law of defamation complied with the underlying values of the Charter and there was no need to amend or alter it. Our courts may well reach a different conclusion: for one thing section 35(3) has no counterpart in the Canadian Charter. But the process of reasoning of the Canadian judges remains instructive.
 What I conclude is that Chapter 3 does not have a general direct horizontal application but that it may and should have an influence on the development of the common law as it governs relations between individuals. I insert the qualification “general” because it may be open to a litigant in another case to argue that some particular provision of Chapter 3 must by necessary implication have direct horizontal application. Section 15(1) is not such a provision. No such implication is necessary. One of the purposes of the section is to give protection against far-reaching censorship laws and other statutes restricting free speech which were common under the regime of Parliamentary supremacy. Accordingly, my response to the second issue referred to this Court by the learned judge would be that Chapter 3 of the Constitution does not in general have direct horizontal application, and more particularly that section 15(1) does not have direct horizontal application. On the other hand, the values which it embodies can and must be taken into account in the development of the common law of defamation.
 I should add that in my opinion the phrase “a court” in section 35(3) means “all courts”, and includes the Appellate Division, notwithstanding the provisions of section 101(5). There is no contradiction as the “application and development of the common law” is not a matter which falls within the jurisdiction of the Constitutional Court under section 98. This is not to say that the Constitutional Court has no control over how the common private law develops. In terms of section 98(2) it has jurisdiction in the final instance over all matters relating to “the interpretation, protection and enforcement of the provisions of this Constitution”. It must ensure that the provisions of section 35(3) in relation, inter alia, to the development of the common law are properly interpreted and applied, otherwise it is not discharging its duty properly in relation to the enforcement of the provisions of the Constitution. The Constitutional Court has jurisdiction to determine what the “spirit, purport and objects” of Chapter 3 are and to ensure that, in developing the common law, the other courts have had “due regard” thereto. It is unnecessary, for the purposes of this judgment, to define the boundaries of its jurisdiction in this regard. Whether the Constitutional Court will exercise review powers along the same lines as the German Constitutional Court is a question for the future.
 What I have said above has implicitly answered the questions put to counsel by the President of the Court, set out in paragraph 11 above.
i. The issue whether the common law of defamation should be developed to make it consistent with the Constitution, was not an issue in the appeal but fell to be considered in relation to the second issue referred to this Court by the judge.
ii. The development of the common law is within the jurisdiction of the Appellate Division, but not of the Constitutional Court, subject to the reservation made in the previous paragraph of this judgment.
iii. Any appeal on such an issue (the development of the common law), once it has been properly raised and dealt with in a provincial division, must be directed to the Appellate Division.
 In the application before him Van Dijkhorst J was not required to consider the development of the common law in terms of section 35(3) and did not do so. That section may in some instances require the Supreme Court to give a new turn to a branch of the common law. It may well follow from the answer I have given to question (a) in paragraph 10 above, that the Defendants cannot derive any assistance from section 35(3) in relation to a defamation which was published before the Constitution came into force. That issue, however, was not argued before us, and is not without its complications. In our courts a judgment which brings about a radical alteration in the common law as previously understood proceeds upon the legal fiction that the new rule has not been made by the court but merely “found”, as if it had always been inherent in the law. Nor do our courts distinguish between cases which have arisen before, and those which arise after, the new rule has been announced. For this reason it is sometimes said that “judge-made law” is retrospective in its operation. In all this our courts have followed the practice of the English courts. Thus in Birmingham Corporation v. West Midland Baptist (Trust) Association; Lord Reid said -
“We cannot say that the law was one thing yesterday but is to be something different tomorrow. If we decide that the rule [i.e. the previously accepted rule] ... is wrong we must decide that it always has been wrong”.
That is why in Geelong Harbour Trust Commissioners v. Gibbs Bright and Co Lord Diplock said that any change in the law made by judicial decision is in effect retrospective. In their well-known book The South African Legal System Professors Hahlo and Kahn say that (save in matters of court practice), there -
“ ... has never been a holding ... by any South African court that the operation of a decision is to be prospective. This solution of certain American courts is unknown.”
This is no doubt correct. It may nonetheless be said that there is no rule of positive law which would forbid our Supreme Court from departing from that practice. Indeed, in England at least two Law Lords have said that the judiciary should seriously consider exercising a jurisdiction to overrule a previous decision prospectively only, as American courts have done.
 It is unnecessary to discuss the American practice in detail. It is described in Tribe, American Constitutional Law and in an article, “Prospective and Retrospective Judicial Lawmaking”, by Professor M.I. Friedland, which also describes English and Canadian practice. It is sufficient to refer to two decisions of the United States Supreme Court, which state the principle that courts may (not must) apply their decisions prospectively if they overrule past precedents. See Great Northern Railway Co. v. Sunburst Oil and Refining Co, and Chevron Oil Co v. Huson. It may be that a purely prospective operation of a change in the common law will be found to be appropriate when it results from the application of a constitutional enactment which does not itself have retrospective operation. But it follows from what I have said above that those are matters which it is for the provincial and local divisions of the Supreme Court to decide as part of their function of applying section 35(3) and developing the common law. I do no more than respectfully draw their attention to the considerations which I have outlined. Whether appeals against judgments on such matters go to the Appellate Division or this Court, need not be decided now and should be left open.
 The Order of the Court is as follows:
1. The appeal is dismissed with costs, such costs to include the costs of two counsel.
2. The two questions referred by the judge a quo, and reformulated by this Court as set out in paragraph 10 above are answered as follows -
a) No: The Defendants in this case are not entitled to invoke the provisions of the Constitution.
b) No: The provisions of Chapter 3 of the Constitution are not in general capable of application to any relationship other than that between persons and legislative or executive organs of state at all levels of government. In particular section 15 is not capable of application to any relationship other than that between persons and legislative or executive organs of the state at all levels of government.
Acting Justice of the Constitutional Court
Chaskalson P, Langa J and O’Regan J concur in the judgment of Kentridge AJ.
 MAHOMED DP: I have had the privilege of reading and considering the main judgment of Kentridge AJ and also the separate judgments of Ackermann, Kriegler, Madala, Mokgoro and Sachs JJ. In view of the fact that I have sometimes different perspectives in regard to some of the issues articulated in those judgments, I have considered it wise to set out briefly my approach on the disputed issues.
In effect, what the amendment sought by the appellants seeks to assert is the proposition that if the disputed articles published by the Pretoria News were unlawful at the time of the publication, the effect of the subsequent enactment of the interim Constitution of 1993 (“the Constitution”) is to render such publication lawful. In my view that is an untenable proposition. The reliance by the Appellants on my judgment given in Mhlungu’s case1 seems to me to be misplaced. Nothing in Mhlungu’s case, in any of the judgments of the majority or the minority, support the proposition contended for. What was involved in the relevant parts of the judgment in Mhlungu’s case was the proper interpretation of section 241(8) of the Constitution. What I did hold was that an accused person in a criminal trial was entitled to rely on any protection of the Constitution in any trial that was taking place after the commencement of the Constitution and even in circumstances where such a trial had actually begun before the commencement of the Constitution. I held that section 241(8), properly interpreted, did not operate as an obstacle in the way of an accused person who sought to assert the protection of the Constitution at a time when the Constitution was already in operation, notwithstanding the fact that the case may already have begun before the commencement of the Constitution.2 Indeed, I held expressly that an accused person could not rely on any of the provisions of section 25(3) of the Constitution in an appeal heard after the commencement of the Constitution in which it was being asserted that a right protected by section 25(3) had not been accorded to the accused at the trial at a time when the Constitution was not yet in operation.3 The lawfulness or unlawfulness of any conduct at the time it took place is determined by the applicable law at that time. The Constitution does not convert conduct which was unlawful at the time when it took place into lawful conduct.
 Notwithstanding this conclusion I would like to make one qualification which might perhaps be important in some future dispute. I would prefer to leave open the question whether or not, following on a declaration of invalidity, this Court has jurisdiction in terms of section 98(6) of the Constitution, to make an order invalidating something which was done (or permitted to be done) at a time when the Constitution was not operative at all. It may arguably be contended in some suitable case that the interests of justice and good government justify an order which invalidates anything previously done or permitted in terms of an invalid law even if the Constitution was not operative at the time when it was so done or permitted. It is unnecessary to pursue this line in the present case. Even if section 98(6) was to be construed as permitting a retrospective order of the kind I have alluded to in certain circumstances, the factual circumstances in the present case would not justify such an order.
 In paragraph 2(a) of the order proposed by Kentridge AJ, it is stated that “[t]he Defendants in this case are not entitled to invoke the provisions of the Constitution” having regard to the fact that the publication of the offending material, the institution of the action and all the relevant facts had occurred before the Constitution came into operation. I am in respectful agreement with that order but I would emphasize the qualification contained in paragraph 20 of the judgment of Kentridge AJ in respect of other cases where the enforcement after the Constitution of rights acquired prior to the Constitution would be plainly inconsistent with our present constitutional values.
 After some hesitation I have come to the conclusion that Kentridge AJ is correct in his interpretation of the section and it therefore follows that the matter was correctly referred to this court by the court a quo.
 Much of the debate before us pertained to the question whether the fundamental rights provisions contained in Chapter 3 are only of “vertical” application against the legislature and the executive or whether they are also of “horizontal” application between private citizens inter se. Having examined the detailed reasons given by Kentridge AJ and those given by Kriegler J, I have come to the conclusion that on any approach the practical consequences are substantially the same.
1. The debate between Kentridge AJ and Kriegler J cannot properly be characterized as a debate on the issue as to whether Chapter 3 of the Constitution is of horizontal or vertical application.
2. Where a statute or an ordinance or a by-law or a regulation is attacked on the grounds that it is inconsistent with the Constitution, there is consensus that this is a competent attack and it matters not whether such an attack is made in litigation between private individuals or whether it is made in litigation between a governmental authority and a citizen. Its effect in such circumstances can therefore be “horizontal”.
3. Even where the attack concerned is made not on any statutory enactment but on the common law, Chapter 3 can be invoked to sustain such an attack if one of the parties to the litigation is a governmental agency.
4. The only residual area of potential disagreement arises in the case where what is sought to be attacked is some or other rule of the common law in litigation between private parties not involving any legislative or executive authority. But even in this limited area the true debate is effectively not whether the rights articulated in Chapter 3 are capable of “horizontal” effect but whether or not such “horizontality” is to arise in consequence of the direct application of the relevant Chapter 3 right or through the mechanism of interpreting, applying and developing the common law by having regard to the spirit, purport and objects of the Chapter, pursuant to section 35(3).
5. On both approaches there is consensus that the power of all the divisions of the Supreme Court (including the Appellate Division) to interpret and develop the common law, having regard to the spirit, purport and objects of Chapter 3, is crucial and unimpaired.
6. The issue which arises in the present case is to be confined to the proper application of section 15 and the answer to that issue may not necessarily be the same as the answer which might have to be given to any other section contained in Chapter 3. (There is some force in the suggestion by Madala J that some of the fundamental rights enumerated in Chapter 3 may apply directly in litigation between private persons.4 It is unnecessary in the present case to determine that issue or to attempt to identify the particular rights in Chapter 3 which might be suitable for such treatment.)
 The differences in the theoretical approaches favoured by Kriegler J and Kentridge AJ therefore seem to me to involve no substantial practical consequences, particularly if regard is had to the fact that whatever may be said about the meaning of the interim Constitution might in the future be of historical importance only, because the interim Constitution will already have been overtaken by a new constitutional text with quite different formulations impacting on the problem.
 In view of the fact, however, that somewhat different theoretical positions have been maintained in the judgments of Kriegler J and Kentridge AJ, I think I should express my views on this debate and the reasoning articulated in the course thereof.
 What is patent from the preamble, the postscript and the substance of the Constitution is a very clear and eloquent commitment to the creation of a defensible society based on freedom and equality setting its face firmly and vigorously against the racism which has dominated South African society for so long and the repression which became necessary to perpetuate its untenable ethos and premises. To leave individuals free to perpetuate advantages, privileges and relations, quite immune from the discipline of Chapter 3, would substantially be to allow the ethos and pathology of racism effectively to sustain a new life, subverting the gains which the Constitution seeks carefully to consolidate. It is for this reason that I have found the approach of Kriegler J particularly attractive, but after some considerable hesitation I have come to be influenced by three important considerations which reduce the cogency of his arguments.
 The first influence is textual. Section 7(1) provides that-
“This Chapter shall bind all legislative and executive organs of state at all levels of government.”
What section 7(1) therefore does is to isolate the bodies who are bound by the Chapter. Section 4(2) and many other Constitutions include the judiciary among the organs expressly bound. Significantly the judiciary appears in section 7(1) to have been deliberately excluded from the organs and bodies which are identified as being the organs bound by Chapter 3. The organs identified as being so bound are simply confined to “all legislative and executive organs of State.” Why was it necessary to isolate such organs? Why was it necessary to exclude the judiciary or other organs or bodies which were not governmental in character? The issue as to whether guarantees on fundamental rights should be of horizontal application in relations between private citizens had, prior to the enactment of the Constitution, been the subject of very considerable public and academic debate, both in South Africa and abroad. Different responses were forthcoming from different constituencies. It was very much a live issue. I find it difficult to accept that if the lawmakers had intended to resolve that debate in the manner contended for by “horizontality” advocates, they would not have said so in clear terms or at least in language which clearly permitted that inference to be made. I am not persuaded that the lawmakers would wish such a crucial issue to be left for discovery and inference by astute judicial craftsmanship and nimble argumentation. The converse would have been understandable. If the language employed in section 7(1) was language which permitted the inference of “horizontality”, but nothing was said as to whether the Chapter also bound organs of government, it might have been easy to infer that because private citizens were bound by the discipline of Chapter 3, it was a fortiori of application against a governmental authority. The controversy which had previously raged was not a controversy pertaining to whether governments should be bound by the fundamental rights articulated in a Constitution but whether citizens should so be bound in their relations between themselves.
 In dealing with the proper force and interpretation of section 7(1) I have not overlooked the argument that section 7(2) applies Chapter 3 to “all law in force... during the period of operation of this Constitution.” I have no doubt that “all law” must include the common law but in my respectful view Kentridge AJ is correct in concluding that what section 7(2) does is to define what law is applicable to the persons bound by Chapter 3 in terms of section 7(1). Nothing in section 4 of the Constitution is inconsistent with that conclusion. The Constitution is, in terms of section 4, manifestly the supreme law of the Republic and binding on all legislative, executive and judicial organs of state at all levels of government. But that fundamental proposition in no way extends the application of section 7(1) to bodies or persons not otherwise bound in terms of that subsection by the fundamental rights articulated in Chapter 3.
 The textual force of section 7(1) appears, in my view, substantially to be reinforced by the provisions of section 33(4) of the Constitution which prescribe that Chapter 3-
“shall not preclude measures designed to prohibit unfair discrimination by bodies and persons other than those bound in terms of section 7(1).”
What this sub-section seems to me to do is to authorize substantive legislation which would extend to other bodies and persons the duties placed on government in terms of section 7(1). The answer favoured by Kriegler J to this approach follows on his understanding of the meaning of section 7(2). Section 7(2), he says, makes the Chapter applicable to all law and it matters not whether the persons seeking to apply such law are private persons or governmental organs but private persons are perfectly entitled to act without regard to the rights articulated in Chapter 3, as long as they do not invoke the law in support of their actions. On this approach section 33(4) becomes necessary to entitle Parliament to enact legislation which would prohibit actions by private persons which are inconsistent with Chapter 3 but in circumstances where the law is not being invoked by private persons in support of such actions. He puts this argument forcefully by stating that-
“.... As far as the Chapter is concerned a landlord is free to refuse to let a flat to someone because of race, gender or whatever; a white bigot may refuse to sell property to a person of colour; a social club may to black-ball Jews, Catholics or Afrikaners if it so wishes. An employer is at liberty to discriminate on racial grounds in the engagement of staff; a hotelier may refuse to let a room to a homosexual; a church may close its doors to mourners of a particular colour or class. But none of them can invoke the law to enforce or protect their bigotry... The whole gamut of private relationships is left undisturbed.”5
 There is force in this approach but I have difficulties with it. The premise is and must be that private persons falling within the examples referred to in this quotation, who perform acts otherwise inconsistent with the rights specified in Chapter 3, are not doing so in terms of law. I think this is an incorrect premise. All the acts performed by such private persons are acts performed in terms of what the common law would allow. A landlord who refuses to let to someone because of his race is exercising a right which is incidental to the rights of the owner of property at common law; this applies equally to the white bigot who refuses to sell property to a person of colour. A social club which black-balls Jews, Catholics or Afrikaners acts in terms either of its own constitution or the common law pertaining to voluntary associations or freedom of contract. I am not persuaded that there is, in the modern State, any right which exists which is not ultimately sourced in some law, even if it be no more than an unarticulated premise of the common law and even if that common law is constitutionally immunized from legislative invasion. Whatever be the historical origins of the common law and the evolutionary path it has taken, its continued existence and efficacy in the modern State depends, in the last instance, on the power of the State to enforce its sanction and its duty to do so when its protection is invoked by the citizen who seeks to rely on it. It is, I believe, erroneous to conclude that the law operates for the first time only when that sanction is invoked. The truth is that it precedes it and is indeed the ultimate source for the legitimation of any conduct. Freedom is a fundamental ingredient of a defensible and durable civilization, but it is ultimately secured in modern conditions, only through the power, the sovereignty and the majesty of the law activated by the State’s instruments of authority in the protection of those prejudiced through its invasion by others. Inherently there can be no “right” governing relations between individuals inter se or between individuals and the State the protection of which is not legally enforceable and if it is legally enforceable it must be part of law.
 The approach of Kriegler J, which I have sought to summarise, also does not explain to me satisfactorily why section 7(1) was necessary at all. If section 7(2) was to be interpreted on the basis that Chapter 3 applied to all law and that it mattered not whether those affected were governmental organs or private parties in litigation between themselves, there would have been scant reason to specially isolate governmental organs in section 7(1).
 The basic premise of section 7(1) is that generally Chapter 3 applies only to legislative and executive organs of State and substantially the same assumption must inform section 98(7) which contemplates restraints only upon governmental authorities performing executive or administrative acts inconsistent with the Constitution. No corresponding machinery is provided for restraints upon any private persons.
 The same temper is reflected by section 35(3) which requires a court to have regard to the spirit, purport and objects of Chapter 3 in the interpretation of any law and the application and development of the common law and customary law. If the law pertaining to relations between private individuals was directly subject to the substantive rights contained in Chapter 3 there would be scant need to provide that the courts should have regard to the spirit, purport and objects of the Chapter in the application and development of the common law. If Chapter 3 was of direct application to such relationships, section 35(3) would appear to me to be only of peripheral value and relevance in the application of Chapter 3. As will appear later, on the interpretation I favour, on the other hand, the role of section 35(3) is crucial in the identification, development and enforcement of the fundamental constitutional values articulated in the Chapter.
 In the main judgment of Kentridge AJ he also relies on the text of section 33(1) in support of his construction of sections 7(1) and 7(2). He argues that if section 33(1) were to apply to private relationships governed by the common law, the problems which would arise appear to be “almost insurmountable.” He contends that-
“The common law addresses problems of conflicting rights and interests through a system of balancing. Many of these rights and interests are now recorded in the Constitution and on any view that means that as a result of the terms of the Constitution the balancing process previously undertaken may have to be reconsidered.”6
I agree that the text of section 33(1) may be some “pointer” in the direction favoured by Kentridge AJ but I would not have regarded this consideration by itself to be sufficient to justify a rejection of the approach favoured by Kriegler J. In the first place, on the approach favoured by Kentridge AJ, Chapter 3 and therefore section 33(1) would in any event apply to the common law where the government or a governmental agency is involved in litigation against a private party and in which some right contained in Chapter 3 is invoked. The lawmaker therefore did not consider the problems in applying section 33(1) in such a situation to be “insurmountable”. I do not think that these problems necessarily become insurmountable in cases where it is not the government which is involved in making or resisting an attack on the common law but private parties inter se (although it must be conceded that where the State is a party in litigation involving the common law it would often not be asserting a “right” which needs balancing against a “right” claimed or asserted by its adversary). In most cases, “the balancing process previously undertaken” in formulating the relevant rule of the common law has involved balancing some of the very rights now detailed in Chapter 3 and it may be perfectly possible, in the application of section 33(1), to do the balancing act by having regard to the other rights articulated in Chapter 3, both in identifying the contours of the particular right invoked in the attack on the common law and in seeking to define the permissible parameters of any limitation.
 Apart from the actual and potential textual difficulties I have in support of the interpretation favoured by Kriegler J, I agree with Kentridge AJ that “there is nothing in the legislative history... which requires the adoption of the horizontal interpretation.”7 I am mindful of the fact, however, that it is theoretically possible for different Parliamentarians and negotiators to support the same formula in an enactment for very different, and sometimes even conflicting reasons, but insofar as the legislative history provides an objective background which is inconsistent with a particular construction which is advanced, it is, I think, permissible to have some regard to that history, although this cannot in itself ever operate decisively.8
 Notwithstanding all these observations I would have remained profoundly uncomfortable if the construction favoured by Kentridge AJ meant, in practise, that the Constitution was impotent to protect those who have so manifestly and brutally been victimised by the private and institutionalized desecration of the values now so eloquently articulated in the Constitution. Black persons were previously denied the right to own land in 87% of the country. An interpretation of the Constitution which continued to protect the right of private persons substantially to perpetuate such unfairness by entering into contracts or making dispositions subject to the condition that such land is not sold to or occupied by Blacks would have been for me a very distressing conclusion.9 These and scores of other such examples leave me no doubt that those responsible for the enactment of the Constitution never intended to permit the privatisation of Apartheid or to allow the unfair gains of Apartheid or the privileges it bestowed on the few, or the offensive attitudes it generated amongst many to be fossilized and protected by courts rendered impotent by the language of the Constitution. For this reason I would therefore have been compelled to ask whether the interpretation favoured by Kentridge AJ is perhaps not flawed in some respect which I might have overlooked or whether I have not perhaps accorded inadequate weight to some of the relevant considerations so forcefully articulated in the judgment of Kriegler J.
 Fortunately, however, none of the distressing consequences to which I have referred in the preceding paragraph, flow from the interpretation which I have now come to favour. I say this because on that interpretation most of the common law rules, upon which reliance would have to be placed by private persons seeking to perpetuate unfair privilege or discrimination, would themselves be vulnerable to invasion and re-examination in appropriate circumstances. What contracts and actions public policy would permit or enforce in the future will have to be re-examined. Such a constitutionally defensible and competent source of invasion would flow not from a direct and literal extension of the provisions of section 7(1) of the Constitution to relations between private persons inter se. It would flow from a source potentially no less richer and creative than such an extension. It would be sourced in section 35(3) of the Constitution which compels the courts to have due regard to the spirit, purport and objects of the Chapter in the interpretation of any law and the application and development of the common law. The common law is not to be trapped within the limitations of its past. It needs not to be interpreted in conditions of social and constitutional ossification. It needs to be revisited and revitalized with the spirit of the constitutional values defined in Chapter 3 of the Constitution and with full regard to the purport and objects of that Chapter.10 Thus approached section 35(3) can, in appropriate circumstances, accommodate much of the concern felt by those like me who are anxious to avoid giving to section 7 of the Constitution an interpretation which would leave the courts substantially impotent in affording the proper protection of constitutional values to those victimized by their denial to them in the past.
 The interpretation which I have come to favour has the advantage of giving to the different divisions of the Supreme Court, including its Appellate Division, a very clear and creative role in the active evolution of our constitutional jurisprudence by examining, and in suitable circumstances expanding, the traditional frontiers of the common law by infusing it with the spirit of Chapter 3 of the Constitution and its purport and objects. Nothing contained in section 101(5), read with section 98, of the Constitution would in any way impede the untrammelled exercise of such powers, but it would leave also to the Constitutional Court the residual power to determine, in suitable circumstances, whether in the application of its jurisdiction in terms of section 35(3) the Supreme Court has in any particular case properly had regard to the spirit of Chapter 3 of the Constitution and its purport and objects.
 In the result, I am in respectful agreement with the order proposed by Kentridge AJ and the answers he gives in paragraph 64 of his judgment.
Deputy President of the Constitutional Court
Langa J and O’ Regan J concur in the judgment of Mahomed DP.
 ACKERMANN J: I concur with Kentridge AJ’s judgment and the order he proposes. I have also had the privilege of reading the judgment of Kriegler J, in which he reaches a different conclusion regarding the effect which the rights guaranteed in Chapter 3 of the interim Constitution have on legal relations between private persons. Because of this disagreement and the importance of the issue at stake, I should like to add briefly certain reasons of my own, some of which have been foreshadowed in Kentridge AJ’s judgment, why I agree with his conclusion on this particular issue. A teleological approach to the construction of the Constitution gives substantial support to this conclusion.
 It is certainly true that our interim Constitution is textually unique and that the historical circumstances in which constitutions are adopted are never identical. In certain cases these circumstances may, for the most part, only differ in degree. Many constitutions, particularly those which come in the wake of rapid and extensive political and social change, are reactive in nature and often reflect in their provisions a response to particular histories and political and social ills.1
 Kriegler J has, in his judgment, referred eloquently to the duration, acceleration and gravity of the human rights denials and abuses to which the interim Constitution is a response and which it seeks, amongst other things, to redress. Without wishing to over-simplify the nature and extent of these abuses and denials it is, I think, fair to say that they related in general to the core values of dignity, freedom and equality. There are other constitutions which have been a response to tragic histories or episodes in the national histories of particular countries during which gross abuses of human rights have occurred.
 I do believe that the German Basic Law (GBL) was conceived in dire circumstances bearing sufficient resemblance to our own to make critical study and cautious application of its lessons to our situation and Constitution warranted. The GBL was no less powerful a response to totalitarianism, the degradation of human dignity and the denial of freedom and equality than our Constitution. Few things make this clearer than Art 1(1) of the GBL,2 particularly when it is borne in mind that the principles laid down in Art 1 are entrenched against amendment of any kind by Art 79(3).3
 The provisions of Articles 1 to 19 of the GBL (which constitute its chapter on basic rights and determine the binding nature of these rights) are no more restricted than the comparable provisions in our Constitution. No distinction is made between public law and private law; statute law and common law; there is no suggestion of their being limited to the relationship between state and persons. Most important of all, Art 1(3) of the GBL states unequivocally that the basic rights “shall bind the legislature, the executive and the judiciary as directly enforceable law”4 (emphasis supplied). By specifically including the judiciary, this provision goes even further textually than section 7(1) of our Constitution.
 Yet in Germany it is today not seriously questioned that in deciding disputes between private persons there is no direct application of the fundamental rights by the judiciary.5 This is particularly noteworthy considering the extensive jurisprudence supporting the view that the basic rights entrenched by the GBL not only establish subjective individual rights but an objective order of values or an objective value system (“eine objektive Wertordnung”).6
The jurisprudence of the Federal Constitutional Court is consistently to the effect that the basic right norms contain not only defensive subjective rights for the individual but embody at the same time an objective value system which, as a fundamental constitutional value for all areas of the law, acts as a guiding principle and stimulus for the legislature, executive and judiciary.7
Von Mhnch/Kunig8 point out that the Federal Constitutional Court and academic opinion are in agreement that the basic rights only apply with indirect horizontality (“nur mittelbaren Drittwirkung”) to the legal relations of private individuals; the basic rights do not apply directly to the private law, but because the basic rights also operate as an objective value system they “influence” (“beeinflussen”) the private law. The learned authors point out9 that the basic rights do not serve to solve disputes in the field of private law in specific cases.
 Against this background, particularly having regard to the explicit wording of Art 1(3) of the GBL, it behoves us to consider carefully why, on textual, teleological and policy grounds, German constitutional jurisprudence has rejected the direct application of the basic rights in the GBL to private legal relationships.
 The impact of Art 1(3) of the GBL on the judiciary is seen as being limited generally to procedural due process, including equal access to the law and equality before the law. Where the law is completely silent, however, the basic rights can be directly applied. Thus, the general equality clause10 requires procedural equality of arms (“Waffengleichheit”), equality “as to the outcome of the case” (i.e. the court must be unbiassed throughout) and above all equality in the application of the law; the courts may not depart from legal norms, existing norms may not be ignored and litigants may not be discriminated against on any of the grounds mentioned in Art 3(3).11
 Direct application of the basic rights by the judiciary in ordinary civil proceedings would make the law vague and uncertain, which is contrary to the concept of the constitutional state.12 Uncertainty is aggravated by the fact that (in contrast to a dispute between citizen and state) in a dispute between two private individuals both sides can invoke the basic rights, calling for a difficult balancing of conflicting rights which could reasonably lead different courts to different decisions.13 Rather, direct application should take place at the law-making level, so that all laws which are being applied by the courts do already comply with the basic rights, obviating the need for direct horizontal application by the courts.14 I consider this to be an equally, if not more compelling, consideration in the context of our own Constitution. As I pointed out in S v Makwanyane and Another,
[i]n reaction to our past, the concept and values of the constitutional state, of the regstaat’... are deeply foundational to the creation of the new order’ referred to in the preamble [of the Constitution].15
Our Constitution moreover commits the state, in different ways, to legislative programmes to rectify past discrimination and denial of human rights and to ensure equality, the equal protection of the law and the protection and advancement of human rights in the future.16
 The fact that Art 1(3) of the GBL explicitly mentions the organs of state, and only the organs of state, as being bound by the chapter on basic rights is seen as a clear pointer that rights are not to be applied horizontally.17 Furthermore, certain provisions18 explicitly state that they are applicable horizontally, which would be superfluous if all provisions were applicable horizontally.19 This supports the approach of and the conclusion reached by Kentridge AJ in paragraph 46 of his judgment that the provisions of sections 33(4) and 35(3) of the Constitution would be superfluous if the Chapter 3 rights were directly enforceable horizontally.
 The German Constitution guarantees a general right to freedom.20 In areas such as the law of contract this would have to be taken into account in deciding whether parties should be bound by other sections of the basic rights chapter since the act of limiting one’s own rights by contract is itself an exercise of the right to freedom. A matter of some controversy is whether the individual can consent to the violation of his/her rights by the state.21 The question whether one may allow another private person to infringe one’s constitutional rights is not dealt with in the constitutional context, because these rights are not considered binding on other private persons in the first place. In this context the objection to the direct application of basic rights to private relations is that it severely undermines private autonomy.22
 On a teleological approach it is argued that the purpose of the chapter on basic rights in the GBL is to confer rights on individuals. The effect of direct horizontal application would be to place duties on individuals instead.23 A related point is made that if the basic rights of the GBL are directly invoked in any dispute between private individuals, the basic rights of both individuals will be at stake, necessitating the balancing of competing rights.24
 A further argument advanced in this context, which is also relevant to our own, is that the basic rights were developed in an historical setting of civil society struggling to free itself from the authoritarian interference of the absolute state.25
 A further problem which militates against the direct horizontal application of the basic rights is the fact that the Federal Constitutional Court would be thrust into the role of an appeal court for large numbers of appeals in what would otherwise be normal commercial litigation.26
 Any attempt at a detailed discussion on the operation of mittelbare Drittwirkung (indirect horizontality) in German constitutional law would be out of place here. There are some features, however, which bear on the construction of our own Constitution. The Federal Constitutional Court refers to the radiating effect (Ausstrahlungswirkung) of the basic rights on private law.27 In the Lhth case28 the Federal Constitutional Court held as follows:
The influence of the scale of values of the basic rights affects particularly those provisions of private law that contain mandatory rules of law and thus form part of the ordre public - in the broad sense of the term - that is, rules which for reasons of the general welfare also are binding on private legal relationships and are removed from the domination of private intent. Because of their purpose these provisions are closely related to the public law they supplement. Consequently, they are substantially exposed to the influence of constitutional law. In bringing this influence to bear, the courts may invoke the general clauses which, like Article 826 of the Civil Code, refer to standards outside private law. “Good morals” is one such standard. In order to determine what is required by social norms such as these, one has to consider first the ensemble of value concepts that a nation has developed at a certain point in its intellectual history and laid down in its constitution. That is why the general clauses have rightly been called the points where basic rights have breached the [domain of] private law...29
 Thus, in private litigation, the German courts are obliged to consider the basic rights in interpreting concepts such as “justified”, “wrongful”, “contra bonos mores” et cetera.30 The basic rights therefore have a radiating effect on the common law through provisions such as, for example section 138 of the Civil Code, which provides that “legal acts which are contrary to public policy are void.”31 Stern32 argues that although general concepts such as the boni mores will be the main entry point into private law for the basic rights, there is no reason why they should not be applied to any other rule whose meaning is unclear. If, in such situations, the court does not have regard to the basic rights in interpreting the law in question, then there is a violation of rights.33
 Stern34 suggests that the major advantage of indirect horizontal application is that, instead of throwing out the good parts of the existing law with the bad and setting off into unchartered waters, it embarks on a cautious reform of the existing law within its own framework. The same learned author also points out35 that even the highest German Federal Labour Court36 which initially supported a direct horizontal application of basic rights had, by 1984, come to accept the position that unless expressly provided (as in Art 9(3) of the GBL), the constitution required that these rights should only be given indirect horizontal application.
 That the drafters of our Constitution had recourse to or were influenced by certain features of the GBL in drafting our Constitution is evident from various of its provisions.37 The marked similarity between the provisions of section 35(3), enjoining courts “[i]n the interpretation of any law and the application and development of the common law and customary law” to “have due regard to the spirit, purport and objects of [Chapter 3]”, and the indirect horizontal application of the basic rights in the GBL in German jurisprudence cannot, in my view, simply be a coincidence. It provides a final powerful indication that the framers of our Constitution did not intend that the Chapter 3 fundamental rights should, save where the formulation of a particular right expressly or by necessary implication otherwise indicates,38 apply directly to legal relations between private persons.
 The implications of direct horizontal application must be logically followed and consistently faced. If direct horizontality is to be applied to section 15(1) then it must be applied to the equality clause, section 8, unless one were to hold (quite impermissibly in my view) that for some reason, not evident in text or reason, freedom of speech and expression is in South Africa a preferred right enjoying a higher status than the right to equality.
 The direct horizontal application of, for example, section 8 would lead to consequences so undesirable and unsupportable that it demonstrates why, absent the very clearest indication to the contrary in our Constitution, we ought not to apply the Chapter 3 rights directly to private legal relations. I have no doubt that a similar direct application of other Chapter 3 rights would lead to similar unsupportable consequences, but examining the consequences in relation to section 8 will suffice.
 The direct application route is the one which the ratio decidendi in Shelley v. Kraemer39 in effect followed. At least two eminent American constitutional scholars, professors Herbert Wechsler and Louis Henkin,40 while applauding the result reached by the Supreme Court in this case, have effectively criticised the ratio decidendi and demonstrated how the route chosen leads to unsupportable consequences.41 Professor Wechsler reasons as follows:
Assuming that the Constitution speaks to state discrimination on the ground of race ... why is the enforcement of the private covenant a state discrimination rather than a legal recognition of the freedom of the individual? That the action of the state court is action of the state, the point that Mr. Chief Justice Vinson emphasizes in the Court’s opinion is, of course, entirely obvious. What is not obvious, and is the crucial step, is that the state may properly be charged with the discrimination when it does no more than give effect to an agreement that the individual involved is, by hypothesis, entirely free to make. Again, one is obliged to ask: What is the principle involved? Is the state forbidden to effectuate a will that draws a racial line, a will that can accomplish any disposition only through the aid of the law, or is it a sufficient answer there that the discrimination was the testator’s and not the state’s? May not the state employ its law to vindicate the privacy of property against a trespasser, regardless of the grounds of his exclusion, or does it embrace the owner’s reasons for excluding if it buttresses his power by the law? Would a declaratory judgment that a fee is determinable if a racially restrictive limitation should be violated represent discrimination by the state upon the racial ground? Would a judgment of ejection? None of these questions has been answered by the Court nor are the problems faced in the opinions.42
Professor Henkin takes the critical enquiry even further:
If some may be prepared to go where Professor Wechsler hesitates to go, even they may hesitate to conclude that the Supreme Court would prevent a state from probating a will leaving money to a group or institution of a particular religious denomination, or from enforcing exclusion, on the basis of religious difference, from church, or church membership, or cemetery, although the state could not make or require these discriminations.
Indeed, the difficulty may lie even deeper, as the testamentary cases would seem to prove. If Shelley v. Kraemer were read to hold that a state cannot enforce a discrimination which it could not itself make, the state could not probate, enforce, or administer many common bequests. The fourteenth amendment forbids discrimination not only on the basis of race or color; it also forbids discrimination on any basis which is capricious or whimsical. But any bequest which favors A rather than B may be capricious or whimsical. In such a case, the state could not by statute require the testator to leave his money to A. Apart, then, from bequests to special categories, to wife and children, for example - where an argument can be made that the category is reasonable - no bequest could be enforced if the enforcement were deemed to make the state responsible for the “discrimination.” Similarly, so long as an individual may capriciously decide who may, and who may not enter upon his property, the enforcement of trespass would not be possible, even where the exclusion had nothing to do with racial discrimination but was based on some other caprice. In the sale of land, too, the fact that the vendor arbitrarily contracted to sell to A rather than to B might be argued to prevent a court from enforcing the sale because the state would thereby make the arbitrary selection its own.43
 It is true that these learned authors are considering equality jurisprudence in the context of the Fourteenth Amendment “state action” doctrine; but their criticism is in fact being directed at the highly undesirable consequences of the direct horizontal application of this constitutional right to private legal relations. I wish to make it explicitly clear that I am not arguing for a situation where the Constitution or the law in general must tolerate a situation where bigotry is perpetuated and apartheid privatized through freedom of contract and other devices of the common law. What I am contending is that the law can deal effectively with these challenges through the very process envisaged by section 35(3), namely, the indirect radiating effect of the Chapter 3 rights on the post-constitutional development in the common law and statute law of concepts such as public policy, the boni mores, unlawfulness, reasonableness, fairness and the like, without any of the unsatisfactory consequences that direct application must inevitably cause. The common law of this country has, in the past, proved to be flexible and adaptable, and I am confident that it can also meet this new constitutional mandate.
 What also needs to be considered carefully is the impact, on the legislative process, of a directly horizontal application of Chapter 3 to private legal relationships. In each case when a final pronouncement of this nature is made through this Court, Parliament will be bound by this Court’s judgment. The Court has after all pronounced on the meaning and application of the Constitution in a particular context. Should Parliament wish to alter the law, resulting from such a direct application of the Chapter 3 rights by this Court, it will have to amend the Constitution. I consider this to be a most undesirable consequence, needlessly inhibiting the normal piecemeal statutory modification of the common law. It is one which directly flows, however, from a holding which in essence constitutionalises the entire body of private law. It could never, in my view, have been the intention of the framers to constitute Chapter 3 as a super civil code, to which the private common law is directly subject.
 I would sum up my views as follows. For the reasons given by Kentridge AJ in his judgment and those advanced above, the text of the Constitution, properly construed, strongly favours the conclusion that the direct horizontal application of Chapter 3 to private legal relations is not intended. Whatever lingering doubts there might be on this score are resolved by teleological considerations. The German experience bears this out. Direct application of the Chapter 3 rights, quite apart from the undesirable consequences already mentioned, will cast onto the Constitutional Court the formidable ultimate task of reforming the private common law of this country, a consequence which could not have been intended by the drafters. It turns the Constitution, contrary to the historical evolution of constitutional individual rights protection, also into a code of obligations for private individuals, with no indication in the Constitution as to how clashes between rights and duties are to be resolved, or how clashing rights are to be “balanced”; section 33(1) was clearly not designed and is quite inappropriate for this purpose. It would also be undesirable in a broader constitutional sense, pre-empting in many cases Parliament’s role of reforming the common law by ordinary legislation.
Justice of the Constitutional Court
 KRIEGLER J: The questions raised by this case are set out in the judgment of my colleague Kentridge AJ and repetition is unnecessary. However, two of them are of such importance that they require emphasis.1 They are indeed definitive of the scope of the Constitution with regard to time and subject matter: (a) Can the provisions of the Constitution be applied to events that took place before it came into force; and (b) Does Chapter 3 thereof2 govern only the relationship between the state and the individual or does it also govern relationships between individuals where law is involved?3
 With regard to the first question I agree with Kentridge AJ that the learned judge in the court below erred in holding that section 241(8) of the Constitution was dispositive of the case. I also agree with Kentridge AJ that the decision in Mhlungu does not avail the defendants in this case.4 Mhlungu’s case lends no support to their startling proposition that conduct which was unlawful when it was committed prior to the advent of the Constitution could be rendered lawful by such advent. The amendment the appellants sought in the court below was aimed at supporting a defence that the supervening protection of freedom of expression, afforded by section 15(1) of the Constitution, rendered lawful a publication which had taken place long before such protection was created.5
 I do not, however, agree with the reasoning of Kentridge AJ regarding the effect of sections 251(1), 7(2) and 98(6) of the Constitution.6 Although section 251(1) fixes the date of commencement of the Constitution, it does not purport to address the issue of retroactivity or retrospectivity. Nor do I believe that section 7(2) is really germane to that issue. Section 7 as a whole is not concerned primarily, if at all, with time, but with the scope of application, the reach, of Chapter 3.7 Certainly subsection (2) does not relate to the possible retrospective or retroactive operation of Chapter 3.
 That issue is, however, specifically addressed in section 98(6) of the Constitution, which has different rules for laws that existed at the time of the commencement of the Constitution and those made thereafter.8 With regard to pre-constitutional laws an order of invalidation does not generally operate retroactively, but the court can make its order invalidating a law to apply retroactively so as to “invalidate anything done or permitted in terms thereof before the coming into effect of such declaration of invalidity”. With regard to post-constitutional laws however, the invalidation, as a rule, does operate retroactively to the time the law was made.
 Kentridge AJ acknowledges that section 98(6) gives the courts the authority “in the interests of justice and good government” to pre-date the operation of a declaration of invalidity and contains “no express limit” to its retroactive power, but concludes that “it could hardly be suggested” that the date of commencement of the Constitution is not nevertheless to be read into section 98(6) as such limit.9 I am not convinced of the correctness of that conclusion and prefer to leave open whether a court could, in the “interests of justice and good government”, invalidate something done prior to 27 April 1994 under a law struck down for its fatal inconsistency with the Constitution.10 Such considerations manifestly do not arise here and I agree that the appeal against the order a quo refusing the proposed amendment falls to be dismissed. I therefore concur in paragraph 2(a) of the order proposed by Kentridge AJ.
 The second main question presented by this case is of greater importance for the future of our constitutional jurisprudence. It is also considerably more difficult to express with neutral precision. In the court below it was formulated too broadly.11 This Court’s refinement of the issues still did not focus the question sufficiently clearly.12 It was only in the course of judicial debate that the real question became starkly apparent. It is, quite simply, this: Does our constitutional law directly enforce the fundamental rights of persons proclaimed in sections 8 to 32 of the Constitution only as against the state or in all legal relationships? The learned judge a quo and the majority of my colleagues in this Court have concluded that the individual’s legal rights guaranteed by Chapter 3 of the Constitution are directly enforceable against the state alone. Their reasoning is cogent; the judicial and academic authority they cite is impressive. Nevertheless I am convinced that they err - and do so fundamentally.
 The legal issues involved are inherently complex. The conundrum is compounded by perceptions of its social, political and economic implications, as also by inarticulate premises, culturally and historically ingrained. It is therefore necessary to strip the problem down to bedrock. To that end two basic points should be made at the outset of the discussion. The first is that the debate is not one of “verticality” versus “horizontality”. As I hope to make plain shortly, it is common cause that Chapter 3 rights do not operate only as against the state but also “horizontally” as between individuals where statutes are involved. It is also common cause that where common law is involved the Chapter has a bearing. The true debate relates to the manner of its “horizontal” operation in common law relationships.
 The second point concerns a pervading misconception held by some and, I suspect, an egregious caricature propagated by others. That is that so-called direct horizontality will result in an Orwellian society in which the all-powerful state will control all private relationships. The tentacles of government will, so it is said, reach into the marketplace, the home, the very bedroom. The minions of the state will tell me where to do my shopping, to whom to offer my services or merchandise, whom to employ and whom to invite to my bridge club. That is nonsense. What is more, it is malicious nonsense preying on the fears of privileged whites, cosseted in the past by laissez faire capitalism thriving in an environment where the black underclass had limited opportunity to share in the bounty. I use strong language designedly. The caricature is pernicious, it is calculated to inflame public sentiments and to cloud people’s perceptions of our fledgling constitutional democracy. “Direct horizontality” is a bogeyman.
 One must of course, acknowledge that there are genuine jurisprudential concerns about the possible ramifications of “horizontality”. Kentridge AJ adverts to a formidable body of foreign and South African learning in support of “verticality”.13 And my colleague Ackermann J, in a characteristically learned and thoughtful judgment, has analysed some of the philosophical foundations supporting the more conservative view held by the majority of this Court. His discussion of the development of the horizontality debate in Germany and of the academic vagaries of Shelley v Kraemer14 provides much food for thought.
 That having been said, I must confess that I still cannot see the dire consequences feared by some jurists, including my esteemed colleague Ackermann J. Indeed, as I see it, it makes no fundamental difference with regard to such consequences whether the horizontal application of Chapter 3 is direct or indirect. It has little, if any, effect jurisprudentially or socially whether the Chapter 3 rights are enforced directly or whether they “irradiate” private legal relationships.15 I stress this point not only because of the impending doom Ackermann J perceives in the direct horizontal application of Chapter 3. I do so also because it would be foolish to ignore the public utterances of political and legal commentators in similar vein.
 But that is beside the point. This Court did not draft the Constitution. Our duty is to interpret, protect and uphold it,16 and to apply it without fear, favour or prejudice.17 Obviously one will not lightly conclude that the drafters of the Constitution intended it to bear a harsh or inequitable meaning, for such would fit ill with their avowed intention declared so solemnly in the Preamble18 and Postscript19 thereto. More pertinently, it would be difficult to reconcile such an interpretation with the spirit of freedom, equality and justice which pervades Chapter 3. It is also trite that the Constitution is to be interpreted purposively and as a whole, bearing in mind its manifest objectives. For that reason one would hesitate to ascribe a socially harmful or disruptive meaning to an instrument so avowedly striving for peace and harmony. One also knows that the Constitution did not spring pristine from the collective mind of its drafters. Much research was done and many sources consulted. It is therefore no surprise that the Constitution, in terms, requires its interpreters to have regard to precedents and applicable learning to be found in other jurisdictions.20 But when all is said and done, the answer to the question before us is to be sought, first and last, in our Constitution.
 In that regard my approach differs radically from that of the judge a quo21 and some of my colleagues who have prepared judgments in this case. The learned judge in the court below conducted extensive review of the constitutions of other countries22 and concluded:
The rule is that fundamental rights and freedoms are protected against State action only. Horizontal protection sometimes occurs to a limited extent but when it is intended over the broad field of human rights, it is expressly so stated.
It would in my view, therefore, be the correct approach to the interpretation of the Bill of Rights provisions in chap 3 of our Constitution to take the view that our Constitution is a conventional constitution unless there are clear indications to the contrary, either in respect of chap 3 as a whole or in respect of individual sections thereof.
There was a pressing need for a bill of rights, given the suppressive State action of the past. The call for a conventional bill of rights was sharp and clear. But there were no such calls for a bill of rights on a horizontal plane. The fundamental rights and freedoms now set out in chap 3 had not been curtailed by our common law. In fact they can be found enshrined therein. The removal of all authoritarian encroachment leads to their resuscitation.23
My colleagues have adopted basically the same approach and came to substantially the same conclusion.
 I make no apology for starting the exercise by examining the document with which we are concerned, the Constitution. Nor do I think one errs in commencing the exercise with an open mind, one untrammelled by subjective perceptions of evils of the past which the Constitution was intended to combat. The way I read the Preamble and the Postscript,24 the framers unequivocally proclaimed much more sweeping aims than those identified by the judge a quo,25 and apparently accepted by some of my colleagues. Our past is not merely one of repressive use of state power. It is one of persistent, institutionalized subjugation and exploitation of a voiceless and largely defenceless majority by a determined and privileged minority. The “untold suffering and injustice” of which the Postscript speaks do not refer only to the previous forty years, nor only to Bantu education, group areas, security and the similar legislative tools used by the previous government. The Postscript mentions “a divided society characterised by strife [and] conflict”. That is not a reference to governmental action only, or even primarily. The “reconciliation and reconstruction” mentioned in the last paragraph relate not so much, if at all, to the oppressed and the oppressive government, but to reconciliation of whites and blacks, to reconstruction of a skewed society. Likewise, when the Preamble speaks of “citizenship in a sovereign democratic constitutional state” the emphasis immediately falls on racial equality.
 The Constitution bracketed by that Preamble and that Postscript is unabashedly egalitarian and libertarian;
The past was pervaded by inequality, authoritarianism and repression. The aspiration of the future is based on what is justifiable in an open and democratic society based on freedom and equality’.26
This is underscored by the very presence of the Chapter 3 rights. It is re-emphasized by the unusual length and detail of the catalogue of rights and freedoms listed, as also by the broad sweep of much of its language. Nor is it coincidental that the very first right enunciated is that of equality before, and equal protection of, the law.27 It is equally significant that such right is coupled with a most detailed, if not vehement, proscription of discrimination.28 Viewed in context, textually and historically, the fundamental rights and freedoms have a poignancy and depth of meaning not echoed in any other national constitution I have seen.
 It is therefore no spirit of isolationism which leads me to say that our Constitution is unique in its origins, concepts and aspirations. Nor am I a chauvinist when I describe the negotiation process which gave birth to that Constitution as unique; so, too, the leap from minority rule to representative democracy founded on universal adult suffrage; the Damascene about-turn from executive directed parliamentary supremacy to justiciable constitutionalism and a specialist constitutional court, the ingathering of discarded fragments of the country and the creation of new provinces; and the entrenchment of a true separation and devolution of powers. Nowhere in the world that I am aware of have enemies agreed on a transitional coalition and a controlled two-stage process of constitution building. Therefore, although it is always instructive to see how other countries have arranged their constitutional affairs, I do not start there. And when I do conduct comparative study, I do so with great caution. The survey is conducted from the point of vantage afforded by the South African Constitution, constructed on unique foundations, built according to a unique design and intended for unique purposes.
 And I respectfully suggest that if one does so logically, applying the interpretational tools with which lawyers are familiar, the Constitution holds no hidden message of so-called verticality. It is common cause that it nowhere says that Chapter 3 governs only the relationship between the state and the individual. On the contrary, section 4(1), which governs the whole of the Constitution, including Chapter 3, pronounces the Constitution to be “the supreme law of the Republic”. Perhaps the word “supreme” has been devalued by overuse and perhaps the word “law” has become mundane. But in combination they have a singular ring, even more tellingly expressed in the Afrikaans text, “die hoogste reg.” By way of emphasis the subsection expressly sets at nought “any law or act [“enige wet of handeling”] inconsistent with its provisions”. Conformably subsection (2) of section 4 bluntly declares the Constitution to be binding on “all legislative, executive and judicial organs of state at all levels of government”. The two subsections, read singly and jointly, are quite unequivocal, I would suggest. First the Constitution is elevated to supremacy over all law, and then all organs of state are enjoined to honour and enforce that supremacy.
 One then turns to Chapter 3, with which we are more specifically concerned here. It is immediately apparent that section 7(2), an integral part of the section which governs the scope of Chapter 3, proclaims that it applies to “all law in force . . . during the period of operation of this Constitution”. Although the Afrikaans equivalents of the word “law” are not used consistently in the Constitution, it is of some significance that in section 7(2) the words used in the Afrikaans text is “reg”, the generic term for all kinds of law. Qualified as it is by the adjective “alle” (“all”), the broad generality of the statement is evident.
 On a reading of section 7(2) alone, the scope of application of Chapter 3 with regard to law therefore seems clear: It governs all law in force during the currency of the Constitution. There is no qualification, no exception. All means all. As the Chapter governs all administrative decisions taken and all administrative acts performed, so it governs all law. And it goes without saying that that is the law the courts of the land, the “judicial organs of state” bound by section 4(2) of the Constitution, are obliged to apply in the performance of their duties. The manifest intention of the drafters of the subsection was to expand its scope to the widest limit their language could express.
 If one then reads the subsection in the context of section 7 as a whole, that conclusion is reinforced. Subsection (1), which deals with executive and legislative organs of state also casts the net as wide as language permits. All of them at all levels of government are bound.29 In turn subsections (3) and (4), which determine the categories of persons entitled to invoke the protection of the Chapter, push back the perimeter. The extension of fundamental rights and freedoms to juristic persons, an unusual step in bills of rights, is consistent with the broad generality of subsections (1) and (2). So is the extension of locus standi in subsection (4) to representative actors, group, class and even public interest claimants.
 Two further points flow from a reading of section 7. Having regard to the clear intention, running throughout the section, to stretch the purview of Chapter 3 to its outermost boundaries, is it at all likely that a limitation as restrictive as that entailed in so-called verticality could have been contemplated? The Preamble of the Constitution proclaims the need to create a new order in which all citizens will be able to exercise their fundamental rights. Chapter 3 spells out those rights and the Postscript envisages a future founded on their recognition. Yet it is contended that, for instance, a woman whose right to human dignity under section 10, or to privacy under section 13, is infringed by a private individual cannot invoke the protection of those sections where there is a legal relationship involved. I find that startling. Secondly, if indeed the drafters had such a major constraint in mind, why did they not say so? Instead they wax expansive, leaving it to the microscope of a “verticalist” to pick up hidden clues.
 But it is not only section 7 that points, I suggest inexorably, to my simple conclusion. It should be read with section 33(2).30 It says that no legal rule of any kind whatsoever which is not saved by subsection (1), or by some other provision in the Constitution itself, limits any right entrenched in Chapter 3. The sweep of section 33(2) harks back to the generality of section 7. If the Chapter were intended to operate only vertically, or only indirectly horizontally, why was it necessary, or indeed appropriate, to declaim the preservation of rights in such unqualified terms? What makes the breadth of section 33(2) more significant is that it follows immediately upon and serves to qualify the general saving clause, section 33(1).
 Subsections (1), (3) and (4) of section 33 are themselves wholly consistent with the plain meaning of section 7(2) for which I contend. Nowhere in those subsections is there any suggestion that Chapter 3 does not apply to all law. Section 33(4) is seen by some as supporting a vertical reading of the Chapter. To my mind it does nothing of the sort. The subsection reads as follows:
This Chapter shall not preclude measures designed to prohibit unfair discrimination by bodies and persons other than those bound in terms of section 7(1).
Section 7(1) applies expressly to organs of state. With or without “measures designed to prohibit unfair discrimination”, they are prohibited from doing so. But there is nothing in Chapter 3 which prohibits any body or person other than an organ of state from discriminating, unfairly or otherwise. What section 33(4) does is to keep the door open for “measures”, whether legislative or executive, prohibiting unfair discrimination in the private sphere. In short, the government is free to introduce a civil rights act, unfair housing practices legislation and the like.
 Interpreting section 33(4) as pointer to verticality illustrates what I regard as a fundamental misconception of the true import of Chapter 3. The Chapter has nothing to do with the ordinary relationships between private persons or associations. What it does govern, however, is all law, including that applicable to private relationships. Unless and until there is a resort to law, private individuals are at liberty to conduct their private affairs exactly as they please as far as the fundamental rights and freedoms are concerned. As far as the Chapter is concerned a landlord is free to refuse to let a flat to someone because of race, gender or whatever; a white bigot may refuse to sell property to a person of colour; a social club may black-ball Jews, Catholics or Afrikaners if it so wishes. An employer is at liberty to discriminate on racial grounds in the engagement of staff; a hotelier may refuse to let a room to a homosexual; a church may close its doors to mourners of a particular colour or class. But none of them can invoke the law to enforce or protect their bigotry. One cannot claim rescission of a contract or specific performance thereof if such claim, albeit well-founded at common law, infringes a Chapter 3 right. One cannot raise a defence to a claim in law if such defence is in conflict with a protected right or freedom. The whole gamut of private relationships is left undisturbed. But the state, as the maker of the laws, the administrator of laws and the interpreter and applier of the law, is bound to stay within the four corners of Chapter 3. Thus, if a man claims to have the right to beat his wife, sell his daughter into bondage or abuse his son, he will not be allowed to raise as a defence to a civil claim or a criminal charge that he is entitled to do so at common law, under customary law or in terms of any statute or contract. That is a far cry from the spectre of the state placing its hand on private relationships. On the contrary, if it were to try to do so by legislation or administrative action, sections 4, 7(1) and the whole of Chapter 3 would stand as a bastion of personal rights.
 To all of that there is the overarching qualification contained in section 33(1), which, it should be noted, draws no distinction between different categories of law of general application (“algemeen geldende reg”). A rule of the common law which, for example, infringes a person’s right to privacy or human dignity, can be saved if it meets the section 33(1) requirements. And it is irrelevant whether such rule is statutory, regulatory, horizontal or vertical, and it matters not whether it is founded on the XII Tables of Roman law, a Placaet of Holland or a tribal custom. That is why section 33(2) makes plain, without any question, that no law of whatever kind or origin can impinge upon any Chapter 3 right unless it is saved by section 33(1) or some other provision of the Constitution.
 For the purpose of the present analysis one can skip section 34, which deals with suspension of rights during a state of emergency. It is not relevant to the question under consideration. That brings one to section 35, the general and final provision of Chapter 3. Section 35(1) is directed at interpretation of the Chapter itself,31 and subsection (2) at the interpretation of the laws to be measured against the template of the Chapter.32 Subsection (3), in turn, says how any statute is to be interpreted and how the common law and customary law are to be applied and developed. Two points should be noted. The subsection applies to the whole body of South African law, any statute and common law and customary law in general. And what is more important, the rules of statutory or other law to which it applies are not limited to those directly struck by the provisions of the Chapter. The intention of the drafters of the Constitution in enacting section 35(3), and in adding it as the last word on Chapter 3, therefore seems clear. Even in those cases where the provisions of the Chapter do not directly apply, the rules of law applicable are to be informed by the “spirit, purport and objects” thereof.
 The point should also be stressed that on any reading of section 35(3) it obliges all courts to interpret every law and to apply all common law in the light of the “spirit, purport and objects” of Chapter 3. Therefore, if the fundamental objection to my reading of Chapter 3 is that it superimposes its provisions on legal relationships between individuals, why is the performance by a court of the duty imposed on it by section 35(3) regarded as any less “undesirable and unsupportable”?33 Also, if the objection is, as was held in the court below, that the application of Chapter 3 to all law introduces uncertainty in what has long been settled, how can it lead to any less uncertainty for courts to “have due regard to the spirit, purport and objects” of the Chapter? In fact, resort to section 35(3) is likely to create even more uncertainty, as the phrase “hav[ing] due regard” is surely a vague concept and “spirit, purport and objects” no less so.
 The internal logic and cohesion of Chapter 3 is manifest. All organs of state in all their decisions and actions are bound by the terms of the rights. So too are any resorts to law by anybody. In deciding whether one or other rights is infringed by such decision, action or law, a court interprets the Chapter in accordance with section 35(1). If the alleged infringement is caused by a statute, the court tries to read the statute down in terms of section 35(2). If, after all that, an entrenched right is indeed infringed, the multiple test contained in section 33(1) is applied. If the infringement fails to pass that test it must be invalidated to the extent that it limits the particular entrenched rights. Over and above that, all courts at all times in the application of any legal rule are obliged to consider the possible effect of the spirit, purport and objects of the Chapter 3 rights on the particular cases before them. Even where a specific right or freedom is not assailed or threatened, the Chapter is still relevant and has to be borne in mind.34
 I wish to dwell a little longer on the internal logic of Chapter 3. Kentridge AJ asks, in relation to section 35(3), “why such a provision would be needed if the Chapter could be directly applied to common law disputes between private litigants.”35 The answer, he finds, is that it is needed to govern that very category of disputes. That, I would respectfully suggest, is begging the question. And that is foundational to his reasoning. I, for my part, agree that section 35(3) is manifestly intended to govern a residuary category not governed by what has gone before in Chapter 3. But I do not find it necessary to make any prior assumption regarding the nature of that category. And I do not think section 35(3) answers the question what is included in the category.
 Section 35(3) answers the question what courts do when there is no direct infringement or claim of an infringement of a right protected under the Chapter. This includes cases dealing with statutory law, common law and customary law. It mandates that all courts - this Court, the Supreme Court including the Appellate Division, the Land Court, the magistrates courts, the labour courts and all other courts - in interpreting statutory law and when applying and developing common and customary law, always have regard to the “spirit, purport and objects” of the Chapter. This includes courts with constitutional jurisdiction and courts without constitutional jurisdiction. The purpose is that this Constitution is to permeate all that judges do just as it is to permeate all that the legislature and the executive do, conformably under section 7(1) of the same Chapter.
 In what situation then is section 35(3) used? It is used in all cases before any court in which there is no direct challenge based on one or more of the rights and freedoms protected in Chapter 3. If a court case involves the interpretation of a statute and there is no claim of unconstitutionality, the judge in that case is bound to have regard to the spirit, purport and objects of Chapter 3. If a court case involves the application of customary law in which there is no claim of unconstitutionality based on a Chapter 3 right, the judge is bound to have regard to the spirit, purport and objects of the Chapter. If a matter before the Appellate Division between two private parties involves the common law, and there is no claim of a violation of a Chapter 3 right, the Appellate Division must have due regard to the spirit, purport and objects of the Chapter. This is the “indirect” application of the Chapter sought to be introduced by Kentridge AJ at paragraph 60 of his judgment. But such indirect application can neither be limited to the situation of common law where both parties are private, nor can it be expanded to include the common law where both parties are private where there is a direct constitutional challenge based on a Chapter 3 right.
 Kentridge AJ, in making the category of cases to which 35(3) applies both over-inclusive and under-inclusive, is forced to argue that the “indirect” application is limited to the jurisdiction of the Supreme Court, but reserving for this Court some overriding review power akin to that of our German opposite number.36 Resort to German jurisprudence or the German constitutional model is particularly unnecessary and unhelpful here.37 There the system is based on a distinction between private and public law, which is not appropriate for us. Second, there is no clear indication that section 35(3) should be read to mean that the Appellate Division as the court of final appeal does the developing and applying of the common law to the exclusion of this Court. On the contrary, section 101(5), which Kentridge AJ seems to dismiss,38 to my mind establishes that such interpretation cannot possibly be correct. Furthermore, section 98(2) gives this Court the authority to apply and develop the common law in accordance with the Constitution.39 This Court applies and adapts the common and customary law while directly applying the Chapter 3 rights. The Appellate Division does it with due regard to the spirit, purport and objects of the Chapter.
 Turning to the particular case before us, in the result, Chapter 3 rights apply to all law, including the common law of defamation. In this regard, I endorse the sentiments expressed by Kentridge AJ in the later part of paragraph 58 of his judgment regarding the manner in which the common law should be developed. Indeed, balancing the competing rights involved in defamation (freedom of expression against the right to human dignity) is not novel. Nor does the advent of the Constitution, which codifies them, warrant the wholesale importation of foreign doctrines or precedents. To be true we are to promote values not yet rooted in our traditions and we must have regard to applicable public international law. We are also permitted to have regard to foreign case law. But that does not amount to a wholesale importation of doctrines from foreign jurisdictions.
 My reading of Chapter 3 gives to the Constitution a simple integrity. It says what it means and means what it says. There is no room for the subtleties and nice distinctions so dear to the hearts of mediaeval theologians and modern constitutional lawyers. The Constitution promises an “open and democratic society based on freedom and equality”, a radical break with the “untold suffering and injustice” of the past. It then lists and judicially safeguards the fundamental rights and freedoms necessary to render those benefits attainable by all. No one familiar with the stark reality of South Africa and the power relationships in its society can believe that protection of the individual only against the state can possibly bring those benefits. The fine line drawn by the Canadian Supreme Court in the Dolphin Delivery case40 and by the US Supreme Court in Shelley v Kraemer41 between private relationships involving organs of state and those which do not, have no place in our constitutional jurisprudence. Nor are we consigned to the hypocrisy so trenchantly excoriated by the authors of the two Canadian articles quoted in Baloro and Others v University of Bophuthatswana and Others.42
 What is more, my reading of the Constitution avoids jurisprudential and practical conundrums inherent in the vertical - but - indirectly - horizontally - irradiating interpretation. One does not need to ascertain whether a question is one of public or private law (wherever the boundary may lie in our legal system); one is not confronted with knotty problems where a private relationship is, wholly or partially, governed by statute; nor where an organ of state is a party to a manifestly private law dispute, for example flowing from contract or delict. There are no anomalies where one sues a policeman and his minister in delict or when an organ of state and a private person are co-plaintiffs or co-defendants. Nor is it of any consequence that a rule of the common law derives from an ancient statute of a former government or from the writings of a legal sage of old. The law is the law; where the Chapter fits, it is applied; where it does not, its spirit, purport and objects are duly regarded.
 I find it unnecessary to engage in a debate with my colleagues on the merits or demerits of the approaches adopted by the courts in the United States, Canada or Germany. That pleases me, for I have enough difficulty with our Constitution not to want to become embroiled in the intricacies of the state action doctrine, Drittwirkung and the like. I must say, however, that I agree with Ackermann J and the eminent authorities he quotes in their misgivings about the decision in Shelley v Kraemer.43 I do not agree, however, that the decision in that case has any bearing on the issue now under discussion. We do not operate under a constitution in which the avowed purpose of the drafters was to place limitations on governmental control.44 Our Constitution aims at establishing freedom and equality in a grossly disparate society. And I am grateful to the drafters of our Constitution for having spared us the jurisprudential gymnastics forced on some courts abroad. They were good enough to say what they mean. The Constitution applies to all three of the pillars of state and Chapter 3 applies to everything they do.
 There is a further pertinent consequence of my reading of the Constitution, more particularly of Chapter 3. That relates to jurisdiction. Section 98 of the Constitution created a new court charged with the duty to exercise jurisdiction “as the court of final instance over all matters relating to the interpretation, protection and enforcement of” its provisions.45 At the same time the Appellate Division of the Supreme Court retained its status and jurisdiction as the court of final instance in all non-constitutional matters.46 Anticipating the jurisdictional confusion that could ensue, the drafters of the Constitution, although conferring constitutional jurisdiction on provincial and local divisions of the Supreme Court, expressly provided in section 101(5):
The Appellate Division shall have no jurisdiction to adjudicate any matter within the jurisdiction of the Constitutional Court.
Of course, that refers back to section 98, and more specifically in this context, to subsection (2)(a) thereof. That means that in case of a claimed violation (or threatened violation) of one or other right entrenched in Chapter 3, this Court has jurisdiction to the exclusion of the Appellate Division. Where no such violation or threat is in issue the Appellate Division retains its erstwhile jurisdiction. And subsections (4) to (7), (11) and (12) of section 102 make detailed provision for the interface between the jurisdictions of the two courts of final instance. The Appellate Division has been deprived of none of its former powers. It has merely not been afforded additional jurisdiction to deal with constitutional issues, a new category introduced by the Constitution. It follows that the Appellate Division is certainly not excluded from the purview of section 35(3). It, like every other court in the land, is not merely empowered but mandated to interpret any statute, and to apply and develop the common law and customary law, with due regard to the spirit, purport and objects of Chapter 3. The appeal in every case not involving one or other of the Constitutional Court’s powers lies to the Appellate Division. Therefore, if a case turns on the alleged infringement or threatened infringement of a right specified in Chapter 3, this Court has exclusive final jurisdiction. Where the issue is whether a statute should be interpreted or the common law should be adopted to meet the exigencies of a case not falling within the protective ambit of an entrenched right, the jurisdiction of the Appellate Division is not ousted.
 In the instant case the appellants tried to invoke the protection of section 15(1). That was clearly a claim that the common law of defamation infringed a specific right protected by Chapter 3. That is an issue falling under section 98(2)(a), in respect of which this Court has jurisdiction and not the Appellate Division. The appellants however must fail on that issue. They are not entitled to invoke the provisions of the Constitution to validate post-constitutionally what was unlawfully done before. They had and have a direct pecuniary interest in the case and pursued it with personal, not public, motives. To mulct them in costs would therefore be just.
 I would therefore make the following order:
1. The appeal is dismissed with costs, including the costs consequent upon the engagement of two counsel.
2. The two questions referred by the judge a quo, as reformulated by this Court, are answered as follows:
(a) No: The defendants in this case are not entitled to invoke the provisions of the Constitution; and
(b) Yes: The provisions of Chapter 3 of the Constitution are capable of application to a relationship other than that between persons and legislative or executive organs of state at all levels.
3. The three subsidiary questions formulated by this Court are answered as follows:
(a) No: It was not competent to raise in this appeal the issue whether the common law of defamation should be developed to make it consistent with the Constitution;
(b) The development of the common law is within the jurisdiction of both courts, but the extent to which - because of the answer to question (a) - has been left open;
(c) Given the answer to questions (a) and (b), it is not necessary to answer this question.
Justice of the Constitutional Court
Didcott J concurs in the judgment of Kriegler J.
 MADALA J: I have had the benefit of reading the various judgments which have been prepared by my colleagues in this case. I agree with the findings of Kentridge AJ on the referral and on retrospectivity. The advent of the constitutional regime did not render, and could not have been intended to render, lawful that conduct which was unlawful prior to the commencement of the interim Constitution (“Constitution”) on 27 April, 1994. Accordingly, I agree that the Defendants in this case are not entitled to invoke the provisions of the Constitution. He has also, correctly in my view, found that in this case the Defendants can place no reliance on the specific provisions of Section 15(1) of Chapter 3 of the Constitution, and that this appeal falls to be dismissed with costs.
 At this point I part ways with my colleague Kentridge AJ, when he states that the provisions of Chapter 3 of the Constitution are not in general capable of application to any relationship other than that between persons and legislative or executive organs of the State at all levels of government (my underlining). At paragraph 62 he states that he has come to the conclusion that:
“... Chapter 3 does not have a general direct horizontal application but that it may and should have an influence on the development of the common law as it governs relations between individuals.” (Footnotes omitted)
He has, however, left open the possibility that a litigant in another case may argue that some particular provision of Chapter 3 must, by necessary implication, have direct horizontal application. It is on the aspect of the so-called horizontality that I wish to explain my position and I shall do so briefly.
 The very incisive and analytical judgment of Kentridge AJ might put to rest, for the time being, at least, and in respect of this case only, the much debated and raging controversy around whether or not Chapter 3 of the Constitution has horizontal application. In my view, Kentridge AJ deals with the question referred to us on too narrow a basis, yet this is one of the most controversial issues in present day South African law.
 It is traditionally accepted that Bills of Rights are intended primarily to correct imbalances between the excesses of government power and individual liberty. This is the so-called vertical application of the Bill of Rights. The proponents of the verticality approach seek to confine constitutional challenges to legislative and administrative actions only, arguing that this is the primary function of a Bill of Rights, that in such a case the legal implications are predictable and that any possible abuse of private powers should be dealt with through legislation. I agree that our Constitution has vertical operation and deal with this aspect no further. However, I do not subscribe to the view that its operation is limited to verticality only. In my view, some of the rights entrenched in Chapter 3 also operate directly in the area of relations between private individuals. Those who would widen the scope of the operation of the Bill of Rights hold the view that the verticality approach is unmindful of the modern day reality - that in many instances the abuse in the exercise of power is perpetrated less by the State and more by private individuals against other private individuals. Our courts are very divided on the issue.
 In Mandela v Falati 1 Van Schalkwyk J came to the conclusion that Chapter 3 did have horizontal application and that the provisions of the Constitution could be enforced in a private dispute. His reasoning for this was based on an analysis of the right in question which, in the particular case, was freedom of expression. He found that any limitation of this right had to be necessary if the right is related to political activity. And he went on to find that political activity was not confined to that which occurred between the State and the citizen only but also between private citizens or individuals. His view was that the spirit, purport and objects of Chapter 3 were to extend fundamental rights beyond those circumstances for which the common law provides. In Kalla v The Master & Others2 Van Dijkhorst J had left open the question whether the Constitution had horizontal application. In another defamation case, Gardener v Whitaker 3 Froneman J, held that the Constitution was obviously primarily concerned with the protection of individual rights against State action. He found, however, that the Constitution was also concerned that the entire legal system, including the common law and customary law, should accord with the broader values of the Constitution. In De Klerk v Du Plessis,4 the forerunner to the present appeal, Van Dijkhorst J found that most constitutions emphasised the curtailment of State powers and argued that ours should be similarly interpreted, unless there were clear indications to the contrary. He considered that an alternative interpretation, one finding that the Constitution had horizontal application, was “an extremely unattractive” one and would result in “legal uncertainty on an unprecedented scale”. In Jurgens v Editor Sunday Times and Another5 the question of horizontality was raised in the context of an application to amend pleadings, but there was no need to decide the issue at that stage. In Motala and Another v University of Natal6 Hurt J held that Sections 8 and 32 applied horizontally. In Potgieter en n ander v Kilian7 McLaren J was faced with the problem of deciding whether the provisions of Chapter 3 of the Constitution apply not only as between the State and the individual, but also between private entities. He came to the conclusion that there was no indication in the history and origin of the Constitution that Chapter 3 was intended to have horizontal application.
 Because of these conflicting decisions, clarity on the issue is needed and the ruling of the Constitutional Court is eagerly awaited. The question involves a consideration of whether or not a private individual can institute action or sustain a defence against another private individual on the basis of the violation of a right contained in Chapter 3. There is no simple answer to the question whether an alleged breach of a fundamental right contained in Chapter 3 could found an action between private individuals, but the answer to this question must be sought in the provisions of the Constitution itself, having regard to the underlying values and objects of the Constitution.
 I begin my analysis of the matter back-to-front’, and consider first the underlying values and objects of the Constitution in general and Chapter 3 in particular. Our Constitution has both a pre-amble and what has variously been called a post-amble, after-amble, post-script or epilogue. Whichever term one uses, both the pre-amble and the post-amble indicate the general purpose for which the people ordained and established the Constitution. The pre-amble envisages the creation of as democratic a climate as is possible in the ushering in of a democratic Constitution for a new legal order and emphasises equality, fundamental rights and freedoms, national unity and a restructuring of society. It is a document that seeks to transform the status quo ante into a new order, proclaiming that -
“... there is a need to create a new order in which all South Africans will be entitled to a common South African citizenship in a sovereign and democratic constitutional state in which there is equality between men and women and people of all races so that all citizens shall be able to enjoy and exercise their fundamental rights and freedoms”;
The post-amble reminds us that the Constitution is a -
“...historic bridge between the past of a deeply divided society characterised by strife, conflict, untold suffering and injustice, and a future founded on the recognition of human rights, democracy and peaceful co-existence and development opportunities for all South Africans, irrespective of colour, race, class, belief or sex.”
Section 35(1) provides that in interpreting the Bill of Rights a court shall promote the values which underlie an open and democratic society based on freedom and equality. Freedom and equality underlie the vision of democracy embodied in the Constitution. The Constitution sets out the values which we must uphold, and in particular enjoins us to recognise a person’s status as a human being.
 My colleague, Mahomed J (as he then was), dealing in S v Makwanyane and Another8 with the values and objects of our Constitution, states that -
“The South African Constitution is different: it retains from the past only what is defensible and represents a decisive break from, and a ringing rejection of, that part of the past which is disgracefully racist, authoritarian, insular, and repressive and a vigorous identification of and commitment to a democratic, universalistic, caring and aspirationally egalitarian ethos, expressly articulated in the Constitution. The contrast between the past which it repudiates and the future to which it seeks to commit the nation is stark and dramatic. The past institutionalized and legitimized racism. The Constitution expresses in its preamble the need for a new order .. in which there is equality between ... people of all races’. Chapter 3 of the Constitution extends the contrast, in every relevant area of endeavour (subject only to the obvious limitations of section 33). The past was redolent with statutes which assaulted the human dignity of persons on the grounds of race and colour alone; section 10 constitutionally protects that dignity. The past accepted, permitted, perpetuated and institutionalized pervasive and manifestly unfair discrimination against women and persons of colour; the preamble, section 8 and the postamble seek to articulate an ethos which not only rejects its rationale but unmistakenly recognises the clear justification for the reversal of the accumulated legacy of such discrimination. The past permitted detention without trial; section 11(1) prohibits it. The past permitted degrading treatment of persons; section 11(2) renders it unconstitutional. The past arbitrarily repressed the freedoms of expression, assembly, association and movement; sections 15, 16, 17 and 18 accord to these freedoms the status of "fundamental rights". The past limited the right to vote to a minority; section 21 extends it to every citizen. The past arbitrarily denied to citizens on the grounds of race and colour, the right to hold and acquire property; section 26 expressly secures it.”
In a nutshell, these are the underlying values and objects of the Constitution. These are the imbalances which the Constitution seeks to redress. The theme of “an open and democratic society based on freedom and equality” runs throughout the Bill of Rights9 - obviously to facilitate the transition from an apartheid society to a democratic society.
 The first question to be asked, it would seem, is whether the provisions of Chapter 3 apply to the common law. Put differently, the question at issue is whether or not a rule of common law can be attacked on the basis of Chapter 3 in litigation between private parties not involving any legislative or executive authority. In my view, there can be no doubt that the provisions of Chapter 3 do apply to the common law. Section 4(1) of the Constitution proclaims the supremacy of the Constitution and the subjection of any law or act to its provisions, stating that:
“... any law or act inconsistent with its provisions shall, unless otherwise provided expressly or by necessary implication in this Constitution, be of no force and effect to the extent of the inconsistency.”
In my view, this language is broad enough to include the common law. To hold otherwise would, in my view, exclude from Chapter 3 application a whole body of the common law which to a great measure, governs the rights and obligations of individuals.10
Section 7 (2) puts the matter beyond any doubt, that Chapter 3 of the Constitution applies to the common law. This Section states:
“This Chapter shall apply to all law in force and all administrative decisions taken and acts performed during the period of operation of this Constitution.” (My underlining)
In interpreting Section 7 (2), read systematically broadly and teleologically in conjunction with the pre-amble, which proclaims “a need to create a new order”, the broad view must prevail so that “all law” includes statutory, common and customary law. That Chapter 3 applies to the common law is further evidenced by Sections 33 (2) and (3) which state:
“(2) Save as provided for in subsection (1) or any other provision of this Constitution, no law, whether a rule of the common law, customary law or legislation, shall limit any right entrenched in this Chapter.
(3) The entrenchment of the rights in terms of this Chapter shall not be construed as denying the existence of any other rights or freedoms recognised or conferred by common law, customary law or legislation to the extent that they are not inconsistent with this Chapter.” (My underlining)
Furthermore, in my view, certain provisions contained in Sections 33(2)-(4) would have been unnecessary as regards private law matters if the fundamental rights provisions applied only vertically. It is, therefore, my view that our Chapter 3 has not gone as far as subjecting the State to its rigours only, but that it has ventured out and colonised the common law.
 The second leg of the inquiry then is whether such application of Chapter 3 to the common law is to arise in consequence of the direct application of the relevant Chapter 3 right, or through the mechanism of interpreting, applying and developing the common law by having regard to the spirit, purport and objects of the Chapter in terms of Section 35(3), or both. It is my view that our Constitution provides for both.
 If the proposition is accepted that the basic concern of the Constitution is to transform the South African society and the legal system into one that upholds democratic principles and human rights, between, inter alia, the State and the individual and between individuals inter se, there must be instances which call, in so far as private relationships between individuals are concerned, for the direct application of the provision of Chapter 3 between such individuals. As a matter of interpretation, certain provisions of the Chapter have direct horizontal application. We should examine every enumerated right and decide whether it can sensibly be applied in the private domain. In support of this approach, it all depends on the nature and extent of the particular right, the values that underlie it, and the context in which the alleged breach of the right occurs.11
As stated above, Chapter 3 has several provisions that specifically regulate labour relations. Section 27 gives every person the right to fair labour practices and further gives workers the right to form and to join trade unions, to organise and bargain collectively and to strike and similar or related rights. Further, Chapter 3 provides for the establishment of private educational institutions on the one hand but prohibits discrimination on the basis of race on the other,12 that every child shall have the right not to be subjected to neglect or abuse,13 and that every person shall have the right to an environment which is not detrimental to his or her health or well-being.14 It is, however, not necessary for purposes of this judgment, nor was it argued before us, that we should provide guidelines for determining which provisions of Chapter 3 apply directly horizontally or to enumerate them. Consequently I concur with Mahomed DP on this score.
 I have had the benefit of perusing the learning of a few foreign jurisdictions on the application of their Bills of Rights to private relations. In most instances, and that is why direct horizontality has been rejected, there were no problems in the relationships of the citizens inter se as they were from relatively homogeneous societies. On this issue our Bill of Rights, because of the peculiar circumstances and the period within which it was drafted, cannot be properly or fairly compared to oft quoted instruments such as the Canadian Charter, the United States Constitution and the German Basic Law. The German Basic Law was a reaction to the genocide perpetrated on the minority by the government of the day. Canadian jurisprudence does not support direct horizontality.15 Canadian authorities supporting this view state -
“Such actions as an employer restricting an employee’s freedom of speech or assembly, a parent restricting the mobility of a child or a landlord discriminating on the basis of race in his selection of tenants cannot be breaches of the Charter, because in no case is there any action by the Parliament or government of Canada or by the Legislature or government of a province. In cases where private action results in a restriction of a civil liberty, there may be a remedy for the aggrieved person under a human rights code, under labour law, family law, tort law, contract law or property law, or under some other branch of the law governing relations between private persons; but there will be no breach of the Charter.”16
 Ours is a multi-racial, multi-cultural, multi-lingual society in which the ravages of apartheid, disadvantage and inequality are just immeasurable. The extent of the oppressive measures in South Africa was not confined to government/individual relations but equally to individual/individual relations. In its effort to create a new order, our Constitution must have been intended to address these oppressive and undemocratic practices at all levels. In my view our Constitution starts at the lowest level and attempts to reach the furthest in its endeavours to restructure the dynamics in a previously racist society.
 I agree with Kentridge AJ on the interpretation of Section 35(3) which explicitly provides that our courts should develop the common law and customary law with due regard to the “spirit purport and objects” of Chapter 3. This, in my view, provides for the indirect horizontal “seepage” in those areas which are not touched directly by the provisions of Chapter 3. As Froneman J put it in Gardener v Whitaker:17
“After all, the past of a deeply divided society characterised by strife, conflict, untold suffering and injustice’ (words used in the unity and reconciliation’ section of the Constitution) is not merely a history of repressive State action against individuals, but it is also a history of structural inequality and injustice on racial and other grounds, gradually filtering through to virtually all spheres of society since the arrival of European colonists some three and a half centuries ago, and it will probably take generations to correct the imbalance. But the development of the law by the courts is by its very nature dependent on litigation and therefore likely to be incremental and perhaps slow, hence the provision for State intervention also, by virtue of section 33(4), to prohibit unfair discrimination by private persons and bodies.” (My underlining)
The provisions of Section 35(3) are in effect an express adoption of the German model of “Drittwirkung”. In my view, it is the task of the Supreme Court to oversee this development. The law is always changing. The Supreme Court has always participated on an active basis in the adjudication of the common law rules. What is now required of it is that in disputes between private individuals it should balance their competing rights as envisaged in the Constitution.
 In the result and as already indicated, I agree with paragraphs 1 and 2 (a) of the order proposed by Kentridge AJ. On the question of horizontality, I am of the view that some of the rights in Chapter 3 lend themselves to direct horizontality while in respect of others, Chapter 3 is indirectly horizontally applicable. However, the defendants in the present case cannot rely on the provisions of Section 15 of the Constitution.
Justice of the Constitutional Court
 MOKGORO J: Having had the privilege of reading the judgments both of Kentridge AJ, and of Kriegler J, I concur in the judgment of Kentridge AJ and in the order he proposes. I also find myself in respectful agreement with the views expressed in the concurring opinion of Mahomed DP.
 The importance of the application question prompts me to write this brief concurring note to emphasise my view that, notwithstanding that Chapter 3 of the Constitution may not be invoked by one private litigant against another to challenge a rule of common or customary law, section 35(3) of the Constitution assigns to courts an affirmative responsibility to apply and develop both common law and customary law in a manner that imbues both systems of law with the values embodied in Chapter 3.
 The unique and stark reality in South Africa is that decades of injustice associated with apartheid gave rise to gross socio-economic inequalities that persist at every level of our society. The disparities between the beneficiaries of state-imposed racial discrimination and its victims, which will doubtless endure for many years to come, makes oppression and discrimination in the “private” sphere both possible and likely. Indeed, in practical terms, the average South African may now be more likely on a day-to-day basis to have her or his human dignity and other fundamental rights threatened by the actions of entities and individuals who are not in any sense organs of state, than by agents clothed with public power.
 But however desirable it might be that Chapter 3 should have what has been termed “direct horizontal” application, the better to swiftly mitigate the worst of those inequities which cannot be ascribed to actions or neglect of the public authorities of the present day, after considering the incisive textual analysis contained in the opinion of Kentridge AJ, I find myself persuaded that the Constitution we are called upon to interpret today simply does not make provision for such “private” application in the ordinary course.
 That said, I would underscore what my brothers Kentridge and Kriegler have both made clear: the Constitution is the “supreme law of the Republic1", and Chapter 3 thereof applies to “all law in force2", which includes common and customary law, as well as statutory law. It follows that the realm of so-called private law, whether embodied in legislation, common law, or customary law, is by no means immunised from the values of the Constitution as a whole or from those articulated in Chapter 3 in particular. And by virtue of section 35(3), which is phrased in the imperative, courts are obliged to bring to bear the spirit, purport and objects of Chapter 3 in the interpretation of all law and in the development of the common law and customary law. While that obligation is imposed upon all courts, I recognise that the primary burden in that regard will fall upon the Supreme Court.
 How precisely courts will go about the business of giving effect to Chapter 3 pursuant to section 35(3) is not spelled out in the text of the Constitution. I note that that section must in any event be read together with Section 33(3), which provides that rights and freedoms “recognised or conferred by common law, customary law or legislation” will continue to apply to the extent they are not inconsistent with Chapter 3. It has always been in the nature of the judicial function to fashion appropriate remedies on a case-by-case basis, having balanced and accommodating competing rights and interests. Courts will no doubt proceed in a similar manner in giving effect to section 35(3).
 In that connection, I would like to draw attention briefly to a matter that has been somewhat neglected in the application debate: the implications thereof for South African customary law in particular. Under the pre-Constitutional order, customary law was lamentably marginalised, and allowed to degenerate into a vitrified set of norms alienated from its roots in the community.3 There is hence significant scope for the dynamic application and development of customary law by the courts in a manner that has “due regard to the spirit, purport and objects” of Chapter 3.4 I note in that regard that the Constitution specifically provides that “[i]ndigenous law shall be subject to regulation by law.”5 As my brother Kriegler points out,6 the Constitution requires that judges be proactive in their application of section 35(3). Indeed, in a matter involving customary law, a court is bound to have regard to the values of Chapter 3 even where there is no claim of unconstitutionality raised.7 I am convinced that the observations of Mahomed J in his concurring opinion in this matter, with respect to the need to actively develop the common law,8 apply a fortiori to customary law.
 It is not within the scope of my brief concurring opinion in this matter to pass upon the precise manner in which a court is to go about applying “indirectly” the fundamental rights provisions in the customary law setting. It suffices for present purposes to observe that, at minimum, where customary law assigns -- as often it does -- an area of discretion to a judicial officer, it is incumbent upon her or him to be fully cognisant of the values articulated in Chapter 3 in exercising that discretion.9 Where a written decision is handed down, and fundamental constitutional values are manifestly implicated, it would be appropriate for the court to articulate in its judgment how it weighed the relevant constitutional considerations.
 This harmonisation exercise will demand a great deal of judicious care and sensitivity. Sections 33(3) and 181(1) of the Constitution acknowledge the continued existence of customary law. Moreover, the guarantee of persons’ right to participate in the cultural life of her or his choice contained in section 31 of the Constitution, even if interpreted narrowly as guarding only the individual’s freedom of cultural affiliation, would appear to require that customary law, which remains integral to the domestic culture of millions of South Africans, be accorded due respect. Although harmonisation will inevitably be an incremental process -- no one should expect customary law to be transformed overnight -- the delicate and complex nature of the task cannot justify courts in avoiding their responsibility to accommodate customary law to the “values which underlie an open and democratic society based on freedom and equality.”10
Justice of the Constitutional Court
 SACHS J: Given a choice between two well-reasoned but conflicting arguments on the question of horizontality and verticality, each with considerable support in the text, I would prefer the one which leads to the outcome I regard as being most consistent with the well-functioning constitutional democracy contemplated by the Constitution.
 Much of the discussion on the question seems, in my view, to conflate two issues that should really be kept separate. The one is the question of the scope of Chapter 3, and the other the matter of how the framers intended the Chapter to be put into operation. By running the two issues into one, an argument in favour of the broadest possible constitutional reach is unfortunately converted into a claim for the widest possible judicial remedy.
 I have no doubt that given the circumstances in which our Constitution came into being, the principles of freedom and equality which it proclaims are intended to be all-pervasive and transformatory in character. We are not dealing with a Constitution whose only or main function is to consolidate and entrench existing common law principles against future legislative invasion. Whatever function constitutions may serve in other countries, in ours it cannot properly be understood as acting simply as a limitation on governmental powers and action. Given the divisions and injustices referred to in the postscript, it would be strange indeed if the massive inequalities in our society were somehow relegated to the realm of private law, in respect of which government could only intrude if it did not interfere with the vested individual property and privacy rights of the presently privileged classes. That, to my mind, is not the issue. I accept that there is no sector where law dwells, that is not reached by the principles and values of the Constitution. If there is indeed an area of human activity exempt from legal regulation in terms of constitutional principles, it is not because the Constitution must be interpreted in a negative way so as to limit its impact, but because the Constitution itself protects such a sphere from legal intervention.
 The real issue, in my opinion, is how the Constitution intends fundamental rights in the broadest meaning of the term to be protected. More particularly, is the Constitution self-enforcing in all respects, or does it require legislative intervention to make it implementable in certain areas, especially as far as positive rights are concerned? The question, then, is not only what balance we should strike between the respective roles of our Court and that of the Appellate Division, but what spheres of decision-making belong in the first place to Parliament, and what to ourselves. This is therefore a question of separate but complementary powers as well as one of separate but complementary judicial functions. Should we be in effect legislating on matters of great social and political concern, leaving it to Parliament to fill in the gaps between our judgments, or should Parliament have the principal task of deciding on appropriate legal rights and duties, with ourselves basically standing as sentinels to ensure that Parliament does not stray beyond the framework within which the Constitution requires it to function?
 A major advantage of following the indirect approach and allowing the Appellate Division to develop the common law in keeping with the soul of the Constitution, is that the decisions of that court would not have the entrenched permanence automatically resulting from our judgments. Parliament could, following normal procedures, opt for amending or even abrogating Appellate Division decisions, provided that it legislated within the range of possibilities permitted by Chapter 3. Such alterations, however, would be severely limited in relation to determinations by our Court, where only a constitutional amendment, or at most, cautious navigation by Parliament around the prescriptive rocks of our judgments, could produce the change.
 The matter is not simply one of abstract constitutional theory. The judicial function simply does not lend itself to the kinds of factual enquiries, cost-benefit analyses, political compromises, investigations of administrative/enforcement capacities, implementation strategies and budgetary priority decisions, which appropriate decision-making on social, economic, and political questions requires. Nor does it permit the kinds of pluralistic public interventions, press scrutiny, periods for reflection and the possibility of later amendments, which are part and parcel of Parliamentary procedure. How best to achieve the realization of the values articulated by the Constitution, is something far better left in the hands of those elected by and accountable to the general public, than placed in the lap of the courts.
 The Constitution contemplates a democracy functioning within a constitutional framework, not a dikastocracy1 within which Parliament has certain residual powers. The role of the courts is not effectively to usurp the functions of the legislature, but to scrutinize the acts of the legislature. It should not establish new, positive rights and remedies on its own. The function of the courts, I believe, is, in the first place, to ensure that legislation does not violate fundamental rights, secondly, to interpret legislation in a manner that furthers the values expressed in the Constitution, and, thirdly, to ensure that common law and custom outside of the legislative sphere is developed in such a manner as to harmonise with the Constitution. In this way, the appropriate balance between the legislature and the judiciary is maintained.
 The above points can well be illustrated by four examples. They deal with defamation, private discrimination, labour law and customary law, respectively.
 The first example relates to the kind of defamation case before us at the moment. If we followed the indirect or diagonal’ approach to applicability, the Appellate Division would remain in the picture. Say, for purposes of argument, it decided to uphold the approach adopted in the carefully articulated judgment by Cameron J2 in terms of which the plaintiff would have to prove negligence on the part of the publisher. Parliament could then examine the Appellate Division’s decision, decide to refer the matter to the Law Commission for investigation, and finally opt for a completely different approach.
 Say that Parliament eventually came to the conclusion that a better approach would be that when publishing defamatory material about someone in the public domain, the media must take reasonable steps to verify the accuracy of the statements, and that the more injurious to the personal as opposed to the political reputation of the person concerned the more stringent should the investigation be; say that the legislators felt that when there is a manifest invasion of the privacy of someone in public life, it is not for the plaintiff to prove negligence or absence of justification on the part of the publishers, but for the publishers to establish that the invasion of privacy was in all the circumstances justified in the interest of the public knowing about the lives of such figures. Legislation could then be adopted to these effects, and if any publishers felt aggrieved, they could approach this Court and ask us to strike down the offending provisions. We would then weigh up the matter, decide whether the legislation conforms to the principles of free speech and respect for dignity and privacy and make an appropriate ruling, bearing in mind a number of factors, such as the powers of reading down, severance and total invalidation subject to the discretionary power granted to us in section 98(5). Furthermore, in determining the justifiability of the legislation in terms of section 33, we would decide whether the path followed by Parliament was one of many reasonably permissible options, not whether we thought it the best one.
 Assume, on the other hand, that the matter was regarded as one of direct, self-enforcing horizontal application, with the result that the Appellate Division was excluded, and our Court came to the very same conclusion as that posited above for the Appellate Division. Parliament would no longer be able to pass the legislation it thought appropriate, unless it was willing to amend the Constitution for this purpose, or, unless, possibly, it could come up with an alternative proposal that met constitutional criteria and did not conflict with the ratio of the Constitutional Court’s judgment. Whatever position we adopted when confronted with the issue, our dilemma would be profound. If we made no reformulation whatsoever and simply left the matter open, the Appellate Division would be out of the picture, and each Division of the Supreme Court could develop its own rulings, with the result that a plaintiff could win in one part of the country and lose in another, the publication being exactly the same in both. If, on the other hand, we reformulated the common law ourselves in the manner we thought most consonant with the Constitution, we would solve the problem of divided decisions, but tie the hands of Parliament until death or a constitutional amendment did us part. There would be little or no scope for Law Commission enquiry, little chance for subsequent amendments in the light of experience and public opinion. Parliament would have to defer to our discretion in the matter, seeking to find some margin of appreciation left in our judgment within which it could dot i’s, cross t’s and seek alternative, not incompatible, solutions.
 Similar problems would arise if we were to attempt ourselves to solve difficult questions which might have to be confronted when dealing with de facto discrimination. Although considerable progress has been made in this field, our country still abounds with inequality and bigotry. It is not just a question of bad and insulting behaviour. People are denied access to jobs, facilities and accommodation on a daily basis purely because of the colour of their skin. It would be a strange Constitution indeed that had nothing to say about such flagrant denials of dignity and equality. I have no doubt that the Constitution speaks to such issues. Yet in my opinion it would be quite inappropriate to say that each and every violation of personal rights in such a situation raised a constitutional question for ultimate determination by our Court. The appropriate manner for such issues to be dealt with would be through legislation pioneered perhaps by the Human Rights Commission. Litigation is a clumsy, expensive and time-consuming way of responding to the multitudinous problems of racist behaviour. Mediation and education could produce results far more satisfactory for the injured person, and considerably more transformatory for the perpetrator. Widespread research and consultation would be needed to decide precisely where to establish the cut-off point in each situation: in many countries, persons employing only a handful of workers in a close and intimate work environment, or a landlady letting one room in her house, or social activities of a genuinely private character, are expressly excluded from anti-discrimination legislation. The problems of sex discrimination might be considerably different from those related to race discrimination, or discrimination on grounds of disability. It is Parliament, and not the courts, that investigates these matters and decides on appropriate interventions and remedies.
 I am not aware of what remedies in the private sphere could be invoked to enforce what are said to be directly enforceable constitutional rights. A purely defensive remedy to someone denied access to a restaurant or promotion at work, would not be very meaningful. Specific performance would not be appropriate where the complaint is refusal to enter into a contract, rather than failure to fulfill a contact. I have found nothing in the Constitution to suggest that the framers envisaged a new form of damages for violation in the private sphere of constitutional rights. In the United States, special civil rights legislation was passed to enable persons to be sued or prosecuted for violation of or conspiracy to violate the civil rights of another. The European Court of Human Rights has express power to order damages in the case of violation of individual rights, but then only against governments, not against private parties. What clearly seems to have been contemplated by Chapter 3 is that persons whose rights have been violated not by the government but by private actors, must find their remedies either through legislation [section 33(4)] or else by means of constitutionally-adapted common law. Thus, even in the absence of anti-discrimination legislation, a person turned away from a hotel because of his or her race might be able to pursue a claim for injuria; report the offender to the licensing authorities; or lay a complaint with the Human Rights Commission. Without such legislation, however, I have difficulty seeing this or any other court finding in the Constitution authority to entertain or develop an action for damages for violation of constitutional rights where the State itself has not been the offending party.
 The constitutionalizing of private relationships in the industrial sphere could also have unacceptable consequences. Much of labour law has a procedural and framework character, leaving it to workers and employers to establish their own agreements in the light of their respective needs and interests. Collective bargaining plays a central role in establishing appropriate balancing of interests. Granting fundamental rights of a constitutional character to individual employees could destroy decades of arrangements, formal and informal, between representatives of employers and employees. Agreements involving closed shop and stop-order facilities for union dues from salary might be regarded by some as controversial and contestable. I do not wish in any way to prejudge the interpretation of constitutional or other provisions relating to labour law. Yet it does seem to me at first sight that the remedy for such persons should be to launch any challenges they may have, either in the legislature or in the many bodies, statutory and otherwise, concerned with industrial relations, not in the Constitutional Court.
 Finally, sooner or later, the question of the relationship between the Constitution and customary or indigenous law will have to be confronted. I have difficulty in seeing how this Court could effectively examine the constitutional propriety of institutions like lobola or bohadi and each and every one of their myriad inter-related rules and practices. Patriarchy permeates many aspects of customary law as it has been developed and applied in the courts over the last century. The direct enforceability of Chapter 3 could require this Court, if asked to do so, to indulge in a wholesale striking down of customary law because of violation of the equality clause in Chapter 3. The indirect approach would permit courts closer to the ground to develop customary law in an incremental, sophisticated and case-by-case way so as progressively, rapidly and coherently to bring it into line with the principles of Chapter 3. At the same time, Parliament could throw the matter open to public debate involving all interested parties, secure investigation by the Law Commission, and come up eventually with what it considers appropriate legislation.
 The issue, then, is not about our commitment to the values expressed by the Constitution, but about which institutions the Constitution envisages as being primarily responsible for giving effect to those values. From the above reasoning, it should be clear that I support the judgment of Kentridge AJ on the question of horizontality/verticality. Since I agree with his approach on the other matters raised, I wish to express my overall concurrence with his judgment and with the order he proposes.
Justice of the Constitutional Court
Case No : CCT 8/95
Counsel for the Appellants (Defendants) : GJ Marcus
Instructed by : Webber Wentzel Bowens
Counsel for the Respondents (Plaintiffs) : TJ Kruger S.C.
J van der Westhuizen
Instructed by : Van der Walt and Hugo
Date of Hearing : 7 November 1995
Date of Judgment : 15 May 1996
As deprecated in many judgments, including Davies and Others v Lombard 1966 (1) SA 585 (W).
This defence was presumably based on the judgment in Zillie v Johnson and Another 1984 (2) SA 186 (W) subsequently overruled by the Appellate Division in Neethling v du Preez and Others; Neethling v The Weekly Mail and Others  ZASCA 203; 1994 (1) SA 708 (A) at 777 - 8; and no doubt foreshadowed a submission that the Neethling judgment required reconsideration in the light of the provisions of the Constitution.
De Klerk and Another v Du Plessis and Others 1994 (6) BCLR 124 (T).
Supra n3 at 127G.
Id per Mahomed J at paras 24 and 30.
Id per Mahomed J at para 46; per Kriegler J at paras 91 and 98.
1995 (2) SA 642 (CC); 1995 (4) BCLR 401 (CC).
Id at para 65.
 I agree fully with both Kentridge AJ and Mahomed J on the question of the non-retroactivity of Chapter 3" - supra n6 at para 132. Chapter 3 ... is not applied retrospectively to undermine the validity of proceedings up to 27 April 1994, or to negate rights which had already accrued at that date - supra n6 at para 144.
The distinction between paragraph (a) and (b) of the sub-section is logical. A statute passed after the commencement of the Constitution and inconsistent with it, must in terms of section 4(1) never have been of any force and effect. A statute passed before the commencement of the Constitution would not have suffered from initial invalidity - inconsistency could only arise as from the commencement of the Constitution. To back-date its invalidity to a time before the Constitution existed would therefore be to deem that which was undoubtedly valid at that time to have been invalid. One would require clearer words than those in sub-section (6) to bring about such a result. See however the reservation in para 20 infra.
Issue (a), set out in para 10.
See the cases in n15.
S v Mhlungu and Others, supra n6 at para 8, per Mahomed J.
Marais J has, however, taken the opposite view in S v Coetzee and Others, unreported decision of the Witwatersrand Local Division, 28 September 1995, case number 70/92.
Cf. Viljoen F A Perspective on Retrospectivity of Fundamental Rights under the Interim Constitution Occasional Paper No. 5, Centre for Human Rights, University of Pretoria, December 1994 at 15.
See any elementary textbook, such as Maasdorps Institutes of South African Law II The Law of Things, 7th edition at 1.
Cf. Logan v. Zimmerman Brush Co., 455 U.S. 422 (1982) in which the United States Supreme Court held that a cause of action was property, protected by the due process clause of the Fourteenth Amendment. See also Hewlett v Minister of Finance and Another 1982 (1) SA 490 (ZSC) at 494, holding that a debt is a right of property protected under the Zimbabwe Constitution.
It follows that the judgment of Cameron J in Holomisa v Argus Newspapers Ltd, unreported decision of the Witwatersrand Local Division, 14 February 1996, case number 19883/95, was wrong on this point. As, however, appears from page 10 of the typed judgment it was not the subject of argument before him.
Constitutional Law of Canada 3 ed (1992) at para 33.10 (Citations omitted).
(1984) 8 C.R.R. 136.
(1988) 33 C.R.R. 107.
Id at 122.
Id at 128.
Id at 131- 2.
Supra n26 at 108.
(1986) 20 C.R.R. 278.
It is right to add that the Court of Appeal observed that there was no simple principle which would govern the result in all cases (Id at 284). The Supreme Court of Canada approved the Lucas and Neely decision in R v. Stevens (1989) 35 C.R.R. 107. See also Jack and Charlie v. The Queen (1986) 21 D.L.R. (4th) 641 where the Supreme Court of Canada held that the freedom of religion right could not be relied upon as excusing an offence committed before the Charter came into force.
Supra para 11.
See Zantsi, supra n34 at para 6.
Supra n6 at para 56.
 1994 (6) BCLR 124 (T) at 128J - 29B; Cf. the dictum of Froneman J in Qozoleni v Minister of Law and Order and Another 1994 (1) BCLR 75 (E) at 81; that the fundamental mischief remedied by the new Constitution is the old constitutional system.
Id at 131C.
1994 (4) BCLR 1 (W).
1995 (3) BCLR 374 (D).
1995 (11) BCLR 1498 (N).
The Thirteenth Amendment (outlawing slavery and involuntary servitude) has by reason of its language been held to impose direct obligations on individuals in private law relationships.
Id at 16.
See Tribe LH American Constitutional Law 2 ed (1988) Chapter 18; Gunther G Constitutional Law 12 ed (1991) at 902 -12.
See e.g. Henkin L Shelley v Kraemer: Notes for a Revised Opinion, (1962) University of Pennsylvania Law Review 473.
We have been furnished only with a typed version of four chapters of this work (perhaps in translation), which itself appears to be part of a larger work on constitutional interpretation, published in 1994.
 (1987) 33 D.L.R. (4th) 174.
Id at 196.
Id at 195.
Lavigne v. Ontario Public Service Employees Union (1991) 81 D.L.R. (4th) 545.
Supra para 36.
(1989) 48 Maryland Law Review 247-346.
See Quint, Id at 263 - 4.
For the facts and arguments in Lhth, Id Quint at 252 - 5; Barak, supra para 36 at 20 - 21. See also Van der Vyver, The private sphere in constitutional litigation (1994) 57 THRHR 378 at 379 - 80.
See Quint supra n61 at 318 ff; on Mephisto at 290 ff. , 302 - 3; on Deutschland-Magazin at 318 ff.
Id at 327.
Although the term state action does not appear in the Fourteenth Amendment to the United States Constitution it is principally around this Amendment that the doctrine seems to have developed.
Supra n61 at 273-4.
Including Cockrell A, Horizontal Application of the Interim Bill of Rights, unpublished seminar paper, U.C.T., 1995; Van der Vyver JD, supra n63; Strydom HA, The private domain and the bill of rights (1995) 10 SAPR/PL 52; Van Aswegen A, The Implication of a Bill of Rights for the Law of Contract and Delict (1995) 11 SAJHR 50; De Waal J, A Comparative Analysis of Provisions of German Origin in the Bill of Rights (1995) 11 SAJHR 1; Marcus G, Freedom of Expression Under the Constitution (1994) 10 SAJHR 140 at 143; Cachalia et al, Fundamental Rights in the New Constitution (1994) at 19-21; Woolman S, Application, in Constitutional Law of South Africa, ed. Chaskalson et al (1996) at 10-1.
Hiemstra and Gonin, Trilingual Legal Dictionary s. v. reg.
On the status of the Afrikaans version of the Constitution see De Waal, supra n68 in n4, at 4.
Such as the audi alteram partem rule.
Traditionally Bills of Rights have been inserted in constitutions to strike a balance between governmental power and individual liberty; to constitute a precaution against State tyranny. That was the reason for its insertion in the United States constitution. - per Van Dijkhorst J in the court below, at 130E.
Section 33(4) presumably envisages legislative measures which would apply the principles of section 8 to relationships between private persons.
Supra para 43 above.
Cf. the Canadian position, set out in paras 37 and 38 above.
This must be read subject to the express provisions of section 98(5) and (6).
As in Brink v Kitshoff NO, CCT 15/95 (argued on 9th November 1995), in which the validity of a section of the Insurance Act 27 of 1943, was contested in proceedings between the executor of an estate and the surviving spouse.
Thus in Shabalala and Others v The Attorney-General of the Transvaal and Another  ZACC 12; 1996 (1) SA 725 (CC); 1995 (12) BCLR 1593 (CC) , the prosecutors common law docket privilege recognised in R v Steyn 1954 (1) SA 324 (A) and relied upon by the state was held by this Court to be unconstitutional.
See para 38, n58 above. Cf Swedish Engine Drivers Union v. Sweden (1979-80) 1 E.H.R.R. 616.
See section 229 of the Constitution.
See Quint, supra n61 at 270-1; Gunther,
supra n50 at 902-912.
Eg Woolman supra n68 at 10-15 to 10-18.
Not in all cases. Cf the Irish case referred to in para 35 above, and cf Shabalalas case, supra n80 where the removal of an unconstitutional accretion to the law of privilege left the previously understood law in place.
Defendants counsel raised the problem of how a private prosecution for defamation would fit into this scheme. Whether a private prosecutor is exercising a governmental power is a point which need not now be decided. It may be argued that the private prosecutor is not vindicating a private right, but is invoking the power of the state to punish crime. Sections 12 and 13 of the Criminal Procedure Act 51 of 1977 reflect the states continuing interest in a private prosecution.
Supra para 49, n80.
As in the dissenting judgment of White J in Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974). See also A Lewis Make No Law (1991). The author in general favours the Sullivan rule, but points out the difficulties and anomalies which surround the concept of public person. He also observes that whereas the confidentiality of sources had been regarded as one of the pillars of freedom of the press, the Sullivan rule requires that a plaintiff may minutely investigate the editorial process (at 201-2). See also Epstein, Was New York Times v Sullivan Wrong? (1986) 53 University of Chicago Law Review 782.
(1995) 126 D.L.R. (4th) 129.
Eg in R v Bunting 1916 TPD 578 at 582-3; S v Gibson NO and Others 1979 (4) SA 115 (D); Government of the Republic of South Africa v Sunday Times Newspapers and Another 1995 (2) BCLR 182 (T) at 188.
As in Farrar v Hay 1907 T.S. 194 at 199, per Innes CJ; Die Spoorbond and Another v South African Railways; Van Heerden and Others v South African Railways 1946 AD 999 at 1013; Argus Printing and Publishing Co. Ltd and Others v Esselens Estate  ZASCA 205; 1994 (2) SA 1 (A) at 25 and the unreported judgment of Eloff JP in Bogoshi v National Media Ltd and Others, Witwatersrand Local Division, 7 February 1996, case number 29433/94. See also Burchell JM, The Law of Defamation in South Africa (1985) at 26.
Perhaps building on the remarks of Ludorf J in Pienaar and Another v Argus Printing and Publishing Co. Ltd 1956 (4) SA 310 (W).
I note in passing that in Gardener v Whitaker 1994 (5) BCLR (2) SA 19 (E) Froneman J arrived at a different reformulation of the law of defamation.
Supra at para 21, n23, and para 59.
Die nachfolgenden Grundrechte binden Gesetzgebung, vollziehende Gewalt und Rechtsprechung als unmittelbar geltendes Recht.
(1992) 8 C.R.R. (2d) 173.
Id at 185 and 189.
(1994) 108 D.L.R. (4th) 178 at 185ff.
Id at 186a-b.
Id at 156.
Including, of course, customary law. The development of customary law in accordance with section 35(3) must be one of the major tasks facing the judiciary.
See para 40 above.
(1968) at 250.
Supra n50 at 27-32.
(1974) 24 University of Toronto Law Journal 170.
2Id at para 48 (paragraph 1 of the order).
3Id at paras 39 and 41.
4See the judgment of Madala J at paras 161 and 165.
5Judgment of Kriegler J at para 135.
6Judgment of Kentridge AJ at para 55.
7Judgment of Kentridge AJ at para 56.
8S v Makwanyane and Another  ZACC 3; 1995 (3) SA 391 (CC); 1995 (6) BCLR 665 (CC) at paras 17 and 19; Westinghouse Brake & Equipment (Pty) Ltd v Bilger Engineering (Pty) Ltd 1986 (2) SA 555 (A) at 562D-563A; Black-Clawson International Ltd. v. Papierwerke Waldhof-Aschaffenburg A.G.  UKHL 2;  A.C. 591 H.L.(E).
10See, for example, Makwanyane, supra n7 at paras 155-6, 220-4, 262, 311-3, 322-3; Mhlungu, supra n1 at para 8; Shabalala and Others v Attorney-General, Transvaal, and Another  ZACC 12; 1996 (1) SA 725 (CC); 1995 (12) BCLR 1593 (CC) at paras 25-6 and see the judgment of Cameron J in Holomisa v Argus Newspapers Ltd (WLD) Case No 95/19883 14 February 1996, unreported.
2"The dignity of man is inviolable. To respect and protect it shall be the duty of all public authority. (For English renderings of provisions of the GBL the official translation of June 1994 issued by the Press and Information Office of the Federal Government has been used).
3It is generally recognised that Art 1 constitutes an unmistakeable rejection of totalitarianism and the ideology of national socialism (you are nothing , your nation is everything). See, for example, von Mhnch/Kunig Grundgesetz-Kommentar Band 1(1992) 4 Aufl Art 1 Rn 6.
4Die nachfolgenden Grundrechte binden Gesetzgebung, vollziehende Gewalt und Rechtsprechung als unmittelbar geltendes Recht. The phrase unmittelbar geltendes Recht (directly enforceable law) was intended to avoid repetition of experiences in the Weimar period where the numerous fundamental rights in the constitution were viewed as nothing more than non-binding guidelines for the state and were thus ignored with impunity. See Stern Staatsrecht III/1 (1988) 1427.
5See, for example, von Mhnch/Kunig supra n3, Vorb Art 1-19 Rn 31; Maunz/Dhrig Grundgesetz Kommentaar (1994) 6 Aufl Art 1(3) Rn 121; Jarass/Pieroth Grundgesetz fur die Bundesrepublik Deutschland (1995) 3 Aufl Art 1 Rn 22 and 24; Hesse Grundzhge des Verfassungsrechts der Bundesrepublik Deutschland (1993) 19 Aufl Rn 355; Hesse in Benda, Maihofer, Vogel Handbuch des Verfassungsrechts (1994) 2 Aufl 138 Rn 22, 152-3 Rn 59; Stern supra n4, 1531-33, 1547, 1553-4, 1561-2, 1578, 1582, 1583, 1586 and the authorities cited at 1531 n118; BVerfGE 7, 198[203-7]; BVerfGE 7, 230[233ff]; BVerfGE42, 143. It is generally accepted that there is direct Drittwirkung in the case of Art 9(3), which specifically provides that agreements aimed at preventing the formation of associations in order to safeguard and improve working and economic conditions are void; BVerfGE 73, 261.
6BVerfGE 7, 198[203-7]; BVerfGE 35, 79[112-114]; von Mhnch/Kunig supra n3, Vorb Art 1-19 Rn 22, 31.
7BVerfGE 39, 1.
8Supra n3, Vorb Art 1-19 Rn 31.
10Art 3(1): All people are equal before the law.
11Maunz/Dhrig supra n5, Art 1(3) Rn 119; Pieroth/Schlink Grundrechte, Staatsrecht II (1994) 10 Aufl 50-51 Rn 192-193; BVerfGE 54, 117.
12Hesse Grundzhge supra n5, 159 Rn 354; Stern supra n4, 1546, 1555-6.
13Hesse in Benda, Maihofer, Vogel Handbuch des Verfassungsrechts supra n5, 153 Rn 60; Stern supra n4, 1513, 1553; Rhfner in Isensee/Kirchhof Handbuch des Staatsrechts V: Allgemeine Grundrechtslehren (1994) 554 Rn 65.
14Pieroth/Schlink supra n11, 50 Rn 191ff.
16See, for example, sections 8(3), 116, 119, and 121 to 123 of the Constitution.
17Pieroth/Schlink supra n11, 49 Rn 186; Jarass/Pieroth supra n5, Art 1 Rn 24 and Vorb Art 1 Rn 5-11.
18Arts 9(3), 20(4), and 38(1) read with 48(2). Although Articles 20(4) and 38(1) are not formally part of the basic rights chapter they are regarded as quasi-fundamental rights (grundrechtsgleiche Rechte) in terms of Art 93(1)(4a).
19Pieroth/Schlink supra n11, 49 Rn 188.
21See Pieroth/Schlink supra n11, Rn 142-152.
22Hesse Grundzhge supra n5, 159 Rn 354; Hesse supra n13, 153 Rn 60; Stern supra n4, 1545, 1554. Direct application of the equality provision would forbid an employer from taking a prospective employees political views into account in deciding whether to employ her; a testator would not be allowed to decide to leave all his property to his sons rather than leaving equal shares to his daughters too. See Hesse Grundzhge supra n5, 159 Rn 356.
23Jarass/Pieroth supra n5, Art 1 Rn 24 and Vorb Art 1 Rn 5-11; Pieroth/Schlink supra n11, 49 Rn 188.
24Jarass/Pieroth supra n5, Art 1 Rn 23.
25Pieroth/Schlink supra n11, 6-13 Rn 18-39; 49 Rn 188; Stern supra n4, 1515-1518; BVerfGE 7, 198[204-5].
26Compare Wassermann (ed) Alternativ Kommentar zum Grundgesetz (1989) 245 Rn 33; Stern supra n4, 1498.
27BVerfGE 7, 198 (the Lhth case). The word radiating seems preferable to the somewhat pejorative term seepage.
28Id at 206.
29Translation by Kommers DP in The Constitutional Jurisprudence of the Federal Republic of Germany (1989) 370-371.
30Pieroth/Schlink supra n11, 51 Rn 193; Rhfner in Isensee/Kirchhof supra n13, 557 Rn 73.
31BVerfGE 73, 261; Maunz/Dhrig supra n5, Art 1(3) Rn 132. See, also sections 157, 242 and 826 of the Civil Code.
32Supra n4, 1557-1558, 1584.
33BVerfGE 7, 198[206ff]; BVerfGE 34, 269; BVerfGE 54, 117. A striking and more recent (1993) example of this approach is provided by BVerfGE 89, 214, instructively commented on by Dr HA Strydom Freedom of Contract and Constitutional Rights: A Noteworthy Decision by the German Constitutional Court 1995 (58) THRHR 696. In this case the German Constitutional Court used the general provisions of section 138 of the Civil Code as well as section 242 (which obliges the debtor to perform in good faith) as a medium through which indirectly to apply Article 2(1) of the Basic Law (guaranteeing a persons private autonomy) to a contract of suretyship. The Court struck down the suretyship, in which the surety had undertaken an exceptionally high risk without obtaining any benefit in the credit supplied, because the bank had failed to inform the surety about the nature and scope of her obligations, thus violating the principle of contractual equality.
36The Bundesarbeitsgericht, whose first president was originally the chief champion of the idea of direct horizontal application.
37Examples which readily come to mind are: the essential content of the right provision in section 33(1)(b); the concept of a separate constitutional court with circumscribed constitutional jurisdiction; the Constitutional Courts jurisdiction in section 98(2)(c),(d)(read with subsection(9)) and (e); the section 102(1) and (2) referral procedures. See further, De Waal J A Comparative Analysis of the Provisions of German Origin in the Bill of Rights (1995) 11 SAJHR 1.
38For example the formulation of the rights in section 27 (labour relations) and section 30 (children).
40Both were in their day Harlan Fiske Stone Professor of Constitutional Law at Columbia University Law School.
41Professor Herbert Wechsler Toward Neutral Principles of Constitutional Law 73 Harvard Law Review 1 (1959) and Professor Louis Henkin Shelley v. Kraemer: Notes for a Revised Opinion 110 University of Pennsylvania Law Review 473 (1962). The latter article was referred to in a different but related context in my judgment in Ferreira supra n20 at para 53 n60.
42Supra n41 at 29-30, footnotes omitted.
43Supra n41 at 477. It is interesting to note that Hesse, supra n22, is concerned about virtually identical applications and for the same reason.
1In paras 10 and 11 of the judgment.
2Which lists, from section 8 to section 32, the fundamental rights and freedoms of every person.
3The first is generally referred to as one of retrospectivity or retroactivity, and the second as one of verticality versus horizontality.
5Set out in para 5 of the judgment by Kentridge AJ. See para 12.16 of the envisaged plea.
6Set out in para 14 of the judgment by Kentridge AJ.
7Section 7 of the Constitution states:
(1) This Chapter shall bind all legislative and executive organs of state at all levels of government.
(2) This Chapter shall apply to all law in force and all administrative decisions taken and acts performed during the period of operation of this Constitution.
(3) Juristic persons shall be entitled to the rights contained in this Chapter where, and to the extent that, the nature of the rights permits.
(4) (a) When an infringement of or threat to any right entrenched in this Chapter is alleged, any person referred to in paragraph (b) shall be entitled to apply to a competent court of law for appropriate relief, which may include a declaration of rights.
(b) The relief referred to in paragraph (a) may be sought by
(i) a person acting in his or her own interest;
(ii) an association acting in the interest of its members;
(iii) a person acting on behalf of another person who is not in a position to seek such relief in his or her own name;
(iv) a person acting as a member of or in the interest of a group or class of persons; or
(v) a person acting in the public interest.
8Section 98(6) of the Constitution states:
(6) Unless the Constitutional Court in the interests of justice and good government orders otherwise, and save to the extent that it so orders, the declaration of invalidity of a law or a provision thereof
(a) existing at the commencement of this Constitution, shall not invalidate anything done or permitted in terms thereof before the coming into effect of such declaration of invalidity; or
(b) passed after such commencement, shall invalidate everything done or permitted in terms thereof.
9In para 14 of his judgment.
10In paragraph 20 of his judgment, Kentridge AJ opines that the consequences of his finding with regard to retrospective application of the Constitution as to action that occurred before the commencement of the Constitution is not necessarily invariable. This I believe to be correct and furthermore believe that section 98(6) of the Constitution provides this Court with such authority to make such a decision in the circumstances of gross injustice and abhorrence as mentioned by Kentridge AJ.
11"[W]hether the Constitution has horizontal application.
12The directions of the Court dated 9 June 1995 directed the parties to address, inter alia, the following issue:
Are the provisions of chapter 3 of the Constitution - and more particularly section 15 - capable of application to any relationship other than that between persons and legislative or executive organs of state at all levels of government?
13At paras 33-41 of his judgment.
15To use the German term of art referred to by Ackermann J at para 103 of his judgment.
16Section 98(2) of the Constitution commences as follows:
(2) The Constitutional Court shall have jurisdiction in the Republic as the court of final instance over all matters relating to the interpretation, protection and enforcement of the provisions of this Constitution, including . . . .
17The judicial oath of office, contained in Schedule 3 to the Constitution, reads, in part, as follows:
. . . [to] uphold and protect the Constitution of the Republic and the fundamental rights entrenched therein and in so doing administer justice to all persons alike without fear, favour or prejudice, in accordance with the Constitution and the Law of the Republic.
18The first paragraph of the Preamble declares:
Whereas there is a need to create a new order in which all South Africans will be entitled to a common South African citizenship in a sovereign and democratic constitutional state in which there is equality between men and women and people of all races so that all citizens shall be able to enjoy and exercise their fundamental rights and freedoms.
19The opening paragraph of the Postscript reads:
This Constitution provides a historic bridge between the past of a deeply divided society characterised by strife, conflict, untold suffering and injustice, and a future founded on the recognition of human rights, democracy and peaceful co-existence and development opportunities for all South Africans, irrespective of colour, race, class, belief or sex.
20See section 35(1) of the Constitution, which provides as follows:
(1) In interpreting the provisions of this Chapter a court of law shall promote the values which underlie an open and democratic society based on freedom and equality and shall, where applicable, have regard to public international law applicable to the protection of the rights entrenched in this Chapter, and may have regard to comparable foreign case law.
22The United States, Canada, Brazil, Germany, Ireland, India, Malaysia, Nigeria, the former Bophuthatswana and Namibia.
23Supra n21 at 48G-I; 131C-E.
24Quoted in part in n18 and n19 supra.
25Supra n21 at 46E; 128J and the passage quoted in para 11 supra.
29The expansive intention is underscored by the definition of organ of state in section 233 of the Constitution as including any statutory body or functionary.
30Section 33(2) reads as follows:
(2) Save as provided for in subsection (1) or any other provision of this Constitution, no law, whether a rule of the common law, customary law or legislation, shall limit any right entrenched in this Chapter.
32Section 35(2) reads as follows:
(2) No law which limits any of the rights entrenched in this Chapter, shall be constitutionally invalid solely by reason of the fact that the wording used prima facie exceeds the limits imposed in this Chapter, provided such law is reasonably capable of a more restricted interpretation which does not exceed such limits, in which event such law shall be construed as having a meaning in accordance with the said more restricted interpretation.
33The epithets used by Ackermann J in para 108 of his judgment.
34I believe it is also safe to say that in such a case a court will have regard to the tenor of section 33(1). If the infringement of a specific right can be saved, the same must apply to an infringement of its spirit. But the point was not argued before us and is not at issue here. No more should therefore be said about it.
35At para 46 of his judgment.
36At para 63 of his judgment.
37See also para 60 of the judgment of Kentridge J.
38At para 63 of his judgment.
39See specifically sections 98(2)(g) and (a) and generally section 98(2).
43In para 109 of his judgment.
44The First Amendment, eg commences: Congress shall make no law . . ..
45See section 98(2), which reads as follows:
(2) The Constitutional Court shall have jurisdiction in the Republic as the court of final instance over all matters relating to the interpretation, protection and enforcement of the provisions of this Constitution, including
(a) any alleged violation or threatened violation of any fundamental right entrenched in Chapter 3;
(b) any dispute over the constitutionality of any executive or administrative act or conduct or threatened executive or administrative act or conduct of any organ of state;
(c) any inquiry into the constitutionality of any law, including an Act of Parliament, irrespective of whether such law was passed or made before or after the commencement of this Constitution;
(d) any dispute over the constitutionality of any Bill before Parliament or a provincial legislature, subject to subsection (9);
(e) any dispute of a constitutional nature between organs of state at any level of government;
(f) the determination of questions whether any matter falls within its jurisdiction; and
(g) the determination of any other matters as may be entrusted to it by this Constitution or any other law.
46See sections 101(2) and (3), read with section 241(1).
9See Sections 26, 33(1) and 35(1).
11See Gardener v Whitaker supra n3 at 684.
15Paras 37-38 of Kentridge AJs judgment.
16Hogg P Constitutional Law of Canada 3 ed (1992) at 848.
17Supra n3 at 31E-G.
2Section 7(2) (emphasis added).
3See Dlamini CRM The Role of Customary Law in Meeting Social Needs African Customary Law (1991) 71 at 74; Currie I in Chaskalson et al (ed) Constitutional Law of South Africa (1996) at para 36.1.
4See S v Makwanyane and Another  ZACC 3; 1995 (3) SA 391 (CC); 1995 (6) BCLR 665 (CC) at para 383 (there are many aspects and values of traditional African law which will . . . have to be discarded or developed in order to ensure compatibility with the principles of the new constitutional order.) (per Sachs J).
6At para 142 of his judgment.
7The application of customary law has long been subject to public policy considerations, under the so-called repugnancy proviso. See Bennett TW Human Rights and African Customary Law (1995) 58-60. Under section 1(1) of the Law of Evidence Amendment Act, 45 of 1988, courts may not apply provisions of indigenous law that are contrary to principles of public policy and natural justice. Needless to say, the legitimacy of such a constraint was dubious during the time when the legal system operated within the framework of a regime of racial domination, but a different approach may be warranted if the values against which customary law is to be tested are derived from a politically legitimate constitution. But see Bennett at 60 (it seems quite wrong to use the repugnancy proviso -- a mark of colonial paternalism -- as the medium for scrutinising the constitutional validity of customary law.); Currie I, supra n3 at para 36.4(c) ([m]ost post-colonial states in southern Africa have removed repugnancy clauses from the statute book, regarding them as a humiliating index of the subordination of indigenous law in the dual legal systems established under colonialism.)
8"The common law is not to be trapped within the limitations of its past. It needs not to be interpreted in conditions of social and constitutional ossification. It needs to be revisited and revitalised with the spirit of the constitutional values defined in Chapter 3 of the Constitution and with full regard to the purport and objects of that Chapter. Per Mahomed at para 86.
9See Bennett supra at 39 (recommending adaptation of the German doctrine of mittelbare Drittwirkung to deal with customary law: [i]n Germany fundamental rights become applicable where rules of private law are general and abstract in their formulation. Customary law, too, is pervaded by generalized norms, usually characterised by a requirement for reasonable behaviour, which provide a starting point for the introduction of fundamental rights.) (footnote omitted).
1I have coined this word to indicate a country ruled by judges - from the Greek word dikastos = judge. The term juristocracy has also been used.
2Discussed in the judgment by Kentridge J at n23, where a full reference is given