Do courts and judges have the institutional capacity needed to settle the kinds of morally and politically controversial issues that arise in constitutional rights cases? Paul Yowell argues that the institutional setting of a court is lacking in basic capacities for moral and empirical reasoning. In doing so he confines the argument to the judicial review of legislation in the adjudication of abstract questions of constitutional rights, because that highlights the contrast between the competing competences of courts and legislatures.

The question is important and the arguments and examples advanced in support of the conclusion powerful, but the further I read the more I wondered whether the genie of entrusting moral reasoning to judges has not long since escaped the bottle. Can we confine the institutional problems of judges and empirical reasoning to the arena of the adjudication of constitutional rights? Even in that restricted environment, can they be resolved merely by enhancing capacity and revising established constitutional structures, itself a task fraught with political and legal uncertainty? Is the judiciary capable of developing other ways to address these issues? It is to these questions that I will address myself.

Moral reasoning

The modern focus on the role of courts in constitutional adjudication obscures the fact that moral reasoning has always played a significant role in the judicial process. The fact that it is demanded of judges under modern constitutional instruments merely increases the extent to which that is so. And if judges must in any event engage in moral reasoning and, I would add, consider issues of policy in performing their ordinary function, there seems little reason to make constitutional adjudication a no-go area.

In any event it is questionable whether, under modern constitutional instruments, judges can avoid moral reasoning and the need to grapple with value-laden terms such as equality, human dignity, privacy and democracy. Constitutions and bills of rights explicitly require judges to engage with these concepts. The courts cannot disregard the constitution when it explicitly requires them to apply indeterminate moral norms. That is so whether they are empowered to strike down legislation on the grounds of constitutional incompatibility or whether they are restricted to making a declaration of incompatibility.

Even without the impetus of a Constitution, courts grapple with moral norms and moral arguments. Our conception of judicial impartiality is itself a moral construct and its requirements are sharply in focus when a judge’s recusal is sought. The judicial oath of office in South Africa requires a judge to uphold and protect the Constitution and the human rights entrenched in it, and to administer justice to all persons alike without fear, favour or prejudice, in accordance with the Constitution and the law. A similar oath is a feature of judicial office in most democratic and constitutional states. Requiring that we do justice invokes possibly the most value-laden term in the legal lexicon. To do it in accordance with the Constitution demands the promotion of the founding values of human dignity, the achievement of equality and the advancement of human rights and freedoms, non-racialism and non-sexism. These tasks are all freighted with a need to make moral judgments, whether we are dealing with a constitutional issue and the validity of legislation or the development of the common law.

Moving from the general to the particular, judges in criminal courts must assess moral culpability in determining sentence. Whether there is a legal duty of care is determined by the standard of what is ‘fair, just and reasonable’, all moral judgments. The mesothelioma exception in the law of causation was expressly founded on the sense of justice of the judges hearing the case. The decision in Bland[1] plainly involves moral reasoning under the common law. In interpreting statutes dealing with housing rights courts have held that ‘family’ is not confined to the nuclear family. These are all examples of the need under the common law to make moral judgments on potentially divisive social issues.

Similarly explicit moral and policy standards under the common law in South Africa governed whether a contract was unenforceable on public policy grounds; whether nutrition could be withdrawn from a patient in a persistent vegetative state;[2] whether a wrongful birth suit should succeed;[3] and whether a doctor who assists a patient to commit suicide is guilty of a crime.[4] Then there are the countless statutory provisions that require courts to determine whether discrimination occurred and was unfair; whether a contractual term was oppressive to the debtor, or simply unfair; whether labour practices were unfair; whether something should be permitted in the interests of justice; or what constitutes equitable redress for land deprivation as a result of past racially discriminatory legislation. In each instance the standard the courts are enjoined to apply is a moral norm and its application demands that the judge engage in moral reasoning.

From the outset the South African Constitution accorded a central role to a Bill of Rights. The jurisdiction of the Constitutional Court was originally limited to constitutional matters, but the court dealt with cases involving the law of delict (tort), vicarious liability, causation and contract, and interpreting legislation having significant private law consequences. Controversial though this was, constitutional amendments effected in 2013 gave it general jurisdiction in regard to questions of general public importance, so that it straddles the Kelsenian and American models as the country’s apex court. The Constitution enjoins all courts to interpret legislation and develop the common law in accordance with the spirit, purport and object of the Bill of Rights. The effect is to amplify the need for moral reasoning at every level of the judicial system.

The genie is out of the bottle and moral reasoning is central to the judicial function. It is not confined to constitutional adjudication although it is prominent in constitutional litigation. Curiously there is a tendency for our courts to disavow it. In both Makwanyane, where it struck down the death penalty, and the Certification Case, in which it was required to certify that the 1996 Constitution conformed to a number of agreed constitutional principles, the court was at pains to claim that it was performing a purely judicial function from which political and moral considerations were absent.

The adjudication of constitutional rights in South Africa shares the features identified by Yowell of constitutional adjudication in other jurisdictions. The limitation enquiry, in terms of s 36 of the Constitution, asks whether the limitation of the right is justifiable in an open and democratic society based on human dignity, equality and freedom, taking into account the nature of the right; the importance of the purpose of the limitation; the nature and extent of the limitation; the relation between the limitation and its purpose and whether there are less restrictive means to achieve the purpose. The limitation analysis tends to focus on the relation between the limitation and the statutory purpose and whether there are less restrictive means to achieve that purpose.[5] One suspects that this may be an analytical device adopted to depict the analysis as an objective, fact-based comparison of competing approaches to achieving the legitimate governmental aim. The right is not defined, but the Court says it is engaged or implicated and asserts an infringement with little explanation. It then proceeds to the limitation analysis. Where the Court thinks that an infringement of rights is justified it may assume that there is an infringement of rights, without a definitive finding to that effect.[6]

This may leave an uncomfortable feeling that a moral choice in favour of, or against, the particular policy underpins the judgment, but it is rare for this to be articulated. Instead one encounters an analysis of the evidence for and against the solution adopted in the legislation This raises the institutional capacity of judges and courts to engage with such evidence and by way of empirical reasoning reach reasoned and legally supportable conclusions. Before turning to that a final word on judges, legislatures and moral reasoning is appropriate.

Legislatures are capable of careful, detailed and respectful debates around highly emotive moral issues about which society at large is deeply divided. Examples are debates that legislatures in various countries have had about abortion, the abolition of capital punishment, gay marriage and physician assisted dying. Such debates tend to produce a compromise decision that is broadly accepted because all views have been heard, although some have necessarily been rejected. This is undoubtedly preferable to judicial fiat, as the ongoing struggle over Roe v Wade and potentially the political fallout over Obergfell v Hodges demonstrates.

But legislatures are equally capable of angry, prejudiced and knee-jerk reactions to social issues such as prisoners’ rights to vote in the United Kingdom;[7] the Defence of Marriage Act in the United States; and the stringent mandatory sentences for various serious crimes in South Africa. Political solutions, especially those enacted in the face of public outrage over social problems, are frequently not the product of careful, detailed and respectful debates in which all moral views are given careful consideration.

Judicial debates, in open courts, with representation of the immediate parties and public interests, through amici briefs, can be more akin to the type of debate of moral issues that legislatures should, but often don’t, provide. Even if judges are drawn from a narrow sector of society efforts to diversify the judiciary are beginning to bear fruit. Where there is an adversarial system judges have a legal training and background, but that does not debar them from a broad knowledge and understanding of society. Experience in practice compels them to learn to understand different views, to respect differences of opinion and to discern the reasons for their adoption. That serves to ameliorate, if not dispel, some of the concerns about their engaging in moral reasoning.

The picture in practice is not as clear-cut as those favouring democratic resolution of all socially difficult issues might wish. Sometimes legislatures get it right, and sometimes they get it wrong. Courts are the same. Both are fallible human institutions, each with advantages and disadvantages. It is a bigger challenge to determine which issues should be left to the legislature and which to the courts. In deciding that, the capacity of each to deal with empirical material bearing upon the problem is a significant factor.

Empirical reasoning

A court approaches evidence in a medical negligence case, construing a patent, the aetiology of a disease or the likelihood that a particular governmental policy will be successful, in the same way. The primary source of the information comes from the parties and any amici. The adversarial system imposes structural limitations on courts seeking out information for themselves. Yet, what else is it to do when the evidential record is inadequate or inaccurate, whether as a result of the deficiencies of the lawyers or the lack of resources of the litigant? There are signs that courts are reluctant to be confined in a straitjacket of the litigants’ making. Recently in considering the constitutional validity of legislation criminalising the cultivation and possession of dagga (marijuana for non-South Africans) the principal technical evidence on the potential harmful effects of consumption of the drug came from a report commissioned by the court from a distinguished academic.[8] This was not entirely novel. In a case involving the withdrawal of nutrition from a patient in a persistent vegetative state the judge demanded additional medical examinations and reports before reaching his decision.[9]

Judges are criticised for undertaking research for themselves, but what is the judge to do when the preparation by the litigants’ lawyers, whether on fact or law, is deficient. Should judges not consult reference sources, readily available at the click of a button, in order to clarify what is obscure, understand technical expressions and provide background material that enables them to do a better job of judging? And we delude ourselves if we think that judges have not always done this. The information we now find from search engines was formerly obtained from the Encyclopaedia Brittanica, atlases, dictionaries and similar reference works. Photocopied photographs and sketch plans are a poor substitute for Google Earth, so why not look at it, provided it reflects the situation at the time. In a case involving historical issues, why be bound by the affidavits of fallible witnesses relying on folk memory, when there are reliable histories available? Even Wikipedia may be reliable as a source depending on what it is used for. It seems a curious rule that prevents judges from looking at the widely used sources of background material and information consulted by the rest of humanity. Are we to go back to asking: ‘Who are The Beatles?’ The taking of judicial notice has moved beyond that.

The potential problems lie with the purpose for which the judge seeks that information, whether it is reliable and whether the parties are afforded a fair opportunity to refute or qualify it. It is unclear why a judge must accept the contents of a report or survey, merely because it is attached to an affidavit, but ignore one from a similar academic source because it is not, but the judge or the judge’s clerk found it on the Internet. Possession of pertinent and reliable information enhances judicial decision-making. Courts prefer facts to predictions,[10] and to prefer accurate and complete facts is in principle preferable to compelling judges to decide cases on an inaccurate or incomplete record.

The greater problem is that judges do not know how to deal with empirical evidence drawn from the social sciences and, I would add, much scientific evidence. The woeful lack of understanding of statistical evidence is legendary. The prosecutor’s fallacy; treating correlation as causation; extrapolating general probabilities to the specific instance; an inability to detect sampling errors; and misunderstanding percentages relating to the enhancement of risk, all feature. Frequently the parties and their advisers have only a superficial understanding of these and many scientific witnesses are unable to explain their science in language comprehensible to laypeople. Causation in criminal and civil cases and the use of market surveys in fields such as competition law, passing off and unfair competition illustrate the problems. Judges deciding as a matter of fact what the likely outcome of a particular policy choice will be, venture into a minefield in assessing evidential material inevitably presented from a partisan perspective.

Legislatures can make use of multiple sources to assess the scope and extent of a problem, weigh up alternative ways of dealing with it and then draft legislation, which is thoroughly debated in the legislature before being implemented This gives them a huge advantage over judges, because court systems cannot undertake a similar exercise. Enhancing judges’ capacity to understand such evidence, or obtain it themselves, bridges, but does not close, the gap. Courts have neither the time nor the resources to undertake similar investigations in determining the constitutional validity of legislation. They may be placed in the invidious position of adjudicating a constitutional challenge to legislation by an organisation, whose submissions to the legislature were overlooked or rejected. The court must then weigh incommensurables to determine whether the legislation is ‘narrowly tailored’ to the legitimate governmental interest. The end result may be that legislative provisions carefully devised are set aside in favour of an approach that the court thinks is a better solution, solely because it has less impact upon the right that is limited thereby.

The remedy

It is too late to redesign our constitution to incorporate the characteristics that suggested in the closing chapter. Given the process whereby it was negotiated I doubt that the parties would have agreed to such a constitution or such a constitutional court. So we have a court that engages in moral reasoning with little equipment to assess empirical evidence.

Enhancing judicial capacity to deal with empirical evidence is obviously desirable at all levels of the judiciary. But no amount of judicial training qualifies a judge to balance permitting people to smoke marijuana in private against society’s interest in inhibiting drug dealers, nor to assess whether vulnerable people will be abused, if physician assisted dying is permitted. A judge is no better qualified to make such decisions than any other thoughtful and informed member of society, so why invest them with the power to make these decisions on behalf of society? Also most countries, especially developing ones like mine, will encounter practical difficulties arising out of resource allocation in the amount it can do in this regard.

One reason for leaving these problems to the legislature is that errors are relatively easily remedied. Determinations by a constitutional court are not, because they are entrenched. They trump all other views of the correct approach to the issue and cannot be lightly overturned. In South Africa stare decisis is a central characteristic of the rule of law and is applied strictly, with departures from previous decisions being rare.

Underpinning Paul Yowell’s book is a sense that certain areas of pressing social and political policy are best left to the legislature and that courts are ill-suited in these areas to determine by an adjudicative process which policies best balance the interests at play and should be adopted. Institutional incapacity has always been a reason for courts to say that issues are beyond their remit. This goes further than saying that courts should respect or show deference to the decisions of the legislature. Nine debates in the Canadian parliament all rejecting changes to statutory provisions criminalising assisted suicide did not affect the Canadian Supreme Court’s decision to overturn the outcome of these debates. Giving substantive content to the separation of powers, by demarcating areas of public life that are beyond the remit of courts, is essential if the areas of concern in this book are to be addressed. Unless that happens judicial power will prevail.

[1] Airedale NHS Trust v Bland 1993 AC 789 (HL).

[2] Clarke v Hurst NO and Others 1992 (4) SA 630 (D).

[3] Stewart and Another v Botha an Another [1999] ZASCA 39; 1999 (3) SA 1065 (SCA) overruled in H v Fetal Assessment Centre [2014] ZACC 34; 2015 (2) SA 193 (CC) on a different view of the moral issues.

[4] There is no equivalent to the Suicide Act in South Africa and the question of criminal liability in such a case depends whether it constitutes the common law crimes of murder or culpable homicide  (manslaughter). Minister of Justice and Constitutional Development and Others v Estate late Stransham-Ford (Doctors for Life and others as amici curiae) [2016] ZASCA 197; 2017 (3) SA 152 (SCA) paras 28-57. In para 61 the court referred to jurisdictions where the legitimisation of patient assisted suicide arose from judicial decision-making rather than constitutional invalidation of legislation prohibiting it.

[5] Teddy Bear Clinic for Abused Children and Another v Minister of Justice and Constitutional Development and Another [2013] ZACC 35; 2014 (2) SA 168 9CC) paras 84-101; Gaertner and Others v Minister of Finance and Others [2013] ZACC 38; 2014 (1) SA 442 (CC) paras 67-75.

[6] Christian Education South Africa v Minister of Education [2000] ZACC 11; 2000 (4) SA 757 (CC) paras 27-28. In such cases the Court invariably holds that the limitation of rights is justified.

[7] The then Prime Minister of the United Kingdom said that it made him ‘physically sick’ to think of prisoners being permitted to vote.

[8] Minister of Justice and Constitutional Development and Others v Prince (Clarke and others intervening); National Director of Public Prosecutions and Others v Rubin; National Director of Public Prosecutions and Others v Acton [2018] ZACC 29 para 29.

[9] Clarke v Hurst NO and Others 1992 (4) SA 630 (D).

[10] Bwilfa and Merthyr Dare Steam Collieries (1891) Ltd v Pontypridd Waterworks Co [1903] AC 426 per Lord Halsbury LC and Lord Macnaghten.