Why big cases make for bad law
“The Supreme Court of the United States has descended from disciplined legal reasoning … to the mystical aphorisms of the fortune cookie.” — Scalia J in dissent, Obergefell v. Hodges 135 S Ct 2584, 2628 (2015).
Perhaps, to start, I need to make the following disclaimer: I am not part of the alleged conspiracy to tarnish the good name of the Constitutional Court, nor that of the Lord’s Chief Justice. But my friend Elisha is. Nor am I part of the undefined group of court-watchers who have long warned us of the Court’s decline post-Chaskalson (which was also incidentally when its demographics began to change).
No. Instead, I serve neither man nor mammon, but only the law. I would like to think of myself as a thoughtful person. And part of being thoughtful is questioning popularly held opinions, rightly or wrongly. I do so here.
When Justice Antonin Scalia, in his usual colourful style, dissented from the US Supreme Court’s majority opinion in Obergefell v. Hodges — the case that declared same-sex marriage legal in all 50 states — he noted that there was no principled basis for the court to do so. The judgment, written by then Justice Anthony Kennedy (a key swing vote in LGBT rights cases), is not a model of clarity. It does not sufficiently ground the right of same-sex couples to marry on any constitutional basis, and is couched in “dignity” language which is foreign to American constitutional law. Even the most liberal of critics have argued that it is not sound in doctrine, which makes it vulnerable to future attack and possible reversal. Do I think that the outcome in Obergefell was right? Yes. But was it correct? Probably not, but I do not know enough about American law to speak authoritatively.
If you are wondering what the distinction between ‘right’ and ‘correct’ is, welcome to my conundrum. A conundrum caused only by big cases for the sole reason that they are too important for the public. Ideally, when a court adjudicates a matter, it does so for all future cases arising from similar facts, i.e it creates a principle. That is why we have the principle of stare decisis — the idea that courts are bound by their own earlier decisions and those of higher courts, unless they are clearly wrong. This denotes a certain level of certainty, neutrality and generality in how the law is applied. The problem that arises in big cases is that there is a temptation to decide them looking at what the outcome might be and what its impact on the broader public would be. Instead of first departing from a particular point, the outcome predetermines the court’s journey. This leaves little room for principled adjudication — adjudication regardless of the outcome — and forces the court to employ a line of reasoning that will lead it to its desired outcome.
As Oliver Wendell Holmes said in Northern Securities Co. v. United States:
“Great cases like hard cases make bad law. For great cases are called great, not by reason of their importance… but because of some accident of immediate overwhelming interest which appeals to the feelings and distorts the judgment.”
And now, for the heresy. There’s a line of three Constitutional Court cases, a troika if you will, whose principle underpinnings are especially dubious. They all have some things in common: they were high-profile, and presented politically-charged questions the answers to which would have immense public impact. The first is Economic Freedom Fighters and Others v Speaker of the National Assembly and Others (Nkandla), the second is United Democratic Movement and Others v Speaker of the National Assembly and Others (Secret Ballot), and the third is Economic Freedom Fighters and Others v Speaker of the National Assembly (Impeachment). This article is primarily about the Nkandla judgment.
By now the events surrounding Nkandla and the subsequent political fallout are etched deep into this country’s collective memory. The eye of that particular storm was of course the Public Protector’s report in which she recommended, among other things, that the President pay back a portion of the public money irregularly spent on upgrading his private residence and that he discipline the Ministers involved in that exercise.
The President argued that her recommendations were not binding, and instead commissioned another report on the Nkandla project, this time to be conducted by the Minister of Police, who subsequently cleared the President of any wrongdoing and found that all the upgrades had been necessary. The National Assembly subsequently set up an ad-hoc committee whose primary role, let’s be real, was to absolve the President. It adopted the Police Minister’s report, purportedly setting the Public Protector’s report aside.
By the time that the EFF had brought the case before the Constitutional Court, the principal question of law the saga presented had crystallized: were the Public Protector’s recommendations and remedial action binding?
However, before the hearing of the matter, the President in his reply to Corruption Watch’s heads of argument conceded that the Public Protector’s remedial action was indeed binding. In effect, there was no longer any disagreement about the effect of the Public Protector’s powers when the case was heard. This is important. As a general rule, the court will not decide a matter that no longer presents a live dispute. This is called the doctrine of mootness. Mootness prevents the court from deciding matters in which the orders granted will be of no immediate effect, and the judgments merely academic. Since the President had accepted that the Public Protector’s remedial action was binding and there was a Supreme Court of Appeal judgment (South African Broadcasting Corporation SOC Ltd and Others v Democratic Alliance) to the same effect, there was no longer a need for the court to decide the question. It was settled law. Or so you’d think.
Mootness is however not absolute. The court still retains the discretion to decide a moot case where it is in the interests of justice to do so. The justification that the court gives for deciding the question despite it being moot is that the President’s concession was made “only in relation to this case, not necessarily as a general proposition.” That is a strange statement. The principal issue in the case related to the powers of the Public Protector to make binding orders generally, not only binding orders against the President and the President’s concession effectively resolved the dispute around that issue.
Another point worth making is that the moment the court decided to entertain the arguably moot question of the nature of the Public Protector’s powers, it had to settle a central dispute between two lower court judgments on the issue. In the high court judgment in the SABC matter (concerning the irregular appointment of one Hlaudi Motsoeneng as COO), the court had held that the Public Protector’s remedial action was not binding in nature, but that the subject of said remedial action could not ignore it without a rational basis for doing so. The Supreme Court of Appeal found otherwise, holding that the Public Protector’s findings and remedial action were binding and could not be ignored absent a review to have them set aside. Both courts engaged in detailed analyses to justify their conclusions. The high court holding that the Constitution did not textually confer the Public Protector with binding powers, choosing instead to question whether the SABC board’s rejection of the Public Protector’s findings and remedial action was rational, and finding that it was not; while the SCA chose to focus the origins of the office, its constitutional and legislative history, and determining the nature of her powers with regard to the purpose of her office.
One would have expected, as is practice, the Constitutional Court to engage with both judgments in detail and methodically, to explain why it prefers one court’s reasoning to the other. It did not do so. Instead, both judgments were briefly dealt with in a total of two paragraphs out of a 105-paragraph long judgment.
The court also suggested that the high court’s preference for testing a subject’s refusal to comply with the Public Protector’s remedial action against the constitutional standard of rationality was “at odds with the rule of law”. How the court arrives at this conclusion is unclear. Rationality is a central part of the rule of law, and requiring the exercise of all public power to be linked to a legitimate governmental purpose, to avoid arbitrary acts. How the High Court could breach the rule of law by requiring that any rejection of the Public Protector’s recommendations have a rational basis is unclear.
Further, the Constitutional Court then curiously described the holding of the SCA in the following terms:
“The judgment of the Supreme Court of Appeal is correct in recognising that the Public Protector’s remedial action might at times have a binding effect. When remedial action is binding, compliance is not optional, whatever reservations the affected party might have about its fairness, appropriateness or lawfulness. For this reason, the remedial action taken against those under investigation cannot be ignored without any legal consequences.”
Are the Public Protector’s powers always binding or only at times? If they are binding, is this the sui generis nature of all Chapter 9 institutions’ powers? If not, whence the source of the binding nature of her powers? It’s clearly not textually in the Constitution, and not the Public Protector Act, which in any event could not give the office more powers than the Constitution originally did, so where does it come from? The Court does not explain.
As if this confusion was not enough, the Court itself seems confused about what the ratio of the judgment is. The ratio decidendi of a judgment — as opposed to the obiter dicta (things said in passing) — is the reason for the decision. It encapsulates a court’s reasoning, which justifies the result.
Enter Oudekraal. A long running saga between the Constitutional Court and the Supreme Court of Appeal, is about what the meaning of the principle in Oudekraal Estates (Pty) Ltd v City of Cape Town and Others is. Shorn of all frills, the gist of the principle is this: an administrative act that is invalid in law but exists in fact has legal effect until set aside by a court of law on review. A wing of the Court led primarily by Jafta J, believes that an invalid administrative act is invalid from its genesis and can never have legal effect and that the principle is at odds with the rule of law and the doctrine of objective invalidity. The other wing, led by Cameron J, holds that the principle is sound in law and allows the courts to regulate exercise of public power and avoids legal uncertainty. This dispute came to a head in Tasima. Tasima was about a state organ trying to set aside a contract it believed to have been unlawfully concluded. This was to be done by means of a collateral challenge (hopefully the subject of a future article) to its own decision. In a judgment concurring in the minority judgment by Jafta J, Zondo J held that the central holding of Nkandla was that—
“First of all, in Economic Freedom Fighters the validity or lawfulness of the remedial action taken by the Public Protector against the President — which was the administrative action under consideration — was not challenged by the President. The President conceded that the administrative action was valid and binding. This is the context in which what this Court said in Economic Freedom Fighters should be understood. It was not called upon to decide what should happen where the person against whom an administrative action has been taken argues that the administrative action is unlawful or invalid. Statements made in Economic Freedom Fighters were made in the context of an administrative action that was accepted by all to be valid.”
As it appears from the text, it is clear that for Zondo J, the President’s concession that the Public Protector’s powers were binding was central to the Court’s ratio in Nkandla. That is, the Court could not have reached a decision other than the decision that it reached had the President not conceded that the Public Protector’s powers were binding.
The disagreement about the ratio of the Nkandla judgment so sharply divided the Court that Froneman J, in his concurrence with the majority, says the following:
“Shortly before the hearing in Economic Freedom Fighters the President made a concession. He stated on record that he accepted the legal and constitutional validity of the Public Protector’s report. That concession is neither the ratio nor the logical underpinning of the ratio of the Economic Freedom Fighters judgment by this Court. Is it seriously intended to state or imply that if the President did not make the concession then this Court’s decision would have been different? And that this Court would have concluded that the President was entitled to ignore the Public Protector’s report without approaching a court of law to have it set aside? Surely not.”
The confusion over the Oudekraal principle and its treatment by the Court in Nkandla is further compounded by two things: (1) Mogoeng CJ, the author of the judgment, relied on a passage from Kirland Investments — a judgment that affirmed the Oudekraal principle — to hold that the Public Protector’s remedial action cannot be ignored without recourse to the judicial process; and (2) Mogoeng CJ concurred in both Jafta J and Zondo J’s minority judgments in Tasima, the latter’s central holding being that the President’s concession was the key that unlocked Nkandla. In addition, both those judgments reject the premise of the Oudekraal principle, and by extension, the Court’s own judgments in Kirland Investments and Merafong.
It is difficult to discern what the holding in Nkandla was, what principle underpins it and how it will apply in future cases. All these are the hallmarks of a decision that is not properly reasoned, or one whose reasoning seeks to justify a predetermined outcome. That the decision was the right one, given the excessive waste of state funds in recent years, is obvious. Whether it was the correct decision is for history, and the court-watchers, to determine.
