Earlier this week former Deputy Chief Justice (DCJ) Dikgang Moseneke recommended that South Africa’s local government election be postponed to February 2022. The Constitution requires these elections to be held before the end of October, which means the only fail-safe legal way to postpone these elections is through a constitutional amendment, something that would be very difficult to do. Unfortunately, the alternative legal mechanism proposed by justice Moseneke to legalise the postponement is both constitutionally undesirable and uncertain to succeed.

At first glance, the decision whether to postpone the local government elections scheduled for October 2021 to February 2022, is an impossible one to make. This is partly because it is impossible to predict with any certainty what the Covid-19 situation will be in late October or in February next year. Moreover, whatever the situation will be, Covid will still be with us, and some Covid restrictions are likely to be in place whether the election is held in October or in February next year. This means, whether the election is held now or early next year, it is unlikely to be perfect.

But even if one assumes that a critical mass of South Africans will be vaccinated by February next year, and that the refusal of anti-vaxxers or vaccine hesitant South Africans to get vaccinated would not have created a more virulent strain of the virus resistant to existing vaccines, and that it will be both safer and fairer to have the election early next year, it will be very difficult to postpone the election in a constitutionally valid and appropriate manner.

As I explained last month, section 159(2) of the Constitution requires that the local government elections be held before the end of October this year. Unless a legal mechanism can be found to change this, local municipal and metro councils will cease to exist on 1 November, or at least, will lose their legal authority to function. Any decisions such councils or mayoral committees may make after this date will have no force or effect, and councillors will not be able to be paid. In short, the democratic part of local government will cease to exist.

As DCJ Moseneke points out in his report, one way to address this obvious problem is to seek to amend the Constitution and the applicable legislation. Although this is not discussed in the report, one obvious problem with this “solution” is that it would take about 3 months (assuming that legislative drafters and MPs act with a speed and efficiency they are not known fore) to amend the Constitution, which means time is close to running out to pursue this option.

 More importantly, as DCJ Moseneke notes, regularity of elections, like our democratic form of governance, is a founding value of the Constitution, which means an amendment would require a super majority of 75 per cent of members of the National Assembly. As some political parties oppose the postponement of the election it is unlikely that this threshold will be met.

This is why DCJ Moseneke suggests an alternative way to legalise the postponement of the local government election, namely to approach a court for an order to this effect. But as Moseneke makes clear, there is no guarantee that a court will grant such an order. It is important to quote extensively from the report to understand why this is so. In this regard, Moseneke writes as follows:

May a court of competent jurisdiction grant or permit the extension or relaxation of a fixed term deliberately set by the Constitution? Happily, our current assignment does not require us to answer that difficult question which we respectfully leave for the courts to decide. It may be argued that a court of competent jurisdiction may want to assume jurisdiction to extend the limited term of office of a municipal council to a finite date if it is shown that exceptional and compelling circumstances warrant the extension. Such circumstances could include elections that are likely to be a nullity because they were not free and fair, or dire circumstances like a pandemic that massively threaten life or limb, or other considerations of necessity that render compliance with the constitutional dictate impossible or exceptionally hazardous.

This proposal is quite extraordinary because a South African court has never before been asked to make such an order. In effect, the court would be asked to order a suspension of a binding provision of the Constitution (section 159(2)) on the basis that this is necessary to ensure free and fair elections guaranteed elsewhere in the same Constitution.

The only suggestion I could find in the jurisprudence of the Constitutional Court that this may be possible, comes from the 1995 judgment of Executive Council of the Western Cape Legislature and Others v President of the Republic of South Africa and Others where the majority had to consider the binding nature of the prescribed constitutional procedures for the adoption of legislation. In an obiter dictum passage (not legally binding as a precedent) the majority suggested that:

There may be exceptional circumstances such as war and emergencies in which there will be a necessary implication that laws can be made without following the forms and procedures prescribed by [the Constitution]… It is possible that circumstances short of war or states of emergency will exist from which a necessary implication can arise that Parliament may authorise urgent action to be taken out of necessity. 

The minority in Executive Council of the Western Cape rejected this suggestion and the Constitutional Court has not raised this possibility since it first did so 25 years ago. On the contrary, in the subsequent 25 years the Court has treated the supremacy of the Constitution and respect for the Rule of Law as sacrosanct, suggesting the passage above was an early, never to be repeated hiccup of a court still finding its feet. In any event, it would be far more extraordinary for a court to order the suspension of a directive provision of the Constitution that gives effect to one of the founding values in the Constitution, than it would have been to allow for a procedural shortcut as suggested by the majority in Executive Council of the Western Cape.

There is another reason why DCJ Moseneke’s proposal to approach the Constitutional Court (I assume) for such an order may be problematic. This is that it might well place the court in an impossible position, where it is forced to choose between two completely unpalatable and constitutionally problematic options.

If the court is required to consider the request in a situation in which the Electoral Commission has failed to meet its statutory obligations to conduct a free and fair election (such as conducting voter registration before 6 August), it could be faced with a choice to endorse an unconstitutional postponement of the election, or face the consequences of later having to nullify an election that was not free and fair. Given the recent political onslaught faced by the Constitutional Court, any decision it may take in such circumstances are bound to be exploited by the charlatans and opportunists wishing to undermine the authority and integrity of the court.

At the same time, if there are overwhelming evidence that the basic requirements for a free and fair election cannot be met, either the legislature (through constitutional amendment) or the court (through an order) would need to intervene to legalise the postponement of the election. This means that if voters are unable to participate in such an election and that parties and candidates are unable to campaign, or if there is credible evidence that the election would “massively threatens life or limb”, a way must be found to postpone the election. The problem is that DCJ Moseneke’s report does not appear to provide clear and overwhelming evidence of this kind.

Many of the experts who were consulted by Moseneke predict that the third wave of Covid-19 would largely have subsided by the time the election were held in late October. If this is correct, the safety concerns rightly raised by DCJ Moseneke might not be as severe as the report suggests. Of course, if these experts are wrong and Covid-19 numbers remain extremely high, thus “massively threaten life or limb”, a way may have to be found to postpone, despite the fact that all the options to do so are unpalatable.

Furthermore, if the experts are correct, we are likely to be in level 1 or 2 lockdown at the end of October. While it will not be ideal to conduct an election during any form of lockdown, and such election might well have a lower participation rate than would otherwise have been the case, the question is whether this would make a more or less free and fair election completely impossible. (If legislation is amended to allow for voter registration after the declaration of the election to ensure everyone who qualifies to vote gets the opportunity to register, it might go a long way to ensure equal participation and therefore a measure of fairness of the election.)

In short, given the fact that all the legal options available to allow for the postponement of the election are difficult to implement or constitutionally problematic, such a postponement should only be considered in two situations. First, if the facts overwhelmingly show that any semblance of a free and fair election would be impossible, to such an extent that the election would have to be nullified if it proceeded, a postponement may be justified. Second, if there is overwhelming evidence that the election would lead to the large scale spread of Covid-19 and would thus pose a serious threat to life and limb, a postponement may also have to be engineered by hook or by crook.

But the suggestion in the report that this would be the case unless there is no lockdown or only a level 1 lockdown, seems open to question. Readers may want to study the Moseneke report themselves to consider whether, on the facts, the report makes out a case that anything but a level 1 or no lockdown would require a postponement of the election.