Log into your member account to listen to this article. Not a member? Join the herd.

To a comparative constitutional lawyer, Article 140 of the Kenyan Constitution is an interesting provision: it sets out, in some detail, the legal route by which a dispute around a presidential election is to be resolved. Read with Article 163(3)(a), it clothes the Supreme Court of Kenya with the exclusive prerogative – as well as the duty – to hear and decide a challenge to a presidential election, within fourteen days. Article 140’s mandatory and time-bound language precludes the Supreme Court from taking two paths, which judiciaries often take, to avoid entanglement in politics: declining jurisdiction to hear a dispute, or simply putting off a decision until the case becomes infructuous. Rather, Article 140 envisages that the Supreme Court will be the first – and final – arbiter of the most contentious of all political disputes.

This puts the Supreme Court in something of a bind. A large part of judicial legitimacy flows from a Court’s ability to stay out of political disputes, or to carefully negotiate political terrain when such questions are thrown up before it. The histories of independent judiciaries around the world have shown us that if a Court deals politicians too many setbacks, a backlash will not be far behind.

This bind is worsened by two things. The first is that complicated electronic technology has become integral to modern-day elections, and disputes around elections will therefore require the Court to assess competing claims around technology, presented by duelling sets of experts. This is a fraught exercise at the best of times, and becomes particularly fraught when a presidential election turns on the outcome. The second – and related – point is that many of the issues that arise in a presidential dispute will necessarily involve high degrees of judicial subjectivity. It is a truism that there is no such thing as a “perfect election”. In any election held at scale, there will be machine errors and human errors – somewhere, somebody will make a mistake, a computer will break down, a rule will be misunderstood or wrongly applied. There is no bright line for determining the point at which these atomised errors coalesce into something that undermines the integrity of an election. It is a matter of judgment, and like all matters of judgment, subject to attack.

To negotiate this bind, a court that is given the kind of task that the Supreme Court of Kenya has been given under Article 140, can do the following things: (a) articulate a set of objective and neutral standards concerning questions of evidence, and the threshold required to invalidate the results of an election; (b) hew closely to the submissions and evidence provided by the parties to the dispute; and (c) set out detailed and transparent reasoning for its decision, so that the losing party has the right to feel aggrieved, but does not feel cheated.

The unanimous judgment of the seven judges of the Supreme Court of Kenya in Odinga and 16 Others vs Ruto and 10 others – the challenge to the 2022 Kenyan Presidential elections, and the certification of William Ruto as the president-elect – reveals both the bind, and the Court’s attempt to negotiate it through the principles set out above. Faced with a series of allegations about the conduct of the 2022 presidential elections – ranging from hacking to physical manipulation of forms, and from voter suppression to technological breakdown – the Court framed its response along two lines: a standard of evidence and a standard of invalidity. With respect to the first, the Court held that allegations of impropriety would have to meet an “intermediate standard” of “clear and cogent evidence” – that is, something between the civil law standard of “balance of probabilities” and the criminal law standard of “beyond reasonable doubt” (the exception to this was when allegations of a criminal nature – such as fraud – were made in the course of the election petition).

With respect to the second, the Court held that where the standard had been met, the next question was: did the improprieties reach a level where they materially impacted the outcome of the election? To an extent, this is a counterfactual question that is difficult to answer with certainty, especially in close elections; what would have happened if the improprieties had not taken place? But it is also an essential question; if an election were to be set aside on the basis of any impropriety, then we would be having election re-runs until the end of time. The standard of invalidity is, to an extent, a compromise, but a necessary one.

With this framework in mind, the Supreme Court’s analysis can be divided into two buckets. In the first bucket were allegations (such as fraud, switching of Forms 34A, and so on) that the Court found were not proven to the required standard. Importantly, in making this assessment, the Court primarily relied upon the competing affidavits of the parties (including upon internal contradictions within some of the affidavits). This is the second principle outlined above: as the Court stressed, in adjudicating the case, it could not travel beyond the quality of evidence provided to it by the respective parties. In the second bucket were allegations (such as printing errors and failure of voting kits) where the Court found that there had been lapses, but that it could not be shown that these lapses had materially altered the outcome of the election.

Perhaps the most significant part of the judgment, however, is the third principle. During the course of the hearings, the Court ordered a scrutiny of the IEBC’s servers – under the supervision of the Court’s registrar – in order to cross-check the veracity of some of the allegations. The results of the scrutiny report are discussed extensively in the judgment, with a candour that is not often found in the adjudication of such disputes elsewhere in the world. Indeed, on most of the issues that it framed, the Court set out its reasoning process – including mathematical calculations in some detail and with great transparency – allowing, in turn, for the foundations of its judgment to be scrutinised by the public.

It is trite to say that one may disagree – on substance – with the Court’s analysis on each of the three steps outlined above. Indeed, this writer believes – for example – that the Court’s holding that spoilt ballots be not counted in the determination of whether the winner of the election has crossed 50 per cent is open to critique. After all, why shouldn’t an individual be entitled to spoil their ballot and have their vote counted accordingly? Such disagreements are in the nature of things; the crucial point, however, is that the Court’s overall analytical framework – that is, the standard of evidence and the standard of invalidity – and the three-step analysis outlined above, is undoubtedly sound, and one of the only routes open to a Court to adjudicate high-stakes political disputes without being dragged down into the mire of political partisanship.

The results of the scrutiny report are discussed extensively in the judgment, with a candour that is not often found in the adjudication of such disputes elsewhere in the world.

It is in this context that the statement of Azimio that the Supreme Court presents a “threat to democracy” is a matter of some concern. As mentioned in the beginning of this article, around the world, clashes between the judiciary and politicians are not uncommon, especially when it comes to high-stakes elections. However, many of those clashes have occurred in contexts of judicial overreach, or where the Court instals a politician or validates an election in highly opaque or secretive proceedings. In the opinion of this writer, two things set apart the Kenyan case: the first is that the Constitution explicitly envisages the Supreme Court as the body that will resolve this dispute, and for good historical reasons (indeed, as the 2017 elections showed, the Supreme Court is capable of – and has – set aside an election in the past). And the second – and more important – thing is that, when you consider the judgment in Odinga and 16 others vs Ruto and 10 others from the perspective of global best practices in adjudication, it stands up to searching scrutiny. The Court set out neutral and objective framework principles to guide its adjudication of the case, reasoned closely and narrowly within those principles, and set out its chain of reasoning in a judgment on record. The Court’s judgment may attract criticism (even stringent criticism), and that is in the nature of things, but – respectfully – it does not warrant an attack. It is important to remember that the dispute resolution process under Article 140 requires an independent and strong Court that can act to invalidate a flawed presidential election (as it did in 2017). If that is gone, then it is an open question how future disputes can ever be resolved without serious problems.