LLB (London), LLM (Zambia)
Doctoral candidate, University of South Africa
Edition: AHRLJ Volume 15 No 2 2015
Pages: 329 - 354
Citation: O Kaaba ‘The challenges of adjudicating presidential election disputes in domestic courts in Africa’ (2015) 15 African Human Rights Law Journal 329-354
Download article in PDF
This article is based in part on my LLD research at the University of South Africa. I would like to express my gratitude to my academic supervisor, Dr Fagbayibo Babatunde, for his guidance and encouragement.
Defective and fraudulent elections are common in Africa. Although there has been some improvement since the democratic wave of the 1980s and 1990s, sham elections are still prevalent across the continent. Where elections have been assailed with anomalies and results are disputed, as is often the case in Africa, aggrieved parties have looked to the judiciary as an institution of last hope to seek redress. The judiciary has, however, almost always decided presidential election disputes in common patterns that militate against the growth of democracy on the continent. The common patterns are that all cases are decided in favour of the status quo; many cases are dismissed on flimsy technical and procedural rules without consideration of the merits; there is misuse of the substantial effect rule to uphold defective elections; there are delays in determining cases; and judges refrain from making any reasonable decisions. The judiciary in Africa may, therefore, be fully complicit in the delayed consolidation of electoral democracy on the continent.
Defective and fraudulent elections are common in Africa.1 Where elections have been assailed with anomalies and results are disputed, as is often the case in Africa, aggrieved parties have looked to the judiciary for redress. The judiciary thus is faced with the unenviable task of determining the ultimate outcome of the poll. Consequently, in order to protect the right to choose in an election, and to promote and safeguard democracy, the judiciary must be competent, honest, learned and independent.2 Such a judiciary plays a transformative role in democracy as an impartial referee or umpire in the democratic game.3
This article looks at how African courts have handled presidential election disputes. The article first presents an overview of the role of the judiciary in resolving disputed presidential elections. It identifies common patterns that characterise the ways in which courts dispose of disputed presidential elections in Africa, all of which are unsatisfactory and a disincentive for the further growth and consolidation of democracy.
Elections affirm the sovereignty of the people.4 Through elections, people constitute a government and hold the government accountable.5 However, the history of elections in Africa is disappointing. Democratic elections have been rare and disputed elections have been the norm. 6
Although there has been some improvement since the democratic wave of the 1980s and 1990s, sham elections are still prevalent across the continent. In Nigeria, for example, all presidential elections since the return to civilian rule in 1999 have been disputed, until 2015 when President Goodluck Jonathan peacefully conceded defeat. The European Union (EU) observer mission to Nigeria, for instance, denounced the 2007 Nigerian elections in the following terms:7
The 2007 state and federal elections fell far short of basic international and regional standards for democratic elections. They were marred by very poor organisation, of essential transparency, widespread procedural irregularities, substantial evidence of fraud, widespread voter disenfranchisement at different stages of the process, lack of equal conditions for political parties and candidates and numerous incidents of violence. As a result the process cannot be considered to have been credible.
In a similar fashion, all election observer missions to Zambia’s 2001 elections, both local and international, concluded that the elections were far from being free and fair.8 The EU, in fact, took a rare stand and waived the immunity of the head of the observer mission, Michael Meadowcroft, to testify for the opposition in the ensuing election petition.9
Huefner classifies the causes of disputed or failed elections in two categories: fraud and mistake.10 Fraud means the deliberate unfair manipulation of the system, often by parties, candidates or their supporters.11 On the other hand, mistake is the unintentional disturbance of the election process, usually caused by those administering the election.12 Whether by mistake or fraud, failed elections deny the people their right to constitute the government according to their will in a transparent way. Distinguished Nigerian scholar, Ben Nwabueze, considers this ‘robbery of the right of the people to participate in their own government’ and ‘therefore the greatest offence that can be committed against the constitution and the people’.13 This is a correct observation because failed elections have the effect of taking away the consent of the people as the basis of the right to govern.
Almost all African constitutions or electoral laws recognise that things can go wrong with elections and provide for the possibility of redress. This is because election wrongs or allegations of wrongs often have a bearing on the legitimacy of the electoral process. A fair and transparent redress mechanism, which commands the respect of the people, lends legitimacy and credibility to the election and ‘serves as a peaceful alternative to violent post-election responses’.14 On the other hand, a failure to put in place an effective electoral dispute mechanism ‘can seriously undermine the legitimacy of an entire electoral process’.15
The article focuses on the post-election redress mechanisms available in the case of disputed presidential elections. It discusses the manner in which courts have handled complaints that seek to correct election results (in whole or in part) or, indeed, to void the whole election. In almost all African countries, this is a task entrusted to the judiciary. The only difference seems to be the stage in the judicial hierarchy at which the litigation is commenced. In countries such as Nigeria,16 Namibia17 and Kenya18 (prior to the adoption of the 2010 Constitution) cases begin in lower courts and are appealed ultimately to the Supreme Court. In other countries, such as Ghana,19 Zambia,20 Kenya21 (since 2010) and Uganda,22 presidential election petitions are heard directly by the Supreme Court, allowing no appeal.
Tanzania seems to be the only African country with a constitutional provision that ousts the jurisdiction of the judiciary from hearing challenges to presidential elections. The Tanzanian Constitution categorically states:23
When a candidate is declared by the Electoral Commission to have been duly elected in accordance with this article, then no court of law shall have any jurisdiction to inquire into the election of that candidate.
Such a provision can only assume that elections will always be impeccable, something which, of course, is at variance with the African experience. This is a blatant denial of the possibility to seek judicial redress in the case of a grievance, and contrary to article 17(2) of the African Charter on Democracy, Elections and Governance (2007), which obliges member states to establish and strengthen national mechanisms that redress election-related disputes in a timely manner. Even where grievances are ill-founded, the offer of a possible judicial remedy provides a peaceful means of venting frustration instead of resorting to violent protests.
In order to be of any significance, the adjudication or judicial determination of election disputes must offer aggrieved persons a genuine possibility of redress for their grievances. In order to do this, Huefer identifies at least three factors that need to be embedded in the adjudication process. First of all, the process must be fair and perceived as fair by litigants and the public.24 This requires that the process treats the parties to a dispute equally and offers them an equal opportunity to present their case. It also requires that the process resolves disputes impartially and meritoriously. A process that only decides in favour of the incumbent or incumbent party, whatever the strength of evidence presented against it, cannot be considered fair. Second, the process must be transparent, that is, when an election is disputed and a court adjudicates on the dispute, it must do so in a way that is understandable (based on prior existing rules) and must make a fair analysis of evidence as it relates to the competing claims.25 Finally, the process must be prompt and cases should be determined with finality.26 As is often said, justice delayed is justice denied.
Before moving to the next section which looks at the challenges that assail the adjudication of presidential election disputes in Africa, it is important to briefly discuss the art of adjudication. The traditional view of adjudication has been that judges simply re-state the law as enacted by the legislature and exercise no discretion. Their decisions, therefore, are nothing more than a discovery of the intention of the legislature.27 This, however, is now recognised as an oversimplification as the adjudication process is inherently imbued with discretion.28
Hart, for example, considers the law to be open-textured.29 This means that ‘when a judge confronts a rule, he is not met by a bloodless category but by a living organism which contains within itself value choices’.30 Hart offers at least three reasons for this discretion: first, that it is due to indeterminacy or ambiguity of language or words; second, that rules usually use only general standards (for example, ‘reasonableness’ and ‘just’) which need to be related to or distinguished from specific circumstances; and third, the indeterminacy inherent in the doctrine of precedents where judges have to relate current decisions with prior decisions.31
Although Dworkin has virulently criticised Hart’s theory of adjudication, for purposes of this article it is sufficient to note that Dworkin still recognises that there is discretion in adjudication, albeit constrained by law. What Dworkin does is to distinguish between ‘weak’ and ‘strong’ discretion.32 Strong discretion is where one is not bound by any standards set by the authority in question, while weak discretion is constrained by standards.33 Dworkin gives as an example the difference between a sergeant who is ordered to pick five men for patrol and another sergeant who is ordered to simply select his five most experienced men for patrol. The sergeant who is ordered to simply select five men for patrol is considered to have strong discretion compared to the one who has to choose five most experienced men, which is weak discretion.34 As can be seen, Dworkin considers that judges only have weak discretion as they are constrained by law.
The point is that adjudication is a value-laden process and judges have to choose between competing claims and values. As will be seen in the next section, African judges in presidential election disputes appear to inhibit the growth of democracy on the continent or, in the telling words of Muhammadu Buhari, have chosen to ‘stunt the growth of democracy’.35
This section discusses the record of African courts in adjudicating disputed presidential elections. Sifting through the judgments, common threads or patterns emerge that disappointingly negate the advancement of democracy. The five patterns discussed here are the following:
One of the most notable trends in decisions on disputed presidential elections is that all decisions of the courts tend to serve one purpose, namely, maintaining the status quo. These decisions are always given in favour of the incumbent,36 the candidate sponsored by the incumbent party37 or the presumptive winner of the election.38 This seems, inter alia, to stem from judges’ misconstrued understanding of their role as that of ensuring political stability rather than deciding cases fairly, according to the facts presented to them, in line with the applicable law.39 This seems to be the overriding driving force in adjudication, impelling judges to uphold all elections that are brought in litigation for their determination. For example, in the judgment following the petition to Ghana’s December 2012 elections, the Supreme Court stated:40
For starters, I would state that the judiciary in Ghana, like its counterparts in other jurisdictions, does not readily invalidate a public election but often strives, in public interest, to sustain it.
It is submitted that ‘striving’ to uphold an election is not a judicial role, but a political decision. This means that, even before the case is presented, the judiciary is only prepared to preserve the election results that have been announced. As a result, any discrepancies are most likely to be explained away as inconsequential or, as discussed below, ‘not substantial’. It is, therefore, hardly surprising that, despite the African continent being replete with sham elections, the judiciary, when called upon to adjudicate, has always (except for the Côte d’Ivoire Constitutional Council in 2010, discussed further below) upheld these disputed elections.
When this happens, the process of adjudication loses meaning. Adjudication is ideally meant to offer litigants a formal method of taking part in the decision making of the court through presenting their evidence and reasoned arguments for a decision in their favour.41 Judges should, therefore, only reach a decision after hearing the presentation of evidence from both sides of the case and make a determination according to the strength of the evidence presented. As Fuller states, ‘participation through reasoned argument loses its meaning if the arbiter of the dispute is inaccessible to reason because he is ... hopelessly prejudicial’.42
Although the Côte d’Ivoire Constitutional Council decision in 2010 is the only available judicial decision to interfere and reverse announced results, its effect is the same as other decisions that uphold disputed elections, as that decision was made in favour of the incumbent, President Laurent Gbagbo, who had clearly lost the election.43 The circumstances of this case are revealing.
After years of conflict and instability, Côte d’Ivoire finally held elections in 2010. The first round did not produce an outright winner, leading to a run-off election, pitting incumbent Laurent Gbagbo and main opposition candidate Alassane Dramane Ouattara against each other.44 The Chairperson of the Independent Electoral Commission, Youssouf Bakayoko, announced Ouattara as the winner, with 54,1 per cent of votes against Gbagbo’s 45,9 per cent. Gbagbo made a prompt appeal to the Constitutional Council to annul the election of Ouattara based on claims that the elections had been rigged in the northern stronghold of Ouattara. Without giving audience to the other party, the Constitutional Council hastily invalidated about 600 000 votes from Ouattara’s stronghold and declared Gbagbo the winner of the election with 51,45 per cent.45 Some of the grounds on which the Constitutional Council based its decision to annul the election of Outtara are clearly spurious. For example, the Constitutional Council indicated the fact that the results were announced from a hotel instead of the offices of the Independent Election Commission seemed suspicious, and that the results were not announced within the prescribed time of three days.46 There was actually no evidence presented to the Council in support of the serious claims of ballot-stuffing and tampering with results. It is, therefore, difficult to see how the Constitutional Council reached a prompt decision to annul Outtara’s election within hours, without affording the other party an opportunity to present its case.
The recent Kenyan presidential petition epitomises a judiciary willing to go to the extent of legislating and re-writing the Constitution in order to uphold an election. This is manifest in the election petition brought by Raila Odinga challenging the election of Uhuru Kenyatta in the March 2013 elections.47 The Kenyan Constitution, enacted in 2010, under which the election was held, required in part that only a candidate who had garnered ‘more than half of all votes cast in the election’ shall be declared President.48 The former Constitution, replaced in 2010, had simply required the winner to be the candidate ‘who receives a greater number of valid votes in the presidential election than any other candidate’.49
Following the election, the Independent Electoral and Boundaries Commission (IEBC) announced Uhuru Kenyatta as the outright winner, with 6 173 433 out of a total of 12 338 667 votes (50,07 per cent of the votes), while Raila Odinga, the main challenger, had received 5 340 546 votes (43,31 percent). However, the percentage by which Uhuru was declared the winner was based on the number of valid votes, contrary to the constitutional provision that required it to be based on ‘all votes cast in the election’. The importance of the difference is that, if the computation is based on the percentage of all votes cast, then that would take into account all votes, that is, including those declared invalid. The consequence would have been that Uhuru would have had less than 50 per cent of overall votes to prevent a run-off and that, therefore, he would not have been declared the winner of the election.50 Stating that it was interpreting the Constitution purposefully, the Supreme Court held that ‘all votes cast in the election’ actually ‘refers only to valid votes cast’, and does not include rejected votes.51 If this were correct, why did the framers of the Constitution, and affirmed by the people in a referendum in 2010, deliberately and consciously change the language of this provision? This approach obviously altered the language of the framers of the Constitution and effectively meant that the judiciary assumed a legislative role. It is an approach which assumes that the change of wording in the Constitution from ‘valid votes’ to ‘all votes cast’ was a mistake and not really what the framers of the Constitution wanted. In other words, the framers of the Constitution did not mean what they wrote. It is respectfully submitted that interpreting the provision as it is written in the Constitution would not have led to any absurdity. There was, therefore, no need for the judiciary to tamper with the language of the Constitution.
Adjudication is a formal and institutionalised method of reasoned (rational) conflict resolution.52 Its goal is to settle disputes fairly and on the basis of applicable laws. In order to decide cases fairly and to render substantive justice, courts need procedural or technical rules to guide the handling of the cases before them. It can be said that courts fly on two wings of rules: substantive and technical or procedural rules. Those rules that apply to the fairness or merits of the case are considered substantive rules, while those that govern the manner of resolving a dispute are considered technical or procedural.53
Procedural rules and technicalities are manifestly ‘handmaids rather than mistresses’54 of substantive justice. These technical rules are instruments available to the judiciary to help it to render substantive justice and are, therefore, not ends in themselves. This was stated by Lord Penzance in 1878 as follows:55
Procedure is but the machinery of the law after all - the channel and means whereby law is administered and justice reached. It strangely departs from its proper office when, in place of facilitating, it is permitted to obstruct, and even extinguish, legal rights, and is thus made to govern where it ought to subserve.
Although in theory this distinction is obvious, it is usually not so in practice, as the British legal historian, Holdsworth, aptly observed:56
One of the most difficult and one of the most permanent problems which a legal system must face is a combination of a due regard for the claims of substantial justice with a system of procedure rigid enough to be workable. It is easy to favour one quality at the expense of the other, with the result that either all system is lost, or there is so elaborate and technical a system that the decision of cases turns almost entirely upon the working of its rules and only occasionally and incidentally upon the merits of the cases themselves.
In modern societies people submit their conflicts to courts in order that courts may look at their merits without being unduly fettered by technicalities, and have the cases decided fairly. Judges, therefore, have a duty to do substantive justice. In some countries, this has been made a constitutional norm. The Constitution of Kenya, for example, requires that ‘justice shall be administered without undue regard to procedural technicalities’.57
A review of presidential petitions across Africa, however, reveals a disappointing record of courts that ‘shy away from this sacred duty by hiding behind technicalities.’58 Often, presidential petitions have been struck out by courts on curable technical grounds, without considering the merits of the case. When an aggrieved petitioner is sent away from the court without consideration of the merits, this often shatters their confidence in the justice system and negates both the rule of law and the consolidation of democracy.59
Where judges render decisions without much regard for substantive justice, as retired Tanzanian High Court Judge James Mwalusanya aptly stated, ‘the people will offer the verdict and the judiciary will find itself without any credible support’.60 This in effect negatively affects the consolidation of democracy in a country. As Julius Nyerere, first President of Tanzania, warned, unless judges do their work properly, ‘none of the objectives of our democratic society can be implemented’.61
3.2.1 Mwai Kibaki v Daniel Toroitichi Arap Moi62
This was a presidential election petition brought by the then main opposition candidate, Mwai Kibaki, against the election of President Daniel Arap Moi, following Kenya’s 1997 elections. According to official results, Mwai had received 1 895 527 votes while Moi had received 2 440 801 votes.63 Mwai brought an action to void the election because of several electoral malpractices violating electoral rules. The petition, however, was thrown out on technicalities relating to the service of the petition.
The relevant rule on serving petitions stated:64
(2) Service may be effected either by delivering the notice and copy to the advocate appointed by the respondent under Rule 10 or by posting them by a registered letter to the address given under Rule 10 so that, in ordinary course of post, the letter would be delivered within the time above mentioned, or if no advocate has been appointed, or such address has been given, by a notice published in the Gazette stating that the petition has been presented and that a copy of it may be obtained by the respondent on application at the office of the Registrar.
The petitioner had served the petition by way of publication in the Government Gazette, since the respondent had not furnished details of his advocates as provided for in the rule. The petitioner did not effect personal or direct service because the respondent, as President, ‘is surrounded by a massive ring of security which is not possible to penetrate’.65
The court held that the rule did not compel the respondent to provide contact details of his advocates. According to the court, a petitioner could undertake service through publication in the Gazette only if the petitioner had attempted and failed to do so through personal service, service through advocates and/or registered mail. Only then could a petition be presented by way of publication in the Gazette and, because this had not been done, the petition failed and the court dismissed it for improper service.66
In Kenya, this was not an isolated incident but a common method the judiciary followed to annihilate presidential election petitions without hearing the merits. For example, following the 1992 presidential elections, the losing opposition candidate, Kenneth Matiba, brought a petition challenging the election of Daniel Arap Moi.67 However, before the elections Matiba became physically incapacitated and unable to write and, therefore, unable to personally sign the election petition as required by the rules of service. The petition was signed by his wife, to whom he had given a power of attorney. The court, however, struck off the petition for failure by the petitioner to sign the petition personally.68
3.2.2 Rally for Democracy and Progress & Others v Electoral Commission of Namibia & Others69
This was an election petition brought by the opposition following the 2009 presidential and parliamentary elections in Namibia. The petition sought to void the presidential election, inter alia, for non-compliance with electoral laws.70 Section 10 of the Electoral Act, 1992, required that election petitions could only be presented within 30 days of the results being announced. The petitioners presented their petition on the thirtieth day at 16:30 and, therefore, within the statutory requirement. The Registrar of the High Court accepted the petition. However, a rule of court did not allow the filing of a process on any day after 15:00. Because the petition was filed after 15:00, the Court held that the petition was invalid for being filed out of time and, therefore, in the eyes of the law there was no valid petition to adjudicate on.71
3.2.3 John Opong Benjamin & Others v National Electoral Commission & Others72
In this case, the petition was brought by the losing opposition leader, John Opong Benjamin, and other opposition leaders against the election of Ernest Bai Koroma during the Sierra Leone elections of 2012.73 Article 55(1) of the Constitution provides that anyone with a grievance in a presidential election should petition the Supreme Court within seven days of the results being declared. The election was held on 17 November and the results were declared only on 23 November.74 The petitioners filed their petition on 30 November, the seventh day after the declaration of results. Further rules of court required that petitioners leave the names of the advocates acting for them at the court registry in a separate notice, and that, within five days of filing the election petition, the petitioners make payment for security of costs.75 The petitioners’ lawyers had indicated their contact details by endorsing these on the petition, but not in a separate notice, and made security of cost payments on 5 December. The Court, however, struck out the petition, holding that it had been filed out of time due to a delay in payment for costs and for not complying with the requirement of lawyers’ contact details to be in a separate notice.
3.2.4 Atiku Abubakar & Others v Umaru Musa Yarsa Ya & Others76
This is the final case in the list of examples of presidential election disputes dismissed on the basis of procedural technicalities without consideration of the merits of the case. The petition arose from the 21 April 2007 Nigerian elections. The petitioner, Atiku Abubakar, had polled 2 637 848 votes against the winner, Umaru Musa Yarsa Ya, who had received 24 638 638 votes.77 Prior to the election, the Independent Electoral Commission of Nigeria (INEC) had disqualified the petitioner from the election and his name had been excluded from the ballot papers. This was based on the INEC’s erroneous view that the petitioner had been indicted for corruption and embezzlement-related criminal offences and was therefore unsuited for presidential office.78 His name was finally printed on the ballot papers, only four days before the election, through a ruling to that effect by the Supreme Court.79
The petitioner sought to challenge the election of Yar’adua on the following grounds:80
The applicable provision, on which the majority based its decision, states:81
The majority reasoned that grounds (a), (b) and (c) above were separated from ground (d) by the use of the word ‘or’, a disjunctive used to express an alternative or choice.82 Since the petitioner’s name ultimately made it onto the ballot paper and he took part in the election, he could not, therefore, plead ground (d) as he had not been excluded from the election. In the view of the majority, the use of the word ‘or’ meant that the petitioner had to choose between the alternatives and could, therefore, only plead one set of grounds.
Having considered the fact that the petitioner’s name was on the ballot paper, the Court declined the invitation to consider whether his initial disqualification may have constituted constructive exclusion from the election as it had left him with barely four days to campaign.83 But for the majority, since the petitioner took part in the election, his petition on the basis of ground (d) collapsed and, since the word ‘or’ denoted alternatives, the rest of the petition collapsed and, therefore, other grounds would not be entertained.84 This decision is surprising, considering that the same court, in a different case, strongly condemned judges occupying themselves with technicalities at the expense of substantial justice and held that judges had a duty to shy away ‘from submitting to the constraining bind of technicalities’.85
The decision of the majority was fraught with many flaws, and at least three of them may be mentioned here. First, the majority noted that the petitioners had pleaded two sets of inconsistent claims. Even if they were true, there was no legal basis for the Court to choose which set of inconsistent grounds to deal with. The Court never explained why they chose one ground on which to dispose of the petition. They could as well have chosen the set of claims which had merit and left the impugned alternative ground.86 Second, the practice of pleading in the alternative and even inconsistent claims has long been established in the common law tradition. This practice is aimed at staving off the possibility of inundating courts with a multitude of successive suits relating to the same facts, and allows courts to deal with all matters in one suit.87 This was, for example, allowed in the Zambian 2001 presidential election petition.88 Similarly, in Uganda, where trials of presidential election petitions are largely by affidavit, the court in 2001 allowed defectively-drafted affidavits, holding that technical weaknesses should not be allowed to vitiate the quality of documents.89 Third, the manner in which the majority dealt with the principal claim of exclusion was nothing more than a trivialisation of the issue and a negation of the right to an effective remedy.
The evidence, accepted by the Court, indicated that the INEC had gone out of its way to exclude the petitioner and had printed the first version of ballot papers without his name. The INEC was forced by order of court to print new ballots four days before the election. This left the petitioner with just four days to campaign and effectively put him at a disadvantage.90 The Court took a narrow and simplistic interpretation of this exclusion, which negated the need to offer candidates an equal opportunity to campaign.
Although all politically-stable African countries have laws and regulations that govern the conduct of elections, these do not of themselves guarantee free and fair democratic elections. Often, election results are affected by honest mistakes, the incompetence of election officials, corruption, fraud, violence, intimidation, cheating and other irregularities. Some of these irregularities may be minor and inconsequential. However, others are significant and have a bearing on the fairness and legitimacy of an election.
When courts are faced with an election petition, there is, therefore, the need for a legal device or mechanism whereby they will determine which irregularities are minor and inconsequential and which are significant and in need of redress. The substantial effect rule does this. For many Anglophone African countries, this is an old rule inherited from the English legal system. The gist of the rule is that elections should not be nullified for minor irregularities or infractions of rules.91
No parliamentary election shall be declared invalid by reason of any act or omission by the returning officer or any other person in breach of his official duty in connection with the election or otherwise of the parliamentary election rules if it appears to the tribunal having cognisance of the question that -
The idea behind the rule is that flimsy mistakes, omissions and commissions should not lead to the annulment of an election, provided that, overall, the fairness of the election was not vitiated. Lord Denning identified three strands to this rule:94
(2) If the election was so conducted that it was substantially in accordance with the law as to elections, it is not vitiated by a breach of the rules or a mistake at the polls- provided that it did not affect the results of the election.
(3) But, even though the election was conducted substantially in accordance with the law as to elections, nevertheless if there was a breach of the rules or a mistake at the polls - and did affect the results - then the election is vitiated.
The substantial effect rule can sometimes be expansive to include criminal acts, such as acts of corruption, cheating and other illegal electoral malpractices. Equally, the rule here has three strands:95
The substantial effect rule produces a major challenge where illegal acts or substantial flouting of electoral laws do not lead to the automatic voiding of the election, unless it be proved that that had a bearing on the results. Such a rule is defeatist and a carry-over, in the case of the British, from the times when electoral corruption and cheating were considered inevitable to the electoral process.96 It seems inappropriate in a modern democracy to saddle a litigant, who has proved a substantial breach of electoral laws and/or corruption, with also proving that this had an effect on the results. Every voter in a modern democracy is surely entitled to an honest, fair and transparently-democratic election. It would certainly not be appropriate for a successful candidate to be heard to say: ‘I accept I was elected following widespread fraud carried out in my favour but, if you cannot demonstrate to a court that the fraud affected the result, my election stands.’97
The substantial effect rule has worked in the most disingenuous way in Africa to uphold elections fraught with major irregularities and fraud. As will be seen from the following case examples, election petitions that manage to survive being thrown out on technicalities are usually decided and dismissed for want for satisfying the substantial effect rule.
3.3.1 Kizza Besigye v Yoweri Kaguta Museveni98
This petition was brought by the main opposition losing candidate, Dr Kizza Besigye, challenging the election of the incumbent, President Museveni, following the February 2006 election. The relevant statutory provision under which the case was decided states: 99
(a) non-compliance with the provisions of this Act, if the court is satisfied that the election was not conducted in accordance with the principles laid down in those provisions and that the non-compliance affected the results of the election in a substantial manner.
At the hearing of the petition, the following issues for decision were framed by the Supreme Court:100
As regards the first two issues, the Supreme Court judges were unanimous that the election had been vitiated by the disenfranchise-ment of voters by unlawfully deleting their names from the voters’ register; the wrongful counting and tallying of results; bribery; intimidation; violence; multiple voting; and ballot stuffing.101 On the third issue, by a majority of four to three, the Court held that the failure to comply with the provisions and principles in statutes as found in issues (1) and (2) did not affect the election in a substantial manner.102 On the fourth issue, by a majority of five to two, the Court held that, although there had been illegal practices and other irregularities, these had not been committed by the respondent or his agents, nor had they been committed with his knowledge or approval.103
The third issue (substantial effect), however, was the main issue around which the petition revolved and was mainly resolved. The majority dismissed the petition, holding that, in determining whether the irregularities and malpractices had affected the results in a substantial manner, numbers were the sole measuring yardstick. That is, the Court could only be persuaded if ‘the winning majority would have been reduced in such a way as to put the victory of the winning candidate in doubt’.104 Since there was nothing indicating that the margin of 1 580 309 votes between respondent and petitioner would have been significantly reduced, the election stood.105
There are many shortcomings to be noted from both the wording of section 59(6)(a) of the Presidential Election Act, which provides for the substantial effect rule, and the manner in which the majority applied it. Four flaws may be noted here.
First, section 59(6)(a) requires that the court should not only be satisfied that there was non-compliance with electoral laws, but also that the non-compliance affected the election results in a substantial manner. This provision is difficult to implement objectively, as the requirement to evaluate whether or not the non-compliance had an effect on election results in a substantial manner is no longer a legal exercise premised on the evidence before court. It basically requires judges to make a subjective evaluation of the consequences of their prospective decision. As Kanyeihamba JSC stated in his dissenting opinion, the provision ‘transports the judge from the heights of legality and impartiality to the deep valleys of personal inclinations and political judgment’.106
Second, the numerical test applied by the majority is manifestly and inherently wrong. While certain malpractices, such as ballot stuffing, voting by ineligible persons and the wrong tabulation of results, may be cured by reference to numbers, others, such as intimidation, violence and deploying the military throughout the country, cannot be captured in a mechanical sense of numbers. If, for example, soldiers kill a person and tell many people that anyone who votes against the incumbent will meet the same fate, how would this be reflected in numbers? It was unanimously accepted by the judges that violence and intimidation were widespread, and the Constitution and other electoral laws were seriously flouted. This should have been enough to assure that there was the necessary deleterious consequence on the elections. It is strange jurisprudence that, after adjudging the election not to have been transparent, free and fair, the majority of the court then backtracked and held that the irregularities were of no substantial effect.
Third, the numerical approach taken by the majority seems at odds with the concept of the rule of law and constitutionalism. The judges were unanimous that the elections had been held in a manner that violated constitutional and other statutory provisions on the conduct of transparent and democratic elections. The decision of the majority effectively means that gross violations of the Constitution and other laws are of no consequence, provided that the petitioners cannot by reference to numbers demonstrate that the gap in results would have been diminished. It goes without saying that this violates the principle of the supremacy of the Constitution.
Fourth, and finally, by overlooking serious electoral malpractices at the expense of numbers, a dangerous precedent for rewarding electoral cheating is entrenched with the full imprimatur of the court. This takes away any incentive for honest behaviour in politics and elections. Ironically, it means that one has to cheat so much that the gap in results should be numerically large to avoid judicial interference with the results. It is hardly surprising that exactly the same pattern of electoral irregularities was handled by the Supreme Court during the 2001 election petition.107 This is because the precedent set by the Court offers no disincentive for committing electoral malpractices, especially by those in power.
3.3.2 Anderson Kambela Mazoka & Others v Levy Patrick Mwanawasa & Others108
This case was brought following the 2001 Zambian general elections. In the Zambian situation, the substantial effect rule was not a statutory requirement, but one effectively legislated into existence by the Supreme Court in the first-ever presidential election petition that followed the 1996 general election.109 The Supreme Court admitted that there had been many flaws in the electoral process, which included the use of the national intelligence service in a partisan way, the unlawful use of public resources by the incumbent party, and the abuse of resources from para-statal companies.110
The Supreme Court held that it could not grant any remedy or interfere with the result of the election because, taking into account the national character of the presidential election, ‘where the whole country formed a single electoral college’, it could not be said that the proven ‘defects were such that the majority of the voters were prevented from electing the candidate whom they preferred’.111 In the view of the Court, the petitioners were supposed to prove that the flaws ‘seriously affected the result’ to such an extent that it could no longer be viewed as a true reflection of the majority of voters.112 To demonstrate this, the petitioners should have proved ‘electoral malpractices and violations of the electoral laws in at least a majority of the constituencies’.113
While this case was decided in a similar way as the Ugandan case discussed above, it differs significantly in that here the substantial effect rule was expanded to include the wide geographical spread of irregularities, in addition to the numbers. The same flaws noted in relation to the reasoning of the Ugandan Supreme Court apply here. However, the subjective nature and arbitrariness of the decision are made clear when one takes into account that the winner and the runner-up in the election were separated by less than two percentage points. It is, therefore, possible that any slight anomaly in even one isolated part of the country could have had an effect on the results. Considering that the election was very close, it seems that the court deliberately added the geographical spread element to the substantial effect rule, knowing that the numerical test would not be easy to sustain considering that the result was very close.
3.3.3 Nana Addo Dankwa Akufo-Addo & Others v John Dramani Mahama & Others114
This case arose from the 2012 Ghanaian elections. The main issue raised by the petitioners included allegations of over-voting; voting without biometric verification as required by law; absence of signatures of presiding officers on some results sheets, contrary to the law; and the occurrence of the same serial numbers for different polling stations.115 The situation was that if votes tainted with these anomalies were deducted, then the president-elect, Mahama, would not have had the 50 per cent-plus-one-vote constitutionally-required majority to be considered the elected President.
Although the majority gave various reasons for upholding the election, the common theme was that, even if there were these noted anomalies, the election itself was ‘conducted substantially in accordance with’ the Constitution and other laws.116 However, such jurisprudence should be worrying. The anomalies were contrary to the Constitution and other laws, and thus could not just be wished away. Taking them into account meant that the declared winner did not really win the election. The decision does not seem to be legally supportable and was probably based on other considerations. Adinyira, JSC, for example, made it clear that in her view, ‘public policy favours salvaging the election and giving effect to the voter’s intention’.117 The decision is in sharp contrast with the guidance of Lord Denning, discussed above, to the effect that even if an election is substantially held in accordance with the law, but is assailed with minor infractions that have an effect on the result, the election is vitiated and voidable.
An effective judicial mechanism for the determination of election disputes should not only be fair, but must also be timely and efficient. Many African states allow the swearing in of the President upon declaration of the results, without waiting for the resolution of election disputes by the courts. In Nigeria, for example, Olusegun Obasanjo proceeded to be sworn in in 2003, despite a court order restraining him and his running mate from presenting themselves for swearing in, pending the determination of the substantive election petition.118 In Zambia, for example, article 34(9) of the Constitution requires the person who has been declared the winner to be sworn in and to assume office immediately and not later than 24 hours after the declaration. Where a president-elect is sworn in even before election disputes are settled by the courts, the need for an efficient resolution of cases becomes even more sensitive. Indeed, the element of time is inherent in the concept of fair adjudication, making justice a time-bound concept.119
There are some African countries that are exemplary with regard to the timely resolution of presidential election petitions. In Uganda, for example, the law requires the hearing and determination of presidential election disputes to be finalised within 30 days of presentation of the petition. The Ugandan Supreme Court in both the 2001 and 2006 elections managed to determine the cases within the set time limit.120 In Kenya, the 2013 election petition, the first since the promulgation of a new Constitution in 2010, was promptly resolved within 30 days of presentation of the petition.121 However, as was noted by the Ugandan Supreme Court in both 2001 and 2006, the pressure to complete election petitions was at the cost that the judges had to rely on mostly evidence by affidavit and, therefore, largely did not have the benefit of having witnesses examined before them so that they could judge their demeanour. The speedy resolution of disputes, while valuable, may mean not giving enough time to the process of fact finding and there is therefore the risk of having important decisions based on inadequate information.
There are, however, still several countries where cases are habitually delayed, rendering the whole adjudication process an exercise in futility. In Nigeria, for example, it is estimated that a presidential election petition takes about two years to finalise, which is actually half of the presidential tenure.122 Perhaps Zambia has the poorest record of inefficiency in the adjudication of presidential election disputes. The presidential election dispute that arose from the 2001 elections, for example, was only determined in 2005, just about a year before another general election.123
The delayed determination of election petitions, where one candidate has already been sworn in, presents numerous challenges. First of all, it raises the issue of the legitimacy of appointments and decisions made by such a president, considering that there is a cloud of uncertainty about his or her election until the court finally determines the matter. Second, delays often increase uncertainty and anxiety in a nation. It is not uncommon that delays in determining election petitions precipitate military coups or coup attempts. In Zambia, for example, the disputed presidential elections of 1996 and the delayed determination of the subsequent election petition is thought to have influenced the 1997 military coup attempt.124 Similarly, the military takeover in Egypt of July 2013 came amidst the delayed determination of the election petition filed by the losing opposition leader, Ahmed Shafiq, during the 2012 elections.125
Third, although there has been no judicial voiding of a presidential election on the continent so far (with the exception of Côte d’Ivoire), where justice is delayed and where an election is overturned, that would lead to a distortion of the tenure of the person who merits to be the President. In the case of Zambia where, for example, it took almost four years to conclude the election petition, had the court determined that the opposition candidate was the rightful winner, that would have left the genuine winner with just a year of office. The illegitimate candidate would then have ruled the country for the better part of the presidential tenure. This, of course, would violate the people’s right to choose their leaders and for the rightful leaders in turn to represent their people.
The Nigerian case of Amaechi v INEC,126 although a governorship case, is illustrative. In this case, the Nigerian Supreme Court held that the person who was declared the winner of the state governorship position, in fact, was not the rightful winner and, therefore, the Court annulled his election and declared the petitioner as the legitimate governor.127 Had this been in relation to a presidential election where, for example, a term runs from one election to the next, it would mean that the wrong person served as president and, consequently, would have had their term unfairly reduced.
Fourth, if the election were to be overturned or nullified, it would mean that the wrong person was allowed to earn a presidential salary and other benefits for a protracted period, to which he or she was not entitled. It is unlikely that such benefits would be reimbursed and, therefore, delayed justice leads to the ‘abuse’ of public resources.
The requirement under the new Kenyan Constitution which requires elections to be held prior to the expiration of the term of office of the incumbent and that election disputes should be resolved prior to the swearing-in of the new President, would seem to be a better alternative here.128
Courts sometimes refrain from making any meaningful decision or simply defer to the executive instead of making a final and binding determination. In the Nigerian case of Buhari,129 for example, the losing candidate, Muhammadu Buhari, sought and was granted an injunction by the court restraining Obasanjo and his running mate from presenting themselves for swearing-in into office pending the determination of the main election petition.130 The respondents, in violation of the court order, went ahead and were sworn in, whereupon the applicants appealed to the Supreme Court for a determination, inter alia, as to whether the President had been validly sworn in when it was done in violation of a valid court order. The Supreme Court held that the appeal was no longer of any relevance since the respondents had already been sworn in and, therefore, the injunction would only be an academic exercise that had no res or status quo to protect. In any case, the Court felt that the injunction was not directed at the Chief Justice not to swear in the respondents.131 The Supreme Court considered that the applicants would not suffer any loss as the courts would still go ahead and determine the main election petition objectively and on its merits.
It goes without saying that such an approach can only lead to cynicism about the commitment of the court to doing justice. It is common knowledge that a person, once sworn into office, can improperly influence the court into passing a favourable decision, considering the state power and resources at his disposal. The then Nigerian chief justice, for example, later revealed that President Obasanjo had made efforts to influence the judges by either bribery of intimidation.132 In this case, the court simply refrained from addressing the consequences of the swearing-in, which had gone ahead despite a court order to the contrary.
The Zimbabwean 2008 election is another example of self-imposed impotence by the judiciary. Following the 29 March 2008 election, the Zimbabwe Electoral Commission (ZEC) inordinately delayed announcing the results, prompting the opposition Movement for Democratic Change (MDC) to seek an order of the court compelling the ZEC to release the results.133 Judge Uchena accepted that the delay had been inexplicable and unjustified.134
However, Judge Uchena decided the case on the basis of section 67(A)(7) of the Electoral Act, which stated that the Commission’s decision on whether or not to order a recount and the extent of the recount ‘shall not be subject to an appeal’. According to Judge Uchena, this provision gave the ZEC a wide discretion and, therefore, its decisions were final, not subject to inquiry by the Court.135 The Court was therefore ‘not entitled to intervene and order the respondents to announce the results’.136
The reasoning of Judge Uchena is defective in many ways, as pointed out by Odhiambo.137 First of all, the action was not an appeal against a decision of the ZEC, but it simply sought an order of mandamus to compel the ZEC to perform its statutory duty. The provision the judge based his decision on was, therefore, completely irrelevant to the case. Second, an ouster clause like section 67(A)(7) should not have been applied literally without ascertaining whether it passed the constitutionality test, especially when important national matters were at stake.138 Judge Uchena had the responsibility to review the consistency of that provision with constitutional provisions that give courts unlimited power of judicial review of administrative action. Third, as stated above, Uchena admitted that the delay had been unreasonable. This finding by the judge, therefore, required him not simply to restate section 67(A)(7), but to inquire into the causes of the delay and the consistency of the causes of the delay with the constitutional obligations of the ZEC to conduct transparent and democratic elections.139
In discussing these shortcomings of judicial decisions in presidential election petitions, it should be noted that the desire is not to impress that all petitions presented before court have merit and, therefore, judges should have found for the petitioners in all cases. There have been some cases that have genuinely lacked in merit, at least in the way the grievance was framed, and that were rightfully dismissed. For example, in the Nigerian case of Chukwuemeka Odumegu Ojukwu v Olusegun Obasanjo,140 the main complaint was that Obasanjo was not qualified to serve another term in office due to the fact that he had served as a military head of state, which the petitioners construed to have been Obasanjo’s first term in office. This, it was argued for the petitioners, was contrary to section 137(1)(b) of the 1999 Constitution of Nigeria, which required that ‘a person shall not be qualified for election to the office of President if he has been elected to such office at any two previous elections’. Obasanjo’s ascent to power as a military ruler was not on the basis of any election as contemplated under the Constitution and, therefore, the term limit set in the 1999 Constitution did not affect him. The petition was, therefore, rightly dismissed.
Cases dismissed for lacking merit are, however, very few. Many cases, as discussed above, usually raise genuine concerns, but judges have routinely passed decisions that seem to inhibit the further consolidation of democracy on the continent.
It is no secret that many presidential elections in Africa have been assailed with major irregularities. When this occurs, it usually falls on the judiciary to protect the rights of people to choose their leaders in a free and transparent atmosphere. The record of the judiciary, however, has been overwhelmingly disappointing. The judiciary has routinely upheld clearly defective elections, erroneously considering it their duty to salvage defective elections as a matter of public policy. To achieve this, the courts have largely applied two techniques. The first is to simply dismiss election petitions on curable procedural technicalities without considering the merits of the case. Second, the courts have wrongly applied the substantial effect rule to uphold disputed elections, even in the face of glaring evidence indicating serious violations of constitutional and other statutory provisions. In other circumstances, judges have simply refrained from making an appropriate decision. Further, while in some countries such as Uganda and Kenya (since 2010), judges have been exemplary in determining cases efficiently, in many countries such cases are still characterised by inordinate delays that negate the whole purpose of adjudication.
2. M Ndulo ‘Rule of law, judicial reform, development and post-conflict societies’ http://inbabyproducts.com/tag/rule-of-law-programs-judicial-reform-development -and (accessed 11 December 2013). See also M Ndulo ‘Judicial reform, constitutionalism and the rule of law in Zambia: From a justice system to a just system’ (2011) 2 Zambia Social Science Journal 1-27.
3. S Gloppen ‘How to assess the political role of the Zambian courts’ http://www.cmi.no/publications/publication/?1927=how-to-assess-the-political-role-of-the-zambian (accessed 12 November 2013).
8. See Coalition 2001 ‘December 27 2001 Tripartite Elections Preliminary Report (2002)’; interim statement by the SADC Parliamentary Forum Election Observation Mission on the Zambian Presidential, Parliamentary and Local Government Elections (2001); EU Electoral Observation Mission to Zambia Final Report 27 December 2001 Presidential, Parliamentary and Local Elections (2002).
13. B Nwabueze ‘Nature and forms of elections rigging’ Niger Delta Congress 21 July 2008 http://www. nigerdeltacongress.com/articles/nature_and_forms_of_elections _ri_htm (accessed 20 May 2014).
14. C Vickery ‘Understanding, adjudicating, and resolving election disputes’ IFES Conference Papers 14 February 2011. See also the Preamble to the document ‘Toward an international statement of the principles of electoral justice (Accra Guiding Principles)’ developed by the Electoral Integrity Group, 15 September 2011, Accra, Ghana.
17. In Namibia, such cases are triable in the High Court. The election petition following the 2009 elections in the case of Rally for Democracy and Progress & Others v Electoral Commission of Namibia & Others Case A01/2010, eg, was heard and determined by the High Court.
35. E Anya ‘INEC leaders corrupt, judiciary partial-Buhari’ http://myondostate.com/w3/inec-leaders-corrupt-judiciary-partial-buhari/ (accessed 23 May 2014).
36. Eg, Akashambatwa Mbikusita Lewanika & Others v Chiluba Supreme Court Judgment 14 of 1998; Kizza Besigye v Museveni Electoral Petition 1 of 2001; Kizza Besigye v Museveni Presidential Election Petition 1 of 2006.
37. See eg Mazoka (n 9 above); Abubakar & Others v Umaru Yar’adua & Others SC 72/2008 Supreme Court of Nigeria Judgment of 12 December 2008; and Nana Addo Dankwa Akufo-Addo & Others v John Dramani Mahama & Others J2/6/2013 Judgment of 29 August 2015.
60. JL Mwalusanya ‘Checking the abuse of power in a democracy: The Tanzanian experience’ in H Kijo-Bisimba & CM Peter Justice and the rule of law in Tanzania: Selected judgments and writings of Justice James L Mwalusanya and commentaries (2005) 587.
67. Kenneth Stanley Matiba v Daniel Toroitichi Arap Moi & Others Civil Application NAI 241 (1993) [Election Petition 27 of 1993]. See also Handbook on election dispute in Kenya: Context, legal framework, institutions and jurisprudence (2013).
96. Simmons (n 95 above) para 36. See also C Kam ‘Four lessons about corruption from Victorian Britain’ http://www.indiana.edu/workshop/colloquia/materials/papers/kam_paper.pdf (accessed 23 August 2014); Eggers & Spirling (n 92 above).
106. Kizza Besigye, dissenting opinion of Kanyeihamba, JSC, 304. See also BK Twinomugisha ‘The role of the judiciary in the promotion of democracy in Uganda’ (2009) 9 African Human Rights Law Journal 1.
122. BA Adejumo ‘The judiciary and the rule of law: Challenges of adjudication in the electoral process in Nigeria’ http://www.nigeriavillagesquare.com/articles/akin-oyebode/the-role-of-the-judiciary-in-the (accessed 21 September 2012).
132. ‘How pressure was brought to bear on judiciary to do Obasanjo’s will concerning Buhari’s 2003 presidential election petition’ http://saharareports.com/news-page/dateline-2005-how-pressure-was-brought-judiciary-do-obasanjo%80%99s-will-co ncerning-buhari%E2%80%99s-20 ? (accessed 1 May 2014).
137. EO Abuya ‘The role of the judiciary in promotion of free and fair elections’ http://www.indabook.org/preview/NTC_NPQZr4ClNhNxrEjoy9F3j8CeiGeBAVzhBlHE_fM ,/THE-ROLE-OF-THE-JUDICIARY-IN-PROMOTION-OF-FREE.html?query=Kenya-Elec tions (accessed 12 December 2013).