Jump to Content Jump to Main Navigation

Part 3 Some Specific Aspects of Corruption and Constitutionalism, 15 Politically Exposed Persons, Corruption, and the Use and Abuse of the Constitutional Power of Pardon: A Comparative Perspective

John Hatchard

From: Corruption and Constitutionalism in Africa

Edited By: Charles M. Fombad, Nico Steytler

From: Oxford Constitutions (http://oxcon.ouplaw.com). (c) Oxford University Press, 2022. All Rights Reserved.date: 26 November 2022

(p. 361) 15  Politically Exposed Persons, Corruption, and the Use and Abuse of the Constitutional Power of Pardon

A Comparative Perspective

1.  Introduction

The power of pardon (which forms part of the prerogative of mercy) is enshrined in constitutions around the world. Its exercise is often controversial and one that can have serious constitutional and political implications. This is especially so in cases involving current or former senior government or political figures—referred to in this chapter as politically exposed persons, or PEPs1—who have been convicted (or are accused) of involvement in corruption or a related offence.2

This chapter explores the scope of the presidential power of pardon and its use and abuse. It focuses on a number of constitutions in Anglophone Africa (collectively referred to as the African Constitutions) which remain influenced by the Westminster export model.3 It also refers to other constitutional instruments where appropriate. It argues that the potential for the abuse of power requires the introduction of effective transparency and accountability mechanisms.

The chapter is divided into ten sections. After this introduction, Section 2 considers the rationale for the presidential power of pardon; Section 3 examines its development; and Section 4, the constitutional constraints on it. Section 5 explores the role of constitutional review bodies. In Section 6, the chapter turns to the problems associated with the pardoning of corrupt PEPs, while Section 7 focuses on the role of the courts in reviewing presidential pardons. Section 8 deals with the collateral consequences of granting pardons to PEPs, and Section 9, with PEPs and pre-emptive pardons. Section 10 provides a summary overview of the chapter and presents its key conclusions.

(p. 362) 2.  The rationale for the presidential power of pardon

The rationale for the exercise of the power has been analysed on many occasions. In his authoritative work Commonwealth and Colonial Law, Roberts-Wray states that the

essence of the Prerogative [of Mercy] is that it enables justice to be done by giving due weight to factors which have not been taken into account by the Courts, because they are not relevant to the legal issues, or not admissible in evidence or have come to light after legal proceedings have terminated.4

This echoes the well-known words of Taft CJ in Ex parte Grossman, who emphasized that such a power exists

to afford relief from undue harshness or evident mistake in the operation or the enforcement of the criminal law. The administration of justice by the courts is not necessarily always wise or certainly considerate of circumstances which may properly mitigate guilt [. . . Accordingly …] it has always been thought essential … to vest in some authority other than the courts the power to ameliorate or avoid particular criminal judgments.5

In President of the Republic of South Africa v Hugo6 Goldstone J in the Constitutional Court of South Africa noted that Holmes J in Biddle v Perovich viewed the exercise of the presidential pardon as being more than merely a private act of grace:

A pardon in our days is not a private act of grace from an individual happening to possess power. It is part of the constitutional scheme. When granted it is the determination of the ultimate authority that the public welfare will be better served by inflicting less than what the judgment fixed.7

The power to grant a full pardon is thus seen as being designed to provide a ‘safety net’ for those convicted of criminal offences in a wide range of circumstances, especially where some error or oversight by the courts or other miscarriage of justice has occurred, or on humanitarian grounds. It is an essential constitutional tool and, as Duker notes, it provides the head of the executive with the gift of ‘benevolence’.8 However, this ‘apparently benign gift’9 has the potential to be a tool for the abuse of power where it is utilized to protect corrupt PEPs and other presidential supporters for political motives.

(p. 363) 3.  The development of the presidential power of pardon: Issues of use and abuse

3.1  Use of the power

The power itself is an ancient one going back at least to the Code of Hammurabi.10 For common law countries, it has an ancient English heritage, with the prerogative of mercy traceable back to Anglo-Saxon times. Thereafter it was used extensively over the centuries by English monarchs, and as the colonial empire expanded, so the pardoning power followed.11 During the colonial period the prerogative of mercy was exercisable by the Governor of the territory in the name, and on behalf, of the British monarch. In the exercise of this power, the Governor was obliged to consider the advice of the Executive Council, which comprised a majority of government officers,12 but was to ‘make his decision according to his own deliberate judgment’.13

It was the practice in colonial legislation to specify the various forms of clemency.14 Thus the colonial Governor had the discretion to

  1. (a)  grant to any person concerned in the commission of any offence for which he may be tried in the [territory] or to any person convicted of an offence in any court in the territory a pardon, either free or subject to lawful conditions;

  2. (b)  grant to any person a respite … of the execution of any sentence …;

  3. (c)  substitute a less severe form of punishment;

  4. (d)  remit the whole or part of any sentence or any penalty or forfeiture otherwise due to Her Majesty on account of any offence in respect of which a person has been convicted.15

The italicized words highlight the fact that the power extended not only to post-conviction pardons but also to pre-emptive pardons—that is, where criminal proceedings have not yet commenced or been concluded—and to the return of property forfeited as part of the sentence.

3.2  Abuse of the power

Concerns about the abuse of the power are not new. Duker notes that in England, ‘[t]he annals of the royal prerogative of mercy are replete with suggestions of the power’s propensity for abuse’.16 Indeed, as early as 1309, Members of Parliament (MPs) made a formal complaint (p. 364) against the use of the pardoning power by Edward II. Similar concerns continued over the centuries, especially with regard to pardons granted by the monarch to his officials and servants, often in highly controversial circumstances.17

There are numerous modern examples of cases where the power has been utilized allegedly for improper purposes. A striking example comes from Vanuatu, where section 38 of the Constitution provides that the President of the Republic ‘may pardon, commute or reduce a sentence imposed on a person convicted of an offence’.

In October 2015, Marcellino Pipite, the then Speaker of Parliament, and fourteen other MPs (each of whom were members of the government) were convicted of corruption and the bribery of officials,18 the Supreme Court of Vanuatu having found that they accepted bribes from the opposition to support a vote of no confidence. At the time, the State President was out of the country and in his absence Pipite was appointed Acting President until his return. In this capacity he proceeded to grant a presidential pardon to himself and ten of the other parliamentarians, claiming it was to ‘maintain stability in the Government of the Republic of Vanuatu’.19 Upon his return, the President immediately rescinded the pardons.

As discussed in Section 7, the case raises three significant issues: first, whether a pardon can be revoked validly once granted; secondly, whether a president can grant himself a pardon; and thirdly, whether, and if so, in what circumstances, a court can review the exercise of the power of pardon.20

In the African context, one example of many comes from Nigeria. In 2013, the then President of Nigeria, Goodluck Jonathan, granted a controversial pardon to a former Bayelsa state governor, Diepriye Alamieyeseigha, who had been convicted of a series of corruption-related offences.21 The ostensible reason was that it was in the public interest and for the welfare of the people of the Niger delta, but others took a very different view, arguing that, among other things, the pardon was granted purely for political gain.22

As these examples suggest, there is a need, given the potential for abuse, for effective constitutional and judicial constraints on the exercise of the power of pardon. How the current constraints operate in practice now needs to be considered.

(p. 365) 4.  Constitutional constraints on the exercise of the presidential power: The position post-conviction

In essence, the granting of a pardon to a convicted person strikes at the independence of the judiciary, as the criminal justice system is nullified when that person is freed by presidential order and the legislature is powerless to reverse the decision. Thus it is always an extraordinary process.

This highlights the need for effective constitutional safeguards on the exercise of the power for otherwise it becomes a significant weapon in the hands of the president. In fact, the issue of the need for a check on the exercise of the presidential power was a hotly debated topic at the Philadelphia Convention in 1797. A motion that any pardon required the approval of the Senate failed,23 however, and as a result article II section 2(1) of the Constitution of the United States provides simply that ‘[t]he President … shall have power to grant reprieves and pardons for offenses against the United States, except in cases of impeachment’.

Thus the power of the President to grant a pardon is a rare example of the failure in the US Constitution to incorporate any check or balance on the exercise of executive power. Furthermore, it appears that the power may be exercised before, during, or after any criminal proceedings; it provides, that is, for both a pre-emptive and post-conviction pardon. This was highlighted in 1974 with the granting to former President Richard Nixon by the incumbent President Gerald R Ford of a ‘full, free and absolute pardon … for all offences against the United States which he, Richard Nixon, has committed or may have committed or taken part in’.24 Moreover, it raises the possibility of the granting of a ‘blanket’ or open pardon.

As noted earlier, the African Constitutions are all based on the Westminster export model constitution. This model provided for the prerogative of mercy to be vested in the President. Section 53 of the Constitution of Botswana provides a typical example. The President may

  1. (a)  grant to any person convicted of any offence a pardon, either free or subject to lawful conditions;

  2. (b)  grant to any person a respite, either indefinite or for a specified period, of the execution of any punishment imposed on that person for any offence;

  3. (c)  substitute a less severe form of punishment for any punishment imposed on any person for any offence; and

  4. (d)  remit the whole or part of any punishment imposed on any person for any offence or of any penalty or forfeiture otherwise due to the Government on account of any offence.

The powers in section 53 are almost identical to those enjoyed by colonial Governors, save for one key exception: the granting of a pardon is limited to those convicted of a criminal offence.

Even today, despite the introduction of new constitutions and the amendment of others, a virtually identical provision to section 53 appears in most of the African Constitutions.25 (p. 366) Seemingly, constitutional drafters have been content to retain the original colonial provisions minus the pre-emptive pardon power, although even this is included in the Constitution of Nigeria.26

In practice, the powers under paragraphs (b), (c), and (d) are frequently used by presidents for a range of purposes.27 However, given its focus on PEPs, this chapter explores in particular the exercise of the power in paragraph (a), that is, the power to grant a full or free pardon and the power to remit any penalty or forfeiture, which is found in paragraph (d).

5.  Reviewing the exercise of the presidential pardon: The role of the constitutional review bodies

Given the potential for an ‘abuse of benevolence’, it is argued that there is a need to introduce effective constraints on the exercise of the power based on the principles of transparency, administrative justice, and accountability. All the African Constitutions provide for some form of review mechanism regarding the exercise of the power. Thus, unlike the situation in the United States, the granting of a pardon is, on the face of it, subject to scrutiny by a review body enshrined in the constitution. Yet the African Constitutions vary widely with regard to both the composition and the role of that body.

5.1  Composition of the review body

The African Constitutions adopt one of two approaches to the review process.

The minority require the involvement of the Cabinet or Council of State. Thus, in Zimbabwe the President may exercise the power ‘after consultation with the Cabinet’,28 whilst in Nigeria and Ghana the President exercises his power after consultation with the Council of State.29 In the case of Nigeria, the President is the Chair of the Council and the Vice-President is the Deputy Chair.30 In Ghana, members of the Council are appointed by the President in consultation with Parliament and comprise a range of serving and former senior public officials, together with eleven ‘other members appointed by the President’.31 This approach seems to offer only a very limited prospect of an independent review.

(p. 367) The majority of the African Constitutions provide for a separate Advisory Committee on the Prerogative of Mercy, although there is a wide variation as to its composition. In most constitutions the membership of the Committee is entirely a matter for the President’s determination. For example, in Uganda the Advisory Committee consists of the Attorney-General and six ‘prominent citizens of Uganda appointed by the President’,32 whilst in The Gambia the membership comprises three persons appointed by the President.33 Remarkably, in Zambia the Advisory Committee consists of members ‘appointed by the President’ who hold office at the pleasure of the President,34 but no details are released as to its membership.35 In Botswana the composition of the Advisory Committee on the Prerogative of Mercy is something of a hybrid, with the President presiding over the committee, which comprises (a) the Vice-President or a minister appointed by the President by instrument in writing under his hand; (b) the Attorney-General; and (c) a person qualified to practise in Botswana as a medical practitioner.36

Again, this approach holds little realistic prospect of an independent and transparent review body.

In Malawi37 and Kenya, a statute sets out the composition of the Advisory Committee. The most detailed consideration on the matter comes from Kenya, the only jurisdiction under review where the constitutional drafters sought to provide for an effective review mechanism. Article 133 of the Constitution of Kenya provides for an Advisory Committee on the Power of Mercy comprising the Attorney-General, the Cabinet Secretary responsible for correctional services, and at least five other members as prescribed by statute. In addition, Parliament is required to pass legislation to provide for the tenure of the members of the Committee, its procedure, and the criteria to be applied in formulating its advice. These provisions are enshrined in the Power of Mercy Act 2011.38

The 2011 Act is notable for its detailed provisions relating to the appointment and tenure of members of the Committee, and its powers and functions. Section 5 sets out its composition and provides that in addition to those prescribed in the Constitution, the Committee is to consist of seven other members representing a cross-section of society. A member must be a citizen of Kenya, possess a recognized university degree, and have ‘had a distinguished career in their respective field’ as well as at least ten years’ proven experience in one of a range of fields including law, human rights, governance, and religion.39 With the exception of the Attorney-General and Cabinet Secretary, members hold office for a single term of five years.40

The procedure for appointment is also noteworthy. A selection panel is established comprising both public officials and representatives of civil society. Applications for membership of the Committee are then invited from members of the public via advertisements in the national press. The names of those shortlisted for interview are also published. The selection (p. 368) panel then recommends to the President ten suitably qualified persons for appointment. Section 9(3) provides that thereafter ‘[t]he President shall … appoint seven members from among the persons recommended by the panel’.

The Kenyan model takes a fresh approach that provides for transparency as regards appointments and independence within the membership of the review committee, and stands in stark contrast to the ‘state secrecy’ approach in Zambia. It provides a useful model for other states.

5.2  Role of the Advisory Committee

The role of the review body is of crucial importance. Once again, the African Constitutions demonstrate a variety of approaches. In the majority of states, the President is merely required to consult with the Advisory Committee (or Council of State or Cabinet) before reaching a decision. For example, in Uganda the President may pardon convicted offenders provided the decision is taken ‘in consultation with the Advisory Committee on the Granting of Pardon’.41 Similarly, article 72(1) of the Constitution of Ghana states that the President, ‘acting in consultation with the Council of State’, may grant a pardon to those convicted of a criminal offence.42 There is no specific requirement to provide the reasons for the decision.

In the case of Malawi, a somewhat different view on the consultative role was expressed by Kgoele J in Chantunya v The State:43

The Presidential Pardon is regulated by section 89(2) of the Malawian Constitution, which states that the President may pardon convicted offenders, but such decision shall only be taken in consultation with an advisory committee on the granting of pardon. It thus stands to reason that the President does not have the prerogative alone and is not the ultimate authority with respect to the granting of pardon.

This view is difficult to understand. Arguably, so long as there is appropriate ‘consultation’ with the Advisory Committee, the decision is entirely one for the President.

This limited consultative role casts doubt on the value of the Advisory Committee. At worst, its members are appointed by, and hold office at the pleasure of, the President. The President is required to consult the Committee, but so long as this takes place, he is under no obligation to follow its advice.44 In effect, there is no realistic restriction on the power of the President to grant a pardon to whomever he chooses and for whatever reason. The role of (and need for) an Advisory Committee in such circumstances is dubious indeed.

A different formulation provides the Committee with an advisory function. For example, article 97(1) of the Constitution of Zambia states that the President ‘may on the advice of the Advisory Committee on the Prerogative of Mercy … remit the whole or part of a fine, (p. 369) penalty or forfeiture’ (emphasis added). Article 133(1) of the Constitution of Kenya provides a stronger formulation. Here, the President ‘may exercise a power of mercy in accordance with the advice of the Advisory Committee’ (emphasis added). This raises the issue of whether under such formulations the President may grant a pardon but only if advised to do so by the Advisory Committee.

In Law Society of Botswana and Motumise v President,45 the Court of Appeal of Botswana considered the meaning of the phrase ‘acting in accordance with the advice of’, albeit in the context of a judicial appointment. Here the Judicial Service Commission had recommended the appointment of the second appellant as a judge of the High Court. Section 96(1) of the Constitution of Botswana provides that ‘[t]he Chief Justice shall be appointed by the President’. Section 96(2) provides that ‘[t]he other judges of the High Court shall be appointed by the President, acting in accordance with the advice of the Judicial Service Commission’ (emphasis added). The President declined to appoint the second appellant as a judge of the High Court. It was accepted by both parties that section 96(1) gave the President the sole power to appoint the Chief Justice. The question for the Court of Appeal was whether ‘it was peremptory under [section 96(2)] for the president not to decline or to refuse to appoint a judge recommended for appointment by the Judicial Service Commission’.46

Crucially, the Court held that the expression ‘shall, acting on the advice of’ required the person advised to follow the advice and act upon it. The Court paid particular attention to principles of constitutional interpretation. In his judgment, Lesetedi JA referred to the well-established principles on constitutional interpretation developed by courts in several Commonwealth countries47 and emphasized the importance of considering the context in which any such construction should be undertaken.48 He concluded:

At the end of the day the duty of the Court in trying to ascertain the meaning of a provision of the Constitution is to look at the totality of the various considerations that come into play in the interpretation; this being the wording of the provision in the context of other provisions of the Constitution having regard to the mores and values of the society as a contextual setting, being mindful that a constitution is not a static document but one that lives with the times and evolves for the better good of the society within the limits of its language.

The analysis of the Court of Appeal as to the interpretation of the phrase ‘acting on the advice of’ provides invaluable assistance in considering the scope of the presidential power of pardon. The wording in the Kenyan provision substitutes ‘may’ for ‘shall’. Arguably, whilst this means that the President is not required to act on the advice of the Advisory Committee, crucially, he is unable to act against the advice of the Committee. This might also apply to the position in Zambia. Provided there is an element of independence amongst the membership of the Committee (which is an unknown in the case of Zambia), this represents a key element to enhance transparency and accountability.

(p. 370) 6.  Politically exposed persons and post-conviction pardons

The Kenyan approach regarding the role of the Advisory Committee on the Power of Pardon is praiseworthy in respect of those convicted of ‘ordinary’ criminal offences. Yet there remains the problem of dealing with the pardoning of corrupt PEPs where political and/or national security considerations may influence the decision. In such cases, the traditional reasons for pardoning offenders to ‘afford relief from undue harshness or evident mistake’ or on humanitarian grounds are often inapplicable.

This poses something of a conundrum in that the pardoning of a PEP may be perceived in very different ways. On the one hand, some may see it as an act of benevolence on the basis that the conviction was politically motivated; on the other, others may view the same pardon as an abuse of benevolence in that a PEP, who was rightly convicted of a corruption offence, has now been pardoned for purely political motives. The differing views on the Alamieyeseigha case noted earlier highlight the point. Another possible (and potentially highly controversial) scenario is the granting of a pardon to a former president or senior political leader convicted of a corruption-related offence where this is considered necessary by the incoming president for the sake of national unity, or as part of a pre-election ‘bargain’,49 or following a period of intense political uncertainty and/or violence, or as part of some other political arrangement.

Given this reality, placing PEPs in a special category has considerable merit. There are several possibilities.

6.1  An untrammelled presidential decision

One approach is to recognize that this is essentially a political decision and hence that the pardon of a PEP is the gift of the president, who has the sole discretion in the matter. Thus, whilst strengthening the role of the Advisory Committees is advisable for ‘ordinary’ pardons, such Committees should play no role in cases involving PEPs. In essence, this is to adopt the US approach of providing the President with the unrestricted power to grant a pardon.50

It is an extremely dangerous approach, however, for it provides the President with excessive and unrestrained power. It is potentially highly controversial, and any (alleged) abuse of benevolence can undermine national anti-corruption efforts. It is unsupportable.

6.2  The ‘no pardon for PEPs’ approach

A second approach is to recognize that the danger of an abuse of benevolence is so strong that a PEP convicted of corruption should never be entitled to a pardon. This approach was suggested by the Malawi Law Commission report on the development of legislation (p. 371) on sentencing guidelines.51 It recommended that any person who commits an offence that causes public harm should not be eligible for a pardon, this on the basis that ‘public anger is stirred whenever perpetrators are pardoned’.52 Based on the unsatisfactory and controversial pardoning of Alamieyeseigha, Adangor has also advocated this approach, arguing strongly that

the enormity of the danger posed to Nigerian society by corrupt practices demands an immediate [constitutional] amendment … by prohibiting the President from exercising the pardoning powers in respect of any person concerned with or convicted of corrupt practices … This will make it impossible for the President to jeopardise the fight against corruption53

However, this is unrealistic for it fails to take into account the variety of situations that can arise with the pardoning of PEPs and unnecessarily restricts presidential power. There is therefore no room for this ‘one-size-fits-fit-all’ approach.

6.3  The legislature as a safeguard

As noted, there was an unsuccessful attempt to include in the US Constitution a requirement that a presidential pardon required the approval of the Senate. This approach is worth revisiting. The involvement of the legislature in the review process is potentially important in that it can review the merits of the decision and reverse it if necessary. This would also require the President to provide the grounds for the decision to pardon and so provide a platform for public debate on the matter.

Such a formal review power hardly appears in many African Constitutions. One such exception is the Constitution of Namibia, a document not based on the Westminster export model. Article 32(3)(d) provides that the President may ‘pardon or reprieve offenders, either unconditionally or subject to such conditions as the President may deem fit’. There is no constitutional provision for a separate review body,54 but article 32(9) provides that such a decision

shall be capable of being reviewed, reversed or corrected on such terms as are deemed expedient and proper should there be a resolution proposed by at least one-third of all the members of the National Assembly and passed by a two-thirds majority of all the members of the National Assembly disapproving any such action and resolving to review, reverse or correct it.

(p. 372) Whilst obtaining the required majorities may well cause problems in practice, and indeed may not be necessary, such an approach brings with it both transparency and a measure of accountability. It recognizes that whilst the pardoning of PEPs is essentially a political issue, it also allows flexibility in that legislators are able to review decisions on a case-by-case basis. It is an approach that merits attention by constitutional drafters elsewhere.

7.  Reviewing the exercise of the presidential power: The role of the courts

Given that the constitutional power to grant a pardon is vested in the President, the decision is not justiciable on its merits but the lawfulness of the process by which it was made is open to judicial review. As Chan Sek Keong CJ in the Court of Appeal of Singapore in Yong Vui Kong v Attorney General, neatly put it:

[N]o legal power, including a constitutional power … is beyond the reach of the supervisory jurisdiction of the courts if it is exercised beyond its legal limits (i.e. ultra vires the enabling law) or if it is exercised mala fide (i.e. for an extraneous purpose).55

In Maru Ram v Union of India, the Supreme Court of India also observed: ‘Pardon, using this expression in the amplest connotation, ordains fair exercise … Political vendetta or party favouritism cannot but be interlopers and will vitiate the exercise’.56 The Court added that the pardoning power ‘can be subject to a review where an executive decision [involved] grounds such as discrimination on the basis of religion, caste, colour or political loyalty’.57

The requirement for administrative justice is specifically enshrined in several African Constitutions. The fundamental principle is that executive power must be exercised in a manner that is consistent with the constitution. For example, article 47(1) of the Constitution of Kenya states: ‘Every person has the right to administrative action that is expeditious, efficient, lawful, reasonable and procedurally fair.’ Similarly, article 68 of the Constitution of Zimbabwe provides a justiciable ‘right to administrative conduct that is lawful, prompt, efficient, reasonable, proportionate, impartial and both substantively and procedurally fair’. This provides the ‘infrastructure within which official power of all sorts affecting individuals must be exercised’.58

The Pipite case, noted earlier, neatly illustrates the effectiveness of judicial review as regards the exercise of a presidential pardon. Two particular issues were in question: first, whether a president can pardon himself, and, secondly, whether a pardon once granted can be revoked.

(p. 373) 7.1  A presidential self-pardon?

The constitutional power of the president to grant a pardon to ‘any person’ convicted of a criminal offence suggests that this can also cover a ‘self-pardon’. The point was considered by the courts in Vanuatu in a case arising from the Pipite self-pardon (see Section 3.2). In Natuman v President of the Republic of Vanuatu,59 the three applicants, who were all MPs, sought a declaration that the granting of a pardon to himself by Mr Pipite, the then Acting President of Vanuatu, was unconstitutional. The Supreme Court held that it had jurisdiction in the matter,60 with Saksak J emphasizing that

[w]ithin any country which is committed to the rule of law, notwithstanding the general terms of the [pardon] article, [this] is not a power which can be exercised except in a way which is consistent with the entire constitutional framework. It is not a power which is beyond the purview of the Court to review and assess its exercise for legality.61

He also held that the applicants had locus standi on the basis both as MPs and citizens of Vanuatu.62

As regards the factual issue, the applicants claimed that the conduct of Mr Pipite in granting a pardon to himself and other government officials had breached his constitutional duties, amongst others, to act with integrity and avoid conflicts of interest.63 Saksak J agreed, holding that Mr Pipite, as Acting President, had exercised his powers of pardon ‘wrongly and unlawfully’. Therefore his action was ultra vires section 38 of the Constitution and the Instrument of Pardon was ‘unconstitutional, void ab initio and of no force or effect’.64

The Court Appeal of Vanuatu in Vohor v President of the Republic of Vanuatu65 upheld the decision of the lower court. Lunabek CJ, giving the judgment of the court, emphasized that Mr Pipite, as the then Acting President

had a duty to conduct himself so as not to place himself in a position in which he had or could have had a conflict of interest, or in which the fair exercise of his public duties might be compromised.66

The Wednesbury principles therefore applied67 and the ostensible reason for the pardon, namely, to maintain the stability in the Government of Vanuatu, was ‘plainly an irrelevant consideration’. The decision to pardon was also a clear misuse of public power and was ‘so outrageous in its defiance of logic [and] accepted moral standards that no sensible person who had applied his mind to the question to be decided could have arrived at it’.68 He added, (p. 374) ‘It is difficult to imagine a more serious and obvious conflict of interest, and a more palpable failure of a leader to recognise his responsibilities to Parliament and his nation.’69

Arguably, this is entirely the correct approach and thus a self-pardon is a constitutional impossibility. Whilst the President has the power to grant a pardon to ‘any person’ convicted of a criminal offence, the act is clearly open to judicial review and must be ruled unlawful. There are a range of arguments here. The action falls foul of the nemo judex in sua causa rule, for here the President is acting as a judge in his or her own case. Further arguments can be based on bad faith, abuse of power, irrationality, and a conflict of interest.70

7.2  Revoking the pardon

There is no specific constitutional mechanism empowering a president to revoke a pardon once granted and therefore the basis for doing so, if one indeed exists, is uncertain. In 2003 a pardon was reversed by US President George Bush the day after he had granted it, following the obtaining of fresh information about the case, but the validity of the revocation was never judicially explored.71 The issue was also considered in Natuman v President of the Republic of Vanuatu.72 Upon his return, the President of Vanuatu had purported to revoke the pardons issued by Mr Pipite via the Pardon (Revocation) Order No. 14 of 2015. In the Supreme Court of Vanuatu, the applicants argued that once a pardon was granted under article 38 of the Constitution of Vanuatu, the President had no power to revoke it. Saksak J firmly rejected this argument:

It is common knowledge that Article 38 does not expressly provide a power of revocation. But the absence thereof does not mean the President seeing the legal and constitutional wrong and error done by a person he appointed as Acting President … [should] sit back and not do the right thing he ought and should do to undo the legal and constitutional wrong. That would be absurd and unacceptable in my view.73

Saksak J noted that the Interpretation Act provided that the power to appoint included the power to remove, and reasoned by analogy that article 38 ‘must be accorded the same interpretation as having that implied power’. The validity of the Revocation Order was accordingly upheld. On appeal, the Court of Appeal in Vohor v President of the Republic of Vanuatu upheld the revocation of the pardon on other grounds (see Section 7.1) and dismissed the appeal concerning the presidential revocation power without considering its merits.

The decision seemingly depends more on political realism than any firm constitutional basis. The principles of interpretation of a constitution are far more complex and detailed than merely prescribed in a statute. However, whilst a constitution is not a static document (p. 375) and moves with the times, it is doubtful that a constitutional ‘granting power’ of this nature can include a constitutional ‘revoking power’. The Bush and Pipite cases illustrate that errors may occur in the granting of a pardon or that it may be granted in bad faith. This does not support an argument that a pardon once granted can be revoked by the president. Arguably, the better view is that the act is final and irrevocable but that it is the responsibility of the courts to determine its legality by way of judicial review.

8.  Politically exposed persons, pardons, and collateral consequences

A criminal conviction may also lead to significant ‘collateral consequences’ that can adversely affect ex-offenders, for instance in terms of their ability to obtain employment or gain access to financial services. A pardoning power is thus useful in removing the stigma of past convictions.74 The issue of collateral consequences has received much attention elsewhere75 but scant notice in Anglophone African states.

As a general principle, Pinard’s view has much merit, namely that a ‘dignity-based approach’ is required such that only those consequences are imposed ‘that directly relate to the underlying criminal conduct and are therefore necessary to minimize the risk of further harm’.76 Yet the issue of collateral consequences requires careful consideration in the case of PEPs, particularly in relation to two issues: first, as regards a possible return to political office, and secondly, as regards the forfeiture or confiscation of the proceeds of corruption.

8.1  A return to public office?

The first issue concerns the effect of a presidential pardon on a provision that bars persons convicted of serious criminal offences from political office either permanently or for a specified period. As discussed earlier, such a provision has significant implications for corrupt PEPs. Indeed, this is a potentially significant deterrent and very much in line with the development of an effective national anti-corruption strategy.

There are two main views as to the consequences of a pardon.77 One is that it ‘wipes out’ or erases both conviction and guilt and places the offender in a position as if he had not committed the offence. As was stated in Ex parte Garland,78

when the pardon is full, it releases the punishment and blots out of existence the guilt, so that in the eye of the law the offender is as innocent as if he had never committed the offence.79

(p. 376) This view was echoed in the Nigerian case of Falae v Obasanjo (No 2) where Musdapher JCA emphasized that ‘[t]he effect of a pardon is to make the offender a new man (novus homo) … His civil rights and liberties are fully restored.’80

A similar view was taken by the Constitutional Court of South Africa in The Citizen 1978 (Pty) Ltd v McBride,81 a case that focused on the issue of the ‘expungement’ of a criminal record in the context of an amnesty granted under the Reconciliation Act to an individual previously convicted of murder. Here the court ruled that

[e]xpungement entitles the grantee of amnesty to full civic status. All civil disabilities are lifted. He is entitled to stand for Parliament. Should he ever be convicted of another crime, he will for sentencing purposes be deemed to be a first offender.82

The removal of collateral consequences thus leads to the restoration of the full constitutional rights of the individual, including eligibility to return to public office. Indeed, article 47(1)(a) of the Constitution of Ghana specifically provides that a disqualification from eligibility for membership of the legislature ends with the granting of a free pardon.83

An alternative view is that whilst the criminal conviction is expunged, this is not equivalent to an acquittal. This is the position in Kenya, where the Power of Mercy Act states that a presidential pardon is not to be construed as an acquittal.84 It means that ‘guilt remains’ and this precludes a return to public office at least for the period specified in the Constitution.85

The other African Constitutions are seemingly silent on collateral consequences, although most provide the president with the power to grant a pardon ‘either free[ly] or subject to lawful conditions’.86 This appears to raise the possibility of the president’s granting a ‘partial’ pardon, which seemingly could include a condition prohibiting the individual from returning to public office.

The different approaches reflect the tension between the gift of ‘benevolence’ and the ‘abuse of benevolence’ discussed earlier. Where the pardon is acting as a ‘safety net’, for example to remedy a miscarriage of justice, it is appropriate that it is deemed to wipe away collateral consequences and thus enable the PEP to return to public office. This is not the case with an ‘abuse of benevolence’. As the Alamieyeseigha case illustrates, the ‘wiping away of guilt’ approach is unsatisfactory as it provides a route for a potential return to public office for corrupt PEPs.

(p. 377) Yet making an appropriate distinction in cases based on benevolence is likely to be extremely challenging in practice. For example, Alamieyeseigha claimed that he was the victim of a political witch-hunt during the tenure of the previous President and argued that it was only right that he should be granted a pardon by the new President as an act of benevolence. In contrast, many viewed the pardon as a blatant political ploy by President Jonathan to protect his former boss and to further his own political ambitions: a case of an abuse of benevolence.

Arguably, the approach of erasing the conviction and restoring full civic status is appropriate for all ex-offenders except PEPs, who are in a special category because of their status, their political influence, the circumstances in which their conviction took place, and the highly political nature of the pardoning decision.

As a general principle, the Kenyan approach is surely preferable, namely that the pardon does not amount to an acquittal and guilt remains. This prevents a corrupt PEP from returning to public office at least for a specified time. It also has a potential deterrent effect in that it acts as a warning to other PEPs as to the consequence of any involvement in corruption. However, provision might be made to address a specific case where it is argued that a PEP was the victim of a miscarriage of justice, including a politically motivated conviction. Here the legislature could be given the power to approve a novus homo approach. This has the advantage of requiring a public debate on the matter and removing the final decision from the president. Such an approach might be useful in Alamieyeseigha-type cases.

8.2  Confiscation or forfeiture of the proceeds of corruption

The second issue concerns the fate of the proceeds of corruption acquired by a PEP. The confiscation or forfeiture of such proceeds is a fundamental tool for taking the profit out of corruption. In practice, most states now have legislation providing a court with the power, following a conviction for a corruption-related offence, to order the forfeiture of assets as part of the sentence.87 However, with the exception of The Gambia, Malawi, and Kenya, the African Constitutions each provide that the president may remit any ‘penalty or forfeiture’ imposed on a person after a criminal conviction.88

This is a separate power and is therefore not dependent upon the granting of a pardon. Such a constitutional provision overrides any assets recovery legislation and it enables a president to override the sentence of a court and to order the return of forfeited assets to a corrupt PEP. This is clearly a matter of concern as it undermines a key anti-corruption strategy whereby corrupt PEPs are ‘be stripped of the proceeds of their crimes, the purpose being to remove the incentive for crime’.89 The reality is that the continued inclusion of this colonial-era provision has no place in modern constitutions since it places an unacceptable amount of power in the hands of the president, which is always susceptible to abuse. The forfeiture of the proceeds of corruption is too important a policy to be undermined by such a provision.

(p. 378) In the absence of a constitutional amendment removing the power, there are alternative approaches. In particular, the presidential power applies only to criminal cases. This raises the possibility of making use of non-conviction-based asset forfeiture (NCBAF). Here the action is brought against property as being ‘proceeds of crime’ and there is no requirement to prove any criminal offence. This is also a very useful approach where a possible prosecution of a PEP suspected of corruption is undermined by political considerations.

In addition, the process can be used in other states to recover the proceeds of corruption. For example, NCBAF proceedings were launched in the United States to recover assets worth more than USD32 million belonging to Teodoro Nguema Obiang, the son of the President of Equatorial Guinea, as allegedly being the proceeds of corruption. Given his status, it was inconceivable that he would face prosecution in his home state.90 Obiang later entered into a Stipulation and Settlement Agreement with the United States Department of Justice to forfeit some of these assets, with part of the proceeds being used for charitable purposes in Equatorial Guinea.91

The NCBAF approach is now widely supported and is an optional provision in the UN Convention against Corruption.92 It is hoped that those African states that have yet to introduce NCBAF provisions will do so at the earliest possible opportunity.

Another possible approach is to bring criminal proceedings for money laundering (or a related offence) against a corrupt PEP in another jurisdiction and to order the confiscation of the proceeds of crime as part of the sentence. For example, James Ibori, a former Governor of Delta State in Nigeria, was convicted in a London court of conspiracy to defraud and money laundering of the proceeds of corruption. He was sentenced to a lengthy period of imprisonment as well as having a criminal confiscation order imposed on him.93

Overall, the African Constitutions fail to address adequately (or at all) the issue of collateral consequences. Given the significance of the issue in relation to PEPs, urgent action is needed to address the omission.

9.  Politically exposed persons and pre-emptive pardons

The power of the president to issue a pre-emptive pardon, that is, one where criminal proceedings have not yet commenced or been concluded is less clear. Its attraction was noted by Alexander Hamilton in The Federalist Papers: No. 74, where he argued that

there are often critical moments, when a well-timed offer of pardon to the insurgents or rebels may restore the tranquillity of the Commonwealth; and which, if suffered to pass unimproved, it may never be possible afterwards to recall. The dilatory process of convening (p. 379) the legislature, or one of its branches, for the purpose of obtaining its sanction to the measure, would frequently be the occasion of letting slip the golden opportunity. The loss of a week, a day, an hour, may sometimes be fatal.94

In the African context, there are a number of examples of such arrangements. In 1983 in Kenya, President Moi announced that he had decided to extend executive clemency to some 412 ex-servicemen and 61 university students who were accused (but never tried) of involvement in the 1982 coup attempt. The basis for such ‘clemency’ is not clear as there was (and still is) no constitutional pre-emptive presidential pardoning power in Kenya. A pre-emptive pardon was also granted by the President of Nigeria to Odumegwu Ojukwu, the former Biafran leader and other political detainees, and there is a move to apply this to members of the terrorist group, Boko Haram.95 The point is of particular contemporary significance where a president wishes to pardon a PEP, such as a former president, who is facing corruption-related charges. The possible cases of former South African President Jacob Zuma and former Malawian President Joyce Banda spring to mind.

The issue becomes more complex where a pre-emptive pardon is offered to a PEP (including close family members) in exchange for the return of assets representing the proceeds of corruption. For example, in 2004 the then President of Zambia, Levy Mwanawasa, reportedly offered a pardon to former President Frederick Chiluba if he returned 75 per cent of the money he allegedly looted whilst in office. Similarly, the then President of Nigeria, Olusegun Obasanjo, ordered negotiations with the Abacha family over the release of Mohammed Abacha, who was being held in prison on multiple money-laundering and corruption charges, in exchange for the return by the Abacha family of more than USD1 billion to the government that had been looted by General Sani Abacha when head of a military government in Nigeria.96

For the state, such a ‘deal’ is attractive in that it potentially allows for the speedy return of a significant proportion of the proceeds of corruption. This is especially so given the reality that such proceeds are often concealed in jurisdictions around the world and that recovering them through the judicial process is frequently a long, expensive, and even fruitless exercise. Whatever the morality of the arrangement, the deal reinforces the primary aim of assets recovery: that of taking the profit out of corruption. Of course, the lack of transparency and accountability surrounding the ‘deal’ is a major concern and the question arises as to the constitutional basis (if any) for its exercise.

9.1  A specific constitutional power

The Constitution of Nigeria is the only African Constitution to deal specifically with this issue. Section 175(1) states that ‘[t]he President may … grant any person concerned with or convicted of any offence … a pardon either free or subject to lawful conditions’ (emphasis added).

(p. 380) 9.2  A residual constitutional power?

As there is no comparable provision in the other African Constitutions, the question is raised as to the legality of any such presidential action. If the power does not reside in the constitution, the question arises as to whether there is a residual prerogative power available to the president, especially bearing in mind that the pre-emptive power of pardon was exercised as part of the royal prerogative and was reflected in the powers of colonial governors. Some guidance on such powers is provided by two cases from Botswana.

In the Law Society of Botswana and Motumise v The President, the respondents argued that the President had the prerogative power to refuse to appoint a candidate for judicial appointment as recommended by the Judicial Service Commission.97 This was on the basis that the exercise of such a power ‘involves a consideration and balance of high policy considerations, for example factors of security and socio-political factors’.98 The Court of Appeal implicitly accepted that the President enjoyed prerogative powers, although the argument was rejected on the ground that ‘once a power is conferred by statute or by the Constitution expressly, it ceases to be a prerogative power’.99 However, as the constitutional provisions on the power of pardon fail to address the ‘no prosecution deal’ scenario, there is an argument that the President continues to enjoy the prerogative power to issue a pardon in such cases.

In Patson v A-G,100 Kirby J considered the exercise of the prerogative in the context of the issuing of a passport. He noted that

the prerogative power still applies, exercised in the name of the President. By section 47(1) of the Constitution the executive power of Botswana is vested in the President. This provides that: ‘The executive power of Botswana shall vest in the President and, subject to the provisions of this Constitution, shall be exercised by him or her either directly or through officers subordinate to him or her.’ In practice he exercises his powers, both statutory and non-statutory (or prerogative) powers, through the relevant ministers of state and their departments.

Kirby J then added:

Many of these principles and concepts have been re-defined and specifically included in the Constitution, such as the prerogative of mercy, but others remain as part of the common law of Botswana, such as the non-statutory prerogative powers of the state …

As noted, the prerogative of mercy contained in the Constitution of Botswana (and, with the exception of Nigeria, the other African Constitutions) applies only to the post-conviction scenario. Thus there is an argument that the granting of a pre-emptive pardon remains part of the non-statutory prerogative powers of the state and is therefore exercisable by the president, albeit that this is subject to any constitutional or statutory provision to the contrary. Yet given the potential for the ‘abuse of benevolence’, a better view is that a pre-emptive pardon is, and should remain, a constitutional impossibility in the absence of a specific power to the contrary.

The practical scenarios suggest that there is no easy answer. The ‘clemency’ shown by President Moi in the case of the coup plotters may well have been necessary for the sake of (p. 381) national unity and national security: any pardon of Zuma or Banda could fall into a very different category. Perhaps the most satisfactory approach is to retain the ‘no pre-emptive pardon’ approach whilst ensuring that there is a constitutional or statutory power to pass an amnesty Act on a case-by-case basis. This will ensure at least a full public debate on the issue.

A more readily available option is to utilize the power of the Attorney-General to terminate a prosecution by entering a nolle prosequi, albeit with the decision subject to judicial review.101 In addition, the common law tradition has always provided prosecutors with the discretion to prosecute a particular case or not, and this enables a decision not to prosecute a case if it is not in the public interest. Of course, the danger of state capture of the prosecution process may rule out an objective assessment of the case and undermine the independence of prosecutors.

10.  Conclusion

A notable feature of many of the African Constitutions, old and new, is the retention of the various forms of clemency derived from the powers of the colonial governors. These include the power to grant a pardon to those convicted of a criminal offence as well as the power to remit any penalty or forfeiture imposed as part of the sentence of the court. Invariably the pardoning decision is exercised by the president and is often surrounded by opaqueness: as Clark neatly puts it, ‘[If] the history of civilisation proves nothing else, it proves that where secrecy cloaks the use of power it also cloaks the abuse of power’.102

The president is the appropriate pardoning ‘agency’, but the danger of an abuse of benevolence requires constitutional drafters to revisit the question of having in place effective and appropriate safeguards on its use.

10.1  Post-conviction pardons

The discussion has focused largely on post-conviction pardons. As an ‘act of benevolence’ designed to provide a safety net for those subject to a miscarriage of justice or on humanitarian grounds, the power of pardon enjoys an important place in every constitution. Yet, as this chapter has highlighted, it also provides an avenue for the ‘abuse of benevolence’, particularly as regards PEPs convicted of corruption and related offences. It has been argued that this calls for a re-examination of the pardoning powers. Two strategies have been identified: first, to strengthen the existing provisions to provide for more effective constitutional constraints based on the principles of transparency, administrative justice, and accountability; secondly, to recognize that, due to their political importance, the pardoning of corrupt PEPs demands separate consideration.

(p. 382) 10.1.1  Reviewing the review mechanisms

All the African Constitutions provide for a formal review mechanism. Thus, unlike the situation in the United States, the granting of a pardon is, on the face of it, not an act of benevolence entirely within the gift of the president. It is at this point that the constitutions vary widely both as regards the membership of the review body and its role.

Whilst the majority of constitutions provide for an Advisory Committee on the Prerogative of Mercy, there is no agreement as to its membership. In most states the president appoints the members and can remove them at will. This approach holds out little prospect of an independent review body. A similar comment applies to those states where the review role is undertaken by the Cabinet or Council of State.103

This calls for a membership of the Advisory Committee that can provide the president with an independent review of each case. Here, Kenya provides the leading example, with the members of the Advisory Committee on the Power of Mercy being appointed via an open and transparent process and with each enjoying security of tenure.

The role of the Advisory Committee is also a matter of concern. At present, in the majority of the constitutions the president is merely obliged to ‘consult’ with it. This is an extremely weak formulation that provides no real possibility of an independent review of cases. It is not suggested that the power of pardon be removed from the president, but rather that the incumbent is required either to ‘act on the advice’ of the Committee or at least be unable to act against its advice.

The chapter has also highlighted the fact that in practice it is the courts that hold out the prospect of an effective safeguard on the exercise of the pardoning decision by way of judicial review.

10.1.2  The position of PEPs

It has been noted that PEPs are unlike other convicted offenders in that they are high-profile individuals whose cases often raise important constitutional and political issues. An abuse of benevolence can result in corrupt PEPs’ being pardoned on exclusively partisan grounds. However, there may be circumstances where the pardoning of a PEP is necessary in the public interest. Given this reality, PEPs fall into a different category from ‘ordinary’ convicted offenders and should be subject to a different constitutional regime. This is the only way to address the variety of situations that can arise in pardoning them.

The chapter has also explored the difficult issue of collateral consequences. In the case of corrupt PEPs, a criminal conviction can lead not only to a possible prison sentence but also to two further consequences, namely, the confiscation or forfeiture of the proceeds of corruption as part of the sentence, and a bar on the holding of public office. As regards the latter, the majority of the African Constitutions are silent on whether a pardon erases the conviction. If this is the case, then the PEP is entitled to return to public office, thus removing a key deterrent in the fight against corruption. It follows that the Kenyan approach is appropriate, that is, that an acquittal does not erase the conviction. This would preclude an automatic right of a PEP to return to public office. However, the variety of reasons for the pardoning of a PEP prevents a ‘one-size-fits-all’ approach; thus, the prohibition on a return to public office could be overturned with the approval of the legislature.

(p. 383) With regard to the return of assets, the scope of the power to remit ‘any penalty or forfeiture’ suggests that the president can overturn a criminal forfeiture order, certainly prior to the order’s being enforced. As yet, this point does not seem to have arisen in the region, but its potential importance in the fight against corruption is such that it must be reviewed. With asset forfeiture being an international priority in the fight against corruption, the presidential power to overturn a forfeiture order is no longer appropriate. This is another area for constitutional reform.

10.2  Pre-emptive pardons

The only significant difference between the pardoning power of the colonial governor and that contained in most of the African Constitutions is the absence of the power to grant a pre-emptive pardon. The sole exception is that of Nigeria where, as noted earlier, the Constitution specifically provides for such a power.

On the face of it, a pre-emptive pardoning power gives the president a degree of flexibility to deal with any situation where national security or interests and the like demands his intervention. The extending of ‘presidential clemency’ by President Moi to those involved in the 1982 coup attempt in Kenya is a good case in point. The point takes on even greater importance in view of the development of ‘deals’ in which PEPs (and/or their family members) agree to return their proceeds of corruption in exchange for a non-prosecution agreement.

At present, in the absence of a residual prerogative power, it is not constitutionally possible for the president to grant a pre-emptive pardon. This would require a constitutional amendment following the Nigerian provision. An alternative is to use the power of the Attorney-General (or Director of Public Prosecutions) to discontinue a prosecution at any stage prior to judgment.104 It is certainly not an ideal situation given the executive influence over the judicial process, but one that is perhaps justified on grounds of public interest.

Overall, there are three key conclusions:

  1. 1.  There is an urgent need for the African states discussed in this chapter to undertake a review of their constitutional provisions relating to the granting of pardons and remitting of sentences, including the forfeiture of the proceeds of corruption. It is remarkable that the current provisions still largely reflect the powers of the colonial governors and that in many jurisdictions have seemingly not been subject to any critical review or analysis since independence.

  2. 2.  The exercise of the presidential power of pardon requires effective constitutional safeguards.

  3. 3.  There is a need for special provisions relating to the pardoning of PEPs. These are required to address situations covering an ‘act of benevolence’, an ‘abuse of benevolence’, or one where there are very different views about the pardoning decision.

(p. 384) Bibliography

  • Adangor Z, ‘The Presidential Pardon Granted Chief DSP Alamieyeseigha: Time to Revisit the President’s Pardoning Power under Section 175 of the Constitution of the Federal Republic of Nigeria, 1999 (as amended)’ (2015) 39 Journal of Law, Policy and Globalization 178
  • Chu V, ‘The President’s Pardon and Legal Effect on Collateral Consequences’ (Congressional Research Service 2016) <http://www.everycrsreport.com/reports/R44571.html> accessed 16 April 2018
  • Clark N, ‘The Impact of Mass Communications in America’ (1968) Annals 68
  • Demone B, ‘LGBT Rights in Malawi: One Step Back, Two Steps Forward? The Case of R v Steven Monjeza Soko and Tiwonge Chimbalanga Kachepa’ (2016) 60 (3) Journal of African Law 365
  • Duker WF, ‘The President’s Power to Pardon: A Constitutional History’ (1977) 18 William and Mary Law Review 475
  • Edukugho E, ‘Alamieyeseigha: Unpardonable Pardon’ Vanguard (Lagos, 23 March 2013)
  • Enweremadu D, ‘Nigeria’s Quest to Recover Looted Assets: The Abacha Affair’ (2013) 48(2) Africa Spectrum 51
  • Financial Action Task Force, International Standards on Combating Money Laundering and the Financing of Terrorism and Proliferation (FATF 2012, updated 2018)
  • Forsyth M, ‘Fourteen Jailed: Politics and Justice in Vanuatu’ (2015) ANU Briefing Paper No. 2015/65 <https://bit.ly/2FiTwc0> accessed 2 May 2018
  • Hamilton A, ‘The Command of the Military and Naval Forces, and the Pardoning Power of the Executive from the New York Packet’ The Federalist Papers: No. 74 (New York, 25 March 1788) <http://avalon.law.yale.edu/18th_century/fed74.asp> accessed 18 September 2018
  • Hatchard J, Combating Corruption: Legal Approaches to Supporting Good Governance and Integrity in Africa (Edward Elgar 2014)
  • Iselin C, Code of Hammurabi: King of Babylon (Louvre Museum 2015)
  • Lardner G Jr, ‘A Test of the Power to Unpardon’ Washington Post (Washington DC, 14 January 2009) <http://www.washingtonpost.com/wp-dyn/content/article/2009/01/13/AR2009011302581.html> accessed 18 September 2018
  • Malawi Law Commission, Report of the Law Commission on the Review of the Constitution, Malawi Law Commission No. 18 (2007)
  • Muhia J, ‘Key Legal Provisions in the Exercise of the Presidential Power of Mercy in Kenya’ (2016) <https://bit.ly/2Qxo1fY> accessed 16 September 2018
  • Murray B, ‘Collateral Issues Resource Center’ <http://ccresourcecenter.org/> accessed 18 September 2018
  • Namadzunda C, ‘Malawi Law Commission Wants President’s Power to Grant Pardon to be Reviewed’ The Nyasa Times (Malawi, 26 September 2016) <http://www.nyasatimes.com/malawi-law-commission-wants-presidents-power-grant-pardon-reviewed/> accessed 10 February 2018
  • Nicholls C and others, Corruption and Misuse of Public Office (2nd edn, OUP 2011)
  • Novak A, Comparative Executive Clemency: The Constitutional Pardon Power and the Prerogative of Mercy in Global Perspective (Routledge 2016)
  • Olivier P, ‘Crime and Punishment (and Presidential Pardons): Does the Constitution Permit the Pardoning of the Corrupt?’ (Helen Suzman Foundation 2016) <http://hsf.org.za/resource-centre/hsf-briefs/crime-and-punishment-and-presidential-pardons/> accessed 3 March 2018
  • (p. 385) Pinard M, ‘Collateral Consequences of Criminal Convictions: Confronting Issues of Race and Dignity’ (2010) 85 New York University Law Journal 457
  • Rai JP, ‘Exercise of the Pardoning Power in India: Emerging Challenges’ [2014] 12(2) NEHU Journal 1 <http://nehu.ac.in/public/downloads/Journals/Journal_Jul_Dec14_Art1.pdf> accessed 28 February 2018
  • Roberts-Wray K, Commonwealth and Colonial Law (Sweet & Maxwell 1966)
  • Sackett L, ‘Vanuatu Constitutional Cases Nos 6 and 7 of 2015: Article 38 Pardons and Multilingual Legal Interpretation Principles’ [2015] (2) Journal of South Pacific Law 1
  • Sebba L, ‘The Pardoning Power: A World Survey’ (1977) 68 Journal of Criminal Law & Criminology 83
  • Thompson RM, ‘The President’s Pardon Power and Legal Effects on Collateral Consequences’ (Congressional Research Service, 2006) <https://fas.org/sgp/crs/misc/R44571.pdf> accessed 20 April 2018

Footnotes:

1  These are individuals ‘who are or have been entrusted domestically with prominent public functions, for example Heads of State or of government, senior politicians, senior government, judicial or military officials, senior executives of state-owned corporations, important political party officials’. See ‘Glossary’ in Financial Action Task Force, International Standards on Combating Money Laundering and the Financing of Terrorism and Proliferation (FATF 2012, updated 2018).

2  For example, bribery, illicit enrichment, embezzlement, abuse of functions, and money-laundering. See the United Nations Convention against Corruption (UNCAC), ch 3, <https://www.unodc.org/documents/brussels/UN_Convention_Against_Corruption.pdf> accessed in June 2019.

3  Botswana, Ghana, The Gambia, Kenya, Malawi, Nigeria, Uganda, Zambia, and Zimbabwe. Unless otherwise stated, this chapter refers to what was the current version of these documents in September 2018.

4  Sir Kenneth Roberts-Wray, Commonwealth and Colonial Law (Sweet & Maxwell 1966) 341.

5  267 US 87, 120–21 (1925).

6  [1997] ZACC 4; 1997 (6) BCLR 708; 1997 (4) SA 1 [24].

7  274 US 480, 486 (1927). Emphasis added.

8  William F Duker, ‘The President’s Power to Pardon: A Constitutional History’ (1977) 18 William and Mary Law Review 475.

9  ibid.

10  See generally Claire Iselin, Code of Hammurabi: King of Babylon (Louvre Museum 2015).

11  For a helpful historical overview, see Andrew Novak, Comparative Executive Clemency: The Constitutional Pardon Power and the Prerogative of Mercy in Global Perspective (Routledge 2016) ch 2.

12  The Executive Council was presided over by the Governor, with the majority being government officers as ex officio and nominated official members. See also Roberts-Wray (n 4) 317–18.

13  Roberts-Wray (n 4) 343. There were no female Governors.

14  As Sebba points out, a ‘pardon’ can cover a free pardon, full pardon, conditional pardon, commutation, remission, reprieve, respite, amnesty, clemency, or mercy. See Leslie Sebba, ‘The Pardoning Power: A World Survey’ (1977) 68 Journal of Criminal Law & Criminology 83, 116.

15  See eg The Gambia Royal Instructions, 1962, cl 12 discussed in Roberts-Wray (n 4) 342–45. Emphases added.

16  Duker (n 8) 478.

17  The issuing of a royal pardon to the Earl of Danby, who was facing impeachment on charges of high treason, by King Charles II in 1689 is an excellent case in point. The case also raises another significant point in that Danby had not been tried at the time of granting the pardon. This indicates that the scope of the pardon was not limited to those convicted of a criminal offence.

18  Contrary to the Vanuatu Penal Code, cap 135, s 73.

19  Pardon dated 10 October 2015 (Gazette No. 87).

20  The saga is set out in Pipite and Others v Public Prosecutor Court of Appeal of the Republic of Vanuatu 7 April 2017, Criminal Appeal Case 583 of 2017 <https://bit.ly/2CWj0Km> accessed 30 April 2018. For background to the case, see Miranda Forsyth, ‘Fourteen Jailed: Politics and Justice in Vanuatu’ (2015) ANU Briefing Paper No. 2015/65 <https://bit.ly/2FiTwc0> accessed 2 May 2018.

21  Alamieyeseigha pleaded guilty to six counts of making false declaration of assets. This is an excellent example of where a relatively obscure criminal offence can be used successfully to convict a powerful corrupt PEP.

22  Jonathan had previously been deputy to Alamieyeseigha when the latter was serving as Governor of Bayelsa State. See eg the critical comment by Emmanuel Edukugho, ‘Alamieyeseigha: Unpardonable Pardon’ Vanguard (Lagos, 23 March 2013); for a more detailed analysis of the case, see Zaccheaus Adangor, ‘The Presidential Pardon Granted Chief DSP Alamieyeseigha: Time to Revisit the President’s Pardoning Power under Section 175 of the Constitution of the Federal Republic of Nigeria, 1999 (as amended)’ (2015) 39 Journal of Law, Policy and Globalization 178.

23  The discussion on the issue is neatly summarized by Duker (n 8) 501 et seq.

24  USA Presidential Proclamation 4311 of September 8, 1974, by President Gerald R Ford 39; Federal Regulation 32601-02. According to President Ford, the reason for doing so was to ‘change our national focus … to shift attention from the pursuit of a fallen President to the pursuit of the urgent needs of a rising nation’. Cited in Duker (n 8) 531.

25  Section 89(2) of the Constitution of Malawi contains a much shorter formulation: ‘The President may pardon convicted offenders, grant stays of execution of sentence, reduce sentences, or remit sentences.’ The Constitution of Kenya, art 133(1), and Constitution of Zambia, art 97(1), contain slightly different formulations but are to the same effect as section 53.

26  See Constitution of the Federal Republic of Nigeria, s 175(1). The Constitution of The Gambia omits any power to remit the forfeiture of property: see section 82(1).

27  The prerogative of mercy is often used to release prisoners under an amnesty or where the beneficiaries have committed minor criminal offences. There are numerous instances of the granting of presidential pardons in such circumstances. For example, in May 2017 President Edgar Lungu of Zambia granted a pardon to 428 prisoners to mark Africa Day, although the main reason for the action was seemingly to reduce prison overcrowding. In cases of individuals, a pardon may be granted on humanitarian grounds. See eg R v Soko and Kapecha [2010] MWHC 2; Bradley Demone, ‘LGBT Rights in Malawi: One Step Back, Two Steps Forward? The Case of R v Steven Monjeza Soko and Tiwonge Chimbalanga Kachepa’ (2016) 60 (3) Journal of African Law 365.

28  Constitution of Zimbabwe, art 112(1).

29  See the Constitution of the Federal Republic of Nigeria, s 175(2), and Constitution of Ghana, art 72(1).

30  The other members comprise former Presidents of the Federation and all former Heads of the Government of the Federation; all former Chief Justices of Nigeria; the President of the Senate; the Speaker of the House of Representatives; all the Governors of the states of the Federation; and the Attorney-General of the Federation: see the Constitution of the Federal Republic of Nigeria, sch 3, pt 1, para 5.

31  Constitution of Ghana, art 72(1).

32  Constitution of Uganda, art 121(1).

33  Constitution of The Gambia, s 89(2).

34  Constitution of Zambia, art 96(1)(3).

35  This was highlighted by a telephone conversation between the author (then a professor at the Faculty of Law, University of Zambia) and a very senior Zambian judge. When a gentle enquiry was made as to the membership of the Advisory Committee, the response was that it was a state secret and the conversation came to an abrupt end.

36  Constitution of Botswana, s 54(1).

37  The Advisory Committee on the Granting of Pardons Act, cap 9:04.

38  For an overview of the Act, see Jane Muhia, ‘Key Legal Provisions in the Exercise of the Presidential Power of Mercy in Kenya’ (2016) <https://bit.ly/2Qxo1fY> accessed 16 September 2018.

39  Power of Mercy Act 21 of 2011, s 6.

40  ibid s 13.

41  Constitution of Uganda, art 121(1).

42  Curiously, the 1957 independence Constitution of Ghana simply vested the power of mercy in the President: see article 48. In Nigeria, the President is required to consult the Council of State except as regards military offences, where he is required to act on the advice of the Council of State (s 175(2)–(3))—a provision which perhaps reflects the fact that the document was drafted at the end of a lengthy period of military rule.

43  [2013] ZANWHC 45 [13]. This was a case involving an extradition request from Malawi heard by the North West High Court in South Africa.

44  What constitutes ‘consultation’ is uncertain. For example, Adangor notes that in the Alamieyeseigha case it was alleged that members of the Council of State ‘were not availed of any information relating to [his] application for pardon until the meeting of March 12, 2013, and that the presentation of the matter by the President was “done in a manner that left no room for discussion” ’. See Adangor (n 22) 189.

45  [2017] 3 LRC 472.

46  [2017] 3 LRC 472 [25].

47  See eg R v Big M Drug Mart Ltd [1985] 1 SCR 295 (Canada); Dow v A-G [1992] LRC (Const) 623 (Botswana); e.tv (Pty) Ltd and Others v Minister of Communications and Others [2016] ZASCA 85; [2016] 3 All SA 362 (SCA); 2016 (6) SA 356 (South Africa).

48  [2017] 3 LRC 472 [40].

49  Ensuring a full royal pardon for Anwar Ibrahim, a key political opposition figure convicted of sodomy and corruption, was a key part of the agreement between the former prime minister of Malaysia, Mahathir Mohamed, and the opposition when it was decided in January 2018 that he would swap sides and run as their leader in the election. This led to the May 2018 election victory for Mahathir and the opposition as well as the immediate pardoning and release from prison of Anwar Ibrahim.

50  However, as discussed below, the manner in which the decision is taken is, arguably, always subject to judicial review.

51  Malawi Law Commission, Report of the Law Commission on the Review of the Constitution, Malawi Law Commission No. 18 (2007).

52  See further Chancy Namadzunda, ‘Malawi Law Commission Wants President’s Power to Grant Pardon to be Reviewed’ The Nyasa Times (Malawi, 26 September 2016) <http://www.nyasatimes.com/malawi-law-commission-wants-presidents-power-grant-pardon-reviewed/> accessed 10 February 2018.

53  Adangor (n 22) 189.

54  Section 2 of the Prisons Act (Namibia), Act 17 of 1998 provides that in exercising these powers, the President may call upon the Minister of Prisons and Correctional Services to recommend to him any offenders for such pardon or reprieve, and may invite the comments of the Minister of Justice thereon. In terms of section 93(2) of the Act, the Minister must give notice in the Gazette of the names of every offender pardoned or reprieved under article 32(3)(d).

55  [2011] SGCA 9; [2012] LRC 439 [77]. This was a case where the appellant sought judicial review of a presidential decision, acting on the advice of the cabinet, to deny a grant of clemency.

56  1981(1) SCC 107.

57  For an analysis of the Indian jurisprudence, see JP Rai, ‘Exercise of the Pardoning Power in India: Emerging Challenges’ [2014] 12(2) NEHU Journal 1 <http://nehu.ac.in/public/downloads/Journals/Journal_Jul_Dec14_Art1.pdf> accessed 28 February 2018.

58  Per Mathonsi J in Telecel Zimbabwe (Pvt) Ltd v Postal and Telecommunications Regulatory Authority of Zimbabwe (HH/446/15, 2015, unreported). See also Constitution of Malawi, s 43(1).

59  [2015] VUSC 148.

60  See also Lee-Anne Sackett, ‘Vanuatu Constitutional Cases Nos 6 and 7 of 2015: Article 38 Pardons and Multilingual Legal Interpretation Principles’ [2015] (2) Journal of South Pacific Law 1, 4.

61  At para 18.13 and adopting the words of the Court of Appeal of Vanuatu in Public Prosecutor v Willie [2014] VUCA 4.

62  An interesting point, particularly for civil society organizations, as it suggests that the locus standi for judicial review in such cases applies widely. The basis was that they had demonstrated that their constitutional rights to protection of the law and equal treatment under the law had been infringed or were being infringed: para 17.5.

63  As enshrined in the Constitution of the Republic of Vanuatu, art 66.

64  ‘Conclusions’ to the judgment.

65  [2015] VUCA 40.

66  [2015] VUCA 40 [23].

67  Associated Provincial Picture Houses v Wednesbury Corporation [1948] 1 KB 223.

68  Para 31 applying the words of Lord Diplock in Council of Civil Service Unions v Minister for the Civil Service [1985] AC 374 (the CCSU case). In a different context, Lestedi JA in the Court of Appeal in Botswana in Law Society of Botswana v President (n 45 at para 48) noted that a presidential refusal to appoint a judge contrary to the recommendation of the Judicial Service Commission was amenable to judicial review on the ground of irrationality as explained in the CCSU case.

69  [2015] VUCA 40 [24].

70  For a helpful discussion in the South African context, see Piet Olivier, ‘Crime and Punishment (and Presidential Pardons): Does the Constitution Permit the Pardoning of the Corrupt?’ (Helen Suzman Foundation 2016) <http://hsf.org.za/resource-centre/hsf-briefs/crime-and-punishment-and-presidential-pardons/> accessed 3 March 2018.

71  After the granting of the pardon, news reports brought it to the President’s attention that a political donation had been made by the beneficiary’s father that ‘might create an appearance of impropriety’. See Associated Press report 24 December 2008; George Lardner Jr, ‘A Test of the Power to Unpardon’ Washington Post (Washington DC, 14 January 2009) <http://www.washingtonpost.com/wp-dyn/content/article/2009/01/13/AR2009011302581.html> accessed 18 September 2018.

72  [2015] VUSC 148.

73  ibid [20(2)].

74  Other strategies for removing the stigma attached to the ex-offender include non-disclosure of a criminal record after a specified period and rules limiting or preventing the disclosure of evidence of ‘bad character’ at the accused’s trial.

75  For example, in the United States a number of states have passed legislation addressing the issue. For details, see generally Brian Murray, ‘Collateral Issues Resource Center’ <http://ccresourcecenter.org/> accessed 18 September 2018.

76  Michael Pinard, ‘Collateral Consequences of Criminal Convictions: Confronting Issues of Race and Dignity’ (2010) 85 New York University Law Journal 457, 529.

77  These are usefully considered, albeit in a mainly US context, by Vivian Chu, ‘The President’s Pardon and Legal Effect on Collateral Consequences’ (Congressional Research Service 2016) <http://www.everycrsreport.com/reports/R44571.html> accessed 16 April 2018.

78  (1866) 71 U.S. (4 Wall.) 333, 380–81.

79  Thompson also notes that the Federal Courts in the USA continue to take the approach that ‘a pardon should generally remove any collateral consequences that may legally attach to a person as a result of the commission or conviction of the pardoned offense’. See Richard M Thompson II, ‘The President’s Pardon Power and Legal Effects on Collateral Consequences’ (Congressional Research Service, 2006) <https://fas.org/sgp/crs/misc/R44571.pdf> accessed 20 April 2018.

80  [1999] 4 NWLR (Pt 599) 476, 495.

81  [2011] ZACC 11.

82  ibid [64].

83  In terms of article 62(c), this does not apply to a candidate for the office of President. In a parliamentary system, it seems to mean that following a pardon, a PEP convicted of corruption can seek re-election to the legislature (or be appointed by the President) and, as a member, be eligible for appointment as a Minister.

84  Section 25(1) provides that the exercise of a pardon under article 133(1)(a) of the Constitution ‘shall not be construed to be an acquittal’.

85  This ‘hybrid’ approach reflects the view of several courts in the USA when dealing with pardons granted to PEPs convicted of corruption-related offences. See Chu (n 77) 6–8. Sebba (n 14, 119–20) notes that civil law systems generally emphasize that a pardon affects only the penalty and does not affect other consequences of a conviction.

86  Chu (n 77) notes that some US states have passed legislation that clarifies the effect of a state pardon. For example, in Colorado, ‘[a] pardon issued by the governor shall waive all collateral consequences associated with each conviction for which the person received a pardon unless the pardon limits the scope of the pardon regarding collateral consequences’: Colo. Rev. Stat. §16-17-103 (2016).

87  This is according to the requirements of the UNCAC, art 31.

88  Article 133(1) of the Constitution of Kenya simply provides that the President may remit ‘all or part of a punishment’. Section 89(2) of the Constitution of Malawi refers to the power to ‘remit sentences’. It is not clear whether the omission of a specific power to remit any ‘forfeiture’ was designed to restrict the presidential power in this respect.

89  Per Ackermann J in National Director of Prosecutions v Mohamed NO [2003] ZACC [15].

90  The case bore the memorable title United States v One Michael Jackson Signed Thriller Jacket and Other Michael Jackson Memorabilia; Real Property located on Sweetwater Mesu Road in Malibu, California; and One 2011 Ferrari 599 GTO. Available at <https://bit.ly/2FdxyYt> accessed 18 September 2018.

91  Obiang agreed to sell a USD30-million mansion located in Malibu, California, a Ferrari automobile and various items of Michael Jackson memorabilia purchased with the proceeds of corruption. Of those proceeds, USD20 million would be given to a charitable organization to be used for the benefit of the people of Equatorial Guinea.

92  Article 54(1) provides that each State Party shall ‘[c]onsider taking such measures as may be necessary to allow confiscation of property without a criminal conviction in cases in which the offender cannot be prosecuted by reason of death, flight or absence in appropriate cases’.

93  John Hatchard, Combating Corruption: Legal Approaches to Supporting Good Governance and Integrity in Africa (Edward Elgar 2014) 297–98.

94  Alexander Hamilton, ‘The Command of the Military and Naval Forces, and the Pardoning Power of the Executive from the New York Packet’ The Federalist Papers: No. 74 (New York, 25 March 1788) <http://avalon.law.yale.edu/18th_century/fed74.asp> accessed 18 September 2018.

95  My thanks to Dr Ernest Ugbejeh of the Nigeria Open University for this information.

96  For a helpful discussion of the case, see David Enweremadu, ‘Nigeria’s Quest to Recover Looted Assets: The Abacha Affair’ (2013) 48(2) Africa Spectrum 51.

97  [2017] 3 LRC 472.

98  Enweremadu (n 96) para 19.

99  Enweremadu (n 96) para 44.

100  [2008] BLR 66 (High Court of Botswana).

101  The British Aerospace saga in the UK is an (unsatisfactory) example of this process. Here, pressure was allegedly brought on the UK government by Saudi Arabia to have an investigation dropped into a major bribery investigation or face the risk of a withdrawal of security and intelligence cooperation. The Prime Minister wrote a ‘Personal Minute’ on the matter to the Attorney-General, and it was later announced that it would not be in the public interest to continue the investigation. An application for judicial review was successful at first instance but failed on appeal. See Colin Nicholls and others, Corruption and Misuse of Public Office (2nd edn, OUP 2011) para 7.208 et seq.

102  Naeemah Clark, ‘The Impact of Mass Communications in America’ (1968) Annals 68 at 73, referred to by Duker (n 8) 504.

103  In essence, reflecting the colonial relationship between the Governor and Executive Council.

104  This may also prove problematic in that in some jurisdictions the permission of the court is required to discontinue a prosecution, eg Constitution of Kenya, art 157(8). It remains to be seen whether a presidential ‘deal’ is viewed as a justifiable reason to do so.