This article explores how conditional amnesties can assist post-conflict societies to recover truth. It examines how such amnesties can be used optimally to achieve the best results as part of transitional justice mechanisms. Thus, a central question is to see how amnesties can be used for truth recovery purposes. For that reason, the status and role of amnesties, and whether such amnesties can be used to learn more about the past and assist in truth recovery is explored. The article explores what amnesties are, how prevalent they are and how amnesties can be used optimally to achieve the best results. An issue that is also explored is whether amnesties are needed for perpetrators to participate in transitional justice mechanisms. The argument that is made, in this regard, is that amnesty is absolutely necessary to persuade perpetrators to testify. If they do not have such legal protection, perpetrators fear the legal consequences that may result if they admit to crimes for which they have not been charged. Another question that is examined concerns whether amnesties, and specifically conditional amnesties, pass international law muster. This article therefore investigates the continual and extensive use of amnesty to determine whether a conditional amnesty violates international law. The article suggests how a conditional amnesty process could be structured and what difficulties such a process should avoid if perpetrators are to enter such a process.
Dealing with the past is now a common occurrence in societies where there has been conflict.1 To deal with the past, such states use a range of processes and tools to try and ensure long-term peace and stability.2 The pillars of a transitional justice process are seen to include justice, truth, reparations, reconciliation and guarantees of non-repetition.3 However, these pillars are not used evenly. Some argue that such societies often go for the supposedly softer options, such as truth recovery methodologies, rather than establish accountability processes to hold people that have committed crimes responsible. Yet, dealing with the past is also necessarily about creating accountability.4 It is about ensuring that the past is not repeated. In this sense, accountability ensures that responsibility is taken for what has been done.5 It has been argued that retributive justice processes create accountability.6
Amnesties should therefore not be seen merely as ‘instruments of impunity’, because ‘amnesties can, at times, instead be seen as important institutions in the governance of mercy, the reassertion of state sovereignty and … the return of law to a previously lawless domain.’10 An amnesty can also have other benefits including truth recovery,11 if a conditional amnesty process is used. Thus, the question of the validity and benefits of amnesty must not be evaluated in isolation.
The traditional concept of amnesty as ‘amnesia’ is increasingly becoming out-dated. Instead, states are finding innovative ways to address past crimes without burying the truth or enforcing widespread retributive justice.9
These processes can play important roles and have a dramatic impact on how much truth can be learnt. At the same time, there have been important recent developments in recognizing the right to the truth at international level. While states often shred documents or adopt practices to limit what emerges about what occurred in the past, individual testimonies in exchange for either an amnesty or to secure employment by the state or government can be important ways in which the past can be resurrected and narrated.
Noncriminal sanctions, such as purges, lustration, and public access to security files, are a critical piece of transitional justice programs and have been featured in one combination or another, in almost every transitional justice case, yet they continue to get short shrift in the research literature.12
Perpetrators often have more truth to tell than anyone about the past. However, they fear prosecution and are therefore reluctant to participate in such processes without some form of protection from legal liability. Thus, an amnesty, as well as other tools, can be helpful in truth recovery processes. This is because perpetrators do not usually participate in such processes due to their concerns about what the legal effect on them will be if they participate and information about what they have done emerges. For that reason, some criticise truth commissions for amplifying victims’ voices over perpetrators and other actors.13 This approach argues that truth commission narratives are skewed towards victims’ memory. This alleged imbalance may reverse the power dynamics between victim and perpetrator. Whether or not this is true, and whether that is important in such practices, the fact remains that perpetrators have little role in such procedures. That reduces the role and effectiveness of these mechanisms, as well as having an effect on processes between perpetrator and victim narratives that may promote peace and stability.14 Not having perpetrators participating, particularly when truth is being sought, reduces the information that is forthcoming as the input that they could bring is excluded. The critical issue is that perpetrators do not come forward to these mechanisms unless they are given some type of immunity or amnesty. The only case where this has been disputed is Sierra Leone, which this article will examine later.
Thus, a central question, particularly in post-conflict societies, is to see how amnesties can be used for truth recovery purposes. For that reason, the status and role of amnesties, and whether such amnesties can be used to learn more about the past and assist in truth recovery is explored. The article explores what amnesties are, how prevalent they are and how amnesties can be used optimally to achieve the best results. An issue that is also explored is whether amnesties are needed for perpetrators to participate in transitional justice mechanisms. The argument that is made, in this regard, is that amnesty is absolutely necessary to persuade perpetrators to testify. If they do not have it, perpetrators fear the legal consequences that may result if they admit to crimes for which they have not been charged. This article therefore examines one case – Sierra Leone – because it has been alleged that the events in this country indicate that amnesty is not needed for perpetrators to come forward and participate in such processes. This seems to be the only country where such a claim is made.
Another question that is examined concerns whether amnesties, and specifically conditional amnesties, pass international law muster. This article therefore explores the continual and extensive use of amnesty to determine whether such processes have been accepted as violating international law. In order to do so, the status of amnesty today and when amnesties can be employed will be examined, because of its legal significance as well as when it can be used to attain benefits for the society and specific victims. Recommendations are made on how such a process could be structured and what problems it should avoid.
2 Obtaining More Truth in Processes of Truth Recovery
Today, the right to the truth15 is accepted almost universally as one of the newest norms of international law.16 It has been specifically incorporated into newer international treaties, such as the 2006 International Convention for the Protection of All Persons from Enforced Disappearance. Oversight bodies of international human rights instruments have more and more often given greater acknowledgement to this right.17 As a result, processes of dealing with the past now recognise that determining what occurred is both a legal necessity and an essential need of victims. Truth recovery is thus a much sought after item in processes to deal with the past. This is especially true for victims if a loved one was killed, disappeared or went missing.
Despite this, the amount of specific truth, on certain events, obtained in transitional processes is often limited. Generally, it is victims who are relied on for truth recovery purposes in processes such as truth commissions. To some degree, their knowledge is useful in constructing a picture of the past. However, the view from victims is usually limited to narrating what occurred to them specifically. They do not have much detail outside of the specific event in which they were involved, and sometimes that might not even be the case. Additionally, there may be no survivors who can testify to what occurred. There is often an absence of information concerning where a victim has been killed. Nonetheless, from victim testimonies, patterns of perpetrator conduct and motivations can be discerned. However, this source is imperfect and very limited in what it usually reveals. Much more could be learned from direct testimonies from those who perpetrated the violations on a range of critical issues. For instance, past perpetrators could disclose who planned, ordered, was involved, prepared, provided information, etc. on specific violations, but also on the way the system worked in general. These matters often remain unknown or only known to very limited degrees. The knowledge that could be revealed through such direct perpetrator narratives includes the motivations of perpetrators, the causes of violence, command structures, who gave what orders, from whence those orders emanated, the connection between security officials and government personnel, and more. In many circumstances, perpetrators are often the only ones alive that have certain information. In other words, there are many cases where it is largely those that perpetrated the harm that have the information about what occurred. There are however major disincentives for perpetrators to give up that information. They are usually too reluctant to say where a victim is buried as they may face criminal or civil accountability. They are concerned that if they reveal the information about what they have done the community will shame them. In this regard, one of the functions of Truth and Reparation Commissions (trcs) is to establish the identities of perpetrators. However, this has occurred to a limited degree. Usually, while trcs collect names, they do not always publish them, and certainly not many of them, for a variety of reasons. This means there is less accountability and less effect on impunity. However, a conditional amnesty where names are made public has the effect of reducing impunity. Yet the problem cannot be resolved, as without amnesty perpetrators have few incentives to tell the whole truth. Therefore, much more information could become available if perpetrators were incentivized to tell the truth. This can be done if they are given amnesty in exchange for truth. Amnesties can thus benefit truth recovery. As has been noted, ‘political opponents may provide the commission with necessary cooperation only if they are assured that they, and those they care about, will receive amnesty.’18 Perpetrators would be more likely to come forward and give information, as they would not fear prosecution because of what they have revealed. However, it has been noted that amnesties violate the right to the truth. In this regard, the Inter-American Court is the institution that has evaluated amnesties more than most others, and it has generally been opposed to such arrangements.19 The Inter-American Commission in the case Massacre of the Jesuits in El Salvador (1999) found that ‘application of the amnesty decree eliminated the possibility of undertaking any further judicial investigations through the courts to establish the truth and it denied the right of the victims, their relatives and society as a whole to know the truth.’20 Additionally, the ‘so-called self-amnesties are, in sum, an inadmissible affront to the right to truth and the right to justice (starting with the very access to justice).’21
However, when finding out the truth, the question to be asked is whether getting perpetrators to testify by giving them amnesty benefits accountability more than prosecuting all of them. This does not mean that a conditional amnesty means that perpetrators do not get prosecuted. As will be discussed below the efficacy of such a process is dependent on prosecutions of some perpetrators. It is necessary at the beginning of the process to shepherd people into the process and certainly necessary at the end of the process for some of them who have not applied for amnesty. As has been noted truth commissions have more positive effects when they are used in amalgamation with trials and amnesties.22 This is not to argue that all perpetrators ought to be, or can be prosecuted. There are however very few cases of countries in the wake of mass atrocity prosecuting most or even many perpetrators. Those examples include Rwanda where prosecutions were more political in nature and only directed at those on one side of the conflict. While those responsible for the genocide were targeted, others who committed serious violations on the other side were not held responsible. Victor’s justice enhances the notion that prosecutions are not for accountability but for political reasons, which in turn enhances the likelihood of future violence and conflict. No doubt an amnesty will affect levels of impunity; but a conditional accountable amnesty should play a beneficial role in this regard. This is particularly true if, as usual, few perpetrators are prosecuted and a conditional amnesty learns names and details of what was done by whom. In these cases there is some form of accountability and some form of justice as these people are named and shamed and possibly excluded from state service as a result of a lustration or vetting process.
3 Understanding Amnesties
Amnesty has been defined as any ‘extraordinary legal measure whose primary function is to remove the prospect and consequences of criminal liability for designated individuals or classes of persons in respect of designated types of offenses irrespective of whether the persons concerned have been tried for such offenses in a court of law.’23
Although amnesty is regularly called different things, it is important to note that there is a conceptual difference between, for example, pardons and amnesty. Pardons are usually individually granted post-conviction, while amnesties are usually for groups of people prior to conviction. Therefore, while they are not the same, the effect is similar since it serves to excuse criminal conduct, either before or after prosecution and conviction. Amnesties do overlap with pardons in the sense that they excuse illegal conduct. They do however differ conceptually as pardons are typically granted to people after a conviction, while an amnesty is usually for groups of people before they have been convicted of a crime.
There are various types of amnesties including blanket amnesties, limited amnesties, and conditional amnesties. Some have argued that self-amnesties ought to be in a category on their own.24 However, all three types of amnesty could be self-amnesties, so they ought not to be a category on its own as it covers all types of possible amnesties.
The blanket variety of amnesty is, as the name suggests, all encompassing. It usually covers all crimes over a specific period of time but there are variations according to the location and the way the amnesty is structured. A limited amnesty is a process that is not a full amnesty for all crimes that have been committed, but one that is narrowed by type of crime, by who committed them or the time period in which the crimes were effected. Thus, in a limited amnesty only certain types of crimes may be included. Such an amnesty may cover crimes that were committed in a specified time period, by persons in general, or from a specific grouping. For example, the law might cover only state security officials. In turn, a conditional amnesty provides that amnesty is only provided if the applicant meets certain conditions. Again, the amnesty may apply to only some crimes (often related to tax or some other specific issue) or linked to a specific historical event or process in the country.
In the context of dealing with past human rights violations, amnesties are commonly found in societies that are transitioning to democracy.28 However, amnesties dealing with politically motivated crimes are also found in societies where there has not been a transition. Such amnesties are in put place to offer the supporters of the regime protection for the future. Thus, amnesties can be used by a range of societies, regardless of whether they are democratic or not, and irrespective of whether they are used for legitimate purposes or not. The type of amnesty for politically motivated crimes adopted by a country frequently reflects the type of change that that specific state has undergone.29 For instance, an amnesty can be approved by a dictatorial regime that wants to excuse the conduct of its own officials, perhaps before a new government comes into power. These types of amnesties are sometimes overturned later because the new government views such a process as illegitimate. However, these types of processes are quite often part of peace agreements wherein it has been agreed that officials on both sides of the conflict will be granted amnesty. It is often the self-amnesty process that is seen to be most problematic. However, a negotiated agreement is less likely to bring about a blanket amnesty. There is however less likelihood of an amnesty where there has been an overthrow of the previous regime either by force or due to the results of an election. A blanket amnesty is more likely where self-amnesty occurs. It is also likely to promote impunity.30
Have been used to express public grace and forgiveness, and to further government corruption and oppression. They have been used to bring law into compliance with an accepted reality, and to exempt a contested reality from public scrutiny and moral and legal accountability. They have been granted at times of great social stability and at times of great social unrest; at the start of and during wars for the purpose of recruiting troops, and at the end of wars to foster peace and reconciliation.27
The question about who gets selected for prosecution ought to be a fundamental question linked to issues of amnesty. The number of people who are prosecuted after mass violations obviously depends on the number of crimes committed and the capacity of that criminal justice system to exercise jurisdiction in such cases. Very few countries have attempted to prosecute thousands of perpetrators, the major exception being Rwanda, but in that case there was a political purpose for doing so. Thus, in cases where there are many that are alleged to have participated in the perpetration of crimes, only some can be held accountable as going after everyone will have major negative effects in the country. At the very least, this includes the economic cost of bringing everyone, or many to book. For that reason, those most responsible ought to be especially targeted but also a few of those lower down on the chain of command to make it clear that everyone is liable for prosecution, and that no one is above the law. However, it is vital that the process is not seen to be victor’s justice and people from all groups who have committed crimes are sought out. Thus, it has been argued that there is ‘practical, legal, and moral justification for dealing with lesser offenders through truth commissions and conditional amnesties, whereas the persons most responsible – i.e., planners, leaders, and those committing the most notorious crimes – should still be held criminally accountable.’31
While the applicability of an amnesty is domestic in nature, and only applies in the state that grants it, an amnesty, hypothetically at least, must be compliant with international law.32 Even though international law is generally clear about what is permissible, in reality many countries ignore what it states, and adopt laws that are not compatible. They do this because the scope to challenge those laws can be limited by where they are in the world. For instance, while overbroad amnesties in Latin-America are usually challenged successfully at the Inter-American Court, in other parts of the world the possibility to challenge such laws remains much more difficult. In this regard, the issue of compliance of amnesties with international law has been extensively debated and studied over the last few decades,33 in part, because amnesties continue to be used significantly.34
One of the most well-known conditional amnesties cases was in South Africa in the mid-1990s. It is the only example of a trc exercising an amnesty power by itself. It emerged from a negotiated agreement. The conditional variety was adopted, as a blanket amnesty was not something that many political parties agreed to. But in fact, not having a blanket amnesty was a major benefit for the society. It meant that only those who applied and met the conditions set out in the legislation obtained amnesty. This meant that the law determined and restricted those who were able to get amnesty. Furthermore, it meant that unlike blanket amnesties, the names of those who got amnesty was known. Additionally, their crimes were revealed, as a person only got amnesty for what they applied for. Crucially, only crimes that were committed with a political objective were amnestied. However, relatively few of the total number of potential applicants applied for the South African amnesty. There were about 7000 applicants, but there should have been many more. Further, about 4500 of the total were not real applicants, as they had mostly not committed political offences. This is dealt with further below. There were multiple contingencies: most people came from the liberation movements, there was collusion between applicants, the process took a long time, and there were problems in verifying the information. Moreover, if the process was to work optimally, the States’ prosecuting authority and the truth commission’s amnesty committee and investigators had to work together in a cooperative way, which did not happen. After all, why would someone apply for amnesty if they do not believe they need it? Perpetrators need to believe that they face the real chance of prosecution if they do not apply. Thus, the state needs to show ability, capacity and will to prosecute. The carrot of amnesty needs to have the stick of prosecution as a real threat hanging over perpetrators. South Africa was not able to fully implement this ideal. The fact that no other country has used a similar means indicates that the model is somewhat tarnished. The issue about how the problematic, bureaucratic process can be dealt with by others, who want to use a conditional amnesty, is dealt with below.
4 The Extent to Which Amnesties are Used around the World
Amnesties have been used extensively for centuries for a variety of reasons,35 and are still widely used by states.36 In fact, between 1945 and 2011, there were 537 amnesty processes in 127 countries. Of these 537 amnesties, 398 of them were in the period after 1979.37 The acceptance of using amnesty is also reflected in the fact that the constitutions of 186 countries refer to either amnesty or pardon. The seven states, which do not mention amnesty or pardons, are Andorra, Australia, Bosnia and Herzegovina, Canada, Syria, Saudi Arabia and Yemen.38 Only 7 states specifically forbid amnesties for certain serious crimes.39 Thus, states are still adopting amnesties as a common practice.40 Various studies show how common amnesties are.41 In terms of amnesties being adopted or in the process of being adopted, in just 2016, countries that have adopted amnesties for political related crimes include Colombia,42 Tunisia,43 Iraq, the Philippines, Ukraine,44 Uzbekistan45 and Venezuela.46 Many others adopted amnesties for tax and other types of non-political crimes.
First, it must be democratic in its creation. … The general involvement of the public and the involvement of more than one branch of a democratic government are two important indicators of the democratic nature of an amnesty. Secondly, it must not apply to those most responsible for war crimes, crimes against humanity, and other serious violations of international criminal law. Third, it must impose some form of public procedure or accountability on its recipients. … Fourth, it must provide an opportunity for victims to question and challenge an individual’s claim to amnesty. Fifth, it must provide some concrete benefit, usually in the form of reparations, to victims. Such a benefit could come either from the beneficiary or from the state. Sixth, … it must be designed to facilitate a transition to a more human rights friendly regime, or as part of a comprehensive program of reconciliation aimed at addressing long-standing and serious societal tensions and injustices.48
5 The Legality of Amnesties
In recent years amnesties have been extensively criticised. Those who review these processes from an academic vantage point usually frown upon states using an amnesty after periods of conflict and human rights abuse.49 Such a process is even more criticised if such mechanism grants amnesty for crimes deemed to be international crimes. Serious international crimes, on the other hand, are ones which states are duty-bound to prosecute (or to extradite the offenders).50
Amnesties are often used for conflict resolution.51 They have been common parts of peace agreements. Various actors such as the United Nations (un) have been often part of such processes. While the un was not always in the past against amnesties,52 today it is generally critical of amnesties.53 This is especially true for granting amnesty for serious international crimes. The un has recently become more critical of amnesties that provide for sexual violations.54 The un has stated that amnesties are ‘inconsistent with States’ obligations under various widely ratified treaties as well as un policy, and may also be inconsistent with emerging principles of customary law.’55 It has also stated that there should be an insistence that ‘peace agreements not grant amnesties for war crimes, crimes against humanity, genocide and gross violations of human rights.’56
However, the issue of amnesties in peace agreements, and their scrutiny during the process of these agreements being drafted, remains a challenge.57 The status of amnesties in peace agreements has often seen great latitude, and enormous discretion being granted to those drafters on the basis that peace is more important than justice.58 However, there is some difference of opinion about the effect of amnesties on such societies. Some see their positive effects,59 while others think that some societies that have used amnesties may be less likely than others to relapse into conflict at some point in the future.60 One position is that the nature of the state makes a difference to whether relapses are more or less likely. However, greater leeway is certainly given to peace agreement authors than to the legal creators of amnesty laws in states.61 Often amnesties are negotiated and agreed to in peace processes, and only later are those arrangements criticised.62 There are pragmatic reasons for doing so as ending the war is seen to be important at a particular point in time and those types of problems can be raised later. There are some recent examples where not as much autonomy as previously has been given to have wide amnesty processes.63
As noted above, many countries use amnesties. It is therefore difficult to make the argument that customary international law does not forbid amnesties in general.64 It is also difficult to do so because a range of states that mediate conflicts discuss, agree, and incorporate amnesties into peace agreements, which denotes their viewpoint that amnesties in general do not breach customary international law.65 Even though various human rights systems often agree that amnesty for certain crimes, such as war crimes, crimes against humanity, genocide and torture, is unlawful,66 many contemporary amnesties do not eliminate such crimes from the scope of such amnesties. In this sense, Afghanistan adopted a general amnesty law in 2007 and which entered into force in 2009, amnestying all crimes, even those that are usually excluded because they violate the dictates of international law.67
While many international treaties are understood to outlaw amnesty, and rather promote compliance with the duty of states to prosecute, there are still questions over which amnesties are forbidden.68 In fact, some treaties, such as the Rome Statute of the International Criminal Court (icc),69 have no provision on amnesty.70 It could be argued that not having a provision on amnesty implicitly means that the icc will not recognise an amnesty and its effects. It could mean, that as the Statute gives the Court jurisdiction if a state cannot or will not deal with a particular matter, an amnesty means that those states fall foul of those provisions.71 In other words, an amnesty in such a state could mean that the icc has jurisdiction to step in, because the state lacks the will to deal with international crimes.72 Hence, the icc will not be bound by an amnesty, unconditional or not, and may choose to examine the situation and prosecute those alleged to be most responsible.
There is however a specific provision in international law that provides for amnesty, although what it means and when it ought to be applied remain controversial. Article 6(5) of Protocol ii to the Geneva Conventions of 1949 explicitly obligates states to grant amnesties upon the conclusion of hostilities in internal armed conflicts.73
A major international law issue therefore is what amnesties are permissible.74 This remains somewhat uncertain. A specific variety of amnesty that may be acceptable and legally valid are types of conditional amnesties.75 A range of experts have argued that while blanket amnesties generally violate the law, there are incidents where amnesties would not do so depending on their scope and the crimes they cover.
Martha Minow has thus stated that, ‘conditional amnesties do not foreclose truth-seeking, but instead promotes it.’76 Juan Mendez argues that conditional amnesties would be valid, unless they include ‘war crimes, crimes against humanity (including disappearances), or torture.’77 In this regard, the Extraordinary Chambers in the Courts of Cambodia decided that: ‘Certain conditional amnesties such as those providing for some form of accountability have also met widespread approval.’78 The South African process of using a conditional amnesty to deal with crimes committed during the Apartheid era is usually referred to when deliberating on the matter of the compatibility of conditional amnesties with international law.79 A range of experts have argued that the South African version of conditional amnesty meets international lawfulness.80 Sadat, for example, has argued that: ‘conditional amnesties, such as those granted in the South African case, are of a different kind, and would not appear to threaten accountability, but to bring it about using a variation of the criminal process.’ She also argued that this could have been different if a blanket amnesty had been adopted.81 Grace Fiddler has noted that while some international laws forbid or limit the scope of some amnesties, she finds that there is no prohibition on conditional amnesties if a process is established to investigate the crimes committed.82
The Guidelines note that although conditional amnesties could see a reduced number of culprits participating in such an arrangement, if conditions are inserted into such an amnesty it could increase the legality and legitimacy of it. The Guidelines therefore find that a conditional amnesty is more likely to comply with international law.84
Amnesties are more likely to be viewed as legitimate where they are primarily designed to create institutional and security conditions for the sustainable protection of human rights, and require individual offenders to engage with measures to ensure truth, accountability and reparations.83
Those that apply within such amnesty arrangements also indicate their willingness to comply with the conditions. In other words, it indicates their willingness to be legally compliant and that therefore impunity will be affected. As far as truth recovery benefits are concerned, such processes need however to make better efforts to ensure that more full and accurate information is collected from those that apply to ensure that victims’ rights to truth are enhanced. McEvoy and Mallinder therefore state that, ‘a lawful amnesty which requires the performance of certain obligations (…) may in fact be preferable to de facto impunity where the vast bulk of perpetrators are untouched by any legal process.’88
[…] attempted to discredit and transform a culture of impunity into one of transparency and truth-telling by making it preferable for a perpetrator to tell the truth rather than cover-up or lie about their human rights violations. No longer are people encouraged to ‘look the other way’ when gross injustices occur. Instead, people are now encouraged to denounce such actions as immoral and impermissible in their society.87
6 Previous Conditional Amnesties
As noted above, not all amnesties are the same. What seems to be the case is that of the different types of amnesties in post-conflict societies, conditional amnesties seem to be the most acceptable. Mallinder argues that there are various categories of conditional amnesties, such as those for ‘surrendering and disarming; repenting and providing information on comrades; fulfilling the conditions within prescribed time limits; telling the truth; repairing the harm; participating in community-based justice mechanisms; and submitting to lustration and vetting procedures.’89
Various countries besides South Africa have used conditional amnesties. In 1974, us President Gerald Ford issued a decree that provided amnesty to Vietnam draft dodgers and military deserters if they confirmed their allegiance to the country and served two years in the public service.90 While more than 200,000 men were charged for draft dodging, many more avoided being charged because of the decree. President Assad of Syria announced a similar type of conditional amnesty in 2015.91 Armenia also has a conditional amnesty for such persons. It was legislated in 2004 and has been extended a number of times since then. Draft dodgers have to pay a fine to receive amnesty that is higher the longer the person has been absent. 5000 men, who have entered into the arrangement with the state, have paid a fine up to $5000.92 Another case was when the Palestinian National Authority in 2001 gave amnesty to collaborators with Israel if they applied voluntarily and revealed how they cooperated with Israel.93 Algeria’s 1999 Law on National Reconciliation gave amnesty on the condition the applicant did not continue terrorism or other types of similar behaviour.94 The Algerian 2005 Charter for Peace and National Reconciliation demanded that those who wanted amnesty had to apply in a specific time period, giving information about what they had done, and agree not to continue with their violent activities.95 In Northern Ireland in 1998, amnesty was given to people who agreed to not return to their former organisations or threaten the public in any way. There have also been a number of amnesties where there was a condition that hostages were released such as in Fiji in 2000. South Korea in the 1980s demanded that those seeking to benefit from an amnesty had to renounce their beliefs in communism. In Ethiopia, re-education was deemed necessary to qualify for amnesty. In Turkey, those to be amnestied had to name and give the location of others in their organisations.96 In Mozambique in 1987, amnesty was granted to those who voluntarily gave themselves up.97
Very common are conditional amnesties for weapons that are handed over.98 For example, the Uganda Amnesty Act of 2000 offered amnesty to those who had been involved in warfare or rebellion against the state at any time from 1986, on condition that they renounced such activities and handed in their weapons.99
A number of trcs have been given powers, in the wake of the South African process, to recommend amnesty.100 These have been conditional amnesties that allow such trcs to recommend to the government of the countries that they are in that certain perpetrators who have met certain conditions should be granted amnesty. Generally speaking these processes have not been very successful and perpetrators have not been willing to trust them enough to apply to them. They have not engendered sufficient certainty to tempt such perpetrators who have when weighting the pros and cons decided that there were too many risks.
The issue of using conditional amnesties remains on the agenda of those wanting to adopt an amnesty in post-conflict states. A variety of proposals suggest that conditional amnesties are to be used. One such proposal was to use a conditional amnesty in Iraq,101 and in Uganda.102 A further proposal is for it to be used in a post-conflict Syria. In this regard, Grace Fiddler has argued that: ‘the implementation of the South African model would be beneficial to the Syrian community.’103
7 Potential Benefits of Conditional Amnesties for Transitional Justice Processes
While amnesty has continually been seen as an obstacle to truth recovery, it can be argued that amnesty can be a tool for truth recovery if used appropriately. This is not to argue that amnesties should always be used for this purpose, or used for all crimes. Rather the argument is that if there will be an amnesty anyway, for example when it is mandated in a peace agreement, there is no reason not to use it in the most productive way. Or if, for whatever reason, there will be few prosecutions, why not maximise the possibility that an amnesty might benefit the society which without it would not benefit from knowing at least what crimes were committed. In this way the right to truth may be enhanced often at little cost. While amnesties might be controversial and their status unsettled,104 a structured conditional amnesty is not the same as a blanket amnesty. A benefit of a conditional amnesty is that amnesty is individualised; on the other hand, blanket amnesties are collective. Thus, much is gained from a conditional process where individuals are specifically identified and what they have done becomes known. Blanket amnesties ensure that those who have qualified for the amnesty and what they have done remains shrouded in mystery. Henrard has therefore stated that blanket amnesties lead to ‘bitterness and deep resentment, thus blocking reconciliation and healing.’105 Therefore, while ‘blanket or unconditional amnesties would not be acceptable, something can be said for discrete amnesties, conditional on the complete disclosure of the atrocities committed. Discrete amnesties furthermore do not preclude the possibility of prosecution for those who chose not to apply for amnesty or whose amnesty applications were refused. Such balanced combination could also achieve the reconciliation between ethical imperatives (the demands of justice) and political constraints.’106 Similarly, it has been argued that: ‘A conditional amnesty, then, can meet peacekeeping, nation-building, and reconciliation objectives of amnesty as well as the requirements of international accountability.’107
Moreover, a conditional amnesty can hold other benefits beyond finding truth. Certainly, finding truth can advance national healing and reconciliation.108 It advances processes of nation building and can assist in creating a cohesive society.109 However, truth needs to be seen as a commodity that should be linked to lustration processes. These processes can also play a truth telling role.110 As Horne has noted, lustration in eastern European countries enforced an evaluation of the past and to some extent composed a reckoning with it through processes of voluntary admissions and, sometimes, involuntary public confessions.111 Thus, lustration processes can be usefully added to processes of amnesties or when they are not available as a means to obtain truth.112 These should not be stand-alone processes but used in conjunction with other transitional justice mechanisms. Where truth is not revealed, or if the information is not accurate, the person should be denied the position they held or seek. There should be a signed acknowledgement by the employee that if it is found later that lies were told by them concerning their past role, the person accepts they will be retrospectively dismissed, and the salary they earned returned. Thus, prospective processes ought to be established to ensure that less lies are told and that there is means to hold people accountable for their least than honest admissions. While there are dangers of using such processes and their human rights implications can be severe, the European Court of Human Rights has upheld their use.113
8 Future Conditional Amnesty Process Procedures
Ensuring that perpetrators make use of conditional amnesties processes is dependent on solid investigations, cooperation between state institutions rather than competition, and committed prosecutors who have the capacity and willingness to pursue those who do not apply for amnesty. Processes between trcs and state prosecutors have not seen good relations. Both the South African and Sierra Leone process saw massive tensions as well as little cooperation and even attempts to undermine each other.114 In Sierra Leone, the idea of having both a court and trc operating alongside each other was a useful one, and one that ought to be replicated elsewhere. However, the trc and Special Court (sc) were involved in a dispute about whether the trc could hold interviews with the detainees held in the custody of the sc. After lots of wrangling, the trc closed its doors and the detainees were not able to testify even though they wished to.
To make sure perpetrators apply for such an amnesty, the process has to ensure that they do apply and their applications achieve what the process sets out to achieve. Therefore, for a conditional amnesty to be viable for truth recovery purposes, there needs to be at least two parts to it. The first part has to ensure perpetrator participation, while the second has to guarantee that participants provide the real and full truth. The first part has to make sure that applicants believe that they need amnesty. Without such a need why would perpetrators participate in a process? That was a failing of the South African process. In fact, many offenders in South Africa did not participate in the amnesty process because they believed they already had amnesty, immunity, a pardon or a range of other acts, which would allow them to escape future accountability. Where this is the case, an amnesty process should revoke all old amnesties, of whatever description or name, in order to prevent perpetrators from escaping future liability. This was done in Sierra Leone by a clause in the Special Court statute that revoked a prior amnesty.115 Such a revocation ensures that perpetrators are liable for prosecution. Hence, in essence, a necessary part of a workable conditional amnesty is to ensure that perpetrators actually need amnesty. Thus, it is necessary that a dedicated prosecuting unit begins to investigate and starts the process of prosecuting perpetrators as soon as possible. This will indicate state commitment, dedication and the necessary political will and means to do so. This again was a failing of the South African process that left this task to the national prosecuting authority. During the apartheid era, prosecuting authority personnel were reluctant to do anything significant for political reasons. Later, when more progressive people were installed in those positions, the national prosecuting authority was once again reluctant to pursue such cases. This was the case because they feared fallout that may have resulted if only people from the former regime were prosecuted and not people from the new ruling party: the African National Congress (anc). Thus, political will has been absent all the way through since 1994 although for different reasons. The second necessary part of an amnesty process, for it to be viable, is to ensure that perpetrators tell the real and full truth. This was another failing of the South African process. A conditional process ought to introduce revocation and perjury clauses. On revocation, this would allow a panel to withdraw an amnesty if it found that the substantial truth was not revealed or that perpetrators cooperated to hide the truth. A perjury clause with severe penalties, including imprisonment, should also be included to try and prevent lies from being told.
Part of the problem with the South African conditional amnesty process was that there were potential applicants who may have wanted to apply for amnesty but did not trust the process fully.116 Trust is critically needed if perpetrators are to apply and brave the system. If there is distrust or uncertainty that the outcomes will be positive for those who apply, there will be reluctance by potential applicants to participate in the process. In South Africa, there was a great deal of uncertainty about how the process would work and thus whether, or which, perpetrators would get amnesty, if they applied. It was unclear whether the process would unfair and biased in its application, and whether the amnesty judges could be trusted. There was some faith, in some quarters, as the judges came from the usual judiciary, which despite their composition, or maybe because of it, inspired some confidence in those from the security force sector. However, the fact that perpetrators had to appear in public, where victims could be present and object to what they were saying and could disagree whether applicants ought to be granted amnesty were other reasons why there was hesitation in participating.
The South African amnesty process had a variety of operational difficulties. The process was cumbersome, bureaucratic, slow and costly. It was costly, as there were eventually 19 judges,117 as well as offices and staff that ate up resources for the six-year period while the amnesty process ran its course. The process dealt with approximately 7000 people, of whom approximately 4500 were simply dismissed on their written applications only because they did not meet the criteria on paper. This chamber process did not take that long. The legislation however demanded a public hearing of the 2500 that had committed what were termed ‘gross human rights violations’ – and those took a lot of time.
Another failing of the South African process was the fact that many perpetrators were able to undermine the system due to the inability of the trc to really investigate the truth that was being told by perpetrators. By the time the South African amnesty process began their public hearings, the rest of the trc had been closed down and few investigators were still employed. When it was still in operation, the trc’s Investigative Unit was the largest of all trc’s that had existed around the world, but still it could not cope with all the investigations needed. This was the case in spite of the international assistance it received. Thus, whether the truth was being told was often uncertain. Indeed, on many occasions this did not occur as there was no way to vet much of the testimony.
The fact that there were different amnesty panels also made coordination problematic as different perpetrators appeared before different panels and judges. Thus, if such a conditional process is to be repeated elsewhere, for it to work optimally then a large investigation unit is needed with the time and resources to adequately be able to verify the veracity of what is being told in the applications. However, the obstacles dealt with above and the costs associated with such a bureaucratic amnesty process make it likely that a conditional amnesty process like the South African process will not be repeated. Elsewhere, a conditional amnesty is likely to be viable if it is simple and if it will ensure that perpetrators apply. Thus, a far more pragmatic approach would be for the process to be designed in such a way to ensure that first, perpetrators apply for amnesty, and second, that if perpetrators meet the simplified guidelines, they get amnesty. There should be very few criteria, and they should be simple to understand. The South African process had too many uncertain criteria, and it relied on a process that took too long. The outcome of applications was far from certain and that uncertainty surely put off some potential applicants. The criteria therefore ought to be simply – whether a crime was committed; was it perpetrated for a political purpose; and have the details about what occurred been fully supplied. The aim should be to ensure that perpetrators see it as simple and straightforward. At the same time, the legislation should ensure the establishment of a dedicated investigative and prosecution unit to pursue those who do not apply for amnesty. Thus, investigating potential people for indictment should not be left to vagaries of the normal prosecuting authority with the usual political processes that accompany them.
The application process should see a form that is full and detailed yet simple. It should have clear straightforward questions. Those administering the process should be able to request more information and clarifications where necessary. Questions ought to be specific and specifically aimed at getting information about the fate and whereabouts of victims. This is essential. Victims will not benefit if such specific information is not obtained. The form should be in the form of blocks that need to be filled in and the form should be an affidavit that applicants have to complete testifying on pain of prosecution that they are telling the truth. The legislation setting up the process, and the form that needs to be completed, should make it clear that any untruths brought to the attention of the authorities at any point in the future would lead to the withdrawal of the amnesty, in addition to severe criminal sanctions. The completed forms ought to be generally available and accessible to the public so as to get the truth out and to prevent lies from being told. It should not be a requirement for applicants to appear in public, but applicants ought to be requested and implored to do so, for reconciliation processes. The first major goal of the process ought to be truth recovery. A secondary objective, where possible, could be reconciliation. There ought to be no requirement of remorse,118 repentance or an apology.119 Again, it ought to be requested and implored.120 It would be best if done in public in order to allow it to impact the reconciliation process.
9 Do Perpetrators Need Amnesty before They Participate in Truth Processes?
Contrary to what Schabas argues, the trc, in its report, notes that:
What the Sierra Leonean experience appears to show is that many perpetrators do not need the promise of amnesty in order to come forward and participate in such a Commission.121
It is true in Sierra Leone that most perpetrators did not participate willingly. There were some who were willing to participate in the trc process but they were willing for political reasons. Others were often cajoled into participation. This is dealt with later. In this regard, the trc notes that:
The reality of the Commission’s work was that most perpetrators were not willing to disclose their involvement in atrocities, at least not in advance of a hearing or interview.122
However, some of them did testify because they believed at the time that they had amnesty. In this regard, it was actually only from July 2002, when the Special Court for Sierra Leone (scsl) began to operate, that it really sank in that there was not a complete amnesty for everyone. Indeed, it was only from then on that some perpetrators began to fear that the trc would hand them over to the sc and began to evade the trc investigators. Thus, until the Statute of the scsl formally withdrew the Lomé amnesty, many perpetrators of even the most serious crimes believed that Lomé granted amnesty to them. In fact, one study has found that there was very little actual knowledge about these institutions in the country.124 This would have impacted the extent or not of the amnesty. Regardless, some believed that while that amnesty no longer applied to everyone, it still applied to them. Many others, correctly, also believed they were not likely to be prosecuted, as they were not in the category that was sought by the Special Court. There was also a belief that whatever perpetrators said in trc hearings or interviews could not be used in a prosecution against them.125 Thus, at least some of these perpetrators who appeared before the Commission believed that they had amnesty as a result of the Lomé Peace accord.126 Some believed that they could not or would not be prosecuted for what they had done, and others were given assurances by the trc in regard to their cases not being taken forward.
Persons who played a central role in the conflict, including Government Ministers, faction leaders, high-level commanders and persons accused of grave criminal conduct, appeared in both public and in closed hearings of the trc. These individuals either sought an appearance of their own accord or were requested by the Commission to make an appearance.123
Even though this was changed later when the Special Court Statute came into force, which permitted prosecutions to occur, many perpetrators believed, at least at that time, that they could not be prosecuted. One author in 2001 wrote in a law journal that: ‘allowing the government to participate in the prosecution of the same people it granted amnesty to creates a sense of ‘victor’s justice.”129
After the signing of the present Agreement, the Government of Sierra Leone shall also grant absolute and free pardon and reprieve to all combatants and collaborators in respect of anything done by them in pursuit of their objectives, up to the signing of the present Agreement.128
The trc to get people to appear at hearings, had to compel some individuals, believed to have played important roles, and have committed violations, to appear before it. It issued six subpoenas: five against serving government ministers and various senior officials, including the Attorney General and the chairman and secretary of the ruling political party, and one against a former coup leader.131 The Sierra Leone trc report also notes that:
The Commission finds that there is evidence to support the conclusion that some people were reluctant to participate in the truth-telling process out of fear of prosecution by the Special Court for Sierra Leone… There were certainly other reasons why some perpetrators did not come forward to tell their stories. Some presumably feared reprisal or simply saw no personal advantage to themselves in speaking publicly about their own actions.130
Thus, perpetrators were generally reluctant to appear before the Commission. While there were some who wanted the stage and benefits from a public platform, many did not want to appear and avoided doing so. The reality is that many of the perpetrators who did appear before the Commission did so after having been located and approached by investigators for the Commission. At least some were notified after being found that that they would be compelled to appear before the Commission if they did not do so voluntarily. So, while the Commission notes that: ‘Although the Commission had the power to compel perpetrators to testify under oath, subject to prosecution for perjury in the case of dishonest testimony and for contempt of court in the case of refusal to testify, it did not exercise these’,133 it did use these powers as a weapon. This meant that they did not have to actually use these powers, because the threat of their use was often sufficient to compel the person concerned.
Many role-players invited to hearings did not appear or request a re-scheduling of their appearances, despite being notified well in advance of the dates and times for the hearings.132
It is important to note that many perpetrators who did testify denied participation or denied responsibility for their actions.134 These people were not at hearings voluntarily to admit guilt, but often to explain their roles and benefit from a public hearing. However, as noted by some, perpetrators did testify but the reason why they testified undermines the perspective that they simply came forward to testify. Many were pressurised to testify on pain of a subpoena being used. It is important to note that the trc interviewed many people, mostly confidentially.135 To some extent the fact that it was the Commission that found perpetrators, and got them to testify, is borne out by the fact that, as Shaw notes, in many districts there were no perpetrators who participated in the hearings of the trc.136 As investigators focused more on the high ranking perpetrators, based around the capital city, few perpetrators in the districts were identified or approached and therefore did not appear before the commission.
Thus, it seems that many only agreed to cooperate once the Commission had located them, rather than them approaching the Commission voluntarily. It seems at on a number of occasions, the threat of a subpoena was used, but none was actually issued. This may have been a reason why some of them participated. The interviews conducted by the trc occurred mostly from June 2003 onwards, which is after the scsl indictments. Hence, the interviewees who were sought and found by the Commission could have been told the indictments had named those to be charged and no one else needed to be concerned. The fact is that the threat of subpoena was available against them by the trc, although eventually not used for that purpose.
From a review of the references in the trc report about 250 interviews can be found, although there could have been more. How many of these were perpetrators is unknown. At least some were with people who were not perpetrators but were interviewed to understand the political situation and the role of the state and various political parties. It must however be noted that it seems that most of these interviews occurred during the time that the trc and sc were at loggerheads about sc detainees appearing before the trc.137 However, the notion that many perpetrators participated is undercut by the fact that only 350 witnesses appeared in public at trc hearings,138 many not being perpetrators.
Therefore, in spite of contrary opinion, the Sierra Leone model seems to support the idea that generally speaking perpetrators do not come forward in large numbers without amnesty unless located and pressurised into doing so. They usually come forward under some type of threat or compulsion. Without an investigation process to determine whom the perpetrators are, where they are, and the ability to force them to come out, few will appear. The fact is that in most situations without an amnesty perpetrators are not forthcoming. This is also borne out by the South African model. Many perpetrators who did apply for amnesty knew their cases were known, because of investigations that had already taken place previously. These investigations were largely done before the trc. The trc itself played almost no role in getting perpetrators to apply for amnesty. Few, if any, investigations were carried out before the cut-off date for amnesty with the intention of getting potential applicants to apply for amnesty. The trc had powers of compulsion but did not use them to get applicants to apply for amnesty. It did use section 29 hearings (named from the section of the act that permitted such types of closed compulsory hearings) to compel perpetrators to disclose their crimes at closed hearings where no amnesty was given.
Within the discipline of transitional justice, amnesty is an issue that is usually seen to undermine one of the pillars of transitional justice – justice or accountability. However, justice is only one of its five pillars, the others being truth, reparations, reconciliation, and guarantees of non-repetition.
The right to the truth is now recognised in international law. Amnesty can assist truth recovery if it is granted in exchange for truth. Conditional amnesty processes have the potential to provide answers and provide substantially more truth than presently occurs in processes that almost exclusively rely on victim testimony. This is not to argue that amnesty should be more regularly provided, or that it should be provided for serious international crimes. It should be exceptional rather than the rule.139 Where amnesty is to be granted, it should be conditional and provided in exchange for truth. Victims have a right to the truth and the state should enhance the ability of victims to learn the truth. Amnesty should not be given without perpetrators at least providing information about the acts for which they need amnesty. Through an examination of the South African conditional amnesty process, lessons can be learnt for other amnesty processes where amnesty can be given in exchange for the truth. Understanding perpetrators and their motivations would also enhance the ability to learn about when and how states and individuals decide to perpetrate human rights violations. This may allow recommendations that avoid repetition of such deeds to emerge.
While amnesties are usually determined to be solely about justice, they can affect other matters besides it, including truth.140 However, while being anti-retributive justice in the formal sense, conditional amnesties do not have to be about the absence of justice if there are conditions to granting it. Simply applying for amnesty and public knowledge about whom has applied for what is already a form of accountability. Similarly, public acknowledgment of a role in the commission of the offences and the shaming that comes with it is also a form of accountability. The crux is: rather than no accountability, there are alternative types of accountability to be brought forward not through retributive justice but through other mechanisms including restorative justice. It must be further noted that additional conditions for granting amnesties, depending on what they are, can further promote and enhance accountability.
R Nagy, ‘Transitional Justice as Global Project: Critical Reflections’ (2008) 29 Third World Quarterly 275.
V Nesiah, ‘Transitional Justice Practice: Looking Back, Moving Forward’  Research Report of Impunity Watch 11.
C Bell, ‘Transitional Justice, Interdisciplinarity and the State of the “Field” or “Non-Field”’ (2008) 3 International Journal of Transitional Justice 5.
O Bakiner, ‘Truth Commission Impact: An Assessment of How Commissions Influence Politics and Society’ (2013) 8 International Journal of Transitional Justice 6.
C Murphy, ‘Transitional Justice, Retributive Justice and Accountability for Wrongdoing’, in C Corradetti, et al., Theorizing Transitional Justice 59.
CF Torne, ‘Truth Commissions and the Accountability Relationships They Generate: A New Framework to Evaluate Their Impact’ (2015) 3 Asian Journal of Peacebuilding 233.
H Cobban, Amnesty After Atrocity? (Paradigm Publishers, 2007).
A Neier, War Crimes (Times Books, 1998).
L Mallinder, Amnesty, Human Rights and Political Transitions: Bridging the Peace and Justice Divide (Hart Publishing, 2008) 12.
K McEvoy and L Mallinder, ‘Amnesties in Transition: Punishment, Restoration and the Governance of Mercy’ (2012) 39 Journal of Law and Society 56.
P Naftali, ‘Crafting A “Right to Truth” in International Law: Converging Mobilizations, Diverging Agendas?’ (2016) xiii, Champ Penal 13, <https://champpenal.revues.org/9245>.
N Kritz, ‘Policy Implications of Empirical Research on Transitional Justice,’ in H Van der Merwe, V Baxter and AR Chapman, Assessing the Impact of Transitional Justice (United States Institute of Peace Press 2009) 13, 19.
NV Nwogu, ‘When and Why it Started: Deconstructing Victim-Centered Truth Commissions in the Context of Ethnicity-Based Conflict’ (2010) 4 International Journal of Transitional Justice 275.
P Gobodo-Madikizela, A Human Being Died That Night (Houghton Mifflin, 2003) 125.
Y Naqvi, ‘The Right to the Truth in International Law: Fact or Fiction?’ (2006) 88 International Review of the Red Cross 245.
D Groome, ‘The Right to Truth in the Fight Against Impunity’ (2017) 29 Berkeley Journal of International Law 175.
V Vriezen, Amnesty Justified? The Need for a Case by Case Approach in the Interest of Human Rights (Intersentia, 2012).
K Greenawalt, ‘Amnesty’s Justice’ in RI Rotberg and D Thompson, Truth v. Justice: The Morality of Truth Commissions (Princeton University Press, 2000) 189.
C Binder, ‘The Prohibition of Amnesties by the Inter-American Court of Human Rights’ (2011) 12 German Law Journal 1203.
Ignacio Ellacuria et al. v. El Salvador, Report No. 136/99, Case 10.488 (1999).
Case of Barrios Altos v. Peru, Judgment of November 30, 2001, (Ser. C) No. 87 (2001), (Inter-American Court).
TD Olsen, et al., ‘When Truth Commissions Improve Human Rights’ (2010) 4 International Journal of Transitional Justice 457.
M Freeman, Necessary Evils (Cambridge University Press, 2009) 13 fn1.
K Allan, ‘Prosecution and Peace: A Role for Amnesty Before the icc?’ (2011) 39 Denver Journal of Internationa Law and Policy 239, 242.
JC Portilla, ‘Evolving 21st Century Constraints Under International Law’ (2014) The Fletcher Forum of World Affairs 38.
M Pensky, ‘Amnesty on Trial: Impunity, Accountability, and the Norms of International Law’ (2008) Ethics & Global Politics 1.
R Slye, ‘The Legitimacy of Amnesties under International Law and General Principles of Anglo-American Law’ (2002) 43 Virginia Journal of International Law 169.
J Gavron, ‘Amnesties in the Light of Developments in International Law and the Establishment of the International Criminal Court’ (2002) 51 International and Comparative Law Quarterly 91 and F Lessa et al., ‘Persistent or Eroding Impunity? The Divergent Effects of Legal Challenges to Amnesty Laws for Past Human Rights Violations’ (2014) 47 Israel Law Review 105.
On a range of various amnesties see M Popkin and N Bhuta, ‘Latin American Amnesties in Comparative Perspective: Can the Past Be Buried?’ (1999) 13 Ethics & International Affairs 99.
M Penrose, ‘Impunity – Inertia, Inaction, and Invalidity: A Literature Review’ (1999) 75 Boston University Journal of International Law 269, 284.
See D Robinson, ‘Serving the Interests of Justice: Amnesties, Truth Commissions and the International Criminal Court’ (2003) 14 European Journal of International Law 481, 494.
FZ Ntoubandi, Amnesty for Crimes Against Humanity Under International Law (Brill, 2007).
A Du Bois-Pedain, Transitional Amnesty in South Africa (Cambridge University Press, 2007) 300.
R Jeffery, Amnesties, Accountability, and Human Rights (University of Pennsylvania Press, 2014.).
G K Young, ‘All the Truth and as Much Justice as Possible’ (2003) University of California Davis Journal of International Law & Policy 209.
F Lessa et al., ‘Persistent or Eroding Impunity? The Divergent Effects of Legal Challenges to Amnesty Laws for Past Human Rights Violations’ (2014) 47 Israel Law Review 105.
L Mallinder, ‘Amnesties’ Challenge to the Accountability Norm? Interpreting Regional and International Trends in Amnesty Enactment’, in LA Payne and F Lessa, Amnesty in the Age of Human Rights Accountability: Comparative and International Perspectives (Cambridge University Press, 2012) 79.
D Galliani, ‘The Reducible Life Imprisonment Standard from a Worldwide and European Perspective’ (2016) 16 Global Jurist 81.
J Close, The International Legal Status of Amnesties Granted for Serious Crimes: Historical and Contemporary Perspectives (PhD thesis, Middlesex University 2016) 7, 11.
F Lessa et al., ‘Persistent or Eroding Impunity? The Divergent Effects of Legal Challenges to Amnesty Laws for Past Human Rights Violations’ (2014) 47 Israel Law Review 105.
See for example on the extensive use of amnesty in Latin America countries R Boed, ‘The Effect of a Domestic Amnesty on the Ability of Foreign States to Prosecute Alleged Perpetrators of Serious Human Rights Violations’ (2000) 33 Cornell International Law Journal 297.
See on amnesties in JC Portilla, ‘Evolving 21st Century Constraints Under International Law’ (2014) 38 The Fletcher Forum of World Affairs 169.
Tunisia’s Legislative Decree n° 2011–1.
R Urueña, ‘Playing with Fire: International Criminal Law, Transitional Justice, and the Implementation of the Colombian Peace Agreement’ (2016) 110 ajil Unbound 364.
TOL TOL, ‘Around the Bloc: Uzbek Senate Vote in Favor of Amnesty for Constitution Day’ Transitions Online (2016) 15.
L Koerner, ‘Venezuela: Supreme Court Rejects Amnesty Bill, Maduro Sets Up Truth Commission’ Green Left Weekly (2016) 18.
LA Payne et al., ‘Overcoming Barriers to Justice in the Age of Human Rights Accountability’ (2015) 37 Human Rights Quarterly 728.
RC Slye, ‘The Legitimacy of Amnesties Under International Law and General Principles of Anglo-American Law: Is a Legitimate Amnesty Possible?’ (2002) 43 Virginia Journal of International Law 173, 245, 246.
See further Y Naqvi, ‘Amnesty for War Crimes: Defining the Limits of International Recognition’ (2003) 85 Revue Internationale de la Croix-Rouge/International Review of the Red Cross 583.
E Schabacker, ‘Reconciliation or Justice and Ashes: Amnesty Commissions and the Duty to Punish Human Rights Offenses’ (1999) New York International Law Journal Review 1.
J Childers, ‘Amnesty, Revenge, and the Threat of Conflict Relapse’ (2014) 14(6) International Criminal Law Review 1095, 1097.
See for example ‘Amnesty Laws and Their Role in the Safeguard and Promotion of Human Rights’, un Doc. E/CN.4/Sub.2/1985/16.
LJ Laplante, ‘Outlawing Amnesty: The Return of Criminal Justice in Transitional Justice Schemes’ (2009) 49 Virginia Journal of International Law 915.
S Aroussi, ‘“Women, Peace and Security”: Addressing Accountability For Wartime Sexual Violence’ (2011) 13 International Feminist Journal of Politics 576.
Office of the United Nations High Commissioner for Human Rights, Rule-of-Law Tools For Post-Conflict States: Amnesties (2009), <http://www.ohchr.org/documents/publications/amnesties_en.pdf>.
un Secretary General, Guidance Note of the Secretary General: United Nations Approach to the Transitional Justice, <http://www.unrol.org/files/TJ_Guidance_Note_March_2010FINAL.pdf>.
SJ Stedman, et al., Ending Civil Wars (Lynne Rienner Publishers, 2002).
A Penman, ‘The Peace-Justice Dilemma and Amnesty in Peace Agreements’ (2007) Conflict Trends 8.
J Snyder and L Vinjamuri, ‘Trials and Errors: Principle and Pragmatism in Strategies of International Justice’ (2004) 28 International Security 5.
E Melander, ‘Justice or Peace? A Statistical Study of the Relationship Between Amnesties and Durable Peace’ (2009) Lund University Working Paper Series 4.
WA Schabas, Unimaginable Atrocities: Justice, Politics and Rights at the War Crimes Tribunals (Oxford University Press, 2012) 177.
M Scharf, ‘The Amnesty Exception to the Jurisdiction of the International Criminal Court’ (1999) 507 Cornell international Law Journal 507.
S Vandeginste, ‘Bypassing the Prohibition of Amnesty for Human Rights Crimes under International Law: Lessons Learned From the Burundi Peace Process’ (2011) 29 Netherlands Quarterly of Human Rights 69, 70–1.
L Mallinder, ‘Amnesties’ challenge to the global accountability norm? Interpreting regional and international trends in amnesty enactment’, in F Lessa and LA Payne Amnesty in the Age of Human Rights Accountability: Comparative and International Perspectives (Cambridge University Press, 2012) 70–71.
CP Trumbull iv, ‘Giving Amnesties a Second Chance’ (2017) 25 Berkeley Journal of International Law 283, 291.
A O’Shea, Amnesty For Crime in International Law and Practice, (Kluwer Law International, 2002).
LJ Laplante, ‘Outlawing Amnesty: The Return of Criminal Justice in Transitional Justice Schemes’ (2009) 49 Virginia Journal of International Law 918.
See further R Slye, ‘The Legitimacy of Amnesties Under International Law and General Principles of Anglo-American Law: Is a Legitimate Amnesty Possible?’ (2002) 43 Virginia Journal of International Law 173.
See further DG Newman, ‘The Rome Statute, Some Reservations Concerning Amnesties, and a Distributive Problem’ (2005) 20 American University Law Review 293.
Ibid, at 304.
A Seibert-Fohr, ‘The Relevance of the Rome Statute of the International Criminal Court for Amnesties and Truth Commissions’ (2003) 7 Max Planck Yearbook of United Nations Law Online 553.
C Stahn, ‘Complementarity, Amnesties and Alternative Forms of Justice: Some Interpretative Guidelines for the International Criminal Court’ (2005) 3 Journal of International Criminal Justice 695.
This has been used in various places to rationalize the use of amnesties. See Azanian Peoples Organization (azapo) and Others v President of the Republic of South Africa and Others, 8 bclr 1015 (South African Constitutional Court Decision 1996).
J Dugard, ‘Dealing with Crimes of a Past Regime: Is Amnesty Still an Option?’ (1999) 12(4) Leiden Journal of International Law 1001.
PB Hayner, Unspeakable Truths (2nd edn, Routledge, 2011) 115.
M Minow, Between Vengeance and Forgiveness (Boston Beacon Press, 1998) 56.
JE Méndez, ‘Foreword’, in F Lessa and LA Payne Amnesty in the Age of Human Rights Accountability (Cambridge University Press, 2012).
Decision on Ieng Sary’s Rule 89 Preliminary Objections (Ne bis in idem and amnesty and pardon)  Extraordinary Chambers in the Courts of Cambodia Trial Chamber (Extraordinary Chambers in the Courts of Cambodia Trial Chamber).
HJ Lubbe, ‘Successive and Additional Measures to the trc Amnesty Scheme in South Africa: Prosecutions and Presidential Pardons’ (Intersentia, 2010) 57.
Dugard, supra n 74.
L Sadat, ‘National Amnesties and Truth Commissions’, in S Macedo Universal Jurisdiction: National Courts and the Prosecution of Serious Crimes Under International (University of Pennsylvania Press, 2004) 193, 194.
G Fiddler, ‘Using A Conditional Amnesty and Truth and Reconciliation Commission as a Transitional Justice Mechanism in Syria’ (2015) 47 The George Washington Institute of International Law Review 893, 901.
The Belfast Guidelines on Amnesty and Accountability (Transitional Justice Institute, 2013), <https://www.ulster.ac.uk/__data/assets/pdf_file/0005/57839/TheBelfastGuidelinesFINAL_000.pdf>.
D Weissbrodt, ‘Review of Andreas O’Shea, Amnesty for Crime in International Law and Practice’ (2003) American Journal of International Law 97, 227.
K Allan, ‘Prosecution and Peace: A Role For Amnesty Before The icc?’ (2011) Denver Journal of International Law and Policy 239.
JW McCarty iii, ‘Nonviolent Law? Linking Nonviolent Social Change and Truth and Reconciliation Commissions’ (2012) 14 West Virginia Law Review 969, 996.
K McEvoy and L Mallinder, ‘Amnesties in Transition: Punishment, Restoration, and the Governance of Mercy’ (2012) 39 Journal of Law and Society 410.
Mallinder, supra n 9, at 54.
Y Bui and JL Jordan, ‘Amnesty and Pardon’ (2014) The Encyclopedia of Criminology and Criminal Justice.
See ‘Syria Grants Conditional Amnesty for Army Deserters’, cbc News, (15 July 2015), <http://www.cbc.ca/news/world/syria-grants-conditional-amnesty-for-army-deserters-1.3167622>.
Asbarez Staff, ‘Armenia Again Extends Conditional Amnesty For Draft Evaders’, Asbarez.com, (2017), <http://asbarez.com/89931/armenia-again-extends-conditional-amnesty-for-draft-evaders/>.
CE Loyle and HM Binningsbø, ‘Justice During Armed Conflict’ (2016) Journal of Conflict Resolution 1.
Y Bouandel, ‘Bouteflika’s Reforms and the Question of Human Rights in Algeria’ (2002) 7 The Journal of North African Studies 23 and V Arnould, ‘Amnesty, Peace and Reconciliation in Algeria’ (2007) 7 Conflict, Security & Development 227.
See G Joffe, ‘National Reconciliation and General Amnesty in Algeria’ (2008) 13 Mediterranean Politics 213.
Mallinder, supra n 9 at 158–162.
V Igreja and E Skaar, ‘“A Conflict Does Not Rot”: State and Civil Society Responses to Civil War Offences in Mozambique’ (2013) 31 Nordic Journal of Human Rights 149, 157.
K Ambos, et al., Building A Future on Peace and Justice (Springer, 2009) 62.
Amnesty Act 2000.
Including Timor-Leste/Indonesia, Kenya, Grenada, Timor-Leste.
RT Coyne, ‘Reply to Noah Feldman: Escaping Victor’s Justice by the Use of Truth and Reconciliation Commissions’ 58 Oklahoma Law Review 11, 16.
S Yarbrough ‘Amnesty or Accountability: The Fate of High-Ranking Child Soldiers in Uganda’s Lord’s Resistance Army (2014) Vanderbilt Journal of Transnational Law 559.
G Fiddler, ‘Using a Conditional Amnesty and Truth and Reconciliation Commission as a Transitional Justice Mechanism in Syria.’ (2015) 47 George Washington International Law Review, 893.
M Pensky, ‘Amnesty on Trial: Impunity, Accountability, and the Norms of International Law’ (2008) Ethics & Global Politics 1; J Close, ‘Crafting an International Norm Prohibiting the Grant of Amnesty for Serious Crimes: Convergences and Divergences in the Case-Law of International Courts’ (2016) Queen Mary Law Journal 109.
K Henrard, ‘The Viability of National Amnesties in View of the Increasing Recognition of Individual Criminal Responsibility at International Law’ (1999) 8 Michigan State University Journal of International Law 639.
Ibid, at 645.
GK Young, ‘All the Truth and as Much Justice as Possible’ (2003) University of California Davis Journal of International Law and Policy 209, 225.
JN Clark, ‘Transitional Justice, Truth and Reconciliation: An Under-Explored Relationship’ (2011) 11 International Criminal Law Review 241.
G Fiddler, ‘Using a Conditional Amnesty and Truth and Reconciliation Commission as a Transitional Justice Mechanism in Syria’ (2015) 47 The George Washington Institute of International Law Review 893, 917.
R David, ‘In Exchange for Truth: The Polish Lustrations and the South African Amnesty Process’ (2006) 33 Politikon 81.
CM Horne, Transitional Justice: Vetting and Lustration, in C Lawther, et al. (eds), Research Handbook on Transitional Justice (Edward Elgar, 2017).
M Los, ‘Lustration and Truth Claims: Unfinished Revolutions in Central Europe’ (1995) 20 Law & Social Inquiry 117.
See for example Rekvényi v Hungary, (App. 25390/94), 20 May 1999, (2000) 30 ehrr 519; Sidabras and Džiautas v Lithuania, (Apps. 55480/00 and 59330/00), 27 July 2004, (2004) 42 ehrr 104; and Ždanoka v Latvia, (App. 58278/00), 16 March 2006, (2007) 45 ehrr 478.
See further AR Lamin, ‘Building Peace Through Accountability in Sierra Leone: The Truth and Reconciliation Commission and the Special Court’ (2003) 38 Journal of Asian and African Studies 295.
K Ainley, et al., Evaluating Transitional Justice (Palgrave Macmillan, 2015).
Interestingly, attitudes towards the amnesty process by victims decreased over time because of how unfair victims found the process and how much truth actually emerged. See B Ugochukwu, Book Review – Pablo de Greiff & Roger Duthie’s (2012) German Law Journal 94, 102 and D Backer, ‘Watching A Bargain Unravel? A Panel Study of Victims’ Attitudes About Transitional Justice in Cape Town, South Africa’ (2010) 4 International Journal of Transitional Justice 443.
DH Foster, et al., The Theatre of Violence (hsrc Press, 2005) 8.
CBN Gade, ‘Restorative Justice and the South African Truth and Reconciliation Process.’ (2013) 32 South African Journal of Philosophy 10.
On the need for this to happen see Eugene Baron, ‘Remorse and Repentance Stripped of its Validity. Amnesty Granted by the Truth and Reconciliation Commission of South Africa’ (2015) 41 Studia Historiae Ecclesiasticae (she) 169.
See Gade, supra n 118 for the deficiencies in this respect in South Africa.
WA Schabas, ‘Amnesty, the Sierra Leone Truth and Reconciliation Commission and the Special Court for Sierra Leone.’ (2004) uc Davis Journal of International Law and Policy 145, 152–3.
Volume Three B, Chapter Six The trc and the Special Court for Sierra Leone, para. 132.
Ibid, para. 72 (emphasis added).
T Kelsall and E Sawyer, ‘Truth vs Justice: Popular views on the Truth and Reconciliation Commission and the Special Court for Sierra Leone’ (2006) Online Journal of Peace and Conflict Resolution 36.
K Gallagher, ‘No Justice, No Peace: The Legalities and Realities of Amnesty in Sierra Leone.’ (2000) 23 Thomas Jefferson Law Review 149.
Government of the Republic of Sierra Leone, and Revolutionary United Front of Sierra Leone. Peace Agreement Between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone, Agreement, Lomé, Togo, (7 July 1999).
K Henrard, ‘The Viability of National Amnesties in View of the Increasing Recognition of Individual Criminal Responsibility at International Law’, (1999) 8 Michigan State University Journal of International Law 595, 639.
Government of the Republic of Sierra Leone, and Revolutionary United Front of Sierra Leone. Peace Agreement Between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone, Agreement, Lomé, Togo, (7 July 1999), Article ix.
DJ Macaluso, ‘Absolute and Free Pardon: The Effect of the Amnesty Provision in the Lomé Peace Agreement on the Jurisdiction of the Special Court for Sierra Leone.’ (2001) 27 Brooklyn Journal of International Law 347, 371.
‘Sierra Leone trc – Volume One, Chapter Four’ (Sierraleonetrc.org, 2017) para. 59, <http://www.sierraleonetrc.org/index.php/view-report-text-vol-1/item/vol-one-chapter-four>.
Ibid, para. 38.
Ibid, para. 70.
T Kelsall, ‘Truth, Lies, Ritual: Preliminary Reflections on the Truth and Reconciliation Commission in Sierra Leone’ (2005) 27 Human Rights Quarterly 361, 372.
See the endnotes in the ‘Sierra Leone trc Report – Volume Three A, Chapter Three, The Military and Political History of the Conflict’.
R Shaw, Rethinking Truth and Reconciliation Commissions: Lessons from Sierra Leone (United States Institute of Peace, 2005) 4, <https://www.usip.org/sites/default/files/sr130.pdf>.
N Boister, ‘Failing to Get to the Heart of the Matter in Sierra Leone-The Truth Commission is Denied Unrestricted Access to Chief Hinga Norman’ (2004) Journal of International Criminal Justice 1100.
J Paulson, ‘The Educational Recommendations of Truth and Reconciliation Commissions: Potential and Practice in Sierra Leone’ (2006) Research in Comparative and International Education 335, 351.
P Krapp, ‘Amnesty: Between an Ethics of Forgiveness and the Politics of Forgetting’ (2005) 6(1) German Law Journal 185, 194.
Naftali, supra n 11.