LLB (Hons) (Malawi), LLM (Cape Town)
Doctoral Candidate; Teaching and Research Assistant, University of Cape Town, South Africa
Edition: AHRLJ Volume 11 No 1 2011
Pages: 75 - 92
Citation: (2011) 1 AHRLJ 75-92
Download article in PDF
In light of the global trend towards the abolition of the death penalty and the stand of the United Nations on the matter, it is not surprising that the maximum penalty available under international criminal law is life imprisonment. However, during the negotiations for the penal aspects of the Rome Statute, some delegates contended that life imprisonment is a violation of human rights such as human dignity and the prohibition against cruel, inhuman and degrading treatment or punishment. On the other hand, some delegates felt that excluding life imprisonment from the International Criminal Court's competence where the death penalty was not available would handicap its mandate to punish gross human rights violators. Adopting a human rights perspective, the article revisits this debate by critically examining the penalty of life imprisonment under international criminal law. It argues that no clear justification has been given for the imposition of life imprisonment and that the release mechanism for lifers needs to be improved. Focusing on the International Criminal Tribunal for Rwanda, the International Criminal Tribunal for the Former Yugoslavia and the International Criminal Court, the article analyses the relevant statutes and rules and the manner in which life imprisonment has been imposed by these tribunals. Further consideration is given to the enforcement of sentences with respect to the prospect of release for 'lifers'. The article concludes by stressing the need for a more focused and cautious approach to life imprisonment and the enforcement of sentences under international criminal law.
The concept of life imprisonment destroys human dignity, reducing a prisoner to a number behind the walls of a gaol waiting only for death to set him free. The fact that he may be released on parole is no answer. [F]or a judicial officer to impose any sentence with parole in mind is an abdication by such officer of his function and duty .... 
Levy J, Namibia High Court
Although most states that have abolished the death penalty have accepted life imprisonment as an appropriate alternative,  the compatibility of the latter with human rights has been an ongoing debate both at national and international levels.  Nevertheless, life imprisonment has been accepted as the maximum penalty under international criminal law. With specific reference to the International Criminal Tribunal for Rwanda (ICTR), the International Criminal Tribunal for the Former Yugoslavia (ICTY) and the International Criminal Court (ICC),  the article examines how this penalty has been applied under international criminal law. It commences with an overview of the human rights debate vis-à-vis life imprisonment, then investigates how the penalty has been imposed and the prospect of release for lifers.
Life imprisonment means different things in different countries. In some jurisdictions, it literally means that a prisoner spends the rest of his natural life in prison without the possibility of parole. In other jurisdictions, prisoners are sentenced to life imprisonment on the understanding that they will be considered for parole after serving a set number of years. In this essay, the emphasis is on life imprisonment as prescribed under international criminal law, that is, life imprisonment with the possibility of parole.
2. Life imprisonment and human rights
Life imprisonment without the possibility of parole is criticised as a violation of human rights. Below is a summary of some arguments against this kind of punishment.
2.1 Human dignity
Human beings should always be treated as ends in themselves;  hence, an offender should not be turned into an object of 'crime prevention to the detriment of his constitutionally-protected right to social worth and respect'.  Even the vilest offender remains possessed of human dignity.  In this context, life imprisonment has been criticised as a violation of the right to human dignity in that it is imposed as a deterrent to potential offenders, hence the instrumentalisation of offenders.
Moreover, it is doubtful whether deterrence can be achieved by life imprisonment or indeed long terms of imprisonment. The underlying causes of gross human rights violations, some of which lie within the political system, cannot be curbed by the threat of imprisonment. Indeed, not even the prospect of death can deter the commission of such crimes.
2.2 Cruel, inhuman and degrading punishment
It has been argued that life imprisonment presents 'an intolerable threat to the human dignity' of the offender because it is a cruel, inhuman and degrading punishment.  At the heart of the prohibition of such punishment lies the concept of proportionality of punishment to the crime.  In addition to being a form of cruel, inhuman and degrading punishment,  disproportionate sentences are generally regarded as violations of other human rights.  The indeterminacy of life imprisonment and the potential loss of liberty until the offender dies lend it to criticism that it is a grossly disproportionate and arbitrary sentence. Certainty is a crucial element of the rule of law as recognised in the principle of legal certainty.  It is in the interests of justice to quantify sentences so that a prisoner knows exactly what his punishment is.  Life sentences leave the quantification of punishment to death itself, hence they are arbitrary.
2.3 Right to rehabilitation
Life imprisonment denies the prisoner any hope of rehabilitation and reintegration into society.  Sentenced prisoners have a right to be given an opportunity to rehabilitate themselves  and re-establish themselves in the community.  Some judges have expressed the view that the mere possibility of parole does not in itself mitigate the fact that life imprisonment infringes the right to rehabilitation. As eloquently put by Levy J: 
When a term of years is imposed, the prisoner looks forward to the expiry of that term when he shall walk out of gaol a free person ... Life imprisonment robs the prisoner of this hope. Take away his hope and you take away his dignity and all desire he may have to continue living ... [E]ven though [he] may be out of gaol on parole, [he] is conscious of his life sentence and conscious of the fact that his ... debt to society can never be paid ... Life imprisonment makes a mockery of the reformative end of punishment.
It is recognised that not all prisoners need rehabilitation. In such contexts, the right argued for here is the right to be returned to a free society.  The essential content of the right to human dignity is seriously compromised 'if the prisoner, notwithstanding his personal development, must abandon any hope of ever regaining his freedom'. 
2.4 A death sentence?
It has been stated that life imprisonment is a death sentence  and that it amounts to 'putting an individual in a waiting room until his death'.  It is therefore akin to death and results in a denial of dignity, because 'a human life involves not just existence and survival, but [also] the unique development of a personality, creativity, liberty, and unfettered social intercourse'. 
Whether or not life imprisonment is a lesser punishment than the death penalty is a 'legal-philosophical question'.  Poor conditions of detention can make it worse than death itself.  Life sentences without parole have also been equated to death sentences.  It is possible to argue that the possibility of parole does not diminish this proximity since there is no guarantee that the maximum sentence for life will not be served. 
3. Life imprisonment under international criminal law
3.1 International Criminal Tribunal for Rwanda
The ICTR is empowered to impose a life sentence pursuant to Rule 101(A) of its Rules of Procedure and Evidence (RPE).  The Trial Chamber in Kayishema attempted to distinguish life imprisonment from a sentence of 'imprisonment for a term up to and including the remainder of [a defendant's] life' as phrased in the Rule.  This resulted in four concurrent 'remainder-of-life' sentences being imposed, with the Chamber holding that the sentence should be given its 'plain meaning' and distinguished from a 'life sentence' as understood in national law.  The Appeals Chamber has also noted that, unlike life imprisonment, 'imprisonment for a term up to and including the remainder of [a defendant's] life' as envisaged in Rule 101A 'is always subject to possible reductions'.  This rather peculiar interpretation is 'suspicious' and 'reveals some of the judges' intention that a full life sentence be imposed', an equivalent of the American life without parole sentence.  Read in this manner, Rule 101A would violate international human rights norms which regard life imprisonment without parole as a violation of human rights. 
The ICTR has not clearly justified its imposition of life imprisonment. Instead, undue emphasis has been placed on the gravity of the offence. It has been held that life imprisonment can be imposed even if there are mitigating circumstances so long as 'the gravity of the offence requires the imposition of a life sentence provided for'.  Alas, there is no threshold of gravity that unambiguously deserves such a sentence. The obscurity of this guideline is further compounded by the assertion that 
mitigation of punishment does not ... reduce the degree of the crime, it is more a matter of grace than of defence. [T]he punishment assessed is not a proper criterion to be considered in evaluating the findings of the court with reference to the degree of magnitude of the crime.
In Kambanda, the gravity of the crimes and the defendant's senior position led the court to impose a life sentence despite his plea of guilty and substantial co-operation with the prosecutor.  A similar sentence was imposed in Akayesu after the court emphasised that a 'heavy penalty' was warranted.  However, there is no knowing why the only appropriate 'heavy penalty' was life imprisonment. In Ndindabahizi, the Trial Chamber specifically mentioned that it had taken into account the defendant's prospects of rehabilitation, but did not state its conclusion on the matter that justified the imposition of a life sentence.  A challenge against this sentence on the grounds that it did not give credit to the mitigating circumstances was unsuccessful.  The ICTR has since held that life imprisonment should generally not be imposed where one has pleaded guilty 'in order to encourage others to come forward'. 
The ICTR also appears to impose life imprisonment in lieu of the death penalty  in the name of 'recourse to the general practice' in Rwanda.  Since the imposition of the death penalty in Rwanda was dependent on the circumstances of the offence,  there can be no absolute certainty as to its imposition in any case. Therefore, while bearing in mind the practice in Rwanda, the court should focus on the circumstances of the case before it in determining the appropriate sentence. The ICTR has noted that life imprisonment is reserved for those who planned and ordered atrocities and who participated in them with 'particular zeal and sadism',  irrespective of the formal position held. 
3.2 International Criminal Tribunal for the Former Yugoslavia
Controversy surrounds the imposition of life sentences by the ICTY pursuant to Rule 101(A) of the RPE because the courts of the former Yugoslavia were not allowed to impose life sentences; they could only impose the death penalty or a maximum term of 20 years' imprisonment.  Under article 24 of the ICTY Statute, the Trial Chamber must 'have recourse to the general practice regarding prison sentences in the courts of the former Yugoslavia'. In its defence, the ICTY has stated that it is not bound by the maximum penalty in the national courts;  that life imprisonment is the natural alternative to the death penalty;  and that its imposition is supported by the travaux préparatoires. 
These justifications are not convincing. Firstly, they reduce article 24 'to a mere statement of general principle of international law, providing only that the death penalty and punishments other than imprisonment may not be imposed'.  This undermines the legal certainty that the drafters of the article sought to create by their reference to the practice in the former Yugoslavia.  Secondly, the ICTY should seriously consider the acceptability of life imprisonment as an alternative to the death penalty; more so since the latter, and not the former, was imposed in the former Yugoslavia.  Thirdly, it is debatable whether reliance on the travaux préparatoires was appropriate.  Had the Security Council intended to make life imprisonment available to the ICTY, it would have been in the interests of legal certainty for the Statute to have stated so explicitly. 
Rule 101(A) is yet another example of the inherent power of the ICTY to regulate itself, a weakness rooted in article 15 of the ICTY Statute that makes it 'alarmingly simple' to amend the RPE.  The ICTY 'simply invoked its power to make' the RPE in including life imprisonment as an applicable penalty.  However, the RPE being subordinate to the Statute, the imposition of life imprisonment is ultra vires the Statute. 
3.2.1 The case of Galic
An example of the manner in which life imprisonment has been imposed by the ICTY is borne out by the case of Galic.  In that case, the ICTY Appeals Chamber held that the Trial Chamber had abused its discretion by imposing a sentence of 20 years on the accused,  which it described as having been taken from 'the wrong shelf' in light of the aggravating factors.  Without further ado whatsoever, the sentence was substituted with life imprisonment. In his dissenting opinion, Pocar J was sceptical about the Appeals Chamber's power to increase a sentence against which there would be no right of appeal,  arguing that the case should have been remitted to the Trial Chamber for reconsideration in order to reserve the right to appeal. 
Meron J, also dissenting, contended that there was no basis for tampering with the sentence. He noted that it was not 'so low that it demonstrably shocks the conscience' and had been imposed after careful consideration of the circumstances of the case.  He concluded thus: 
The majority's decision to increase Galic's sentence to life imprisonment may satisfy our sense of condemnation. But this increase disserves the principles of procedural fairness on which our legitimacy rests. As the highest body in our court system, we are not readily accountable to any other authority and thus have a particular obligation to use our power sparingly. We should not substitute our own preferences for the reasoned judgement of a Trial Chamber. A sound method for assuring that we have not fallen prey to such preferences is to measure our choices fully and comprehensively against those made in prior cases. Although precise comparisons may be of limited value, the radically different approach adopted by the majority in this case requires at least some explanation. Rather than undertaking such an analysis, however, the majority simply offers conclusory statements. I cannot accept the majority's approach. No matter what he has done, Galic is entitled to due process of law — including a fair application of our standard of review.
It is submitted that by imposing such an extreme punishment without proper justification, the Appeals Chamber itself abused its discretion. It is unfortunate that life sentences are imposed in this manner without due regard for an accused person's right to appeal.  Surely if it falls within the Trial Chamber's discretion to impose life imprisonment,  it also falls within its discretion to impose a lighter sentence provided all circumstances are considered? Even more unfortunate is the view expressed by Shahabuddeen J that the ICTY is not a state, hence it cannot be bound to uphold all fair trial rights. 
3.3 International Criminal Court
Article 77(1)(b) of the Rome Statute of the ICC (ICC Statute) restricts the imposition of life imprisonment to cases where it is 'justified by the extreme gravity of the crime and the individual circumstances of the convicted person'. Although the ICC is expected 'to try nothing but crimes of extreme gravity' and 'the most heinous offenders',  the restriction implies that life imprisonment should be the exception rather than the rule. Clear distinctions will have to be made to justify a life sentence.  It is hoped that, unlike the case in the ad hoc tribunals, this restriction, coupled with the provision for distinct sentencing hearings in article 76(2) of the ICC Statute, will bolster a more comprehensive approach to sentencing. Of some concern, however, is the manner in which the Statute limits the Court's discretion to a choice between a fixed term of 30 years' imprisonment or less and life imprisonment. While the limitation may influence the ICC to reduce its sentence in order to avoid a life sentence, it may also tilt in favour of a life sentence where a longer prison term would have otherwise sufficed.
4. Prospect of release
In order to pass the test of human dignity, life imprisonment should offer a 'real and tangible prospect' of release.  It should be noted that the decision to release a prisoner under international criminal law is final and irreversible.  None of the tribunals have supervisory powers over the offender after his release, a feature which can be attributed to the lack of an international police force. The inevitable reliance on states for the enforcement of sentences is therefore a setback in the development of a universal system of international criminal sentencing and enforcement proceedings.  This section considers the release system in the courts under discussion.
4.1 The ad hoc tribunals
The ICTR and ICTY cannot grant early release proprio motu. Their Statutes provide that pardon or commutation of sentences will only be considered '[i]f pursuant to the applicable law of the state in which the convicted person is imprisoned, he or she is eligible for pardon or commutation of sentence'.  The state of enforcement is required to notify the tribunal if the prisoner is so eligible. However, it is the president of the tribunal, in consultation with the judges, on 'the basis of the interests of justice and the general principles of law',  who is to make the final decision. Factors that must be taken into account include the gravity of the crime, the treatment of similarly-situated offenders, the demonstration of rehabilitation by the offender, and any substantial co-operation with the prosecutor. 
4.1.1 Evaluating the tribunals' release system
It would be interesting to see how these tribunals would consider reducing a sentence on the basis of a factor which is in fact a mitigating factor in sentencing but which was 'negated' during sentencing. It is submitted that it would be paradoxical to grant release on the same factors that were evident during trial but which, for one reason or another, were considered insufficient to warrant a lenient sentence in the first place. For instance, would Kambanda benefit from his co-operation with the prosecutor at this stage? 
A major weakness of this system is that the eligibility for pardon or commutation lies primarily with the state of enforcement.  This does not only put prisoners on an unequal footing due to differences in national laws,  but also renders the system susceptible to manipulation by states.  There is no guarantee that a lifer, or indeed any other prisoner, has a real prospect of release. Secondly, some of the criteria cannot be applied equally to all prisoners. For instance, co-operation is not possible in some crimes.  In addition, not all prisoners may have the same potential for co-operation, such as being witnesses in other cases. This factor may also pressurise prisoners to co-operate with the prosecutor in the hope of increasing their chance of early release. Thirdly, consideration of the gravity of the crime at this stage may amount to 'double jeopardy' against prisoners whose crimes are very grave.  Conversely, prisoners who committed less serious offences may benefit twice, both at sentencing and release.
While appreciating the need for equality in the treatment of prisoners, taking into account the treatment of 'similarly-situated prisoners' is not appropriate. It is unrealistic to expect all international prisoners to be considered for pardon at the same time, more so since the criteria for such pardon are subject to national law. Furthermore, it is not clear whether the phrase 'similarly-situated' is confined to prisoners serving in the same country or those who committed similar crimes or indeed those tried jointly. It would be unfair to deny release to a prisoner on the ground that his 'counterpart' has not been given an opportunity for the same; when there is no uniform or alternative release system.
Fifthly, the Statute does not clarify what is meant by the 'interests of justice' or 'general principles of law'. The former notion gives too much discretion to the Court.  It is a fluid concept capable of accommodating a wide variety of factors. This can, of course, act either in favour of or against the prisoner. However, it would have been better to explicitly list the factors. 
Of great concern, in the context of this paper, is the fact that the system does not have special considerations for lifers. It is the possibility of release that saves life imprisonment from being a cruel, inhuman, and degrading punishment.  It is ironic that the ad hoc tribunals which have no release mechanism of their own have the power to deny release contemplated by the state of enforcement, leaving no guarantee that release may be considered again. In the absence of an independent release system, it is best that the tribunals not impose life imprisonment at all. 
A solution to some of these problems may be the introduction of a rule providing for release pursuant to articles 27 and 28 of the ICTY and ICTR Statutes respectively, which vest the tribunals with supervisory powers over the enforcement of sentences.  Alternatively, review may be provided for in the sentencing judgment.  However, the latter measure would not satisfy the principle of equality. Moreover, whether such measures would in fact be intra vires the Statutes is debatable. 
4.2 International Criminal Court
The ICC Statute reserves the right to reduce sentences to the Court itself.  The hearings on reduction of sentence are mandatory and are to be heard by three judges of the Appeals Chamber or a judge delegated by it.  Article 110(3) mandates the ICC to review the sentence after 'the person has served two-thirds of the sentence, or 25 years in the case of life imprisonment'. Subsequent hearings on reduction may be scheduled every three years or at any time stipulated in the first hearing. 
Reduction may be granted if the ICC is satisfied that the offender was either willing to co-operate with the Court from an early stage and continues to be so willing;  or rendered voluntary assistance in the enforcement of Court judgments and orders;  or if there are other factors 'establishing a clear and significant change of circumstances sufficient to justify the reduction of sentence'.  Rule 223 of the ICC RPE expounds the last condition by listing the following criteria: genuine dissociation from the crime; prospect of resocialisation; impact of release on social stability and victims; significant action taken by the prisoner for the benefit of the victims; and the individual circumstances of the prisoner, such as age or poor physical or mental health.
4.2.1 Evaluating the International Criminal Court release system
The ICC release mechanism attempts to balance the interests of the offender, the victims and society at large.  While is worth applauding as a considerable improvement on the ad hoc tribunals' system, it is not without its flaws. Firstly, there is concern with the set minimum term of 25 years for lifers. Though this period is reasonable because the maximum fixed term is 30 years,  it is rather harsh as a minimum and is in fact sufficient for retribution.  Further, it is questionable how two-thirds of a fixed sentence has been parallelled to 25 years.  The indeterminacy of life imprisonment means that 25 years can be any fraction of the sentence, indeed more than two-thirds of it.
The inclusion of article 110(4)(c) mitigates legal certainty of release.  The change of circumstances was initially intended to cater for a change in the political circumstances under which the original offence was committed.  However, it now gives room for a fragmented release mechanism where one set of factors is fixed in the ICC Statute itself while the other is articulated in the RPE, hence subject to change at any time.  To make matters worse, the Statute gives room for the application of different criteria in subsequent hearings.  Consequently, the criteria in Rule 223 may be applied independently.
Interestingly, the factors set out in the ICC Statute and RPE do not bear a direct connection to the purposes of punishment. On the contrary, some of them render the reduction process an award ceremony. Reduction may be awarded for co-operating with the Court or assisting victims. There is no clarity as to when such co-operation or assistance should be made. However, from the wording of the Statute, which states 'early and continuing willingness' to co-operate, it can be said that co-operation before conviction may also be considered. This clearly puts offenders who plead guilty at a great advantage. Since there is nothing to preclude such an offender from benefiting from his plea both at the sentencing and reduction stages,  the ICC will have to distinguish its treatment of such factors at the two stages.  Other factors, such as assisting victims and the prisoner's dissociation from the crime, erroneously assume that prisoners generally accept their being guilty after conviction by a court of law. 
The extent to which a prisoner may assist victims is greatly limited by the very fact of imprisonment itself. Prisoners with good connections outside the prison and sufficient finances would undoubtedly be better placed to offer such assistance. This would influence a positive attitude from victims, hence it would increase the chances of release for the concerned prisoner even more. It is not clear what action should be taken for the benefit of victims to qualify as 'significant'. Would an apology suffice? The Court may look to the impact of the action on the victims in order to assess its significance. The major challenge with such an assessment is the identification of victims, who may be countless.
The problem of taking into account the acknowledgment of the offence or dissociation therefrom is illustrated by the case of Leger v France, where the applicant's claim of innocence throughout his incarceration following a conviction of murder was seen as evidence of a lack of 'serious effort to readjust to society'.  This factor may also face evidential challenges. Would it be appropriate for the Court to look to the prisoner's communications with other people? I think not. It would be improper for a prisoner's friends or therapists to testify against him on this score based on his conversations with them. It is therefore submitted that acknowledgment of the offence should not be a major consideration, if at all.
The prospect of resocialisation and successful resettlement can be criticised as having no relation with the punishment of the prisoner in the first place, save perhaps for its remote connection to rehabilitation. Resocialisation is dependent on a wide variety of factors, most of which are beyond the control of the prisoner. These include the attitude of the community towards the prisoner and the prospect of securing a job. Chances for success in these matters for the class of perpetrators to be tried by the ICC are indeed slim.
Continued detention based on the perceptions of victims would tend to be arbitrary. Some victims are more vengeful than others, while some take more time to heal. This is in turn related to the impact of the atrocities in the first place. Such considerations may also lead to the unequal treatment of prisoners. It would be inappropriate to detain a person simply because the victims concerned are still aggrieved, hence against the idea of his early release.
It is foreseeable that the determination of social instability would pose a great difficulty to the Court. It would be an affront to justice to deny release because it may cause social upheaval,  a situation over which the prisoner has no control. Such incarceration would not be 'punishment', but rather an abuse of imprisonment for convenience. This would result in arbitrary detention and degrading treatment. After all, the reactions of society are not considered in assessing the guilt of an accused person.
The individual circumstances of the prisoner, as envisaged in Rule 223, appear to be more appropriate. However, the problem is that such consideration is only possible after the prescribed period. A prisoner may be in a critical health or mental condition before the minimum period has elapsed. It is submitted that a more reasonable position would be achieved by the inclusion of a general proviso to article 110(3), to the effect that a reduction of sentence may be considered at any time after sentencing on medical, humanitarian or other compelling grounds.  There is no reason for consideration of age at this stage when the age of the prisoner at that point can be deduced during sentencing. A prisoner can be spared the turmoil of being a lifer if due account of his age is taken during sentencing.
4.3 Final observations on release systems
The ICC has tacitly set the preventive element of life imprisonment at 25 years. Once this has been served, the grounds for continued detention must be based upon considerations of risk and dangerousness; any other grounds would necessarily carry the risk of arbitrariness.  While considerations of dangerousness can be accommodated into the 'interests of justice' category in the ad hoc tribunal system, they can hardly be read into the ICC's article 110(4) or Rule 223. In any case, predictions of recidivism are 'notoriously hard to make', especially in politically-motivated crimes.  Indeed, do convicts like Rutaganda, Akayesu, Kambanda and other lifers pose a danger to society? Can they reoffend in the absence of the social conditions and political climate that facilitated their actions that have been characterised as international crimes? It is unfortunate that criminal law at times decontextualises offences from the circumstances in which they were committed.  Gross human rights violations are essentially group crimes; one can hardly commit them singly. Neither are they necessarily connected to previous criminal conduct that may predict future dangerousness.  Therefore, the real question should not be whether the prisoner will be able to live a law-abiding life, but whether it is abusive to continue to detain him further.  In the long run, despite the possibility of release, continued detention beyond a certain period will raise other issues of fundamental rights. 
This article has examined life sentences under international criminal law. Save for rehearsing aggravating and mitigating factors,  the tribunals have paid little attention to justifying the imposition of life imprisonment. No wonder their sentencing judgments have been criticised as 'repetitive and ground-clearing exercise[s]'.  There is a need for a more cautious approach to life imprisonment because of its potential to deny liberty indefinitely, especially where no independent release system is in place.
The uncertainty of release weighs heavily on lifers,  and its denial has 'as much an impact on an offender as the initial sentencing decision'.  Legal certainty as to the possibility of such release is therefore imperative. It is recommended that the criteria for release should be clearly spelt out in the Statutes of the ad hoc tribunals. The ICC regime promises better treatment of life imprisonment. However, its release mechanism is quite inflexible, with the Court only able to consider reduction of a sentence after 25 years in the case of lifers. Further, the manner of its implementation, particularly considerations of eligibility for release, raises significant challenges.
The position taken by international criminal law with regard to life imprisonment is undoubtedly in need of revision in light of the challenges raised in the article. It is hoped that this will be done in the near future to promote the protection of the human rights of offenders facing trials before international tribunals.
 S v Tjijo 4 September 1991 (NM), unreported – but reproduced substantially in S v Tcoeib (2001) AHRLR 158 (NaSC 1996), previously reported as 1996 1 SACR 390 (NmS), para 16.
 R Stokes 'A fate worse than death? The problems with life imprisonment as an alternative to the death penalty' in J Yorke (ed) Against the death penalty: International initiatives and implications (2008) 282.
 D van Zyl Smit 'Life imprisonment as the ultimate penalty in international law: A human rights perspective' (1999) 9 Criminal Law Forum 5 26-45.
 The Special Court for Sierra Leone is excluded because art 19(1) of its Statute demands imprisonment for a specific number of years.
 The second premise or maxim in Emanuel Kant's Categorical imperative.
 BVerfGE 45, 187 227-228 (1977), translated in R Miller 'The shared transatlantic jurisprudence of dignity' (2003) 4 German Law Journal 925 930.
 Gregg v Georgia 428 US 153 (1976).
 Van Zyl Smit (n 3 above) 29.
 S v Dodo 2001 1 SACR 594 (CC) para 37.
 D van Zyl Smit Taking life imprisonment seriously in national and international law (2002) 198.
 See D van Zyl Smit & A Ashworth 'Disproportionate sentences as human rights violations' (2004) 67 Modern Law Review 541.
 Van Zyl Smit (n 3 above) 29.
 Tjijo (n 1 above).
 G de Beco 'Life sentences and human dignity' (2005) 9 The International Journal of Human Rights 411 414.
 Art 10(3) of the International Covenant on Civil and Political Rights, 1966 (ICCPR) states that 'the penitentiary system shall comprise treatment of prisoners the essential aim of which shall be their reformation and social rehabilitation'.
 Lebach BVerfGE 35, 187 235-236 (1973).
 Tjijo (n 1 above).
 Van Zyl Smit (n 3 above) 34.
 BVerfGE 45, 187 245.
 See Tjijo (n 1 above) reproduced in Tcoeib (n 1 above).
 De Beco (n 14 above).
 LS Shellef Ultimate penalties: Capital punishment, life imprisonment, physical torture (1987) 138.
 GA Knoops Theory and practice of international and internationalised criminal proceedings (2005) 275.
 Stokes (n 2 above) 288.
 D van Zyl Smit 'Life imprisonment: Recent issues in national and international law' (2006) 29 International Journal of Law and Psychiatry 405.
 Rule 101A reads: 'A convicted person may be sentenced to imprisonment for a term up to and including the remainder of the convicted person's life.'
 The Prosecutor v Kayishema and Ruziandana, ICTR-95-1-T, judgment, 21 May 1999 para 31.
 Kayishema (n 28 above), para 32. On appeal, a single life sentence was imposed. See Prosecutor v Kayishema, ICTR-95-1, appeals judgment (Reasons), 1 June 2001.
 Kambanda v Prosecutor, ICTR-97-23-A, appeals judgment 19 October 2000, n 144 (emphasis in original). See, generally, JD Mujuzi 'Is there a need for the ICTY to clarify the difference(s) between life imprisonment and imprisonment for the remainder of the offender's life? The Galic and Lukic decisions' (2010) 10 International Criminal Law Review 855.
 Van Zyl Smit (n 10 above) 186 187. See also Kigula v Attorney-General Constitutional Petition 6 of 2003, Constitutional Court of Uganda (unreported) 140-142, where Twinomujuni J remarked: 'Life imprisonment is a realistic alternative to a death penalty and it can only be a viable alternative if it means imprisonment for life, and not a mere 20 years.'
 Tcoeib (n 1 above) para 22: Life sentence without parole treats a prisoner as a 'thing' instead of a human being; S v Bull 2001 2 SACR 681 (SCA) para 23; Kafkaris v Cyprus, application 21906/04, judgment 12 February 2008, ECHR; Van Zyl Smit (n 10 above) 183-185; JD Mujuzi 'Why the Supreme Court of Uganda should reject the Constitutional Court's understanding of imprisonment for life' (2008) 8 African Human Rights Law Journal 163-185.
 Musema v Prosecutor, ICTR-96-13-A, appeals judgment 16 November 2001, para 396.
 Prosecutor v Akayesu, ICTR-96-4-T, sentencing judgment 2 October 1998 8.
 Prosecutor v Kambanda, ICTR 97-23-S, judgment and sentence, 4 September 1998, paras 44 & 61-62. This sentence was upheld on appeal. See Kambanda v Prosecutor, ICTR-97-23-A, appeals judgment 19 October 2000, para 126. See also Prosecutor v. Ferdinand Nahimana, Jean-Bosco Barayagwiza, Hassan Ngeze, ICTR-99-52-A, appeals judgment 28 November 2007, para 1038(3). (The existence of mitigating circumstances does not automatically preclude the imposition of life imprisonment.)
 Akayesu (n 34 above) 8; see also Rutaganda v Prosecutor, ICTR-96-3-T, judgment and sentence, 6 December 1999, paras 455-473.
 Prosecutor v Ndindabahizi, ICTR-2001-71-I, judgment and sentence 15 July 2004, para 498.
 Ndindabahizi v Prosecutor, ICTR-01-71-A, appeals judgment 16 January 2007, paras 124-142; see also Prosecutor v Ngiyitegeka, ICTR-96-14-A, appeals judgment, 9 July 2004, para 267; Ngiyitegeka, appellant's brief, 23 December 2003, para 215.
 Prosecutor v Serugendo, ICTR-2005-84-I, judgment and sentence, 12 June 2006, paras 57 & 89.
 See eg Akayesu (n 34 above).
 Art 23(1) Statute of the International Criminal Tribunal for Rwanda (ICTR Statute).
 Simba v Prosecutor, ICTR-01-76-T, judgment and sentence 13 December 2005, para 434. Rwanda bolished the death penalty in 2007.
 Simba (n 42 above), para 434; See also Prosecutor v Nchamihigo, ICTR-01-63-T, judgment and sentence, 12 November 2008, para 388; Prosecutor v Muvunyi, ICTR-00-55A-T, judgment and sentence, 12 September 2006, para 538; Prosecutor v Seromba, ICTR-2001-66-I, judgment 13 December 2006, para 403; Prosecutor v Ntagerura, ICTR-99-46-T, appeals judgment 25 February 2004, para 815; Ngiyitegeka v Prosecutor, ICTR-96-14-T, judgment and sentence 16 May 2003, para 486; Prosecutor v Muhimana, ICTR- 95-1B-T, judgment and sentence 28 April 2005, paras 604-616; Prosecutor v Kamuhanda, ICTR-99-54A-T, judgment and sentence, 22 January 2004, paras 6, 764 & 770; Ndindabahizi (n 37 above) paras 505, 508 & 511; Musema (n 33 above) para 388.
 Karera v Prosecutor, ICTR-01-74-T, judgment and sentence 7 December 2007, para 583; Prosecutor v Musema, ICTR-96-13-T, judgment and sentence 27 January 2000, paras 999-1008; Rutaganda (n 36 above) paras 466-473.
 WA Schabas 'Perverse effects of the nulla poena principle: National practice and the ad hoc tribunals' (2000) 11 European Journal of International Law 521 525.
 Prosecutor v Tadic, IT-94-1-T, sentencing judgment, 14 July 1997, para 21.
 Prosecutor v Erdemovic, IT-96-22-T, sentencing judgment 29 November 1996, paras 33-39; Tadic (n 46 above) para 9.
 The statement relied on was that of Madeline Albright, the United States representative at the Security Council. See Provisional Verbatim Record of the 3217th meeting.
 Van Zyl Smit (n 10 above) 181.
 Van Zyl Smit (n 10 above) 181. In Erdemovic (n 47 above) para 38, the ICTY recognised that the reference to 'general practice' arose because of concerns about legal certainty but refused to give effect to the intention of the drafters, stating that its application did 'not recognise the criminal nature universally attached to crimes against humanity'; MC Bassiouni & P Manikas The law of the International Tribunal for the Former Yugoslavia (1996) 701-702 hold the opinion that Rule 101A also violates the prohibition of ex post facto laws; See also WA Schabas 'Sentencing and the international tribunals: For a human rights approach' (1997) 7 Duke Journal of Comparative and International Law 461 482, who avers that the principle of legal certainty can be respected without the need for the tribunals following 'in a strict sense' the practice of Yugoslavia or Rwanda.
 Van Zyl Smit (n 10 above) 181-182.
 D van Zyl Smit 'Punishment and human rights in international criminal justice' (2002) 2 Human Rights Law Review 1 8.
 As above.
 See J Laughland Travesty: The trial of Slobodan Milosevic and the corruption of international justice (2007) 90-91 who argues that the RPE can even be amended by an exchange of e-mails by the judges.
 Van Zyl Smit (n 52 above).
 Van Zyl Smit (n 52 above). This question does not arise in the ICTR since life imprisonment is imposed in Rwandan courts.
 Prosecutor v Galic, IT-98-29-A, appeals judgment 30 November 2006; Prosecutor v Stakic, IT-97-24-T, appeals judgment 31 July 2003; Prosecutor v Lukic, IT-98-32/1-T, judgment 20 July 2009. Stakic's sentence was reduced to 40 years' imprisonment on appeal. See Prosecutor v Stakic, IT-97-24-A, appeals judgment 22 March 2006, para 428.
 See Prosecutor v Krištic, IT-98-33-T, judgment 2 August 2001 (46 years) – reduced to 35 years in Prosecutor v Krištic, IT-98-33-A, appeals judgment 19 April 2004, para 275; Prosecutor v Blaškic, IT-95-14-T, judgment 3 March 2000 (45 years) – reduced to nine years in Prosecutor v Blaškic, IT-95-14-A, appeals judgment 29 July 2004, 258; Prosecutor v Jelišic, IT-95-10-T, judgment 14 December 1999 (40 years). In his partially dissenting opinion to the confirmation of Jelišic's sentence in Prosecutor v Jelišic, IT-95-10-A, appeals judgment 5 July 2001, para 2, Wald J stated that the 40-year term imposed on the accused, who was 31 years old at the time of sentencing, was 'in effect a life sentence'.
 Galic (n 57 above).
 Galic (n 57 above) para 456.
 Galic (n 57 above) para 455.
 See the partially dissenting opinion of Pocar J in Galic (n 57 above) paras 2-4 186-187.
 The question of whether or not an Appeals Chamber can competently increase a sentence is beyond the scope of this paper. Suffice to say that art 14(5) of ICCPR guarantees the right to appeal against sentence.
 Separate and partially dissenting opinion of Meron J in Galic (n 57 above) paras 9 & 10 205-208. The gist of Meron J's argument was that the Trial Chamber had not committed a 'discernible error', hence the Appeals Chamber had no power to intervene on the sentence. (See Prosecutor v Tadic IT-94-1-A and IT-94-1-Abis, judgment in sentencing appeals 26 January 2000, para 22.) The Trial Chamber had in fact noted that a 20-year sentence was the maximum prison term available in the former Yugoslavia Courts. See Prosecutor v Galic, IT-98-29-T, judgment 5 December 2003, para 761. It is arguable, therefore, that it was imposed as a comparably heavy penalty, short only of the death penalty itself.
 Separate and partially dissenting opinion of Meron J (n 64 above) para 13.
 See also Gacumbitsi v Prosecutor, ICTR-2001-64-A, appeals judgment 7 July 2006, where the ICTR Appeals Chamber increased a 30-year sentence to life imprisonment.
 Prosecutor v Jelišic, IT-95-10-A, appeals judgment 5 July 2001, para 100.
 See separate and partially dissenting opinion of Shahabuddeen J in Jelišic (n 67 above) paras 25 & 26 55-56, asserting that the ICTY Statute is equivalent to a reservation to ICCPR.
 WA Schabas An introduction to the International Criminal Court (2003) 317.
 Van Zyl Smit (n 52 above) 14.
 De Boucherville v The State of Mauritius (2008) UKPC 37, para 23; Kafkaris (n 32 above) para 6 of the joint dissenting judgment of Tulkens, Cabral Barreto, Fura-Sandström, Spielmann and Jebens JJ.
 Schabas (n 69 above) 142.
 Knoops (n 23 above) 274.
 Arts 27 & 28 of the ICTY and ICTR Statutes respectively.
 As above.
 Rules 125 & 126 of the ICTY and ICTR RPE respectively.
 Kambanda provided 'invaluable information' to the prosecutor and agreed to testify in other cases. See Prosecutor v Kambanda, ICTR 97-23-S, judgment and sentence 4 September 1998, para 47.
 Van Zyl Smit (n 52 above) 9. The criteria for pardon are not always public knowledge. See JD Mujuzi 'The evolution of the meaning(s) of penal servitude for life (life imprisonment) in Mauritius: The human rights and jurisprudential challenges confronted so far and those ahead' (2009) 53 Journal of African Law 242-244.
 Van Zyl Smit (n 3 above) 51.
 A Hoel 'The sentencing provisions of the International Criminal Court' (2005) 30 The International Journal of Punishment and Sentencing 37-66.
 Van Zyl Smit (n 10 above) 195.
 Van Zyl Smit (n 10 above) 184.
 Schabas (n 50 above) 513.
 The listing in the RPE (n 76 above) is not exhaustive.
 Bull (n 32 above).
 Schabas (n 50 above) 516.
 Schabas (n 50 above) 510.
 As above.
 As above.
 Art 110.
 Rule 224(1) of the ICC RPE.
 Rule 224(3) of the ICC RPE. A sentenced person may also apply for an earlier subsequent hearing.
 Art 110(4)(a).
 Art 110(4)(b).
 Art 110(4)(c). Art 27 expressly proscribes the consideration of official capacity as a factor 'in and of itself'.
 Hole (n 80 above).
 Art 77(1)(a).
 Van Zyl Smit (n 52 above) 16.
 Van Zyl Smit (n 3 above) 52 observes that the determination of this period is 'inevitably arbitrary'.
 Van Zyl Smit (n 10 above) 195.
 As above.
 As above. As at 10 December 2009, the ICTY had amended its RPE 44 times.
 Art 110(5) allows the Court to apply 'such criteria as provided in the Rules' in subsequent hearings.
 Hoel (n 80 above) 65.
 See the discussion at 4.1.1 above.
 An offender can have less reason for assisting victims if he or she does not accept responsibility for their victimisation.
 Leger v France, Aapplication 19324/02, judgment 11 April 2006, ECHR. For a detailed analysis of this case, see Stokes (n 2 above).
 Van Zyl Smit (n 3 above) 48.
 Van Zyl Smit (n 3 above) 52.
 Stokes (n 2 above) 293.
 Van Zyl Smit (n 10 above) 194; see generally N Padfield (ed) Who to release? Parole, fairness and criminal justice (2007).
 P Allot 'Deliver us from social evil' Guest Lecture Series of the Office of the Prosecutor, 11 August 2004, The Hague, para 2.68.
 AL Smeulers 'What transforms ordinary people into gross human rights violators?' in SC Carey & SC Poe (eds) Understanding human rights violations: New systematic studies (2004) 239-256.
 De Beco (n 14 above). There can be no guarantee that any individual will live a law-abiding life.
 Fura-Sandström J, dissenting, in Leger (n 107 above) para 14.
 Pursuant to arts 23 & 24 of the ICTR and ICTY Statutes respectively.
 Van Zyl Smit (n 52 above) 9.
 United Nations Crime Prevention and Justice Branch Life imprisonment (1994) 20.
 Van Zyl Smit (n 3 above) 52.