1.   Introduction

In February 2013, the Appeals Chamber of the International Criminal Tribunal for the former Yugoslavia (ICTY) acquitted the former chief of the general staff of the Yugoslav army, Momcilo Perisic, of aiding and abetting crimes committed in Sarajevo and Srebrenica.[1] The Appeals Chamber found that the Trial Chamber had erred in not establishing that the accused had specifically directed his assistance to the commission of the crimes. Some months later, in September 2013, the Appeals Chamber of the Special Court for Sierra Leone (SCSL) considered, in the Taylor case, that the Perisic Judgment did not adequately state the law on aiding and abetting liability and that ‘specific direction’ was not an element of it.[2] Following that, in January 2014, a differently composed bench of the ICTY Appeals Chamber in the Sainovic Judgment attempted to reverse Perisic by equally finding that it had erred in finding that the assistance needed to be specifically directed to the commission of the crimes.[3]

The Perisic Judgment and subsequent judicial reactions to it in Taylor and Sainovic have sparked a lot of debate. The legal question at the heart of the controversy is well known by now: whether it is a legal requirement of the actus reus of aiding and abetting liability that the assistance must be ‘specifically directed’ at the commission of particular crimes. The object of this commentary is not to discuss the substance of this question, to which a number of academic discussions have been devolved already,[4] nor, as a consequence, to take a position on the accuracy of the findings in the three judgments. Rather, the idea is to take a step back and consider what the debate can tell us about the unity of International Criminal Law (ICL). Indeed, this string of contradictory decisions has raised the spectrum of the fragmentation of ICL and therefore seems to deserve some attention.

It should be noted as a preliminary note that disagreements and controversies are regular features of discussions within ICL. Specific issues in this field that have sparked heated debates in the past decades would be too numerous to list here, as they essentially cover the full range of ICL matters, from the definition of crimes themselves, the modes of liability, defences, sentencing, the rights of the accused and the possibility for victim participation, down to more technical procedural rules such as admissibility of evidence or witness proofing. This state of affairs is normal, and in fact, a natural feature of any epistemic community.

In this context, the first reaction one might have when observing the frenzy that surrounded the Perisic judgment and its (judicial) aftermath in Taylor and Sainovic is surprise that it reached such a level of intensity and debate, even beyond the courtroom and the circle regular commentators of ICL. This raises the question of what is so specific about this issue that is has sparked so much discussion?

This commentary claims that the reason for the controversy lies not so much in the technical discussion of the scope of aiding and abetting liability, but rather in the existing fragmentation of ICL. More accurately, I believe that it in fact shatters some mythical unities of this field of law that have been its foundations and may have to be rethought in the future.

The following developments will more particularly focus on three unities that have been directly or indirectly challenged by the Perisic judgment: the historical unity of ICL (2), the unity of ICL as a body of law (3) and the unity of the epistemic community of ICL (4). The conclusion will then propose a more humble unifying ambition of the field: the unity of professional competence (5).


2.   The Historical Unity of ICL

One way in which the Unity of ICL is alleged, is through the adoption of a common historical narrative. While some trace back the origins of ICL to events such as the trial of Peter van Hagenbach[5] or more habitually to the attempt to prosecute the former Prussian Emperor after World War I,[6] the most commonly accepted starting point of modern ICL is of course the Nuremberg and Tokyo Trials.[7] Following this starting point, the narrative of ICL takes us through a series of events that inevitably lead us to the development of an increasingly comprehensive system of ICL, which includes the creation of the ad hoc tribunals, a number of hybrid tribunals as well as domestic prosecution of international crimes, aimed at fighting impunity for the crimes that ‘deeply shock the conscience of mankind’.[8] This narrative is rarely questioned in any meaningful way.

The consequence of this narrative which is most relevant to understand the ‘specific direction’ debate, is the possible over-reliance on the Nuremberg trials and subsequent prosecutions under Control Council Law n. 10[9] as relevant legal precedent for determining the content of ICL today. Indeed, while Perisic essentially relied on a discussion of ICTY (and ICTR) cases[10] to establish whether they had identified ‘specific direction’ as an element of aiding and abetting, the Sainovic judgment relies heavily on discussion of WWII cases.[11] This reliance on world war two precedents in Sainovic that were ignored in Perisic can be seen to be discussed in two ways.

First of all, from a historiographical perspective, it should be kept in mind that writing the history of a discipline is not a neutral exercise. It involves a normative choice in deciding to connect the dots between various historical events. This choice carries with it certain assumptions about the actual existence (and unity) of ICL that would have such a clear and identifiable history. In other words, it might be a desire for having a unified ICL that determines what commentators claim is part of its history, rather than the history grounding in a neutral way the unity of ICL. In this sense one can possibly question to what extent there is a necessary historical continuity between Nuremberg and modern examples of ICL, such as the ICC and other international(ized) tribunals. Of course, there is definitely some narrative continuity. Indeed, in 1946, the United Nations General Assembly affirmed ‘the principles of international law recognized by the Charter of the Nürnberg Tribunal and the judgment of the Tribunal’.[12] This was followed by decades of international discussion on international crimes, which led to the adoption, for example, of the Genocide Convention in 1948, the Convention on the Non-Applicability of Statutory Limitations to War Crimes and Crimes Against Humanity in 1968 and the adoption by the International Law Commission of a draft code of crimes against the peace and security of mankind in 1996. But there is another narrative that is possible, beyond the recourse to ICL discourse, which sees particular instances of internationalized or hybrid tribunals as contextual responses to specific situations of mass violence. In that perspective, political opportunity rather than historical genealogy, or the progressist pressure of ICL, most certainly might play a far bigger role in explaining the creation of a given institution.

Second of all, even assuming that the Nuremberg Judgment (and the trials that followed within Germany) and Tokyo are indeed part of the history of ICL, there is a difference between considering these examples as an interesting historic precursor to current efforts of international criminal justice, and considering them as legally relevant for understanding the actual scope of ICL today. In fact, putting aside the emotional attachment for Nuremberg, the opposite assumption might actually make more sense. Why should a handful of cases decided 70 years ago on the basis of hastily written legal instruments and hardly developed legal principles, in a particular historical context, be of any relevance to tribunals created in different circumstances, by different legal instruments? The fact that the ‘principles’ from Nuremberg were adopted in the ensuing years, does not provide a sufficient justification to the contrary for a number of reasons. First of all, the 1946 GA Resolution refers to the Nuremberg Tribunal, not the other trials that took place in the various occupied German zones, which are mostly the ones that are ‘imported’ as precedent in modern case-law. More importantly, the recognition of non-identified ‘principles’ in a four paragraph non-binding General Assembly Resolution is certainly not a sufficient basis for claiming a customary law status of the outcome of either the Nuremberg Judgment or subsequent trials which would justify use as precedent at the ICTY. This is particularly true of modes of liability, which is the question at the heart of the Perisic controversy. The WWII trials were operating under a unified theory of commission,[13] which is not the case at the ICTY, nor in subsequent international tribunals.[14] It is somewhat difficult in those circumstances to accept that the WWII case law be automatically of relevance to modern discussions on modes of liability.

In light of this, while the actual reasoning of the Perisic judgment might be questionable,[15] there might be reason to approve its departure from a tradition that gives too much judicial weight to the Nuremberg precedents. This reliance might have made sense in 1993 when the body of law of ICL was near to inexistent, it is much less justified after 20 years of practice and development of the law, especially in light of the increased specificity of the Rome Statute.

As an aside, it is interesting to note that Perisic actually was in line with a more recent, but no less traditional, genealogy of modern ICL: the Tadic case. Indeed, while the Tadic Appeal Judgments are arguably less known to the general public than the Nuremberg one, they are foundational for most international criminal law experts that were born to the practice in the early 90s. It is precedent, among other things, for the disappearance of the link between armed conflict and crimes against humanity, for the way in which to deal with legality challenges, for the recognition of the commission of war crimes in non-international armed conflicts, for the ‘discovery’ of Joint Criminal Entreprise, for providing a more flexible and all-encompassing definition of an armed conflict or for challenging the traditional test of attribution under state responsibility, and for making a number of obiter comments on a number of issues, such as the one that Perisic relied on in relation to aiding and abetting.

Despite the support of this prestigious ancestor for the Perisic judgment, both Sainovic and Taylor move away from that attachment to Tadic. The Sainovic Judgment considered that the ‘specific direction’ comment in Tadic was merely made to contrast aiding and abetting liability with JCE liability, and that ‘consequently, the Tadic Appeal Judgement […] does not purport to be a comprehensive statement of aiding and abetting liability. The Appeals Chamber […] therefore considers that the analysis of the previous case law conducted in the Perisic Appeal Judgement relied on the flawed premise that the Tadic Appeal Judgement established a precedent with respect to specific direction’.[16]

This relatively lesser influence of Tadic is a general trend in recent years. Indeed, JCE III was not explicitly recognised as forming part of customary international law by the ECCC,[17] while JCE altogether never found its way in the Rome Statute. The STL recently ‘reversed’ the Tadic approach to legality challenges,[18] and the ICJ took the opportunity in 2007 to reaffirm the Nicaragua test for attribution.[19]

As a whole, while one should be careful not to extrapolate a general trend from three judgments, it would be a welcome development that judges moved away from an over-reliance on the historical judgments that founded modern ICL. Indeed, Nuremberg and Tokyo, but also Tadic, might arguably be better suited for history books than for use as legal precedent in current judgments.


3.   The unity of ICL as a body of law

The second unity that was possibly shattered by the Perisic Judgment is the unity of the applicable law in ICL.

Indeed, the perception of unity of ICL as a body of law is usually linked back to the use of customary international law as a preferred source of law. This ensures that, beyond the particular mandates of various tribunals, one can possibly identify, outside of the specific institutional framework, a body of law that could universally be called ‘International Criminal Law’.

It should be noted that recourse to customary law in this field, while widely accepted today, was not necessarily a methodological necessity. If we go back to the founding judgment of modern international criminal law in Nuremberg, one can see that, even if the Judges went into some arguably unconvincing discussions on the customary law recognition of crimes against peace,[20] this was in fact only obiter given that the first element the Court puts forward is the idea that the judges are simply bound by their statute.[21]

It is the ICTY that established the importance of customary international law for ICL legal reasoning. Despite the fact that its statute does not contain a provision on the applicable law, and therefore does not refer to customary international law,[22] the Yugoslav tribunal was famously invited by the Secretary General to apply ‘rules of international humanitarian law which are beyond any doubt part of customary law’ to satisfy the requirements of the principle of legality.[23] This invitation was taken seriously by the judges, as illustrated by the thorough discussion of the customary law status of the crimes contained in the Statute of the ICTY in the seminal Appeals Chamber decision on jurisdiction in the Tadic case.[24]

Both the Taylor and the Sainovic Judgments took issue with the Perisic judgment on this point. In Taylor, the Appeals Chamber of the SCSL noted that, in its reasoning, the Perisic bench did not consider the question of whether ‘specific direction’ as an element of the actus reus of aiding and abetting was part of customary law. Consequently, it found that ‘in the absence of any discussion of customary international law, it is presumed that the ICTY Appeals Chamber in Perisic was only identifying and applying internally binding precedent’.[25] As a result, it conducted its own determination of the content of customary international law and concluded that the element identified by Perisic was not a part of it. As for Sainovic, as a Chamber from the same tribunal as Perisic,[26] it could not just follow the Taylor reasoning and discard the judgment as being the binding law of another tribunal. It had to explain in another way why it was departing from it. In order to do so, it not only, as mentioned previously, disagreed with the reading that Perisic did of previous case law of the tribunal, but also endeavored to discuss whether the ‘specific direction’ criteria had its source in customary international law.[27]

While both the Taylor and the Sainovic judgments seem to revert to the traditional method of law ascertainment in ICL, it is useful to take a step back and question whether it is in fact a sound method. Indeed, there are several reasons why this reliance on customary international law as the basis for a unified applicable law can actually be challenged.

On a more general level, one can question whether the uncertain nature of the content of customary international law can ever make it a satisfactory source of international criminal law, given that the principle of legality requires that the law be certain and specific.[28] The complex and lengthy academic and judicial debates that surround the content of most international criminal law customary rules makes the idea that defendants should have known about it in advance rather fanciful.[29] More specifically, by definition, the process of formation of customary law is an ambiguous process in which the border between law and non-law is difficult to determine and where, in effect, such rules are condemned ‘to being dangerously indeterminate, at least as long as they have not been certified by a law-applying authority’.[30] In other words, until a judge says so, there is no certainty that the rule was customary or not. In those circumstances, the fiction that the rule was out there waiting to be discovered is in reality masking de facto reactive punishment.

In relation to that, there is some irony in the fact that while customary law was called upon to be applied by the Secretary General in order to respect the principle of legality, it has, in reality, been a way for international judges to justify what would appear to be clear violations of that principle. Indeed, customary law has often been a way for judges to reintroduce considerations of morality into their judicial reasoning, under the guise of a customary law discussion.[31] For example, in Kupreskic, the Trial Chamber found that the Martens Clause[32] illustrated the fact that ‘principles of humanitarian law may emerge through a customary process under the pressure of the demands of humanity or the dictates of public conscience, even where State practice is scant or inconsistent’[33] and then applied it to the question of reprisals against civilians. Also, In Tadic, the Appeals Chamber, when considering the expansion of the prohibition of certain means and methods of warfare to non-international armed conflicts, famously referred to ‘elementary considerations of humanity and common sense’ to justify it.[34] In essence, this illustrates the fact that international tribunals, while adopting a natural law perspective, usually veil it in positivist reasoning.[35]

In addition to that, the claim that the existence of a customary law norm came to justify that prosecutions could be initiated by tribunals even created after the facts. However, one could argue that the principle of legality requires foreseeability of both criminalization of the conduct and of prosecution. This reliance on substantive law to justify retroactive assumption of jurisdiction by certain international tribunals (all of them, but for the ICC in fact), gives the impression that criminal law might exist in an institutional void. However, I would argue that while the source of incrimination must be somewhere, it cannot be anywhere, it should be linked to a particular prosecution mechanism that is tasked to prosecute the crimes.[36]

Specifically, in relation Perisic, it has been argued by some that because the ‘specific direction’ element of aiding and abetting liability comes to limit criminal responsibility, there is actually no obligation to resort to customary law, which is only there to safeguard that criminal responsibility is not unduly expanded.[37] There is certainly some logic to that, but it does raise the question, if customary law is not the source of law to be referred to, then what is?[38] Indeed, the principle of legality does not operate in a legal void, it operates within the context of an applicable law of a tribunal, which judges should in theory be bound by. If it is customary law, then we are back to square one.

In conclusion, it can be argued that customary law, on principle, should be avoided as much as possible rather than promoted as a source of ICL, as is often the case. Indeed, it might have been unrealistic, 20 years ago, to reject customary law entirely as a source of ICL, in the absence of the substantive body of law that we now have. However, the progressive codification of ICL and the adoption of the Rome Statute with increasingly specific provisions in relation to the definition of the crimes and modes of liability, should naturally lead to a lesser role in the future for customary international law. In that respect, one can note that the Rome Statute for the ICC contains, for the first time, a provision on the applicable law[39] which not only provides that the judges should apply, ‘in the first place, this Statute, Elements of Crimes and its Rules of Procedure and Evidence’,[40]but, in addition does not even mention customary international law as a subsidiary source of law. While some have argued that ‘there is little doubt’ that the provision refers to customary law,[41] one cannot help but conclude to the caution of the drafters against it, given that it is such a traditional source of international criminal law and when so many other categories are mentioned explicitly.[42]


4. The Unity of the Epistemic Community of ICL

I introduced this commentary by apparently stating the obvious: disagreements within a given epistemic community are common and unoriginal. One can wonder, however, whether this claim actually fully applies to what happened in the case of the Perisic judgment. In fact, when considering the reactions to the judgment, it would appear that what it revealed is the lack of unity of the epistemic community of ICL in the first place.

Indeed, the practitioners (in a broad sense) in fact belong to very different communities with possibly strongly conflicting agendas and ideologies. To put it simply, for a number of people, ICL is merely the enforcement arm of the broader human rights movement. In that context, ICL is perceived as a tool to ensure reparations for human rights violations, with a strong victim-oriented bias. In that context, the idea that ICL needs to provide justice for victims takes precedence over the respect of the rights of the accused. There are numerous examples of this approach. For example, a number of commentators approved the Taylor judgment as a ‘victory for justice’. However, technically, if the system is to be considered as fair as a whole, then any verdict should be seen as a victory for justice, even an acquittal.[43]

Along the same lines, it appears that Prosecutors of international tribunals fall within this category. It is interesting to note that the office of the Prosecutor of the ICTY, in its motion for reconsideration of the Perisic judgment in light of Sainovic, claimed to be representing the victims of the crimes.[44] The former Prosecutor of the ICC, Luis Moreno Ocampo, made similar statements, referring to the victims he represented when talking about the Kenya case.[45] This is notable shift from the traditional idea that Prosecutors act in the name of the community as a whole, rather than in the name of specific victims.

The pervasiveness of this human rights ideology in ICL, which calls for an ever expanding scope of criminal liability, can often clash with certain principles of criminal justice, such as the principle of legality. This tension has been theorized by Darryl Robinson as an ‘identity crisis’ of ICL, when human rights approaches to interpretation or modes of liability comes to contradict what can be called liberal justice.[46]

The Perisic judgment and its aftermath illustrate ideally this epistemic tension, as exemplified by one of its most famous outcomes: the demise of an ICTY Judge. The episodes of this have been commented on in detail elsewhere,[47] but it can be noted that following the acquittal of Perisic, Judge Harhoff, then a judge at the ICTY, expressed in a private email his disappointment in a way that was considered to show bias towards convictions. As a consequence, following a request for disqualification by Seselj, he was removed from that case.[48] Interestingly, it appears that a number of staff members at the ICTY share Judge Harhoff’s concerns about the acquittal of Perisic.[49]

In sum, beyond the legal debates on the nature of aiding and abetting liability or the differences in legal reasoning in relation to customary international law, it is the fragmentation of the epistemic community of ICL that explains that the Perisic Judgment has received so much attention. More precisely, had Perisic been found guilty on a faulty use of customary law or an over-reliance on an unsubstantiated obiter dictum in the Tadic judgment, the case would never have been so publicized and criticized.


5. Conclusion: Promoting a unity of competence

It appears from the previous developments that rather than promote the fragmentation of ICL, the Perisic, Taylor and Sainovic judgments have in fact made apparent mythical unities on which ICL is constructed. These unities, whether historical, substantial or epistemic need to be but in perspective if the exact nature of ICL is to be understood.

The question that remains is whether this absence of unity is a problem for the coherence of the field of ICL. We would argue that the need for unity should not be over-exaggerated, depending on the scenario.

The first scenario is where different tribunals don’t apply the same law. In that case, there is no a priori need for unity. Judges of each international tribunal should first and foremost apply their own law, in order to satisfy the requirements of the principle of legality. There is no particular harm in the fact that this law might be different from one judicial institution to the next. In application of the principle of legality, what is important is that a defendant appearing before a particular court be able to ascertain the content of the law applicable by that court. If it happens to be based on customary law and be the same than other courts, that is not a problem, of course, but if it happens to be different from one could to the next, it is not a problem and judges of different institutions should not desperately try to ‘unify’ ICL, for example through a systematic reference to customary law, even when their founding document might be clear enough. In fact, there could be more danger in applying the ‘unified’ customary international law of ICL, because it leads to the false premise that, irrespective of the existence of a pre-existing court and the foreseeability of prosecution, alleged criminals anywhere are expected to know the content of sometimes ethereal customary law norms and abide by them. The ‘unity’ of ICL becomes in that case a tool for expanding its scope and a weapon against defendants, creating a tension with the principle of legality.

The second scenario is where different Chambers within one institution have different interpretations of the law. In that case, ‘self-fragmentation’[50] could arguably be seen as more problematic, especially from the point of view of the rights of the accused, because defendants will not know on the basis of what law to argue their case. This is not a desirable situation and should of course ideally be avoided. But even in that case, the ‘risks of fragmentation’ should not be exaggerated. For one, while not an excuse, it should be noted that even within well structured and hierarchized domestic systems, there are often discrepancies in the interpretation of the law between appeal chambers or even between different decisions of the supreme court. In France, for example, the various chambers of the Cour de Cassation do not always agree on the interpretation of the law and there have been examples of disagreement between the highest courts of the country (Conseil d’Etat and Cour de Cassation) on important issues.[51] Second of all, and more importantly, beyond the specific example of Perisic, it is necessary to look for the reasons for such a self-fragmentation in order to remedy it.

In that perspective, it is necessary to recall that ultimately, judges bear the responsibility to ensure the consistency of the applicable law.[52] They are the ones who make the decisions to consider or not a particular legal issue and address it in a particular way and they need to find the right balance between judicial activism and judicial restraint.[53] For example, it is interesting to note that in the Sainovic Judgment, Judge Tuzmukhamedov dissented in relation to the Chambers finding on ‘specific direction’, not on the substance, but because he considered it was inappropriate to consider the issue, as it did not arise from the facts of the case.[54] In other words, he considered that ‘it would be prudent to exercise some restraint in addressing such rifts in the jurisprudence of a respectable and authoritative judicial institution so as to preserve as much as possible, judicial harmony in the case law that impacts the development of international criminal law and international humanitarian law, as well as legal certainty, stability and predictability, in particular, for the benefit of the parties to proceedings before the Tribunal’.[55]

A second tool for the judges to avoid self-fragmentation is to adopt a clear and understandable legal reasoning. The fact is that when reading the Perisic and Sainovic judgments with no preconceived idea of what the law says, both of them seem somewhat unconvincing. Indeed, Perisic, as pointed out by both Taylor and Sainovic, merely attempts to reconcile vague pronouncements on the issue from other chambers and ultimately relies on an unsubstantiated pronouncement in the Tadic decision on the need for ‘specific direction’. Moreover, the reasoning appears unconvincing in relation to determining whether the ‘specific direction’ should be part of the actus reus or the mens rea of the mode of liability. Indeed, while the judgment itself concludes that it is part of the actus reus, the three majority judges appended separate opinions where they considered that it should more adequately be dealt with under the mens rea heading. This does not send a very good signal about the strength of the judgment itself.

However, even if the reasoning seems more elaborate, the Sainovic judgment is hardly more convincing. While it is beyond the scope of this commentary to analyze in detail the failings of the judgment, two examples can be given. First of all, the Appeals Chamber criticizes the Perisic bench for not grounding the ‘specific direction’ requirement in customary law, while at the same time relying on a definition of the mens rea of aiding and abetting liability that was established in other cases where no customary law basis existed.[56] This is also true of the Taylor Judgment.[57]

Second of all, it is striking to see that while the Appeals Chamber in Sainovic devolves 10 pages to discussing second world war cases, it only discusses the Rome Statute in one paragraph at the end of its reasoning, as a sort of afterthought. This is problematic in two ways. First of all, methodologically, one could possibly say that the customary law applicable by the ICTY should be the one that existed at the time of the events, which predate the adoption of the Rome Statute. So technically, the reference to it might be considered inappropriate. However, this might not be such a problem, because if the Rome Statute is going to be an indication of states’ opinions, it is unlikely to have fundamentally changed between 1993 and 1998. More Second of all, if the judges are going to refer to the Rome Statute, it should be done in a serious manner. However, despite the fact that the language of Article 25(3)(c) of the Rome Statute might indicate a ‘specific direction’-type requirement for aiding and abetting (albeit as part of the mens rea of the mode of liability), the Appeals Chamber did not engage with it, considering that that ‘while the ICC Statute may be in many areas regarded as indicative of customary rules, in some areas it creates new law or modifies existing law. The adoption of an international treaty, by itself, does not necessarily prove that states consider the content of that treaty to express customary international law’.[58] In other words, a handful of cases decided nearly 70 years ago by a small number of judges from a select group of victorious western countries in a casuistic way when ICL did not even really exist, had more weight for the Appeals Chamber than a treaty negotiated for years after careful thought, signed and ratified by 122 states representing all continents. This is clearly unsatisfactory. While it is certainly true that the wide adoption of a treaty is no guarantee of its customary law status, the issue deserved at least some attention on the part of the judges.

As a result, and maybe a little provocatively, it appears that in fact, the only unity of ICL that the Perisic, Taylor and Sainovic judgments illustrate is the inadequacy of the legal reasoning of international judgments and possibly the inadequate reasoning skills of a number of international judges, irrespective of their agenda or ideology. This finding is important, because it brings the conceptual debate on unity and fragmentation down to the reality of the sometimes poor quality of judicial decision-making.

This in turn creates a serious methodological difficulty for those trying to understand the nature and logic of ICL, or conceptualize it under headings such as ‘unity’ or ‘fragmentation’. Such theorisation is premised on the possibility to identify certain trends and is therefore premised on some predictability of judicial decision making which is based on the assumptions that judges will be competent. If these assumptions and premises turn out to be unfounded, no theory of unity, however elaborate, will be able to avoid the chaotic fragmentation of ICL.


[1] Prosecutor v Perisic (Judgment) ICTY-04-81-A (Appeals Chamber, 28 February 2013) (‘Perisic Judgment’).

[2] Prosecutor v Taylor (Judgment) SCSL-03-01-A (Appeals Chamber, 26 September 2013) (‘Taylor Judgment’).

[3] Prosecutor v Sainovic (Judgement) ICTY-05-87-A (Appeals Chamber, 23 January 2014) (‘Sainovic Judgment’).

[4] See for example, A Coco and T Gal,‘Losing Direction: The ICTY Appeals Chamber’s Controversial Approach to Aiding and Abetting in Perisic’ (2014) 12 J Intl Criminal Justice 345.

[5] On this see G S Gordon, ‘The Trial of Peter Von Hagenbach: Reconciling History, Historiography, and International Criminal Law’ (February 16, 2012) available at SSRN: <http://ssrn.com/abstract=2006370>.

[6] A Cassese and P Gaeta, International Criminal Law (3rd edn, OUP 2013) ch 14.

[7] See typically, P Sands, From Nuremberg to The Hague: The Future of International Criminal Justice (CUP 2003).

[8] Rome Statute, Preamble.

[9] On this, see K J Heller, The Nuremberg Military Tribunals and the Origins of International Criminal Law (OUP 2011).

[10]Perisic Judgment (n 1) paras 25-36.

[11]Sainovic Judgment (n 3) paras 1627-1642.

[12]Affirmation of the Principles of International Law recognized by the Charter of the Nurnberg Tribunal, UNGA Res 95(I) (11 December 1946).

[13]Art II(2), Control Council Law n 10.

[14] For example, art 25(3) of the Rome Statute clearly distinguishes principal from accessorial liability.

[15] See infra, section 5.

[16] Sainovic Judgment (n 3) para 1623.

[17] Co-Prosecutors v Duch (Judgment) ECCC-001/18-07-2007 (Trial Chamber, 26 July 2010) para 513.

[18] M Nikolova and M J Ventura, ‘The Special Tribunal for Lebanon Declines to Review UN Security Council Action: Retreating from Tadic’s Legacy in the Ayyash Jurisdiction and Legality Decisions’ (2013) 11 J Intl Criminal Justice 615-641.

[19]Case Concerning Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia v Serbia) (Judgment) [2007] ICJ Rep 43 paras 396-415.

[20]T Weigend, ‘‘‘In general a principle of justice”. The debate on the “Crime against Peace” in the Wake of the Nuremberg Judgment’, (2012) 10 J Intl Criminal Justice 41-58.

[21] Trial of the Major War Criminals Before the International Military Tribunal, vol XXII, 461.

[22] With the exception of art 3 which refers to the ‘laws and customs of war’.

[23] ‘Report of the Secretary-General Pursuant to paragraph 2 of Security Council Resolution 808 (1993)’ UN Doc S/25704 (3 May 1993) para 34.

[24]Prosecutor v Tadic (Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction) ICTY-94-1-A (Appeals Chamber, 2 October 1995) (Tadic Decision).

[25] Taylor Judgment (n 2) para 476.

[26] It is even technically the same Chamber (the Appeals Chamber) with simply a different composition.

[27] Sainovic Judgment (n 3) para 1626.

[28] D Jacobs, ‘Positivism and International Criminal Law: The Principle of Legality as a Rule of Conflict of Theories’, in J d’Aspremont and J Kammerhofer (eds), International Legal Positivism World (CUP, forthcoming) available at SSRN: <http://ssrn.com/abstract=2046311> 18-24.

[29]ibid 27.

[30] J d’Aspremont, Formalism and the Sources of International Law: A Theory of the Ascertainment of Legal Rules (OUP 2011) 164.

[31] Jacobs (n 28) 24-26.

[32] This clause was introduced in the 1899 Hague Convention on the Customs of War on Land and states that: ‘Until a more complete code of the laws of war is issued, the High Contracting Parties think it right to declare that in cases not included in the Regulations adopted by them, populations and belligerents remain under the protection and empire of the principles of international law, as they result from the usages established between civilised nations, from the laws of humanity and the requirements of the public conscience.’

[33] Prosecutor v Kupreskic (Judgement) ICTY-IT-95-16-T (Trial Chamber, 14 January 2000) para 527.

[34] Tadic Decision (n 24) para 119.

[35] R Cryer, ‘The Philosophy of International Criminal Law’ in A Orakhelashvili (ed), Research Handbook on the Theory and History of International Law (Edward Elgar 2011) 232–267, 250.

[36] Jacobs (n 28) 21-24.

[37] See for example, Kevin Jon Heller, ‘The SCSL’s Incoherent – and Selective – Analysis of Custom’, Opinio Juris, 27 September 2013, available at <http://opiniojuris.org/2013/09/27/scsls-incoherent-selective-analysis-custom/> (last accessed 26 May 2014).

[38] M J Ventura, ‘Guest Post: Thoughts on Kevin Jon Heller’s two thoughts of my critique of specific direction’, Spreading the Jam, 20 January 2014, available at <http://dovjacobs.com/2014/01/20/guest-post-thoughts-on-kevin-jon-hellers-two-thoughts-of-my-critique-of-specific-direction/> (last accessed 26 May 2014).

[39] Rome Statute, art 21.

[40]Art 21(1)(a), Rome Statute (emphasis added).

[41] A Pellet, ‘Applicable Law’, in A Cassese, P Gaeta and J Jones (eds), The Rome Statute of the International Criminal Court (OUP 2022) 1051-1084, 1071.

[42]While beyond the scope of this contribution, possibly the way to go, as a middle ground, is to use customary law as a tool of interpretation rather than as a source, as the Special Tribunal for Lebanon did when interpreting the domestic definition of terrorism in light of the customary international law definition of terrorism.

[43] D Jacobs, ‘Taylor Judgment: a “victory for justice”?’, Spreading the Jam, 26 September 2013, available at <http://dovjacobs.com/2013/09/26/taylor-judgment-a-victory-for-justice/> (last accessed 26 May 2014).

[44] On this, see D Jacobs, ‘The Return of the Sequel to the Specific Direction Saga: Prosecutor files for reconsideration of Perisic Appeals Judgment’, Spreading the Jam, 3 February 2014, available at <http://dovjacobs.com/2014/02/03/the-return-of-the-sequel-to-the-specific-direction-saga-prosecutor-files-for-reconsideration-of-perisic-appeals-judgment/> (last accessed 26 May 2014).

[45] RNWAfrica, ‘Ocampo Exclusive: The First ICC Prosecutor Talks About the Kenya Trials’, 6 February 2014, available at: <www.youtube.com/ watch?v=tHBwQwUzgYo> (last accessed 26 May 2014).

[46] D Robinson, ‘The Identity Crisis of International Criminal Law’ (2008) 21 Leiden J Intl L 925-963.

[47] D Jacobs, Spreading the Jam available at: http://dovjacobs.com/category/harhoff/ (last accessed 26 May 2014).

[48] Prosecutor v Seselj (Decision On Defence Motion For Disqualification Of Judge Frederik Harhoff And Report To The Vice-President) ICTY-IT-03-67-T (Chamber convened by order of the Vice-President, 28 August 2013).

[49] ‘Morale Slump at the Tribunal’, Sense Tribunal, 22 August 2013, available at <www.sense-agency.com/icty/morale-slump-at-the-tribunal.29.html?news_id=15246&cat_id=1> (last accessed 26 May 2014).

[50]M Milanovic, ‘The Self Fragmentation of the ICTY Appeals Chamber’, available at <www.ejiltalk.org/the-self-fragmentation-of-the-icty-appeals-chamber/> (last accessed 26 May 2014).

[51] A prime example is the recognition of the superiority of EU law over domestic law. While the Cour de Cassation recognised this in 1975 (arrêt Société des cafés Jacques Vabre, 24 May 1975), the Conseil d’Etat took 14 years longer (arrêt Niccolo, 20 October 1989).

[52] For a general discussion of this idea in international law, see P-M Dupuy, ‘Unité d’application du droit international à l’echelle globale et responsabilité des juges’ European J Legal Studies, vol 1, issue 2, available at <www.ejls.eu/2/21FR.pdf> (last accessed 26 May 2014).

[53] S Darcy and J Powderly, Judicial Creativity at the International Criminal Tribunals (OUP 2010).

[54] Sainovic Judgment (n 3), Dissent of Judge Tuzmukhamedov, para 43.

[55] ibid para 45.

[56] My gratitude to Kevin Jon Heller for pointing this out to me.

[57]Heller, ‘The SCSL’s Incoherent’ (n 37).

[58] Sainovic Judgment (n 3) para 1648.