Category Archives: International Criminal Justice

Part 2 – Rohingya Deportation: Deportation is a Distinct Crime in the ICC Statute

Saif Ansari is a recent graduate of New York University, School of Law. He has previously served under Judge Nonkosi Mhlantla at the Constitutional Court of South Africa and Judge Nicola Lettieri at the Special Tribunal for Lebanon.

1. Introduction

In April 2018, the Prosecutor at the International Criminal Court filed a motion seeking guidance from the Pre-Trial Chamber about whether the court had jurisdiction over the deportation of Rohingya Muslims from Myanmar, a non-state party, to Bangladesh, a state party. In earlier discussions, commentators argued that deportation and forcible transfer are combined into a unified offense in the Rome Statute, and because crossing an international border is an element of deportation but not forcible transfer, the unified crime was complete in Myanmar. Consequently, no essential element of the crime occurred in Bangladesh, depriving the ICC of jurisdiction over the acts of deportation.

In our first post, we showed how the International Criminal Court has jurisdiction over these acts of deportation, regardless of whether or not deportation and forcible transfer are a single crime. To do this, we contested whether an “essential” element is the requisite standard. We used the crime of torture as an analogy, showing how the essential element requirement used in these discussions is of mysterious origins, poorly defined and overly demanding. Applying what we believe are more defensible tests for jurisdiction, we concluded that even if deportation and forcible transfer are a single offense, their unification does not deprive the ICC of jurisdiction over acts of deportation to Myanmar.

In this second post, we argue that deportation and forcible transfer are distinct offenses, even though they are listed in a single provision in the Rome Statute and Elements of Crimes. In effect, deportation and forcible transfer are more akin to Article 8(2)(a)(vii) of the Rome Statute (which prohibits the distinct offences of “[u]nlawful deportation or transfer or unlawful confinement”), than torture (which prohibits the infliction of physical or mental suffering as alternate means of perpetrating the same offense of torture).

On one level, our argument offers an alternate basis for the court to find that the acts of deportation in question took place in Myanmar if it disagrees with our first post. If deportation is a distinct, separate offense, the distinction turns on victims being expelled across a border “to another State,” which is highly germane to the Rohingya situation. On another level, we believe that there are important reasons to preserve the continued existence of deportation quite apart from the issue of jurisdiction in this case. In particular, we are concerned about the interpretative coherence of the Rome Statute, the distinct expressive value of deportation as a separate crime, and for sentencing.

In what follows, we propose five mutually reinforcing arguments as to why deportation remains a single, distinct offense in the Rome Statute and Elements of Crimes, despite the fact that it is listed under the same heading as forcible transfer. First, we show how a literal reading of the Rome Statute and the Elements of Crimes preserve the historical distinction between deportation and forcible transfer. We accept that the interpretation we offer is not the only interpretation available, but we believe it is analytically preferable as a matter of pure hermeneutics, and the only interpretation that squares with the other material we review. Second, we show how the drafters clearly intended for there to be a distinction between deportation and forcible transfer based on whether the displacement was within a state or across an international border. Third, we demonstrate how the ICC itself has consistently retained the distinction in its own practice to date, notwithstanding some curious language in one decision. Fourth, we marshal evidence of state practice to argue that the distinction is also firmly entrenched in customary international law. Fifth, we show why there is a distinction in the first place: because deportation and forcible transfer protect similar but different interests. We conclude that conflating deportation and forcible transfer would be a mistake in law and principle.

2. Article 7(1)(d) preserves the distinction between deportation and forcible transfer

“Deportation or forcible transfer of population” is a crime against humanity under Article 7(1)(d) of the Rome Statute. According to Article 7(2)(d), for purposes of Article 7(1)(d), “‘[d]eportation or forcible transfer of population’ means forced displacement of the persons concerned by expulsion or other coercive acts from the area in which they are lawfully present, without grounds permitted under international law.” Article 7(1)(d) of the Elements of Crimes lists the elements of the crime as follows.

Article 7(1)(d)

Crime against humanity of deportation or forcible transfer of population


  1. The perpetrator deported or forcibly transferred, without grounds permitted under international law, one or more persons to another State or location, by expulsion or other coercive acts.
  2. Such person or persons were lawfully present in the area from which they were so deported or transferred.
  3. The perpetrator was aware of the factual circumstances that established the lawfulness of such presence.
  4. The conduct was committed as part of a widespread or systematic attack directed against a civilian population.
  5. The perpetrator knew that the conduct was part of or intended the conduct to be part of a widespread or systematic attack directed against a civilian population.

Footnote 12 of the Elements of Crimes says

[t]he term forcibly is not restricted to physical force, but may include threat of force or coercion, such as that caused by fear of violence, duress, detention, psychological oppression or abuse of power against such person or persons or another person, or by taking advantage of a coercive environment.

Likewise, footnote 13 says

deported or forcibly transferred is interchangeable with forcibly displaced.

There is some basis to conclude that this definition combines the crimes, but overall, we find these bases unpersuasive. First, if the two offenses were distinct, one might anticipate that they merited separate definitions in the Statute itself, especially if deportation has a cross-border element that marks its particularity. This, we argue, is answered by the terms of the Elements of Crimes we discuss below. Second, the Ruto decision refers to deportation and forcible transfer as “labels” for a single “unique crime.” Nevertheless, we show below how the same paragraph in Ruto also indicates that “[t]he factor of where they have finally relocated as a result of these acts (i.e. within the State or outside the State) in order to draw the distinction between deportation and forcible transfer is thus to be decided by the Trial Chamber.”[1] We come to each of these issues momentarily, but for now, it is sufficient to observe that a coherent literal interpretation of deportation and forcible transfer support the ongoing distinction between the two normatively overlapping offenses.

To begin, just because they are grouped under the same heading, does not mean deportation and forcible transfer are the same crime. Article 7(1)(d) is an example of a disjunctive crime, or a crime that can be committed in more than one way. Disjunctive crimes are usually denoted by the insertion of the word “or” in the name of the crime and/or an element of the offense. But as the OTP shows, the Elements of Crimes lists disjunctive crimes in different ways. For example, sexual offenses under Article 7(1)(g) are listed under different headings, but “using, conscripting and enlisting children” under Article 8(2)(e)(vii) are listed under one heading. In terms of substance, the Pre-Trial Chamber in Lubanga held that the very use of “or” in the first element shows that “conscripting or enlisting” child soldiers “or using them to participate actively in hostilities” are “separate offenses” under Article 8(2)(e)(vii).[2] We believe the same is true of deportation and forcible transfer.

A systematic interpretation of the definition of deportation or forcible transfer in the Elements of Crimes supports this view. The reference to “deportation or forcible transfer” in the headings of the Elements arguably initiates a sequencing that the entire structure of the definition subsequently follows. The word “deportation” comes first in the heading, then “deported” precedes “forcibly transferred” in Element 1. “[T]o another State” is also the first location listed by Element 1. Pursuant to the same logic, “forcible transfer” follows “deportation” in the heading, marking a sequence that is matched by the words “forcibly transferred” following “deported,” and “location” coming after “to another State.” Through this sequence, “deportation” corresponds to “deported…to another State” and “forcibly transfers” corresponds to “another… location.” Systematic structure is, of course, only one attribute of good drafting, but we argue that it weighs in favor of treating deportation and forcible transfer as distinct crimes, especially when other material we review unequivocally support this reading.

Despite some lamentable ambiguity in drafting, it is clear here that footnote 13 of the Elements of Crimes does not conflict with the systematic reading we offer here. According to footnote 13, “‘[d]eported or forcibly transferred’ is interchangeable with ‘forcibly displaced.’” Some believe this language suggests that deportation and forcible transfer, taken separately, are identical. We are not persuaded by this view and believe that much discredits it. Footnote 13 says that the terms “deported and forcibly transferred” and “forcibly displaced” are “interchangeable,” not that deportation and forcible transfer are the same crime. The difference is important. In effect, footnote 13 implies that forced displacement is another description for “deported or forcibly displaced,” which is not to say that “deported,” “forcibly transferred,” and “forced displacement” are legally synonyms. Mathematically, the interpretation we offer based on the positioning of the quotation marks around “deported or forcibly transferred” as a pair is the difference between (A + B) = C and A = B = C. Put another way, it would not be safe to deduce from the fact that, because apples and oranges are fruit, apples are oranges. So, we favor the view that footnote 13 indicates that “forcibly displaced” is a term of convenience that can be used to refer to both crimes together.

Whether by coincidence or design, courts have used the closely related term “forced displacement” in precisely this fashion, especially in circumstances where the distinction between the two offenses is of no factual significance.  Although the negotiating history of the Rome Statute is silent as to the intended meaning of footnote 13, it was likely inserted in light of the ICTY’s jurisprudence, which had employed the term “forcible displacement” as an umbrella in these circumstances. In 2000, two years before the ICC Elements of Crimes were first drafted, the trial judgment in Blaškić used “forced displacement” as a synonym for “deportation or forcible transfer of civilians” when assessing acts that could constitute persecution as a crime against humanity.[3] This use of “forced displacement” as a means of referring to both crimes together instigated a major trend, even as the same courts decided that deportation was distinguishable from forcible transfer based on whether the victim was displaced across established borders. We come to the caselaw about cross-border transfers momentarily, but for now, we pause to provide slightly more evidence of the popularity of “forced displacement” as an umbrella term.

The habit of using “forcible displacement” in this manner has spanned the better part of two decades. In the appeal judgment in Krnojelać, for instance, the ICTY held that for purposes of the crime against humanity of persecution, it does not matter whether a given act of “forcible displacement” constituted deportation or forcible transfer.[4] As the ICTY subsequently held in the appeal judgment in Simić, what matters is whether the act of “forcible displacement” was carried out with the requisite discriminatory intent.[5] Most recently, the trial judgment in Milutinović held that “[a] number of elements of these offences are the same and are discussed herein under the heading ‘forcible displacement.’”[6] Therefore, a long line of cases at the ICTY refer to deportation and forcible transfer together as “forcible displacement,” usually for the purpose of proving the distinct crime of persecution. As we will see below, this caselaw also insists that there is a formal legal difference between deportation and forcible transfer. Thus, this body of jurisprudence reinforces our view that footnote 13 of the Elements of Crimes merely creates a linguistic label for the two crimes, without altering their separate legal meaning.

Finally, the ICC has also referred to “forcibly displaced” as a generic label that serves as a placeholder until such time as the distinction between deportation and forcible transfer is established at trial. In Ruto, the Pre-Trial Chamber said that it is not necessary to distinguish between the two “labels” at the pre-trial stage, not because the distinction is irrelevant, but because it is not yet apparent which of the two was committed. The Pre-Trial Chamber delegated the task of determining whether acts of forcible displacement constituted deportation or forcible transfer to the Trial Chamber based on where victims were finally relocated to—“i.e. within the State or outside the State.”[7] But in Ruto, it was apparent even at the pre-trial stage that people were “forcibly displaced” somewhere, wherever they ended up.[8] Therefore, in spite of the seemingly contradictory reference to “labels” of a “unique crime,” Ruto shows that the ICC itself has referred to deportation and forcible transfer together as “forcible displacement,” in line with the meaning we ascribe to footnote 13. For all these reasons, a literal interpretation of the provision suggests that the offenses remain distinct.

3. The drafters intended for there to be a distinction between deportation and forcible transfer

The literal interpretation we offer is, to our minds, the only interpretation that can be squared with the negotiating history of the Rome Statute and Elements of Crime. Most importantly, numerous sources indicate that the drafters of the Rome Statute intended for there to be a legal distinction between deportation and forcible transfer, and we have unearthed no evidence that suggests otherwise. We set out some of these sources here in chronological order:

  • The International Law Commission’s commentary on the predecessor to the Rome Statute explicitly references the distinction between deportation and forcible transfer as well as the fact that transfer across a border demarcates the two crimes. According to the ILC, “[w]hereas deportation implies expulsion from the national territory, the forcible transfer of population could occur wholly within the frontiers of one and the same State.”[9] Of course, the drafters could have written Article 7(1)(d) in generic terms so as to encompass both, but they did not. They retained the distinction between deportation and forcible transfer out of a desire to ensure that the two remained separate crimes.
  • At the Preparatory Committee in 1996, there were three drafts of the relevant provision prepared, all of which mentioned deportation but only one of which mentioned forcible transfer in brackets.[10] Again, this suggests that the drafters considered whether there should be a distinction between forcible displacement within a state and across an international border. As the Members of the Canadian Partnership for International Justice argue in an amicus brief submitted to the ICC in support of the OTP’s request in the present case, “[t]his reflects the drafters’ awareness that they were dealing with separate crimes.”
  • Several authors who were present at the negotiating of this provision argue that states intended to maintain the distinction. According to Hebel and Robinson, “‘forcible transfer of population’ was added as an alternative to ‘deportation’ so as to encompass large-scale movements within a country’s borders.”[11] This addition and the basis for it, suggest that the language of the Rome Statute and Elements of Crimes we review above sought to create an additional complementary offense, not deliberately collapse deportation into a broader joint crime.
  • Similarly, Hall and Stahn report that that early versions of the section show that the drafters distinguished between “mass deportations or forced transfer of persons from the territory of a State [or from an area within a State].”[12] Apparently, the drafters were also given a paper by Amnesty International indicating that “forcible transfer” was inserted in order to create a distinction between forcible displacement within a state and across an international border. According to Hall and Stahn, no one objected to the distinction.[13]

All in all, we believe that these sources support the literal interpretation we offer. They provide this support by reinforcing the co-existence of two separate crimes and by providing no indication that States intended the fairly radical step of unifying two offenses into one such that deportation’s unique character disappeared.

4. The ICC usually treats deportation and forcible transfer as separate crimes

The ICC itself has distinguished deportation and forcible transfer on a number of occasions. As we have shown, in Ruto, the ICC says that “deportation” and “forcible transfer” are different “labels” for what is a “unique crime” based on the final destination of the victim.[14] But as we have also noted, the same paragraph in Ruto goes on to distinguish between deportation and forcible transfer precisely on the basis of a cross-border element by stating that “[t]he factor of where they have finally relocated as a result of these acts (i.e. within the State or outside the State) in order to draw the distinction between deportation and forcible transfer is thus to be decided by the Trial Chamber.”[15]

From this language, we deduce that “deportation” and “forcible transfer” are not merely descriptions of the underlying conduct. Indeed, it would be curious for the Pre-Trial Chamber to delegate the task of distinguishing between deportation and forcible transfer to the Trial Chamber unless there was a legal distinction between the two concepts that required consideration. Thus, despite the reference to a “unique crime,” the Court actually treats deportation and forcible transfer as separate crimes based on whether the victim crossed an international border. Importantly, a large number of cases before the ICC are consistent with this reading even though the rationale for their approach is not articulated explicitly.

In a host of other cases, the Pre-Trial Chamber has referenced either deportation or forcible transfer separately, undermining the notion that “deportation or forcible transfer” is a single offense. More precisely, in a range of arrest warrants and confirmations of charges, the ICC has described the conduct in question as “forcible transfer” only, without even mentioning deportation. The failure to reference the entire unified offense of “deportation or forcible transfer” suggests that, as we suspected, the two elements come apart and can stand alone. These cases are not rare—there appear to be at least nine such decisions.[16] For example, in the court’s arrest warrant for Omar Al Bashir, only forcible transfer is alleged.[17] There is no deportation charge because it was clear even at the pre-trial stage that any forcible displacement occurred “throughout the Darfur region” only.[18] If there was no difference between the two, or Article 7(1)(d) referred only to a joint crime, the ICC would have referenced deportation and forcible transfer together.

So in cases in which there is no allegation of a border-crossing, or it is apparent at the pre-trial stage that any forcible displacement must have taken place within national borders, the ICC confines itself to an analysis as to whether there are substantial grounds to believe that only forcible transfer was committed. But as decisions like Ruto show, to the extent to there is an allegation that deportation was committed or it is uncertain whether there was a border-crossing, the ICC references both deportation and forcible transfer. In this light, the ICC clearly treats deportation and forcible transfer as separate crimes in practice. This corroborates a literal interpretation of the elements of Article 7(1)(d), our analysis of footnote 13, and the intent of the drafters we have discussed.

5. Customary international law establishes that deportation is a distinct offense

To the extent that Article 7(1)(d) is ambiguous, Article 21 of the ICC Statute provides for the application of “applicable treaties and the principles and rules of international law.” These rules of international law are clear: deportation is a long-standing offense in its own right. The crime predates WWII, having been a war crime long before the Allies created the corresponding crime against humanity. In the wake of the war, the crime against humanity of deportation was codified in Article 6(c) of the Nuremberg Charter, Article 5(c) of the Tokyo Charter, and Article II(1)(c) of Control Council Law No. 10. As a result, deportation came to protect German nationals, stateless people and other people not previously protected by the war crime of deportation.

Decisions like the trial judgment in Krnojelac[19] and the decision on motion for judgement of acquittal in Milošević[20] at the ICTY draw on WWII caselaw to show that deportation is established as a separate, standalone offense in customary international law. For example, in Baldur Von Schirach, the Nuremberg Tribunal convicted the Nazi politician of deportation for the forcible displacement of 60,000 Jews from Vienna to concentration camps in occupied Poland.[21] And in Milch, the tribunal convicted a member of the German forces of deportation for purposes of slave labour both as a war crime and a crime against humanity.[22] In a concurring opinion, Judge Philip argued “[d]isplacement of groups of persons from one country to another is the proper concern of international law in as far as it affects the community of nations.”[23]

Moreover, deportation is clearly distinguished from forcible transfer, which is treated as an “other inhumane act.” The trial chamber in Krstić held that “[b]oth deportation and forcible transfer relate to the involuntary and unlawful evacuation of individuals from the territory in which they reside. Yet, the two are not synonymous in customary international law. Deportation presumes transfer beyond State borders, whereas forcible transfer relates to displacements within a State.”[24] The trial chamber drew the same distinction in Krnojelac, holding that deportation required forcible displacement across a national border, whereas forcible transfer required forcible displacement within national borders.[25] As a result, the trial chamber in Naletilić held that the jurisprudence of the court had established that deportation and forcible transfer were different crimes based on whether the victim had been forcibly displaced within a state or across the border.[26]

Admittedly, there was one rogue decision that dispensed with the cross-border element of deportation as a way of collapsing the distinction between forcible transfer and deportation, but this reasoning was quickly overturned then consistently rejected ever since. In an outlier judgment, the trial chamber in Stakić declared that crossing a border was not an element of deportation after all. Disregarding precedent, the trial chamber also held that deportation protects “the right and expectation of individuals to be able to remain in their homes and communities without interference by an aggressor.”[27] Thus, according to this judgement, the final destination of the victim was of no relevance. The novel interpretation of deportation in Stakić was born of Judge Schomburg’s distaste for the open-ended scope of “other inhumane acts” as a residual category of crimes against humanity. However, his attempt to remove forcible transfer from “other inhumane acts” then force it into deportation was quickly rejected.

In the Stakić appeal, the ICTY clarified that deportation is forcible displacement “across a de jure state border or, in certain circumstances, a de facto border.”[28] According to the appeals chamber, crossing an international border became an element of deportation with the adoption of the Geneva Conventions and Additional Protocols after WWII.[29] Every decision since that time has cited the appeals judgment in Stakić as establishing once and for all the requirement that deportation be across a border of some kind. As a result, the appeal judgment in Stakić reflects the current consensus that deportation is a distinct offense in its own right, that deportation and forcible transfer are separate offenses, and that deportation is forcible displacement across an established border. Thus, customary international law seems to provide yet another basis for treating the two offenses as distinct.

6. Deportation and forcible transfer protect different legal interests

Deportation and forcible transfer as crimes against humanity protect an interest in peaceful residence or the rights and expectations that ordinarily inhere in a person’s abode or home. In the appeal judgment in Stakić, the ICTY holds that “[t]he protected interests underlying the prohibition against deportation include the right of the victim to stay in his or her home and community and the right not to be deprived of his or her property by being forcibly displaced to another location.”[30] This is reflected in the very definition of “displacement.” In the trial judgment in Prlić, the ICTY holds that displacement or “removal” for purposes of deportation occurs when a person travels so far away from home that they are no longer able to effectively enjoy these rights.[31] But as the OTP shows, it is one thing to be expelled from one’s home, and forced to flee to another part of the same country, but it is quite another to be forced to flee across the border to another country altogether. According to the OTP, “deportation also protects a further set of important rights: the right of individuals to live in the particular State in which they were lawfully present—which means living within a particular culture, society, language, set of values, and legal protections.”[32]

In other words, a person forced to flee a country incurs an additional harm—the violence of being thrown into a completely new socio-cultural milieu in which he or she has a radically different legal and political status. Of course, there are any number of ways a court might characterize the interests that underlie deportation and forcible transfer. Reasonable people will differ about what these interests are, but this disagreement around the fringes does not camouflage the widely held view that there is an important normative difference here—the difference between being forced to flee from London to Manchester, and being forced to flee from London to Moscow. Obviously, in any one case, the harm incurred by the victim might vary. Say a person lives near the border, and is forcibly displaced far away to the other side of the same country, depriving them of their culture and familial connections. In certain circumstances, this may be more serious than the harm they would have incurred were they forcibly displaced just across the border. This exception to the rule, however, does not mean that the interests that underlie deportation and forcible transfer are the same, or that deportation is not worse than forcible transfer in the majority of cases. We believe that the Rome Statute as well as the ICC itself appreciate this important difference.

7. Conclusion

Although the definition of deportation and forcible transfer in the Rome Statute and the Elements of Crimes cannot be held up as an exemplar of legislative clarity, the factors we have reviewed suggest that deportation and forcible transfer remain distinct offenses. This view is supported by a systematic interpretation of the Elements of Crimes, by evidence that the drafters clearly intended this result, and by customary international law. What is more, notwithstanding the Ruto decision which appears to have attempted to have things both ways, the ICC itself has consistently treated deportation and forcible transfer as separate crimes in its decisions thus far. Of course, there may be factual scenarios in which treating these two crimes as if they were the same is legally innocuous and practically helpful, as the ICTY cases on “forced displacement” show. In the case of the Rohingya Muslims, however, it matters a great deal whether they were forcibly displaced within Myanmar or across the border to Bangladesh, so one cannot brush aside legal formalities. Ultimately, whether or not the Court intervenes in Myanmar and Bangladesh is a difficult moral, political and practical question, but we do not believe the law we have addressed here precludes it.



[1] The Prosecutor v. William Samoei Ruto and Joshua Arap Sang, Decision on the Confirmation of Charges Pursuant to Article 61(7)(a) and (b) of the Rome Statute, para. 268.

[2] Lubanga Trial Judgment, para. 609.

[3] Blaškić trial judgment, para. 234.

[4] Krnojelac appeal judgment, para. 224.

[5] Simic appeal judgment, para. 172 (“The Appeals Chamber recalls that for the purposes of a persecutions conviction, it is not necessary to distinguish between the underlying acts of “deportation” and “forcible transfer” because the criminal responsibility of the accused is sufficiently captured by the general concept of forcible displacement.”). Early cases at the ICTY also distinguished between “forcible displacement” within or across national borders. In Krstić, the trial chamber held that “[s]ince the Srebrenica civilians were displaced within the borders of Bosnia-Herzegovina, the forcible displacement may not be characterised as deportation in customary international law.” See Krstić trial judgment, para. 531. The trial chamber also cited Kupreškić, in which it held that “forcible displacement” was another inhumane act for purposes of article 5(i) of the ICTY statute, whether it took place within a state or across an international border. See Kupreškić trial judgment, para. 523.

[6] Milutinovic trial judgment, para. 163.

[7] Ruto, supra fn. 1, para. 268.

[8] Id.

[9]  Draft Code of Crimes Against the Peace and Security of Mankind with Commentaries, 1996, Article 18(g), commentary, p. 49. See also footnote 27 of the OTP brief, which makes the same reference.

[10] Christopher K. Hall and Carsten Stahn, The Rome Statute of the International Criminal Court: A Commentary, 3rd ed (Oxford: Hart, 2016), “Article 7: Jurisdiction, Admissibility and Applicable Law,” p. 194, fn. 293.

[11] H von Hebel and D Robinson, The International Criminal Court: the making of the Rome Statute: Issues, Negotiations, Results, Crimes Within the Jurisdiction of the Court,” p. 99.

[12] [12] Christopher K. Hall and Carsten Stahn, The Rome Statute of the International Criminal Court: A Commentary, 3rd ed (Oxford: Hart, 2016), “Article 7: Jurisdiction, Admissibility and Applicable Law,” p. 96, fn. 300.

[13] Id.

[14] Ruto, supra fn. 1, para. 268.

[15] Id.

[16] See, (1), Prosecutor v Harun and Kushayb, ICC PT. Ch. I, Decision on the Prosecution Application under Article 58(7) of the Statute, ICC-02/05-01/07-1.Corr, 27 April 2007, pp. 45, 48, 56 (Deportation is not discussed at all. There is only a forcible transfer charge, discussed in paras. 69 and 74, under counts 9, 20 and 51, for forcible displacements throughout Sudan.); (2), Prosecutor v Harun and Kushayb, ICC PT. Ch. I, Warrant of Arrest for Ahmad Harun, ICC-02/05-01/07-2, 27 April 2007, pp. 7, 10, 15-16 (There is no discussion of deportation. There are only forcible transfer charges for forcible displacements throughout Sudan, under counts 9, 20 and 51.); (3), Prosecutor v Harun and Kushayb, ICC PT. Ch. I, Warrant of Arrest for Ali Kushayb, ICC-02/05- 01/07-3-Corr, 27 April 2007, pp. 8, 10, 16-17 (Counts 9, 10, 20, and 51 discuss forcible transfer only. There is no discussion of deportation, the warrant being only for forcible transfers in Sudan.), (4), Prosecutor v Al Bashir, ICC PT. Ch. I, Decision on the Prosecution’s Application for a Warrant of Arrest against Omar Hassan Ahmad Al Bashir, ICC-02/05-01/09-3, 4 March 2009, p. 92 (There is only a forcible transfer charge—no deportation charge. The pre-trial chamber concludes that there are reasonable grounds to conclude that people were merely displaced “throughout the Darfur region.” See para. 100.); (5), Prosecutor v Al Bashir, ICC PT. Ch. I, Warrant of Arrest for Omar Hassan Ahmad Al Bashir, ICC-02/05-01/09-1, 4 March 2009, pp. 7-8 (The arrest is issued for forcible transfer, among other things. Deportation is not included and not mentioned in the warrant. See pp. 6-7.); (6), Prosecutor v Al Bashir, ICC PT. Ch. I, Second Warrant of Arrest for Omar Hassan Ahmad Al Bashir, ICC-02/05-01/09-95, 12 July 2010, p. 6 (There is only a forcible transfer charge. There is no discussion of deportation. See pp. 6-7.); (7), Prosecutor v Hussein, ICC PT. Ch. I, Public redacted version of “Decision on the Prosecutor’s application under Article 58 relating to Abdel Raheem Muhammad Hussein”, ICC-02/05-01/12-1-Red, 1 March 2012, pp. 29-30 (Counts 8, 18 and 41 concern forcible transfer. The allegation made by the prosecution is that people were forced from their homes in the Wali Salih locality in West Darfur, to other places in the same locality. Deportation is not mentioned at all.; (8), Prosecutor v Hussein, ICC PT. Ch. I, Warrant of Arrest for Abdel Raheem Muhammad Hussein, ICC-02/05-01/12-2, 1 March 2012, pp. 8, 11 (There is a forcible transfer charge, but no deportation charge. Deportation is not even mentioned.); and, (9), Prosecutor v Ntaganda, ICC PT. Ch. II, Decision Pursuant to Article 61(7)(a) and (b) of the Rome Statute on the Charges of the Prosecutor Against Bosco Ntaganda, ICC-01/04-02/06-309, 9 June 2014, paras. 36, 64-68, p. 63 (Counts 12 and 13 concern forcible transfer of population and displacement of civilians as a crime against humanity and as a war crime. There is no charge or mention of deportation.).

[17] Prosecutor v Al Bashir, ICC PT. Ch. I, Warrant of Arrest for Omar Hassan Ahmad Al Bashir, ICC-02/05-01/09-1, 4 March 2009.

[18] Id., p. 6.

[19] Krnojelac trial judgment, paras. 472-5. See also fn. 1429.

[20] Prosecutor v Milošević, Decision on Motion for Judgement of Acquittal, paras. 49-52.

[21] International Military Tribunal Judgment, Vol I, p. 319.

[22] Trials of War Criminals Before the Nuremberg Military Tribunals Under Control Council Law No.

10, Vol. II, pp. 773-879.

[23] Id., p. 865.

[24] Krstić trial judgment, paras. 521

[25] Krnojelac trial judgment, para. 474;

[26] Naletilić trial judgment, para. 670.

[27] Stakić trial judgment, para. 677

[28] Stakić appeal judgment, para. 278

[29] Id., paras. 292-4.

[30] Stakić appeal judgment, para. 277.

[31] Prlić trial judgment, para. 49.

[32] OTP request, para. 17.

Part 1 – Rohingya Deportation: Whether Deportation and Forcible Transfer are a Single Crime is Irrelevant to ICC Jurisdiction

Saif Ansari is a recent graduate of New York University, School of Law. He has previously served under Judge Nonkosi Mhlantla at the Constitutional Court of South Africa and Judge Nicola Lettieri at the Special Tribunal for Lebanon.


In April this year, the Office of the Prosecutor at the International Criminal Court filed a motion seeking something akin to an advisory opinion from the Pre-Trial Chamber on whether the Court enjoys jurisdiction over the deportation of Rohingya Muslims from Myanmar to Bangladesh. The issue has given rise to significant debate and inspired no less than eight excellent amicus briefs, as well as a consolidated response by the OTP, largely because it raises important principles for the Rohingya crisis as well as for the functioning of the Court more generally.

According to the OTP, the ICC may assert jurisdiction over the deportation of Rohingya Muslims from Myanmar to Bangladesh despite the fact that Myanmar is not a party to the Rome Statute, “because an essential legal element of the crime—crossing an international border—occurred on the territory of a State which is a party to the Rome Statute (Bangladesh).”[1] In contrast, critics have argued that deportation and forcible transfer are fused into a single crime in the Rome Statute, such that crossing an international border is not an essential element of the offense, depriving the Court of jurisdiction.

In this, our first of two posts on the topic, we wonder whether an “essential” element that all sides adopt in this discussion sets too high a standard for determining jurisdiction. We argue that an element need not be essential to a crime in order to ground the Court’s jurisdiction over an offense. Not only does Article 12(2)(a) not require that an essential element of an offense take place on the territory of a state party, adopting this standard would deprive the Court of the ability to prosecute what we call disjunctive crimes in many very ordinary situations.

By “disjunctive crimes,” we mean offenses, like deportation and forcible transfer, that are either included as alternate ways of perpetrating a single offense, or multiple, distinct offenses contained in one and the same provision of the Rome Statute. Torture, for instance, offers an example of a single offense that can be perpetrated in more than one way, since it criminalizes the infliction of mental or physical pain. The prohibition on “[u]nlawful deportation or transfer or unlawful confinement”[2] provides an illustration of multiple crimes in a single provision. We call these crimes disjunctive because their use of “or” creates alternatives as to how they can be perpetrated.

In a subsequent post, we argue that deportation and forcible transfer remain distinct and separate crimes within the Rome Statute. We will not rehearse the various arguments for that view presently, principally because we believe that the issue may be irrelevant for the purposes of jurisdiction in the Rohingya case. To our minds, far better substitutes for the “essential” element test employed throughout these discussions include: (a) whether any element of the offense was perpetrated within a state party; or, (b) whether a particular manifestation of a crime was completed within a state party. Said differently, if the “essential” element requirement leads to plainly perverse outcomes for disjunctive crimes in the Rome Statute, we should adopt a standard focused on any element of the crime or the element that completes a particular manifestation of the offense. As we will see, if we adopt either of these alternative tests, debates about whether deportation and forcible transfer are one offense or two fall away in significance.

Several caveats are necessary before we begin our discussion. First, we have not addressed the term “conduct” in Article 12 of the Rome Statute, partly because discussions about essential elements have focused on the deportation/forcible transfer dyad exclusively, but mainly because we profess to have no real expertise on questions of jurisdiction. By contrast, we are now completing a multi-year project on deportation and forcible transfer, so thought to offer these thoughts on that aspect of the wider legal issue. Second, we take no position here on whether deportation and/or forcible transfer might be ongoing crimes as advanced in several amici briefs.[3] Although we do not explore this argument here, we note that it would also transcend much of the debate. Finally, we remain agnostic about whether the ICC should intervene in Myanmar/Bangladesh. That decision raises complex moral and political issues, which extend well beyond our expertise.

Deconstructing the Essential Element Standard

We begin by noting our lack of certainty about the origins of the essential element standard. The Rome Statute does not require that an “essential” element of an offense take place on the territory of a state party in order to establish jurisdiction. Article 12, which defines jurisdiction, is silent on the issue. Of course, the Statute does not indicate that an essential element is not necessary or that a lesser standard, including those we posit as more coherent here, would be sufficient. Nevertheless, one might anticipate that a requirement that would restrict the court’s jurisdiction over and above the terms already set out in the Statute would require explicit legal authority.

As we say, we are far from expert in questions of jurisdiction in public international law or before the ICC, but from our preliminary inquiries, we have found no basis in statute, custom or caselaw for this added “essential element” requirement. On its face, then, there is real ground for skepticism that an element of an offense must be “essential” to the commission of the crime in order for the ICC to acquire jurisdictional authority over the crime. Moreover, as we argue below, we are concerned that the requirement would have a major and unjustifiable impact on the Court’s ability to try disjunctive crimes.

Before we engage with that concern more squarely, we should also point out our uncertainty about what an essential element is. Up until this point, the term has served as fulcrum for much of the debate about jurisdiction in this case, although it has escaped explicit definition throughout. From these discussions, we take it to mean that an element must be necessary to the commission of the crime for the ICC to enjoy jurisdiction over that offense. According to this argument, in the case of deportation/forcible transfer of the Rohingya Muslims to Bangladesh, crossing a border is not necessary because the offense could be legally and was factually completed at the point the victims were forced from their homes in Myanmar, well before they reached Bangladesh.

Thus, given that crossing a border is a contingent and not necessary element of what some view as a fused crime of “deportation and forcible transfer,” the essential element standard arguably operates to preclude jurisdiction when Myanmar is not a state party. So, because of this essential element requirement, whether deportation and forcible transfer really are a single unified offense or two separate crimes, and whether a cross-border element is required for one or both has real world implications for both the Rohingya Muslims and the Court.

We contest the salience of those questions by doubting that the essential element requirement can bear the weight placed upon it. To illustrate our thinking, we use the first disjunctive offense we pointed out by way of introduction, namely, torture. We employ torture because it is clearly a single offense, so by selecting it as an analogy we isolate the debate about whether deportation and forcible transfer are separate or unified crimes. As mentioned, our subsequent post argues that forcible transfer and deportation are separate offenses as a matter of law, but for argument’s sake, we assume the contrary here. The torture analogy is also useful because, to reiterate, torture is made disjunctive by the possibility of perpetrating the offense by inflicting only mental or only physical pain (for analytical purposes, we leave aside the truism that in real life, neither disjunct in torture can likely be perpetrated without impacting the other).

If the essential element standard deprives the ICC of jurisdiction where only a contingent element of a crime is satisfied, the standard would preclude the prosecution of just mental torture in a state party. This is true even if one strips the Myanmar/Bangladesh example of the controversial non-state party/state party aspect—the essential requirement standard would have this effect if all of the acts constituting mental torture occurred in a single state party, say New Zealand. Analytically, because mental suffering is not a necessary element of torture (physical suffering will suffice), the Court could not have jurisdiction over the offense because the case of purely mental suffering involves a contingent, not essential, element of the crime. Consequently, if there was widespread mental torture constituting a crime against humanity in New Zealand, the ICC would not have jurisdiction because mental suffering is not essential.

This position is both absurd and far reaching. It would preclude jurisdiction over disjunctive crimes in many far simpler scenarios. This is because even in instances where all elements are satisfied in a single state party (i.e. physical and mental suffering was inflicted in New Zealand), the disjunction makes each of the disjuncts unnecessary for the commission of the crime.

We return to Myanmar/Bangladesh to observe different illustrations of the absurdity an essential element requirement brings about. Employing the torture analogy in the Myanmar/Bangladesh situation, if a group of victims were physically tortured in Myanmar, then mentally tortured in Bangladesh, one could hardly argue that the ICC’s jurisdiction over the mental torture in Bangladesh is precluded because these victims suffered a different manifestation of the same crime in Myanmar. The mental torture in Bangladesh may not be legally essential to the commission of the crime, but it is likely sufficient to ground jurisdiction over a distinct manifestation of the crime that takes place there.

Or, to explore a different variant by employing the crimes(s) deportation/forcible transfer, if an essential element of the crime(s) were the standard in a situation where Myanmar was a state-party to the Rome Statute, the ICC might still not be able to prosecute acts of forcible transfer in Myanmar because even forcible transfer is not a necessary element of the crime(s) of “deportation or forcible transfer”—the Elements of the Crimes indicate that “deported…to another State” is an alternative means of establishing the same offenses.

We believe that the foregoing shows how all sides should dispense with the essential element standard. In its place, we posit that any element of the offense should be sufficient for purposes of jurisdiction, or alternatively, an element that completes a particular manifestation of the offense. For reasons that follow, we think it clear that the deportation of the Rohingya Muslims from Myanmar to Bangladesh satisfied either of these two alternative standards.

The Rohingya Muslims were Unequivocally “Deported…to Another State”

Having dispensed with the essential element requirement, it remains to be determined if and how deportation to Bangladesh satisfied: (a) any element of the crime(s); or (b) whether it completed a particular manifestation of an offense in Bangladesh. Here, we begin by noting that much of the argument thus far has referred to the “cross-border” element of deportation, “crossing a border” and “crossing an international border.” This language may be implicit from wording in the relevant aspect of the Elements of Crimes, but it is not a verbatim replication of the provision that is binding on the Court in this case, so we set out then analyze the literal language in the Elements of Crimes that define deportation and forcible transfer. These Elements of Crimes read as follows:

Article 7(1)(d)

Crime against humanity of deportation or forcible transfer of population


  1. The perpetrator deported or forcibly [footnote 12] transferred, [footnote 13] without grounds permitted under international law, one or more persons to another State or location, by expulsion or other coercive acts.
  2. Such person or persons were lawfully present in the area from which they were so deported or transferred.
  3. The perpetrator was aware of the factual circumstances that established the lawfulness of such presence.
  4. The conduct was committed as part of a widespread or systematic attack directed against a civilian population.
  5. The perpetrator knew that the conduct was part of or intended the conduct to be part of a widespread or systematic attack directed against a civilian population.

[footnote 12]  The term “forcibly” is not restricted to physical force, but may include threat of force or coercion, such as that caused by fear of violence, duress, detention, psychological oppression or abuse of power against such person or persons or another person, or by taking advantage of a coercive environment.

[footnote 13]  “Deported or forcibly transferred” is interchangeable with “forcibly displaced”.

If we focus on Element 1 above, we notice immediately that the structure of the offense is divided into multiple disjuncts. We also know from our previous discussion that requiring an essential element of an offense as a prerequisite to jurisdiction will defeat disjunctive offenses like this, because none of “deported”, “forcibly transferred,” “to another State”, “location”, “expulsion” or “other coercive acts” are necessary for the realization of the crime. In other words, there are multiple combinations of these contingent elements that are sufficient to establish the offense(s), even though none of them by themselves are essential for the commission of the crime.

Thus, if we back away from the essential element standard, we soon see that any element, or perhaps the element sufficient to complete a particular manifestation of the crime, are far more compelling bases for thinking about jurisdiction where there are multiple disjunctions within an offense.

This brings us to the permutation of deportation/forcible transfer whereby a perpetrator “deported” victims “to another State.” Even if we concede that deportation/forcible transfer is one unified crime (again, we contest this in our next post), one sufficient means of perpetrating this arguably unified crime would involve  a perpetrator who has “deported” victims “to another State.” Clearly, in the Rohingya situation, these phenomena take place in Bangladesh. The Elements of the Crimes mention “deported” in the past tense. This implies having crossed a border already. Plus, the element explicitly states that one of the contingent ways of committing deportation/forcible transfer involves rendering victims “deported…to another State.” Given that this crime is not inchoate, the Rohingya are only “deported…to another State” once they enter Bangladesh. Thus, it seems fairly uncontroversial that one of the ways of perpetrating deportation/forcible transfer occurs in Bangladesh.

And, recalling our earlier torture discussion, we know that this logic holds regardless of whether the disjunctions in the Elements above come together to form a single unified offense (like mental or physical suffering in torture) or whether they constitute separate, distinct crimes housed within one and the same provision of the Statute and Elements (like “[u]nlawful deportation or transfer or unlawful confinement”). We surmise, therefore, that whether deportation and forcible transfer are unified or distinct is irrelevant for purposes of jurisdiction.

This reasoning holds regardless of how we interpret footnote 13 of the Elements of Crimes. In our next post, we explain how equating “deported and forcibly transferred” with “forcibly displaced” in footnote 13 likely borrowed from a habit of using the umbrella term “forced displacement” as a catchall when the distinction between deportation and forcible transfer was factually immaterial. The practice developed first in the Blaškić trial, prior to the drafting of the Elements of Crimes, where the court used “forced displacement” as a term of convenience for deportation and forcible transfer in assessing allegations of persecution. For better or worse, that practice caught on. As we will show in our next post, there is much supporting this view and nothing in the history of the provision suggesting a contrary interpretation. But even if others disagree with this reading of footnote 13, it does not eviscerate the element “deported…to another State,” which is not essential for the commission of the offense(s), but likely adequate to ground jurisdiction.


This initial post acts as a precursor to and qualification of our next post. Momentarily, we will argue that deportation and forcible transfer are separate crimes, more akin to the provision in the Rome Statute that criminalizes “[u]nlawful deportation or transfer or unlawful confinement” than to torture, a single unified offense with disjunctive parts. Nevertheless, this initial post has queried whether the distinction matters for the purposes of jurisdiction, largely by assuming the single unified crime thesis we disagree with, then by critiquing the essential element standard it relies on.

From this analysis, we believe that being “deported…to another State” is an element of a crime against humanity that very clearly takes place in Bangladesh. Likewise, being “deported…to another State” completes a particular manifestation of the deportation/forcible transfer crime in Bangladesh. To our minds, the argument that the crime was complete in Myanmar beforehand misses that many international crimes have different disjunctive elements, and that a particular campaign of terror can and will often involve multiple variations of one and the same offense.


[1] Prosecution’s Request for a Ruling on Jurisdiction under Article 19(3) of the Statute, 9 April 2018, para 2.

[2] Art. 8(2)(a)(vii) ICC Statute.

[3] For a helpful summary, see Prosecution Response to Observations by Intervening Participants, paras. 27-31.

Legal Pluralism and Justice: A Reply to Critics

This symposium has embodied everything I sought to promote in establishing this blog. First and foremost, it has housed frank but respectful criticism. In our piece The Ahistoricism of Legal Pluralism in International Criminal Law, we saw an implicit deference to extant law in prescriptive accounts of Global Legal Pluralism that we did not think could be justified normatively. In pointing this out, we spent much time discussing how to frame the argument so as to communicate respect for a set of excellent scholars who had done so much to inform our thinking about these problems, then sought out their criticism once our piece was finally complete as a mark of this respect. In turn, they have offered equally courteous and frank responses to our arguments. In all, I am so pleased that the exercise has conveyed a commitment to the primacy of ideas, a recognition of the great intellectual value of critique, and an inclination on all our parts to metabolize whatever impersonal emotions arise from scrutiny. I hope this reply registers in that spirit and promotes that scholarly culture.

I begin by offering a set of clarifications, which I suspect the piece itself should have made clearer. Ours was never a total assault on Legal Pluralism. I continue to believe that it plays a crucial role in a number of areas, in particular through its ability to undermine what James Sákéj Youngblood Henderson calls “the colonial contrived superiority of European law.”[1] Anyone with even the most rudimentary understanding of law’s role in colonialism will know that withholding recognition of other pre-existing normative systems was a key element in this contrived superiority. In this light, Legal Pluralism is particularly important because it reasserts the authority of jurisgenerative communities other than the colonizing state while undermining rigid, hegemonic conceptions of what it means to have law. In many respects, this point explains our focus on state-made criminal law, even though this is deliberately a caricature of Legal Pluralism, as Paul Berman correctly points out. Nonetheless, as I will explain, our caricature eschews even more difficult problems for Global Legal Pluralism in an attempt to focus on a core of state law that Global Legal Pluralism’s deference to the legal requires it to embrace.

Reduced to its essence, our article was largely an attack on law as a necessary repository of moral, political, epistemic or cultural variation a cosmopolitan vision of global justice might seek to promote. Thus, we argued that Legal Pluralism’s method did not count against universal norms in the ways it claimed, in large part because it overvalued law. This point warrants emphasis. We do not argue that universal norms always solve the problems we point to across the board, which would be patently absurd. Instead, we mean to reverse the argumentative onus and place the burden onto the prescriptive account of Global Legal Pluralism, insisting that just managing whatever we consider to be law globally seems overly deferential to law. Our brief histories were, in the grand scheme of things, relatively superficial attempts to “scratch the surface” to see what lies beneath law. Having peered below the surface very briefly, we sought to point out how Legal Pluralism was not necessarily counter-hegemonic, since for a large set of what would constitute law, the hegemon had beat Legal Pluralism to the punch by imposing the law in the first place. Consequently, our aim was to warn that in prescriptive guise, the concept risked entrenching unjust norms.

We were, of course, conscious that history would never be able to “distinguish this superficial transplantation of domestic criminal law into the international realm, from potentially legitimate and useful reliance on national doctrines,” as Neha Jain rightly points out. Instead, we argued that the history of much of the world’s criminal law doctrine, both national and international, should undermine confidence that Legal Pluralism was necessarily respectful of genuine cultural variation, such that it could ground some concept of justice. Our first example of Argentine criminal law procedure operated as a null hypothesis where, by “scratching the surface,” we quickly found evidence of autonomous domestic ownership of and influence over criminal law doctrine. But in all our other examples, we found nothing similar. Far from stating that there was no congruence between local values and criminal law doctrine in any of these states, our argument was merely that these histories suggest real reason for caution against an idea that Legal Pluralism is, without more, worthy of veneration. We also felt that generic attempts to circumscribe Legal Pluralism’s over-enthusiasm for law, perhaps by citing human rights as an exception, came too late in the day because too much was already smuggled in through the initial deference to whatever law might be at the descriptive stage.

To develop this point, let me begin by recounting the descriptive and prescriptive variants of Legal Pluralism, since Global Legal Pluralism depends on both. Initially, Legal Pluralism was purely descriptive, tracing its origins to anthropological inquiries into the interactions between displaced social orders and formal colonial law. To the extent that the field drew on normative ideas within this descriptive mode, the need for this engagement largely grew out of challenges to methodology; i.e. objections to what should figure as “law” within the wider sociological inquiry. I circle back to this difficulty with defining law momentarily, since I have come to believe that it must be most acute in international criminal justice. For now, I again point to a major prescriptive shift for Legal Pluralism, where many scholars have begun to argue that diversity of social phenomena we are prepared to call law is not only empirically observable, but that this state of affairs is normatively desirable for the world. Global Legal Pluralism necessarily adopts both elements, first accepting a wide array of norms as constituting law, then recommending various institutional and procedural mechanisms for managing their interface. The caricature of Legal Pluralism the article offers was an attempt to hive off some of the most difficult problems at the descriptive stage, to highlight conceptual concerns with the transition from a descriptive to a prescriptive mode.

To explain this, let me start by agreeing wholeheartedly with Mireille Delmas-Marty’s eloquent statement that “choosing an exclusively state-centric perspective while the world is moving, the authors risk confining themselves to a state-centric, modern representation of LP which identifies with the rights of the state and makes the concept necessarily oxymoronic.” A theory of Legal Pluralism focused on state law alone would be oxymoronic. As our essay highlighted, “an over-emphasis on domestic criminal doctrine is anathema to true pluralism, whose very program involves looking beyond positivistic state-centered law.” Nevertheless, focusing on a segment of a field to elucidate conceptual concerns strikes me as entirely defensible. In our piece, we focused our analysis on state-centered criminal law in an attempt to avoid the wider jurisprudential problem involved in deciding what constitutes law at all. As Brian Tamanaha and many others have pointed out, that problem is a perennial thorn in Legal Pluralism’s side,[2] so we sought to bypass the conceptual impasse by criticizing a segment of the legal we believed was unambiguously so. I have since come to think that the positivistic assumption that informed this method was unjustifiable on our part, but in ways that radically strengthen our argument not detract from it.

Understandably, the rules crafted to bring about the atrocities international criminal justice seeks to address are the very subject matter legal theorists use to debate the nature of law. Since completing the paper, it has dawned on me that whether Nazi law was law at all is the paradigmatic debate for a large segment of jurisprudence; indeed the question goes to the heart of the relationship between law and morality and therefore operates as something of an acid test for so many aspects of legal theory. To cite what is probably the most celebrated example (in the Anglophone tradition), the Hart/Fuller debate is a long and in places heated dispute between the positivist tradition that would see law as identified through social facts quite apart from its moral value, and a natural law tradition that was shocked by the positivists’ willingness to confer normativity on a set of legislative enactments in Nazi Germany that were clearly morally perverse.[3] In many respects, the histories we discuss in the paper replicate these dilemmas. It stands to reason, therefore, that much of what we treat as law in the article might not be at all. This thought strikes me as very significant: it suggests that international criminal justice is not just a pleasant illustration of the normative interactions Global Legal Pluralism seeks to understand and manage. The field is at the heart of what Global Legal Pluralism means.

So, let me circle back to our caricature, showing how the natural law critique of our positivistic assumption only magnifies the concerns we raise. In our example of the inchoate crime of association de malfaiteurs in the Democratic Republic of Congo (see here, pp 33-41), we sought out a modern illustration of abhorrent criminal law doctrine in national systems comparable to the law “that allowed Joseph Stalin to sign 3,167 judicially-imposed death sentences in a single day, and Adolf Hitler to make being Jewish a criminal offence.” Our assumption, which I now think is highly disputable, was that this forcibly imposed inchoate crime of French then Belgian origins was unequivocally law; it survived the sieving off of law from religious, moral and social norms on the one hand, while distinguishing itself from the full range of non-obligatory propositions on the other. Once parsed out in this way, association de malfaiteurs was a law that both descriptive and descriptive theories of Legal Pluralism had to take seriously. Our project was to question whether Legal Pluralism should take it seriously at all, based only on its ability to pass this formalistic test. We used the apparent misalignment between local values and legal doctrine as our foil.

But Fuller, I suspect, would have doubted that association de malfaiteurs was law at all, then rebuked our positivistic attempt to avoid thorny jurisprudential problems as failed. But if our caricature fails for these reasons, it does so in ways that only galvanize our deeper point, namely, that “doctrinal pluralism is unsafe as a measure of diverse values and interests in the international community”. Perhaps Legal Pluralism is only about managing just law, such that the concept is instantly insulated from our criticism, but I see no evidence of this position in the literature with which I am familiar, and it would mark a sea-change for thinking about these problems. Then, to follow this jurisprudential line one step further in abstraction, the question becomes what the relationship is between Legal Pluralism and Justice. And here too, I now see this theme as a central but tacit pre-occupation in our paper as well as the literature more broadly. It is this unstated concern for justice, for instance, that leads scholars like Martti Koskenniemi to warn that Legal Pluralism “ceases to pose demands on the world;”[4] and Boaventura de Sousa Santos to assert that “there is nothing inherently good, progressive, or emancipatory about Legal Pluralism.”[5]

Justice is also useful in responding to criticisms based on the Rule of Law. In employing a Rule of Law frame, Kevin Davis’ excellent and thought-provoking critique of our article argues that “[d]octrines that confirm to people’s expectations are desirable, all else being equal, because they contribute to legal certainty and avoid situations that are tantamount to ex post facto lawmaking.” While there is certainly weight to this criticism, I am not sure how it squares with other principles espoused in the Rule of Law. As Waldron points out, avoiding contradictions in the law is also one of Fuller’s eight elements in the internal morality of law, and others like Dicey viewed legal equality as indispensable too.[6] Universality appears better able to achieve these competing values. But more fundamentally, as critical traditions from Feminisim to Marxism teach us, Rule of Law arguments are all vulnerable to deeper normative commitments that take some notion(s) of justice as the paramount goal. To offer an illustration from our paper, even if we do assign legal certainty primacy as an Rule of Law value, I am doubtful that it could ever ground a norm like association de malfaiteurs in the Congo. I hold this view since a norm imposed by force as part of a brutal campaign of subjugation and plunder, which operates to inhibit political participation and freedom of expression now, seems plainly unjust.

Markus Dubber’s response takes us down a slightly different track on this justice path, where he argues that “[p]luralistic international criminal law thus becomes international criminal law kind and gentle enough to mollify its protagonists’ latent concerns about its apparent lack of legitimacy”, and then later that, “[i]ncapable of facing its legitimacy challenge head-on, international criminal law instead makes do with professions of concern about ‘pluralism’.” These types of normative concerns are slightly different to those that animated our paper, but they also play off the relationship between law and a conception of justice. So, while I view Legal Pluralism as a helpful concept to ward off “the colonial contrived superiority of European law” and agree with Sasha Greenawalt that, descriptively speaking, Legal Pluralism is inherent, I remain convinced that history is a useful mechanism to illustrate the concept’s shortcomings as a guarantor of justice. I have also become convinced that, far from raising questions that are peripheral to Global Legal Pluralism, international criminal justice poses problems at the concept’s very heart.

I am so thankful for all of the excellent criticism we have received, which has enriched my thinking considerably. I hope that some of the foregoing is useful to others as they grapple with problems of law in the global order.


[1] James (Sákéj) Youngblood Henderson, Postcolonial Indigenous Legal Consciousness, 1 Indig. Law J., 2 (2002), (last visited Apr 19, 2018).

[2] Brian Z. Tamanaha, Understanding Legal Pluralism: Past to Present, Local to Global, 30 Sydney L. Rev. 375, 375 (2008).

[3] H. L. A. Hart, Positivism and the Separation of Law and Morals, 71 Harv. Law Rev. 593–629 (1958); Lon L. Fuller, Positivism and Fidelity to Law: A Reply to Professor Hart, 71 Harv. Law Rev. 630–672 (1958).

[4] Martti Koskenniemi, The Fate of Public International Law: Between Technique and Politics, 70 Mod. L. Rev. 1, 23 (2007).

[5] Boaventura de Sousa Santos, Toward a New Legal Common Sense: Law, Globalization, and Emancipation 89 (2002).

[6] Jeremy Waldron, Legal Pluralism and the Contrast Between Hart’s Jurisprudence and Fuller’s, in The Hart-Fuller Debate in the Twenty-First Century (Peter Cane ed., 2010).

Legal Pluralism and History: A Reply to Critics

Asad Kiyani is an Assistant Professor at Victoria Law School. He is a recipient of the 2017 Antonio Cassese Prize for International Criminal Law Studies for his article Group-Based Differentiation and Local Repression: The Custom and Curse of Selectivity.

One of the many pleasures of developing The Ahistoricism of Legal Pluralism in International Criminal Law has been the opportunity to engage with the work of leading scholars in international criminal law, legal pluralism, and comparative law. In an academic environment that increasingly turns on metrics and particular forms of scholarship and productivity, it is incredibly generous of all of the participants in our mini-symposium to share their thoughts. It represents the best of the tradition of academic engagement that they have taken the time to comment (and many have done so on earlier drafts as well). I should also thank James for being such an excellent partner on this project; I have benefited from his advice on my doctoral work, and it was a pleasure to join him as co-author here. It’s fair to say we both learned a terrific amount over the course of this project, and I enjoyed it immensely.

In what follows, I offer three overlapping responses, addressing whether historicism as a concept is relevant to pluralism in ICL today; whether the examples we study overemphasize certain values and under appreciate others; and finally, what direction the study of pluralism in international criminal law might also take us.

The insights graciously offered by Alexander Greenawalt and Paul Berman, who have written extensively and thoughtfully on pluralism in ICL, and pluralism more broadly, overlaps to an extent with Neha Jain’s comments. Put broadly, the three query whether our work impacts pluralist scholarship (Berman wonders whether ahistoricism as an analytic is relevant to the work of contemporary international legal pluralists), and whether there is a more pragmatic justification for pluralism that is important in its own right even if it does not equate to the idea of ‘value pluralism’.

As a starting point, it should be noted that the umbrella-type frameworks of Drumbl, Greenawalt, and van Sliedregt (amongst others) see pluralism as a feature of a legal structure that allows us to add in different laws and norms around the settled ‘core’ of ICL. Where gaps present themselves, we can find our answers in domestic law, either iteratively by reference to specific rules in particular instances, or comprehensively, by developing an international position on the basis of these surveys. But the questions of what is ‘core’ and how it came to be so are as important as the questions about identifying what should be filled in around that core and how. Part of our work then is to explain and critique this development of this ‘core’; that is ahistoricism coming to bear on pluralism.

The main thrust of this commentary is that our critique is overstated not because it misapprehends the work of legal pluralists, but because our focus on critiquing the aspirational ephemerality of ‘pluralism’ overlooks somewhat different rationales for incorporating domestic rules into ICL. In short, those rules offer something other than ‘value pluralism’ that is of importance: rules that have been tested in (and held up in) various legal systems, and thereby demonstrated their durability and relevance for international law. I will deal with this critique further in the next section, but note here that while such functionalism is not inherently flawed, it may also represent a particular vision of pluralism that is shallower and formalist. That vision is susceptible to papering over the underlying histories that may either be constrictive of the development of ICL, or that give lie to the claim of benefit to be derived.

One of the primary ways in which pluralists see benefits deriving from preserving legal diversity is the idea that it allows for multiple possible rules to be tested in multiple possible systems (a point Berman makes in his work on Global Legal Pluralism, and which Greenawalt cites). This is the laboratory idea: that rules can experimented with in the laboratory of global legal systems, and the testing reveals what rules are appropriate either as universal norms or in specific but circumscribed conditions. What the paper contests is  not the idea that experimentation is possible, or that diversity is valuable, but that when put into practice the experimentation rationale often exhibits an indifference to sources and an indifference to the context in which these rules are implanted. Exposing the colonial history of so many domestic criminal systems says something about both the context of the law, but even more fundamentally something about the idea that there is meaningful diversity being tested.

As Jain points out, the literature on legal transplants is already highly sensitive to the context in which facially similar rules are applied and develop over time, in ways that might serve to reflect local needs and perhaps even local values. From our point of view, this sensitivity to context and history is often absent when engaging in such exercises in ICL. There, the determinative factor often seems to be whether the rule is present, without regard to how it is embedded, even though it is the system along with the rule itself that conditions the outcome of the experimentation. For all the talk of laboratory testing, very little attention is paid to the experiment itself: how the law was designed, what it was designed to do, and what effect it has had. Instead, when the ‘experimentation’ rationale is put into practice, the Darwinian persistence of a law is taken as proof of its success and therefore relevance to international tribunals. To the extent that these concepts of a legal core and legal experimentation are central to explanations of pluralism in ICL, then ahistoricism is also relevant.

*          *          *          *          *

While Berman wonders if our historical enquiries are relevant to the present, it is also suggested that perhaps we take our historical analysis too far. For example, what of complementarity, Greenawalt asks? Should DRC’s legal history disqualify it from accessing complementarity at the ICC? There is no absolute answer to be offered he says, and is wary that we might propose such (presumably in part because this would be what Kevin Davis describes as a pedigree-based distinction). His concern is further important to us because of the danger that our historical analysis will be used to justify the accessibility of complementarity procedures (and therefore an affirmation of sovereignty) to only Western states whose domestic systems procedurally and substantively replicate international ones. This marginalization of Third World states and their legal agency would seem to run against what Jain rightly identifies as our concern with ICL’s continuing exclusion of the Global South as a norm-generator.

We are not of the view that we have fully resolved the balancing exercise Greenawalt rightly says needs to be engaged when decisions are being made about whether to permit or utilize domestic laws instead of international ones. However, that balancing exercise needs to be more attendant to the experimentation that elsewhere seems to justify greater pluralism. When that experimentation process — i.e. the practice of the domestic law —  reveals extensive state-sourced violence (say through association de malfeiteurs in DRC), those factors ought to be considered.

Is this too obvious? Are the examples we rely upon extreme because they express concern about the remote possibility that obvious rights violations might somehow be tolerated in ICL under the guise of pluralism (recast here as an extreme relativism)? Greenawalt suggests that the examples we focus on shed little light because they clearly offend universal values. Davis takes a different tack, suggesting in his intriguing commentary that our suspicion of local law that is based on the substantive values they (fail to) uphold ignores extremely important factors. We overemphasize the values these local laws offend he says, rather than appreciate the principles they support, and in particular we overlook core rule of law precepts.

Davis suggests our pessimism leads to an assessment of national laws that is overly onerous, and queries whether any national law would pass muster from our view. He is concerned with our focus on doctrinal pedigree, which leads to national laws being suspect on their origin (often colonial), evolution (often undemocratic), or application (repressive). This, he says, “is a stringent test, probably too stringent.”

This is a problem in particular because in his view we overlook the rule of law values that accrue when the law is predictable and stable. Disrupting local norms on the basis of international standards represents its own form of imperialism we would do well to shy away from. He suggests that in critiquing contemporary international lawmaking, we bypass a more obvious answer – representative democracy – that might assuage our concerns about the legitimacy of local laws and, by extension, pluralist methodology. We share some of the concerns raised by Davis, but for slightly different reasons which leads to important and different conclusions.

First, as important as predictability and stability may be, international tribunals arguably have a special obligation to not validate illiberal laws under the umbrella of pluralism, complementarity or some other diversity-based argument given the centrality of international human rights norms to their functioning. It is worth noting then, in response to the suggestion of several of our helpful interlocutors that ICL would not tolerate such obvious rights violations and that we are building a case against a problem that does not exist, that the intersection of competing visions of fundamental rights remains unresolved in both national and international criminal contexts.

Whether it be association de malfeiteurs, the conditions of detention of international criminals (including the estimated 10,000 who died while awaiting trial in Rwanda), the culpability of child soldiers, the difference in punishment that may attach to those tried in the Hague versus their collaborators tried in national courts, or how the Akayesu definition of sexual violence was later constricted by the ICTY’s use of comparative analyses of national law, it remains the case that international criminal law struggles to respond to the thorny questions that arise in defining human rights norms.

On that point, it is worth raising the question of universality again (and repeating a reply in a slightly different register): that the examples we attend to are clearly infringements of universal values, and thus we are arguing against no one in particular. Left unaddressed here though is that the sense of what is a universal norm to be protected remains deeply contested, as we show through our analysis of the Draft Code of Crimes Against Peace and Security of Mankind, and with the Apartheid Convention and indeed in the important regional variation between human rights regimes.  Contestation remains on fundamental issues, and the history of human rights doctrine in the post-war era resists the triumphalism often associated with human rights. There is an inextricable historical link to power associated with the concept of universality, and pluralist schemas that seek to declare and then build around those universals must reckon with those histories and dynamics as well.

Second, as suggested earlier, local correspondence — whether it be to cultural values or the expectations of a polity — necessarily says nothing about the supposed twin pragmatic benefits of pluralism that concern themselves with whether local law is suitable as part of international law. The first of these is the idea that thorny questions of international criminal law may be resolved through experimentation with a variety of different possible solutions that are offered by diverse local legal regimes; the second is that we can develop universalist international law by surveying national regimes and identifying nodes of commonality the diverse legal cultures have independently struck upon. Our position is that even where these regimes correspond with local values or expectations, the underlying justificatory factor of diversity may be absent simply because the vectors of history and legal imperialism are such that these various national regimes are largely duplicative of one another and their Angle-European origins. Thus both experimentation and justification on the basis of independent agreement lose resonance given the erasure of diversity that predates contemporary pluralist moves.

Davis further suggests that a consequence of our stringent test is that it leads inexorably towards a demand for international drafting from first principles, which is a process that is likely to be deeply unsatisfactory to all. We agree that such a process would be imperfect, and recognize in particular the inherent imbalances in negotiation and drafting that often replicate international power imbalances . However, without abandoning the possibility of an inclusive, equal negotiating process, we note that there are other possible approaches.

One episode that is missing from our published article in the American Journal of Comparative Law, but included in the longer draft available on SSRN (which was shared with all our invited commentators), is the development of more liberal criminal procedure code in Argentina over an extended period of time. That federal code borrowed from multiple legal systems, including a quite important influence from the criminal procedure code of Cordoba province. Conscious efforts to liberalize Argentine procedure involved legislators setting an agenda and legal experts providing advice as to how to reshape the federal code, and the eventual dispersion of this code as a model for other Latin American countries who similarly sought to liberalize their criminal procedure. This fusion of legal principles, and deliberate consideration and adoption, neither demands rewriting the law from scratch (as Davis suggests is a necessary corollary of our approach) nor does it forgo the important principles Davis articulates in his reflection: that there is value in a law that is stable but not ossified, and which adheres to the expectations of those it binds. In this respect, we find common ground with Davis and so many of our other thoughtful and generous interlocutors from which to argue for a more inclusive pluralism.

*          *          *          *          *

Where then does the study of pluralism in ICL take us? For Dubber, studying pluralism is in part a way to reflect on the legitimacy of ICL: “Pluralistic international criminal law thus become international criminal law kind and gentle enough to mollify its protagonists’ latent concerns about its apparent lack of legitimacy.” Similarly, Mireille Delmas-Marty’s typically profound reflections highlight the ambivalence of pluralist approaches as both resistance and reconciliation.

The propriety of the concern arises first with the very real problem that, per Dubber, ICL purports to judge individuals based on the questionable premise that it is only “their capacity for autonomy, or self-government” that is relevant, “rather than their place on some status hierarchy”. It is compounded by a second insight, that ICL is not law per se, but really a system of policing in the sense of coercive and discretionary governance that adjudicates over and punishes the morally inferior (if not evil) wrongdoers who commit international crimes. The turn to pluralism, Dubber suggests, recognizes and is primarily a mode for managing this legitimacy deficit by softening the Western imperialist legacy of ICL by infusing it with local, values.

Of great interest here is that central debates about how to manage pluralism and complementarity and related diversity-focused concepts only sharpen the hegemonic origins of much of international criminal law. It is only when there is a question of incorporating the law of a non-Western state that the issue becomes particularly thorny, given that first there is a clear familiarity between Western domestic and international law, and that most of the attention of international criminal tribunals is directed at non-Western states.

But we do not need pluralism to tell us that the practice of international criminal law seems imbalanced in the legitimacy-threatening sense that Dubber describes; that story can be told through critical reflections that focus on the hypocrisy of international criminal practice. That being said, understanding ICL as a system of coercive discretionary governance may help us recognize the limits of pluralism in ICL, where the idea of ‘legal’ pluralism seemingly presupposes a structure of largely unified and hierarchical formal law of the state or of international tribunals. What this concept of legal pluralism leads to – and I do not claim that Delmas-Marty endorses this outcome – is the exclusion of non-state legal orders from  the realm of possible responses to international crime. She rightly warns that our historical analysis risks suggesting that legal pluralism must be state-centric.

One of the concerns with our paper, and with other approaches to legal pluralism in ICL, is the difficulty with finding spaces for non-state law. Though criminal law is often conceptualized as necessarily state-sourced, the context of transitional justice suggests that some flexibility may be needed: it is not necessarily the case that the law to be applied in respect of international crime be ‘criminal law’ per se. I do not propose to carve out that space at this juncture, but only to note that we are alive to those concerns. When we refer to legal pluralism’s descriptive origins (in contrast to more prescriptive contemporary modes), it is precisely that history of legal pluralism to which we refer: the history of Sally Engle Merry and John Griffiths and the classical sense of legal pluralism as identifying and describing non-state normative orders as ‘legal’.

Rather than carve out that space, let me take what space remains and sketch out what I argue in a work-in-progress is a fundamental limit on the possibility of non-state law becoming part of pluralism in ICL. I have argued elsewhere that the selectivity problem of international criminal tribunals is most acute in respect of the partiality shown within conflicts, where only  certain political actors are prosecuted, and not others, even though multiple parties are responsible for comparably grave crimes. This political-prosecutorial alignment is a function of the gatekeeper role played by local political authorities in respect of international criminal prosecutions: state authorities control access to witnesses and evidence, and can thus force tribunals to make compromises on which cases to pursue.

In so far as the legal norms to be applied are part of the tacit arrangement that permits international tribunal involvement without threatening the current arrangements of domestic political power, non-state law finds itself on the outside looking in. In several of the conflicts that international criminal tribunals seek to exercise jurisdiction, part of what is at stake is the modes by which competing parties are to be governed. Should the state be built in the Western, liberal, enlightenment model that Dubber suggests attends to international criminal law? Or should there be a prioritizing of customary non-state legal orders and traditions?   For state authorities to defer to non-state legal orders as a means of response to international crimes would arguably translate into a recognition that its political opponents have some political legitimacy. If international criminal law is not the end of political conflict, but only its continuation in a slightly more decorous forum, then such concessions in the legal sphere may well amount to concessions in (and inflammations of) the underlying conflict as well.

In this light, accepting pluralism in its classical sense, as recognizing non-state legal orders as valid normative structures that have salience in contemporary states, requires not that scholars be willing to challenge the histories of international law, but that states themselves be willing to challenge their own histories – to move beyond the ahistorical narratives of conflict that they often promote – in order to give effect to the aspirational qualities that animate international criminal law and practice. Ahistoricism remains germane.

*          *          *          *          *

I thank again James for his collaboration on this extensive project over the last several years, and our extremely insightful commentators – Mireille Delmas-Marty, Paul Berman, Neha Jain, Alexander Greenawalt, Markus Dubber and Kevin Davis — whose analyses have provoked much reflection. It has been a privilege to engage with all of their ideas.


Pluralism and the Alegitimacy of International Criminal Law

Markus D. Dubber is Professor of Law and Director of the University of Toronto’s interdisciplinary Centre for Ethics. Much of Markus’s scholarship has focused on theoretical, comparative, and historical aspects of criminal law.

Stewart & Kiyani’s sweeping article, The Ahistoricism of Legal Pluralism in International Criminal Law, makes an important contribution to the literature on international criminal law, and many other topics besides. It covers so much ground that I’m in no position to comment on much, or even most, of it, but I’ll spell out a few of the thoughts that arose while reading this stimulating paper.

Historical analysis as critical analysis. Given my own (mis-)adventures in historical analysis of law and historical methodology,[1] it’s no surprise that I wholeheartedly agree with the authors that international criminal law, and in fact all of criminal law (comparative and domestic, regional and parochial) needs more history, to paraphrase Christopher Walken. The authors not only insist that it’s worth looking into the history of legal norms but also aren’t shy about putting that historical inquiry to critical use. It’s not just that historical analysis is worth doing for its own sake (I certainly learned a lot from their article) but this historical analysis then allows them to formulate a critique of a position(s) they describe as “pluralism.” And so we see, for instance, that Japanese criminal law is German criminal law and that, in fact, Japanese criminal law swallowed German criminal law whole for a specific (power-political) reason in a specific historical context, as a kind of exercise in self-colonization or preemptive imperialization. I have no idea whether this story holds water, but it makes sense on its face, and certainly enough sense to suggest the need for further inquiry and, more important for the authors’ purposes, the need to take a breath before treating Japanese criminal law as specifically (and characteristically? genuinely?) Japanese criminal law, whatever that would mean, and whyever that would matter.

Pluralism. Why it would matter, according to the authors, is: pluralism! “Pluralism,” in the authors’ telling, comes in a descriptive and a normative flavor. Their paper takes issue with both versions, though the second—normative—one takes the brunt of their critical question-raising (which, in the end, politely pulls its punches and, as a result, never quite evolves into an all-out critique). The descriptive version is “ahistorical” and, more important, misleading because international criminal law isn’t always as pluralistic as it might appear: if historical analysis shows that, to stick with our example, Japanese criminal law is German criminal law then what looks like pluralism is really just duplicatism: German criminal law by different names. But, leaving that aside, even if we maintain—descriptively—a distinction between Japanese and German criminal law (they are, after all, not identical!), then the supposed fact of pluralism doesn’t translate into the desideratum of pluralism. For it turns out that local doctrinal norms may be poor indicators of local social norms because (ignoring a more basic question regarding the relationship between—artificial?—doctrinal norms and—real?—social norms) they may not be truly local, all the way down; upon closer inquiry of the historical kind, they instead turn out to be foreign, alien, “irritating,” “partial,” “dysfunctional.”

International criminal law. But why does pluralism matter? Why is it a good thing that pluralism is a good thing in international criminal law? I take it the apparent interest in, and professed preference for, legal, social, cultural, etc. pluralism (or “diversity”) has something to do with the existential anxiety of international criminal law. Is it legitimate? Is it imperialistic? Is it racist? Socially, cultural, politically, economically, legally hegemonic? Since self-government (most obviously by personal consent), the familiar mode of legitimation of modern (liberal) law, including—at least in theory—modern criminal law, isn’t available in the self-made sui generis, stateless, asovereign, and nonhierarchical realm of international criminal law, its oppressive potential, I suppose, is thought to be mitigated (rather than acknowledged and justified) by showing respect for a vague notion of “pluralism.” “Pluralism” here might mean nothing more than not universalism or not Western/Northern imperialism (leaving conveniently unclear what it is, rather than what it is not). Pluralistic international criminal law thus becomes international criminal law kind and gentle enough to mollify its protagonists’ latent concerns about its apparent lack of legitimacy.

This anxiety about legitimacy turns out to be both appropriate and inapposite, at the same time. Appropriate because there is a very real problem with a system of international criminal law that threatens and inflicts penal violence on persons, insofar as that system holds itself out a system of law in a very particular, liberal, Western, enlightenment sense, namely in the sense of a then-radically new mode of governance grounded in the conception of the subjects and objects of penal power as persons marked by, and only by, their capacity for autonomy, or self-government (rather than their place on some status hierarchy or other, e.g., patriarchalism). This enormous, and prima facie unmeetable, legitimacy challenge is not unique to international criminal law; it applies to any (but only any) political-legal regime that regards itself as committed to this conception of law, domestic, regional, international, global.

Inapposite insofar as international criminal law is, in fact, a misnomer.[2] Instead of conceptualizing international criminal law as a system of law, it may be useful to regard it as a regime of police, where police is understood as that long-pedigreed patriarchal mode of governance on which the modern conception of law cut its teeth during the long turn of the eighteenth century. In this (historicizing…) light, international criminal law appears as an alegitimate regime of discretionary governance in which technically and supposedly asovereign subjects exercise penal power over categorically distinct—and inferior—objects (pirates, ex lex, outlaw, vogelfrei, hostis humani generis, all vanquished, none victorious, evil creatures beyond the pale, for whom punishment under international criminal law is a merciful alternative to a far worse fate, etc.).

Incapable of facing its legitimacy challenge head-on, international criminal law instead makes do with professions of concern about “pluralism.” Sensitivity to pluralism, in this light, appears as part of a general attempt to obfuscate and thereby to manage, rather than to address, the fundamental legitimacy crisis of international criminal law as law…ironically, by highlighting the superior character and benign intentions, if not the benevolence, of its subject-protagonists, however unconstrained by, say, formal “principles of legality” their penal actions may be.

[1] E.g., Markus D. Dubber, The Police Power: Patriarchy and the Foundations of American Government (2005); Markus D. Dubber, “Legal History as Legal Scholarship: Doctrinalism, Interdisciplinarity, and Critical Analysis of Law,” in Oxford Handbook of Legal History (Markus D. Dubber & Christopher Tomlins eds., forthcoming 2018) [SSRN].

[2] See Markus D. Dubber, “Common Civility: The Culture of Alegality in International Criminal Law,” 24 Leiden Journal of International Law 923 (2011) [SSRN]; see generally Markus D. Dubber, The Dual Penal State (forthcoming 2018).

The Inherent Pluralism of International Criminal Law

Alexander K. A. Greenawalt is a Professor of Law at the Elisabeth Haub School of Law at Pace University. His research focuses on criminal law, international law, and the laws of war.

I am grateful to James Stewart for inviting me to participate in this symposium on the provocative, deeply researched, and illuminating article that he and Asad Kiyani have authored. I have enjoyed engaging Stewart’s work over the years—including in a recent article on complicity which greatly benefitted from his writings—and I look forward to reading more of Kiyani’s work.

Let me start by identifying some points of agreement. I share the authors’ conviction that it is worth looking for right answers in criminal law, and I agree that those answers can transcend sometimes superficial doctrinal differences among legal systems (indeed, my above-linked article on complicity advances precisely such a claim). I also agree that it is misguided to romanticize difference for the sake of difference and that one must avoid the casual assumption that national criminal laws always have claim to deep cultural authenticity within the societies they regulate.

As Stewart and Kiyani note, I have used the word “pluralism” to defend a particular approach to international criminal law (“ICL”). Stewart and Kiyani are skeptical about pluralist approaches to ICL, but I struggled when reading their article to understand how exactly their position might impact my own, or indeed the actual practice of ICL. In part, this is because I find their account of pluralism somewhat elusive. Sometimes the authors seem to suggest that pluralism means blind deference to every doctrine of domestic criminal law no matter how oppressive. Elsewhere (among other approaches), they associate pluralism with a method by which ICL might arrive at consensus international standards—namely, by embracing rules derived from national criminal law that reflect a “real degree of inclusive, plural, cosmopolitan values” and are “doctrinally plural within a diverse, conflicting, sometimes inter-penetrating system of criminal law.” I am not aware of any scholar who has advanced the first position and the second claim is not one that I would associate with pluralism in the sense that I have used the term, although it does resemble a standard methodology embraced by international criminal tribunals attempting (sometimes misleadingly as the authors observe) to marshal support for universalist claims about the content of ICL.

I am curious, as well, about how exactly Stewart and Kiyani’s broad rejection of pluralism informs the actual practice of international criminal justice institutions. Take, for instance, the International Criminal Court’s (“ICC”) complementarity requirement which sometimes requires the Court to defer its jurisdiction in favor of criminal justice at the national level. Suppose the ICC is pursuing a suspect in the Democratic Republic Congo (“DRC”) who is being investigated at the national level for the same conduct. Do the authors believe that complementarity can or should apply in a case like that, or should the systemic, historically rooted deficiencies they identify in the DRC’s criminal law categorically preclude deference in all cases? Or perhaps complementarity itself should be abandoned because it inherently introduces the problems of pluralism that the authors explore? How then should a global system of ICL proceed if it rejects domestic prosecutions of international crimes?

The most obvious solution I can glean from Stewart and Kiyani’s article is that the world must coalesce around a single, best, cross-cultural and universal approach to all questions of criminal law and procedure, one that is devoid of power politics or cultural bias. If the world can achieve that, then I agree that the normative defense of pluralism becomes quite unconvincing indeed. But if that is the claim, then the authors must establish far more—and overcome far greater hurdles—than they do in this article.

For example, even if the world were to coalesce around the unitary theory of perpetration that the authors advocate (rejecting formally differentiated modes of criminally participation) I don’t see how the general choice of a unitary versus differentiated model is especially important in comparison to the host of other choices affecting guilt, innocence, and degrees of culpability that must be decided upon under either model.

Perhaps, however, one should read Stewart and Kiyani to advance a more modest claim that universalism is something to work for, that some universal answers are accessible, and that the blind embrace of pluralism is both threatening to this project and harmful given the colonial origins and distortions evident in many criminal law systems. If that is the claim, then I agree with the central thrust of the argument, but my own defense of pluralism does not proceed from the assumption that national criminal law demands blind deference or that it necessarily has some claim to deep cultural authenticity. To me, the central question is not one of universalism versus pluralism, but of how to manage the pluralism that unavoidably has accompanied the establishment of ICL. I also believe that my own framework accommodates many of the concerns that the authors raise, and so I will attempt in the remainder of this post to sketch out some of my own thoughts on this issue and attempt to see how they may accommodate at least of some of the concerns raised by Stewart and Kiyani.

I begin with the fact that there is variety in the criminal law. I agree (even outside the colonial context) that variety need not have any deep cultural basis. While differences among states may sometimes reflect important cultural fault lines, they are just as likely to reflect arbitrary, and sometimes pernicious, historical vestiges or the influence of judicial interpretation.

International law, in its current state, is hardly agnostic about this diversity. The body of international law that most broadly regulates these choices is international human rights law (“IHRL”). Many of the examples that Stewart and Kiyani invoke—judicial bias after World War I, restrictions on freedom of association, anti-blasphemy laws—reflect straightforward human rights violations. As a matter of IHRL (as the authors themselves acknowledge), these examples already offend universal values.

ICL, by contrast, presents only a limited intervention by international law into matters of criminal law. ICL evolved, in my view, not to harmonize national approaches to criminal law, or even to define and regulate uniquely “international” offenses in some qualitative sense, but instead to enable institutional interventions that counteract unique obstacles to the prosecution of certain especially grave offenses. The point of prosecuting Holocaust crimes at the International Military Tribunal at Nuremberg, for example, was not to reject German criminal law’s general judgments about how to define and assign liability for murder in ordinary cases. The point was to counteract and ensure accountability for the specific ways in which Nazi Germany had made an exception to those judgments by converting the state into a system of domestically authorized mass extermination. I believe that ICL speaks the most strongly when it works to justify and define the scope of such interventions, and it speaks more tentatively (although not entirely without authority) when it addresses more general questions about what it means to be a criminal.

How does this way of looking at things interact with Stewart and Kiyani’s specific claims? In previous work I have explored the example of the Erdemović case at the International Criminal Tribunal for the former Yugoslavia (“ICTY”) involving a soldier who was forced under threat of death to participate in a firing line that massacred Bosnian Muslim civilians.   The ICTY Appeals Chamber rejected the defense on (I believe unpersuasive) policy grounds, notwithstanding the unresolved nature of the question under international law and the fact that many states, including all the republics of the former Yugoslavia, take a more permissive approach to duress.

I do not know whether Stewart and Kiyani have access to an optimal, universalist approach to duress. My position is that the ICTY should have looked to Bosnian law under the circumstances. In advancing this argument, I do not assume that the Bosnian criminal code reflects a deeply embedded and culturally specific approach to duress. Rather the Bosnian law reflects a fairly standard civil law approach that the state inherited from its prior membership in communist Yugoslavia. I think that the law of duress presents unavoidably hard questions, that all the dominant legal approaches to the issue are problematic in some respects, and that the Bosnian law falls within a range of reasonable disagreement that neither offends core human rights value nor the specific purposes of ICL. In a case like that, I don’t think that ICL has a strong interest in overriding the local law to deprive the accused of a defense which his society affords to other similarly situated persons. But even if one disagrees with that specific conclusion, the same problem arises in other guises. Suppose that the Bosnia’s own courts were prosecuting Erdemović for international crimes. Must those courts also prefer the ICTY’s approach to duress over their own? And what about ordinary domestic prosecutions for non-international crimes? Wherever one draws the line, the basic problem remains: absent universal, global agreement on how to handle duress, some defendants must be treated differently from others based on perhaps arbitrary, non-culpability driven reasons. Similar issues arise with respect to other standards of responsibility, principles of sentencing and so forth. Must ICL take an absolute position on every one of these questions, no matter how tangential the doctrinal issue is to ICL’s core mission?

At the same time, I agree (and have argued) that is desirable to have a single, comprehensive set of ICL rules that are available for use by a global court like the ICC and in other contexts where reliance on domestic legal principles proves problematic or undesirable. There are many reasons to favor this restult, including problems that Stewart and Kiyani explore as well as concerns having to do with clarity and administrability. But I do not think that those reasons exhaust all the procedural contexts in which ICL finds itself enforced. In the end, implementation of ICL involves a number of situation-specific considerations that cannot be resolved by broad appeals to either universalism or pluralism. Instead, there are a balance of factors to consider. Stewart and Kiyani convincingly elucidate some important factors that may sometimes arise. But they do not, in my view, eliminate the need for such balancing.




What Does it Mean for International Criminal Law to Conform to Local Values?

Kevin E. Davis is the Beller Family Professor of Business Law at New York University School of Law. His current research is focused on contract law, anticorruption law, and the general relationship between law and economic development.

Does international criminal law conform to local values? How can we tell? Why should we care? In their fascinating and illuminating article, The Ahistoricism of Legal Pluralism in International Criminal Law, James Stewart and Asad Kiyani focus on the first of these three questions. Inevitably though they end up addressing the other two questions as well. I say “inevitably” because it would be impossible for them to answer the first question about conformity without answering the methodological question of how to identify it, at least implicitly. And to choose among the variety of ways of assessing conformity they have to determine the purpose of the assessment, again, if only implicitly.

Stewart and Kiyani’s main interest is in debunking the view that doctrines drawn from national criminal laws necessarily conform to local values. The promised payoff is insight into the important issue of whether international criminal law ought to be pluralistic or universal — or as I prefer to put it, should it be uniform or not? They seem to believe that their debunking exercise undermines the case for non-uniform approaches. However, this only follows if a) they have an accurate test for conformity with local values and b), the case for a non-uniform international criminal regime depends in a significant way on the results of that test.

The way that Stewart and Kiyani test for conformity suggests they believe that the main justification for non-uniform approach to international criminal law is political legitimacy. I believe they are mistaken about what political legitimacy demands in this context. I also believe they neglect at least two other plausible justifications for a non-uniform regime: respect for individuals’ rights to be governed by familiar doctrines and effectiveness. For all these reasons I think that Stewart and Kiyani’s historical analysis does less damage to the case for non-uniform international criminal law than they make out. Let me explain.

Political legitimacy

The Ahistoricism of Legal Pluralism in International Criminal Law is organized around a series of case studies that examine whether various criminal law doctrines do or do not conform to local values. In each of the negative cases the reasons for concluding that there is a lack of conformity are slightly different. In the Democratic Republic of Congo the problem is that the colonial doctrine has remained in force without being amended by any local actors. In Pakistan the concern is that local amendments to a colonial doctrine were made by repressive dictators. Etc.

Curiously, nowhere do they test for conformity by referring to direct observations of local values. Stewart and Kiyani could have tried to ascertain the amount of popular support for individual doctrines by examining, say, survey data or commentary in the popular press. It is unclear why they eschew this seemingly obvious approach.

Instead, Stewart and Kiyani focus on assessing the pedigree of the doctrines. When a doctrine is damned it is because: it was originally drafted by a colonial power; if it was revised by local actors they were working at the behest of an un-democratic government or for some other reason failed ensure popular participation in the lawmaking process; or, it has been used, either locally or elsewhere, either as a tool of repression or in a partial fashion.

This is a stringent test, probably too stringent. Given the prevalence of legal transplantation and legislation by non-participatory, un-democratic or repressive regimes, most national criminal law doctrines seem likely to fail. More importantly for their ultimate argument, I believe that even the universalistic doctrines Stewart and Kiyani seem to favor would shrivel under this kind of scrutiny. What international drafting process is likely to ensure participation from all affected people, without intermediation by undemocratic or even repressive governments?

Focusing on whether the formulation of a doctrine involved democratically elected leaders and public participation seems to be a way of determining whether the doctrine in question is politically legitimate. If that is the goal then I think Stewart and Kiyani go astray because they underestimate both the costs of participation and the appeal of an alternative method of taking into account the views of the relevant public, namely, representative government. Drafting legislation from scratch and in full view of the public is a time-consuming exercise whose results may well leave everyone unsatisfied. Participation by all affected individuals is unrealistic, unnecessary and inefficient. Representative government, meaning participation by officials who are accountable to the people, may be sufficient to ensure that the public’s views are respected, and is far less costly.

Civil rights

Another element of Stewart and Kiyani’s test for whether a doctrine conforms to local values is whether it has been used to facilitate repression. That is what I take from their suggestion that the doctrine of association de malfaiteurs fails to conform to local values in France—where it originated—because it has been used repressively against people suspected of terrorism.

By suggesting that even contemporary French criminal law might be a problematic source for international criminal law, Stewart and Kiyani shift the debate away from political legitimacy to a new terrain, individual civil rights. They clearly believe that a test which excludes doctrines that violate civil rights will tend to be biased in favor of universalistic doctrines.

Stewart and Kiyani neglect a countervailing factor though: a non-uniform approach may result in adoption of doctrines that conform to local expectations, assuming that exposure tends to breed familiarity. When it comes to legal doctrine, familiarity is intrinsically desirable, even if, as Stewart and Kiyani might point out, it can also breed contempt. Doctrines that conform to people’s expectations are desirable, all else being equal, because they contribute to legal certainty and avoid situations that are tantamount to ex post facto lawmaking.


Finally, Stewart and Kiyani neglect another type of justification that might weigh in favor of a non-uniform approach to international criminal law: effectiveness. Suppose the effectiveness of criminal law doctrines varies depending on the circumstances. For example, harsh punishment for inchoate offenses might be reasonable in a society where evidence of completed offences is difficult to obtain, but less reasonable where evidence is more accessible. Similarly, a law against blasphemy might be reasonable in a society riven by religious tensions, but unacceptable in a society where religious toleration is better entrenched. In other words, different doctrines might be optimal in different circumstances. A non-uniform approach to international criminal law can accommodate this kind of diversity. Note that if effectiveness is the primary concern the critical issue will not be whether a given doctrine conforms to local values, but rather whether it fits local conditions, which is a slightly different test.

I develop this argument in favor of non-uniform “contextualist” approaches to lawmaking at greater length elsewhere, including in my writing about transnational – and international – anti-corruption law.[1]


For all these reasons, tracing the tainted history of national criminal law doctrines is an interesting and valuable exercise but does not provide compelling reasons to reject non-uniform approaches to international criminal law.

[1] See Legal Universalism: Persistent Objections, 60 University of Toronto Law Journal 537 (2010) and Between Impunity and Imperialism: Rethinking Transnational Anti-Corruption Law (forthcoming).

Legal Pluralism as a Dynamic Process in a Moving World

Mireille Delmas-Marty is the Professor of Comparative Legal Studies and Internationalisation of Law at the Collége de France. She is the author of Ordering Pluralism, A Conceptual Framework for Understanding the Transnational Legal World (Hart, 2009) and Towards a Truly Common Law – Europe As a Laboratory for Legal Pluralism (CUP, 2002).

I fully agree with the starting point: « International criminal law is legally plural, not a single unified body of norms ». ICL is an excellent example to illustrate what is called “Legal Pluralism” (LP), especially when considering « the complex dance between international law and domestic criminal law ». I also agree with the authors’ insight that this dance “takes place on at least five levels: (1) within national courts, trials involving international crimes often employ their own local criminal law standards rather than the international law equivalent; (2) international courts sometimes follow a particular national system in interpreting ICL rules; (3) international courts often survey then synthesize a wide selection of national rules to demonstrate widespread support for their favored approach; (4) at times, international statutes, treaties and national legislation define the same ICL concept differently and finally (5) judicial bodies that interpret ICL occasionally disagree amongst themselves about the interpretation of the same body of law”.

Before concluding that the dance is “improvised  rather than choreographed”, I propose considering diverse observable dynamics as a “work in progress for a world in process,”[1] leaving the question of “what might count as law” open.  It seems to me that by choosing an exclusively state-centric perspective while the world is moving, the authors risk confining themselves to a state-centric, modern representation of LP which identifies the rule of law with state law and makes the concept necessarily oxymoronic.  To be “legal”, pluralism must be integrated into a state order which presupposes a structure that is more or less unified and hierarchical (unified because it is hierarchical) whereas pluralism implies (as traditional international law) the plurality of systems of law and their independence; it separates or juxtaposes different systems but does not organize them.

Beginning in the post-war period, globalization has opened a “post-modern” perspective which does not necessarily imply the disappearance of states, stimulating a somewhat paradoxical sovereign withdrawal, but increasing state interdependence. It leads to an unprecedented development of international law at supra and trans-state levels and even more surprisingly at the infra-state level. It has resulted in a whole that is plural, complex and evolving; a true challenge to the notion of the rule of law.  To address such a challenge, the goal of bringing order to pluralism involves a choreography which combines three interacting processes attempting to reconcile universalism and pluralism.

Three interacting processes

One evokes, in increasing order of integration: horizontal coordination, for example, across “circular models” of one legislator to another, or judicial dialogue; harmonization, which brings differences together rather than abolishing them; and lastly unification as a fusion when differences are incompatible.

The process of coordination facilitates spontaneous and horizontal reconciliation of differing systems.  It is without a doubt the most widely practiced process, from unilateral borrowing to cross-exchange, with examples of “cross fertilization”; but it is also the most uncertain since it relies on the good will of state actors.

The process of harmonization is more restrictive because of its reliance on a hierarchy between the national and international levels, which is itself complex because the hierarchy is flexible depending on its origins at the national or international level.

From the national level to the international level, an upward dynamic of reconciling differences demonstrates the importance of history and memory in illuminating comparative law.  We can only understand it when using critiques of “ahistoricism of legal pluralism” from preceding authors as a starting point.  Neglecting the historical dimension may have the effect of “condoning illegitimate or dysfunctional law”.  But it is also necessary to take into account the ambivalence of comparative law, in the sense that it can be invoked equally to resist as to reconcile differences.  As such, the example provided by the authors of Argentinian criminal procedure as being the “exception which confirms the rule” can also be understood as a reflection of this ambivalence.

It is also crucial to consider the downward dynamic that is at work from the international level to the national level. As discussed, international law is no longer simply inter-state and takes a supra-state status in its universal calling, but pluralism may reappear across diverse international law techniques allowing for “contextualising” the universal.  From this point of view, international criminal law is particularly interesting. Even to demonstrate the notion of a unified principle such as crime “against humanity”, the requirement of “complementarity” can be reintroduced at the national level.  We know that the International Criminal Court can only function competently in contexts where states in question lack the willingness or capacity to judge themselves. To exercise their competence, they must transpose international criminal law to domestic law and this transposition does not exclude a certain margin of national interpretation.

In noting that in both cases, whether ascendant or descendant, harmonization calls for “fuzzy logic”, cleaning up national margins precisely by replacing a requirement of conformity with one of compatibility.  This softening is not always possible.

Whereas the third process, that of unification, solidifies the hierarchy of norms by imposing a single norm, but risks abolishing any hope of pluralism.  Barring the renunciation of the unilateral process of transplantation – the pernicious effects of which have been demonstrated in colonialism –, the benefit of introducing a process of hybridization, which is multilateral and reciprocal.  In fact, it is rare to see such a practice across all scales on the planet.  Even at the level of the ICC, decisions that surround defining attempt, complicity and criminal procedure are informed directly by Western culture.  As such, hybridization is often limited in procedure to combining accusatory and inquisitory models.  On the other hand, crimes against humanity are broadened by notions such as apartheid, and criminal procedure can be enriched with alternative such as truth and reconciliation commissions.  Even more broadly, international law can contribute to the duty of remembrance by creating a duty to anticipate the intentions of future generations (for example, by creating, across cultures, a crime of ecocide).

Reconciling universalism and pluralism 

In a time when globalization reinforces interdependence from one end of the earth to the other, we must discuss the UNESCO Declaration of Cultural Diversity (Nov. 2001).  This declaration (extended through a 2005 convention) made clear the tension between two apparently contradictory principles of international law: on one hand, the universality of values proclaimed by the “Universal” Declaration of the Rights of Man of 1948; and on the other, the principle of diversity of culture and religions, thus of pluralism, making up a “common heritage of humanity” per the UNESCO declaration.  It is not an accident that this principle was posed in the days following September 11th 2001, which is sometimes characterized as a collision of cultures.

Some have interpreted the principle of diversity of “setting the fox among the chickens”.  But neither the human rights, nor international sanctions against the most serious crimes are comparable to a chicken coup; rather to a winding and unpaved road at risk of becoming an impasse if it leads to a complete relativism where all values ​​are equal. The 2005 Convention does not assert that all values are equal.  It poses the principle of “equal dignity and respect of all cultures” but attempts to push back at the risk of relativism by affirming as a first principle the respect of the rights of man “as set out by the universal declaration or guaranteed by international law”.

While this objective is set out, neither the Universal Declaration of Human Rights nor the UNESCO convention provide a method for reconcile pluralism and universalism.  Our hypothesis is that by rationalizing processes of coordination, harmonization and reciprocal hybridization, it is possible to put a coherent and diversified dynamic in place.  In a moving world, this dynamic must serve not only to construct a “truly common law of humanity”[2] based on the model of the rule of law, but also to avoid the pluralism which divides and universalism which leads to uniformity. We aim to create a path which can be called “ordering pluralism” or “contextualized universalism”, somewhere between the relative and the universal.



[1] See M. Delmas-Marty, « The ICC: a Work in Progress for a World in Process », in M. M. de Guzman and D. M. Amann, Arcs of Global Justice: Essays in Honour of William A. Schabas, Oxford University Press, 2018.

[2] M. Delmas-Marty, Towards a Truly Common Law – Europe As a Laboratory for Legal Pluralism, CUP, 2002. See also her current research project “Vers un jus commune universalisable?”:

Understanding Global Legal Pluralism and International Criminal Law

Paul Schiff Berman is the Walter S. Cox Professor of Law at the George Washington University Law School. He is the author of Global Legal Pluralism (CUP, 2012)

James G. Stewart & Asad Kiyani argue that the core human rights norms embodied in international criminal law (ICL) should not be lightly pushed aside in the name of cultural diversity or purported deference to the criminal law norms of the local communities where crimes might have occurred. First, they suggest, whatever these local norms might be, they might not actually be embodied in the communities’ formal criminal law doctrine because that official doctrine may instead reflect historical contingency or may be the product of hierarchy. Second, they contend, even if domestic criminal law doctrine does sometimes accurately reflect local norms, we might nevertheless choose to ignore those local norms and insist on the primacy of ICL norms in order to make sure that those who commit heinous crimes are sufficiently punished.

I do not disagree with either of these points, but I confess that I am a bit mystified that the authors think that by embracing these arguments they are somehow challenging theories of Global Legal Pluralism or revealing the “ahistoricism” of such theories. I readily admit that I am not an ICL expert, so perhaps there are scholars who invoke legal pluralism to reject the imposition of international norms in all circumstances or who celebrate the “local” even when what is truly local is itself contested and is perhaps the product of hierarchy. I am not aware of such scholars, but if they exist then Stewart & Kiyani are right to offer their critique.[1]

But given that the authors name me in their introduction as one of the scholars of Global Legal Pluralism to whom they are responding, I think it is necessary to point out that nothing in my approach contradicts the arguments Stewart & Kiyani make. In this brief response, I will try to explain.

To begin, it is important to recognize that the focus in my book Global Legal Pluralism: A Jurisprudence of Law Beyond Borders was not on the substantive norms embodied in legal doctrine at all and certainly not substantive criminal law norms. Instead, I argued for procedural mechanisms, institutional designs and discursive practices that might take into account the multiple communities that might have a stake in a given decision and therefore better manage the inevitability of legal pluralism. Accordingly, to the extent I discussed ICL at all, I focused on the design of criminal law institutions and procedures, not the content of criminal law doctrine or norms. For example, I discussed the possibility that hybrid tribunals comprising both domestic and international actors might sometimes be preferable to either a fully international tribunal, on the one hand, or a fully domestic one, on the other. But I never addressed the content of the norms that might be applied by such tribunals. And I certainly never argued that the ICL norms regarding, for example, genocide, war crimes, or crimes against humanity should be jettisoned in favor of “local” criminal law.

This distinction between substantive doctrine and procedure seems crucial in considering the points Stewart & Kiyani make. For example, the authors rightly note that we might not want to defer to the substantive law of communities that themselves received their substantive law through colonial imposition. That seems perfectly correct. But even so, we might still choose a hybrid court—with both local and international judges and local and international prosecutors—in order to maximize local participation while still protecting against the possibility of local capture. Or we might choose a trial process that incorporates certain local forms of testimony or practice within it, regardless of the derivation of those practices.

More broadly, my version of Global Legal Pluralism is emphatically not a celebration of the local in all circumstances and under all conditions. Instead, I staked out a position that was both cosmopolitan and pluralist and argued for procedures that would strike a middle ground between universalism on the one hand and localism on the other. As such, far from rejecting all international norms, I provided numerous examples of so-called “local” actors using international norms to gain leverage within their own domestic legal or political systems that they would not otherwise have had. Further, I made clear that even if pluralism is a potential value to be considered in designing institutions, procedures, and practices, it is certainly not the only potentially relevant value. Thus, I wrote that whatever value there might be in trying to defer to a norm of a community, that value must be weighed against competing values, for example the values embodied in universal human rights. Accordingly, I made clear that a procedural mechanism that gave voice to pluralism might still be rejected on other grounds. My point was only that the value of pluralism should always be factored into the institutional design decision, not that it should always win.

So how might my approach translate into ICL during a period of transitional justice after mass atrocities? First, as mentioned above, I believe it is useful to consider multiple kinds of procedures and institutions and think about which sorts of procedures and institutions might best generate buy-in from among the many communities that might have a stake in the new society being formed. Second, I believe it is also at least worth thinking about how one might incorporate local norms and procedures into any given transitional justice mechanism to the extent possible, while still remaining faithful to the other core substantive values embodied in ICL.

But under no circumstances would my vision of Global Legal Pluralism so reify the “local” that it would require international norms to give way to local norms just because those norms were local. After all, Global Legal Pluralism not only recognizes that the content of norms is always contested by multiple communities; it also recognizes that the international community is one of those communities participating in the contestation. Moreover, pluralists of all people recognize that there is no single “local” norm anyway and that multiple local actors have a wide variety of interests and are therefore always contesting the content of local norms. In addition, as Stewart & Kiyani rightly point out, the “local” can be as much the product of hierarchy (or colonialism) as the international. So there is nothing inherently good about the local or evil about the international. Nor should we assume that the value of giving voice to plural voices always trumps every other possible value. After all, a lynch mob may be an expression of local norms, but it need not be celebrated or deferred to for that reason.

So, in the end, from my perspective Stewart & Kiyani end up arguing against a caricatured portrait of Global Legal Pluralism that I for one have never embraced and certainly have no wish to defend. To the contrary, if the focus shifts from doctrine to procedure, then I think my vision of Global Legal Pluralism actually comes out pretty close to where Stewart & Kiyani do: emphasizing diversities of values rather than doctrines in designing international criminal law mechanisms. But a pluralist approach would be sure also to include the value of broad-based participation by multiple communities and to think carefully about how best to foster such participation in all its forms.

If Stewart & Kiyani want to argue that local participation and buy-in is completely irrelevant to ICL, then I think we can have a legitimate debate. But I don’t see them to be arguing that point. If instead they simply want to make sure we don’t treat the codified criminal law of a community as an automatic proxy for the actual norms of that community, one would think they should get no argument from pluralists, who after all have always started from the premise that formal codified law is not the only relevant determinant of community norms. Alternatively, if Stewart & Kiyani want to argue that we shouldn’t always reject international norms in favor of local ones, again it seems hard to imagine they will get much argument from those who espouse Global Legal Pluralism. Global Legal Pluralism recognizes that what is local and what is global tend to be mutually constitutive anyway, as norms seep back and forth from local to global and vice-versa, multiple actors import and export norms for strategic purposes, and international law becomes “vernacularized” and transformed in local settings.

In short, Stewart & Kiyani have done wonderful research and have provided important insights regarding ICL. But it is not at all clear that their work has much to say about Global Legal Pluralism, at least not the Global Legal Pluralism with which I am familiar and to which I subscribe. Thus, it seems to me that the article would be better framed simply as a contribution to a discussion about the normative content of ICL, rather than as a response to Global Legal Pluralism. Re-framed in that way, I suspect that it is an important contribution to the theory of ICL and can be celebrated on its own terms. And as a pluralist, I would be happy to join in that celebration.

[1] I note that my reading of the scholars cited by Stewart & Kiyani in their introduction do not suggest that those scholars make the sort of broad ahistorical claims Stewart & Kiyani are criticizing either. For example, Alexander Greenawalt does not argue that ICL must always give way to local domestic law norms, only that such deference may sometimes be appropriate. Likewise, Elies van Sliedregt actually argues for maintaining a universalist core to ICL even while trying to accommodate some pluralism in application. Thus, it is difficult to see which pluralists Stewart & Kiyani are criticizing.

A Comparatist View on Doctrinal Diversity in International Criminal Law

Professor Neha Jain is an Associate Professor of Law and a McKnight Land-Grant Professor (2016-18) at the University of Minnesota Law School. Her scholarship focuses on public international law, criminal law, and comparative law.

In their provocative article on The Ahistoricism of Legal Pluralism in International Criminal Law, James Stewart and Asad Kiyani challenge the increasingly popular view that legal pluralism, as reflected in doctrinal diversity, is a value that ICL should strive towards. For Stewart and Kiyani, in embracing doctrinal heterogeneity as a marker of cultural cosmopolitanism, ICL scholars, practitioners, and courts, have been barking up the wrong tree and have ignored the socio-historical context of criminal law doctrines in both domestic and international law. Instead, universal norms that represent a multitude of cultural values and political interests might be better vehicles for ICL to become a genuinely value plural enterprise.

Stewart and Kiyani assemble an arsenal of theoretical arguments and case studies to demonstrate their thesis. One of the main strengths of the article is the astonishing variety and depth of national and international doctrines relied on to illustrate the perils of using doctrinal diversity as a proxy for cultural pluralism. Moreover, in a refreshing departure from the bulk of ICL and comparative criminal law scholarship, several of their national examples, such as blasphemy laws in Pakistan, are from typically unrepresented regions of the world and are analysed with rigour and sophistication.

A work of such sweeping knowledge and bold assertions nonetheless comes with its own set of challenges, some of which I can only address in broad strokes in this short commentary. First, Stewart and Kiyani’s main focus, as they acknowledge, is on state law. More specifically, the primary target of their criticism is doctrinal state law and the borrowing of this law by international criminal law actors without regard to its history and context. This analysis could have benefited from more engagement with comparatist scholarship. A vast and, by now, fairly standard, body of comparatist scholarship is dedicated to the need to move beyond doctrine. Indeed, much of the literature of legal traditions, legal formants, and variants thereof, has repeatedly emphasized the limitations of doctrine in ways that are entirely compatible with  Stewart and Kiyani’s thesis. This leads to a question – given that almost no serious comparatist today would adopt such a narrow approach to legal systems, who exactly is championing such a blinkered vision of “diversity” in ICL?  ICL scholars, as Stewart and Kiyani point out, are usually “more considered in their pluralism”.

Are the true culprits then international tribunals? The authors discuss, for example, the adoption of conspiracy as an inchoate mode of liability at the Nuremberg and Tokyo trials. However, the only reference to municipal legal surveys comes in the form of the document prepared by the American Chief Prosecutor at Tokyo to refute the argument that the conspiracy doctrine was unique to Anglo-American law. The authors do not dispute that the Prosecutor was correct in his geneaology of the national laws he cited on conspiracy – their objection is to his omission of their violent history in some jurisdictions. This is an entirely sympathetic position, but it does not serve to further the thesis of the paper. Whether introduced through consent, indifference, or violence, a rule or doctrine that forms part of the criminal law of a legal system can hardly be said to be completely unknown to it. Neither do Stewart and Kiyani claim that the Tokyo tribunal ultimately went on to rely on this survey to make the further leap that conspiracy therefore embodied value pluralism. Instead, they rightly criticize the way in which the court applied the doctrine to the facts of the case, which had little to do with whether or not conspiracy was a culturally cosmopolitan legal construct.

Moreover, even if international courts are conducting national surveys of criminal law doctrine in order to formulate ICL principles, are they truly doing so in the name of cultural pluralism, or simply because national doctrine is an easily accessible source of legal concepts and ideas that can be plugged into an international criminal law regime ridden with gaps and contradictions? In Stewart and Kiyani’s analysis of the ICTY’s jurisprudence on Joint Criminal Enterprise, there are a few references to the reasoning adopted by the Appeals Chamber in the Tadić case that would support the former position. However, this remains a mostly isolated example in an extensive article that is premised on the argument that ICL practitioners treat doctrinal diversity as a substitute for legal pluralism.  In other words, while the authors draw on numerous instances where ICL practitioners rely on domestic legislation and case law to formulate ICL principles, it is far from clear that the practitioners themselves look upon this exercise as furthering value pluralism or justify it in those terms.

Relatedly, at the outset of the article, Stewart and Kiyani recognize “the inevitability of the reliance on foreign domestic criminal law by international lawyers and institutions.” However, since their underlying assumption is that foreign national doctrine is being misappropriated by international criminal law institutions to tout ostensibly pluralistic values, they fail to distinguish this superficial transplantation of domestic criminal law into the international realm, from potentially legitimate and useful reliance on national doctrines. Yet again, comparative and international criminal law scholarship could have provided a rich source material to explore this possibility. ICL scholars have, for example, criticized the practice of international criminal tribunals whereby courts conduct cursory surveys of domestic criminal laws to propose “general principles of law”. However, this does not have to entail a complete abandonment of references to domestic legal rules found in statutes and case law. These formal sources of law may still be useful for an international court, not as embodiments of cultural pluralism, but as models of legal principles that have been tested in the laboratory of domestic legal systems, and that can serve as an inspiration for rules and principles that are tailored to the requirements of the international criminal law regime.

Finally, while Stewart and Kiyani are clearly in their element in highlighting and exploring case studies from different domestic jurisdictions and international courts, at times, the link between their abstract thesis and the case study is quite tenuous. For instance, the authors claim that the failure of the domestic German war crimes trials after World War I can be explained in part by their adoption of a German criminal procedure that was in accord with native German values, but alien to other legal systems and to foreign audiences. The experience of the Leipzig war crimes trials is adduced to demonstrate that “[e]ven when criminal doctrine is a safe proxy for social and cultural values within the community it governs… this fact alone is not a sufficient condition for privileging it in a contest between normative orders.” This is a perfectly reasonable proposition except that on the authors’ own account, it is difficult to say that the criminal procedural innovations that were controversial at Leipzig should in fact be regarded as a proxy for German social and cultural values. The two specific procedural rules that Stewart and Kiyani cite were both introduced as deviations from normal German criminal procedure in 1920 and 1921. Given the care that the authors otherwise take to refrain from equating the acceptance of every single doctrinal rule with a country’s culture, it is a stretch to then argue that a recent procedural amendment introduced in the wake of a devastating war loss, and in circumstances of exceptional political tension, was reflective of German community values more broadly. Equally, with the intense resentment caused by German actions in World War I, not least in France and Belgium, it is quite likely that that no matter what procedural model had been adopted at Leipzig, acquittals in large numbers, especially of high profile defendants, would have been unlikely to secure their approbation.

None of these critical questions, however, should be taken to undermine the importance and urgency of Stewart and Kiyani’s central claim: an ICL that continues to be biased, discriminatory, and myopic will have little claim to global legitimacy and the burden of rectifying the many parochialisms of ICL is a task that falls upon all of us who care about and practice this enterprise. The article is a welcome and impressive contribution to this vital conversation.