Content uploaded by Frédéric Mégret
Author content
All content in this area was uploaded by Frédéric Mégret on Oct 22, 2015
Content may be subject to copyright.
Electronic copy available at: http://ssrn.com/abstract=1537213
1
TOO MUCH OF A GOOD THING?
ICC IMPLEMENTATION AND THE USES
OF COMPLEMENTARITY
CONTENTS
I. The Scope of Implementation ................................................................................................................................... 3
II. Implementation as a Trojan Horse? .................................................................................................................. 6
A. Abusively transposing ICC specific features to Domestic Jurisdiction ........................................... 7
B. Mandating what should only be an option ................................................................................................ 8
C. Requiring that which is not strictly connected to complementarity ........................................... 10
D. Requiring too much identity with the Rome Statute ........................................................................... 10
E. Going further than the ICC .............................................................................................................................. 12
III. The reality of complementarity ........................................................................................................................ 14
A. The Practice of complementarity ................................................................................................................ 14
B. Theoretical Tensions ........................................................................................................................................ 15
IV. Some Illustrations ................................................................................................................................................... 17
A. Failure to Exercise Jurisdiction .................................................................................................................... 17
B. Failure to Secure a Conviction ...................................................................................................................... 19
Conclusion ................................................................................................................................................................................ 21
Electronic copy available at: http://ssrn.com/abstract=1537213
2
Complementarity has emerged in the last decade as a formidable leverage to obtain
implementation. In fact, one could argue that the virtual effect of complementarity via
implementation has outweighed, at least so far, the actual jurisdictional operation of
complementarity. Complementarity has become part of the way in which international criminal
lawyers project a sense of the “international criminal law acquis,” a sort of global package of norms
that have to be adopted by states who become part of the ICC club.
As such, complementarity has been a boon to international civil society, a disciplining device that
makes it possible to raise the prospect of real world consequences with states if they do not
implement norms which may in some cases have been languishing for implementation for decades.
Darryl Robinson was one of the first to theorize the “intrinsic value” of implementation as forcing
“policy-makers, officials, legislators, politicians, and members of civil society … to confront the
persistent problem of impunity.” States would be forced to “strengthen their own legislative
capacity to prosecute perpetrators of genocide, crimes against humanity and war
crimes.”1“Effective implementing legislation,” suggests Amnesty International, “will demonstrate
that the state is aware of its primary responsibility under international law to ensure accountability
for these crimes and will make certain that national courts, not the Court, will undertake these
tasks.”2
In this paper, I will be interested not in implementation per se as much as in what the debate on
implementation reveals about different views of the requirements of complementarity. Proper
implementation is meant to preempt an adverse finding under the complementarity regime before
the ICC, rather than simply to flow from explicit implementation obligations contained in the Rome
Statute. The implementation debate thus mimics and anticipates hypothetical receivability
challenges in the future. Implementation is a way of thinking about complementarity, by asking
complementarity’s questions virtually and long in advance, when all that is at stake is legal and
institutional reform. The particular vision that is propounded of the requirements of
implementation, then, may end up influencing the actual practice of complementarity, even as
complementarity as a somewhat theoretical concept continuously feeds into the debate about what
implementation should entail. Implementation, in short, is the long shadow of complementarity.
Long before the ICC prosecutes a single case, some of the beneficial efforts of
implementation may be felt, not only in terms of domestic prosecution of international crimes, but
more generally in terms of improved accountability and rule of law.
But there is a risk at times that the shadow will become disconnected from the real thing, that
implementation will develop a life of its own under the pressure of various other exigencies. In this
chapter, I want to express a specific concern with a tendency to overstate the case about what
complementarity actually mandates for the purposes of implementation. The concern is that some
of the elements that are being passed as necessary under complementarity, in particular by NGOs
involved in lobbying for implementation, have really little to do with it. Whilst I do not particularly
take issue with the politics of this sort of stratagem and whilst it may be part of a well-intentioned
1 1 D. Robinson, The Rome Statute and its Impact on National Law, In A. Cassese, P. Gaeta and J. R. W. D.
Jones (eds.), The Rome statute of the international criminal court : a commentary, (Oxford ; New York 2002),
pp. 2 v. (cxl, 2018 p.).
2 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33.
3
drive to help states be “safe rather than sorry”, this does raise strategic issues about what is fair to
bundle as part of implementation. At the very least, I believe the NGO community has not always
been as clear (forthcoming?) as it could be about what is strictly needed, and what simply reflects a
broader human rights or criminal law reformist agenda in its own right.
In particular, I am concerned that the push to “mirror” the Rome Statute domestically is part of a
universalizing drive that overstates the homogeneizing requirements of the struggle against
impunity beyond and even in contradiction with complementarity. My main hypothesis is that
implementation is one of the early battle grounds between two visions of international criminal
justice at the ICC, one universalist and centripetic, the other legally pluralist and centrifugal.
According to the first vision, international criminal justice is not simply about broadly ensuring that
individuals are prosecuted for their crimes but, increasingly, that they be prosecuted with “la
manière,” i.e.: in ways that conform with the international community’s own standards of the best
form of criminal justice. According to the second vision, means matter less than the goal, so that a
greater tolerance is adopted of a diversity not only of legal traditions, but even approaches to the
problematique of accountability. Clearly these are ideal-types: no one in the field is so “universalist”
as to suggest that states should reform every aspect of their criminal law to conform exactly with
the Rome Statute; and few are so relativist as to suggest that any form of “justice” will do for the
needs of the international community. But there is no mistaking that this is one of the founding
tensions of international criminal justice, and one that is, for our purposes, strikingly alive in the
implementation debate
This chapter will thus reflect on the tension between these paradigms and the extent to which the
debate over implementation foreshadows the operation of complementarity and international
criminal justice. In order to frame a critique of some of the features of the implementation debate, I
begin by delineating the broad scope of complementarity for the purposes of implementation as it
has occasionally been presented publicly (I). I then suggest that the depth of implementation that
has sometimes been presented to states as necessary makes implementation into a sort of “Trojan
horse” for legal reform that is not strictly required or necessary under the Rome Statute (II). I then
suggest that if anything both the practice and theory of complementarity suggest that lack of
implementation is unlikely by itself to lead to a finding that a state is unwilling or unable to
prosecute crimes (III). I then illustrate this with possible scenarios, distinguishing between
situations where lack of implementation entirely prevents states from exercising jurisdiction and
those where it leads to a failure to convict (or “imperfect” convictions) (IV). I conclude that
complementarity should be understood as allowing for a broad diversity of approaches to
delivering on the promise of international criminal justice.
I. THE SCOPE OF IMPLEMENTATION
Complementarity has allowed “packagization” of a wide diversity of items –procedural, substantive,
jurisdictional, institutional- a bit like an omnibus legislation, leftovers from previous ratification
campaigns, but still on the menu long after they were put there. Quantitatively, perhaps most
4
implementation obligations have to do with states’ obligations in relation to cooperation with the
Court. But in terms of the functioning of the Court’s jurisdiction the most important ones (although
paradoxically not the most explicitly binding and the most vague) are those that relate to the
admissibility issue, and exercise of jurisdiction by the Court in cases where states are found
unwilling or unable.
It was not clear originally that becoming a party to the Rome Statute involved an obligation to
implement it in ways that would safeguard a state from assertions of jurisdiction of the Court.
However, NGOs and scholars were prompt to spot the existence of “one very significant, and
possibly unexpected, benefit of the Rome Statute negotiations”, namely “a substantial overhaul by
State Parties to the Statute of their domestic penal legislation”.3
In some ways, it is a little as if the occasion was almost too good to be true, and civil society jumped
on the opportunity to effect substantial domestic legal reform. The fact that the Court will try very
few individuals and will often face quite reluctant states, made the shift to implementation even
more alluring, as something that could be accomplished here and now, and which could have
substantial ripple effects beyond simply the Court’s operation (for example, once a state has
overhauled its criminal procedure, nominally to guard against exercise of ICC jurisdiction, that will
obviously benefit all defendants in that country, and the rule of law). Implementation is in fact one
of the highpoints of complementarity, a very early and dispassionate opportunity to structurally
prepare legal system for what may one day befall them. The beauty of complementarity, in addition,
is that it makes it possible to frame implementation – which in ordinary circumstances one might
expect states to see as onerous and disconnected from their immediate political priorities – as an
issue of national interest of the highest order, on which hinges retention of one’s monopoly of
jurisdiction.
Although initially this was very
much focused on the need to incorporate the crimes within the Court’s jurisdiction, the trend
quickly developed for NGOs to package such implementation with a series of domestic legal
reforms. The idea was, assuming states would not want to relinquish jurisdiction under
complementarity, that they should reinforce their domestic ability to prosecute international
crimes to preempt any finding that they were “unwilling” or “unable”.
For some, the language used was that of an “opportunity”, whilst for others it was that of
“obligation”. In all cases, however, and over time, there has bee a notable inflation of what
complementarity is held to require in terms of implementation. At conferences over the years I
have seen almost everything put into the “wheelbarrow” of complementarity, under the generic
argument that “national law must be consistent with international law”.4
First come substantive requirements. The most obvious and most often mentioned is that states
adopt the definitions of the crimes over which the Court has jurisdiction domestically. In addition,
The typical demands of
implementation, as they can be gleaned from a variety of NGO sources, can be presented in the
following way.
3 3 K. Doherty and T. McCormack, "Complementarity as a Catalyst for Comprehensive Domestic Penal
Legislation" (1999) 5 UC Davis J. Int'l L. & Pol'y 147.
4 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33.
5
the idea has increasingly emerged that “principles of criminal responsibility in national legislation
should be at least as strict as in Part 3 of the Rome Statute”.5 For example, AI’s checklist suggests
“all crimes of accessory criminal responsibility, such as aiding, abetting and direct and public
incitement as contained in Article 25 should be punishable under national law”.6 There is also the
requirement that “defenses in national law to crimes under international law should be consistent
with customary international law”.7 In other words, there is a clear extension from the special part
of international criminal law to the general one. For the FIDH, for example, the Cambodian
definition of constraint and state of necessity should be changed to better reflect the meaning given
to these concepts in the Rome Statute.8 Human Rights Watch adds the interesting idea that all acts
constitutive of crimes against humanity be separately prosecutable even if the chapeau
requirements are not met.9
Second, come a number of procedural requirements. According to Amnesty International, “Trials in
national courts of persons suspected of crimes under international law must be consistent at all
stages of the proceedings with international fair trial standards”.
10 Mark Ellis also makes the case
that State Parties will have to provide a whole panoply of fair trial rights taken verbatim from the
ICC Statute.11 Moreover, states should protect victims and witnesses and create a trust fund for
victims.12
Third, come a number of jurisdictional requirements whose goal is to avoid bars to prosecution.
According to Amnesty International, states should essentially incorporate article 27(1) of the Rome
Statute: “It will be essential for states wishing to avoid the Court asserting jurisdiction over cases
they are investigating or prosecuting to ensure that any existing immunity under national law with
respect to crimes under international law based on official capacity be eliminated.”
13
5 2 Ibid.(
Moreover,
6 2 Ibid.(
7 4 The International Criminal Court: Summary Checklist for Effective Implementation, ( Also, 5
International Criminal Court: Manual for the Ratification and the Implementation of the Rome
Statute: 3rd edition, (2008). (“all States wishing to rely upon the complementarity principle should review
existing defences under their national criminal law in order to ensure that these defences do not potentially
shield persons from criminal responsibility for acts amounting to ICC crimes”).
8 6 Report: International Criminal Court, Implementation of the Rome Statute in Cambodian Law,
(2006).
9 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
10 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33.
11 8 M. S. Ellis, "International Criminal Court and Its Implication for Domestic Law and National Capacity
Building, The" (2002) 15 Florida Journal of International Law 215.
12 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
13 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33.
6
states should remove statutes of limitation.14
Fourth, are a number of requirements dealing with sentencing, amnesties and pardons. Amnesty
International, for instance, suggests that the death penalty should not be imposed.
Finally, comes the quite ambitious requirement
expressed by some NGOs that “national courts should be able to exercise universal jurisdiction in all
cases of crimes under international law”.
15 The FIDH
considers that amnesty and pardon provisions should be amended to reflect the particular
necessities of dealing with international crimes.16
II. IMPLEMENTATION AS A TROJAN HORSE?
As can be seen therefore what is contemplated involves almost all aspects of criminal jurisdiction.
However, the key aspect for our purposes is not the scope but the depth of implementation that is
recommended.
Typically, the reasoning is that states, although not strictly “obliged,” are strongly advised to adopt
extensive implementing legislation if they want to be “safe rather than sorry”. It is in states’ interest
to reinforce their ability to prosecute international crimes if they want to maximize the odds of
retaining jurisdiction.17 Moreover, although this does not flow directly from Statute obligations, it is
often an obligation of states’ more generally to prosecute international crimes within the
jurisdiction of the ICC. However in practice the languages of “interest” and “obligation” are often
used interchangeably in ways that put more rather than less pressure on states. Amnesty
International, for example, makes the distinction between changes that states parties are “required”
to do and those that it “recommends that they should do to ensure that the Court is an effective
complement to national courts.”18
14 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
But these fine distinctions tend to be quickly lost in the process
of advising states and emphasizing the loosely defined dangers of not fully implementing the Rome
Statute. “State parties,” it has been suggested, in an appealing but characteristically vague formula,
“are bound by the principles of complementarity and decentralized justice underlying the
15 9 F. Jessberger and C. Powell, "Prosecuting Pinochets in South Africa-Implementing the Rome Statute of
the International Criminal Court" (2001) 14 S. Afr. J. Crim. Just. 344.
16 6 Report: International Criminal Court, Implementation of the Rome Statute in Cambodian Law,
(2006).
17 1 D. Robinson, The Rome Statute and its Impact on National Law, In A. Cassese, P. Gaeta and J. R. W. D.
Jones (eds.), The Rome statute of the international criminal court : a commentary, (Oxford ; New York 2002),
pp. 2 v. (cxl, 2018 p.).
18 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33.
7
Statute”.19
A. ABUSIVELY TRANSPOSING ICC SPECIFIC FEATURES TO DOMESTIC
JURISDICTION
Not all of what is suggested under implementation as a result is very controversial but
there is a suspicion that some typical demands are only loosely connected to complementarity.
One demand often made on states is that they abolish all sovereign immunities of foreign heads of
state when it comes to international crimes, including incumbent heads of state.20 This is clearly
based on the fact that the ICC itself is not bound by any immunities.21 Certainly the ICC would have
jurisdiction when it comes to the head of state of a state party that is under the jurisdiction of that
state party or another state party and which that state is failing to prosecute.22
What is acceptable for the ICC to do is not necessarily legitimate for states to be doing by analogy.
There are arguably very specific reasons why the ICC can override immunities of current heads of
state. The ICJ has confirmed in the Yerodia case
State parties to the
Rome Statute can be seen as having committed to abolishing internal immunities, and even
bracketed the application of foreign immunities between themselves. But what if the head of state is
the head of a non-party state that simply happens to be present on the territory of a state party?
23 and the Special Court for Sierra Leone in the
Taylor case24
The question when it comes to state parties, then, might be formulated in the following way: should
state parties exercising jurisdiction over third-state incumbent heads of state be seen as acting
primarily as “states” (i.e.: normal states), or should they be seen primarily as state-parties? In the
latter case, one might argue that they were endowed, almost by osmosis, with the ICC’s powers to
that certain “international courts” can override immunities in the case of international
crimes, precisely because they are “international”. The rationale is that international tribunals will
not create the sort of inter-state tensions that assertion of jurisdiction by states would. The
corollary, however, is that as far as the ICJ is concerned at least, states still have to respect the
immunity of incumbent heads of state on the horizontal level.
19 9 F. Jessberger and C. Powell, "Prosecuting Pinochets in South Africa-Implementing the Rome Statute of
the International Criminal Court" (2001) 14 S. Afr. J. Crim. Just. 344.
20 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33. (“Immunity
of officials from prosecution for crimes under international law should be eliminated”).
21 Article 27 of the Rome Statute.
22 10 P. Gaeta, Official Capacities and Immunities, In A. Cassese, P. Gaeta and J. Jones (eds.), The Rome
Statute of the International Criminal Court: a commentary, (Oxford; New York 2002).
23 Case Concerning the Arrest Warrant of 11 April 2000 (Democratic Republic of the Congo v. Belgium).
International Court of Justice (ICJ). 14 February 2002, p. 26.
24 Prosecutor V Charles Ghankay Taylor, Case Number SCSL-2003-01-I, Decision on Immunity from
Jurisdiction, 31 May 2004.
8
ignore incumbent head of state immunities, because they are exercising ICC jurisdiction “by
procuration”? This is undoubtedly a difficult question, but the reasoning does not weigh in the
direction of transposing the quite exceptional ability of the ICC to override immunities to states.
When states are exercising jurisdiction, be it under the watchful gaze of the ICC, they are first and
foremost exercising it as states, who could have done so before and independently of becoming a
party to the Rome Statute. Technically, as long as a case is not admissible because it is being
prosecuted domestically, it is the state party that is primarily exercising jurisdiction, not the ICC.25
Some of the same issues of inter-state tension and interference with sovereignty will arise in this
primarily horizontal set-up that would arise if the ICC were not involved, and which are still
considered by the ICJ to militate in favor of a conservative approach to immunities. Saying that “the
ICC made me do it” will not particularly alleviate these tensions. This is why article 98 of the Rome
Statute stipulates that the Court “may not proceed with a request for surrender … which would
require the requested State to act inconsistently with its obligations under international law with
respect to the State … immunity of a person … of a third States”. To require states to equip
themselves with the same sort of jurisdiction as the ICC would arguably involve doing just that.
B. MANDATING WHAT SHOULD ONLY BE AN OPTION
There are several areas where implementation is presented as obligatory by some NGOs that
should really only be an option and where what is typically at stake is a particular preference about
international law. A good example is the death penalty. According to Amnesty International, “It
would be inappropriate for national courts to impose a more severe penalty for a crime under
international law than the one chosen by the international community itself.”26 It is quite hard,
however, to see why that should be the case. However desirable the elimination of the death
penalty may be, and whatever the positive evolutions towards more systematic abolition, the truth
of the matter is that international law on the matter is unsettled, except insofar as it tolerates the
imposition of capital punishments in some circumstances.27 In fact, the Rome Statute itself makes it
very clear that “Nothing in (the part on penalties) affects the application by States of penalties
prescribed by their national law”.28
25 Jan Kleffner makes the interesting argument that the “admissibility procedure before the ICC could provide
the necessary safeguards to ensure that national courts only retain their competence to investigate and
prosecute an incumbent senior State official as long as these proceedings are conducted in accordance with
internationally recognized standards of due process”. 11 J. Kleffner, "The Impact of Complementarity on
National Implementation of Substantive International Criminal Law" (2003) 1 Journal of International
Criminal Justice 86-113. This might be the case as between two state parties (i.e.: where a state party is
prosecuting the head of the state of another state party), but it is much more doubtful that a non-state party
would feel that ICC supervision provided a guarantee against unwarranted interference in its internal affairs.
Moreover, many states will impose more severe penalties than
26 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33.
27 ICCPR, article 6.
28 Article 80.
9
the Court, including longer prison sentences than anticipated by the Statute,29
Another example is the requirement expressed by some NGOs that states should be in a position to
assert universal jurisdiction. This seems to go significantly beyond what is unambiguously
mandated under general international law. Universal jurisdiction is probably better seen as an
option rather than an obligation. The Genocide Convention specifically indicates that the state in
whose territory the crimes were committed is primarily responsible. The practice of states
certainly does not evidence anything like a general practice and opinio juris about universal
jurisdiction. It is remarkable that universal jurisdiction has been recognized as a possibility and not
in itself contrary to international law, but there is a sense that organizations like Amnesty
International and Human Rights Watch are pushing their luck when they try to present its adoption
as compulsory.
yet this is normally
not something that would be considered problematic or that Amnesty International suggests should
be changed. The maintenance of the death penalty for international crimes may be unsatisfying
from a human rights point of view, but it is hardly something that should be changed merely on
account of the ICC’s complementarity regime.
Of course, one could argue that a state that has ratified the Rome Statute has already opted for a
higher degree of commitment to international criminal law so that it has to behave in every other
way according to the “state of the art.” But that is a mere construct. In some ways, states might
precisely want to join the ICC and its overwhelmingly vertical architecture to escape the dangers
and limitations of horizontal universal jurisdiction. One simply cannot presume that commitment to
the ICC does and should coincide with commitment to universal jurisdiction, an institution that may
seek to achieve the same anti-impunity goals but that reflects a quite different logic.30 In fact, some
have even argued that one of the significant progresses inaugurated by the ICC is that it does away
with the need for universal jurisdiction.31 Hence a number of states, such as the Netherlands, called
the bluff and opted to not implement universal jurisdiction, arguing that the “trend in international
practice” did not support an obligation to do so.32
29 The regime adopted by the ICC, at the insistence of some states for whom it raised constitutional issues, is
in fact quite indulgent.
30 This is quite evident from some declarations made in the course of parliamentary implementation debates.
See, in particular, the passage by Baroness Scotland quoted in 12 D. Turns, Aspects of National
Implementation of the Rome Statute: The United Kingdom and Selected Other States, In P. J. R. Dominic
McGoldrick, Eric Donnelly (ed.), The Permanent International Criminal Court: Legal and Policy Issues, (2004).
31 13 G. Bottini, "Universal jurisdiction after the creation of the international criminal court" (2003) 36
New York University Journal of International Law and Politics 503. 14 L. Reydams, Universal
jurisdiction: international and municipal legal perspectives, (2003).
32 15 A. Cinar, Implementation of the Rome Statute in the Netherlands, SSRN, (2007).
10
C. REQUIRING THAT WHICH IS NOT STRICTLY CONNECTED TO
COMPLEMENTARITY
There has been a tendency to require that states fully implement fundamental guarantees of due
process, in a way that suggests that the Court might exercise jurisdiction in cases where the rights
of defendants have been violated. For example, HRW argues that “states should guarantee the
highest international standards for fair trials at the national level”, something which is seen as
requiring the incorporation of the Rome Statute due process guarantees, and which will “be
important in the determination of the admissibility of a case by the ICC”.33 Scholars have followed
suit and whilst Otto Triffterer concedes that “it may not be necessary that States ‘copy’ the
procedure applicable in cases before the Court” he does emphasize that the “basic pillars of a fair
trial” should be respected.34
Reforming procedural criminal law is all well and good, but locating this under the
complementarity header is more than a little misleading. The references to “independence and
impartiality” and to the principles of fair trial in the Statute, were always misnomers. If anything,
the Court will be attentive to manipulations of the procedure that evidence “too generous” a trial,
i.e.: one designed to shield the accused. The Court will not be involved in reviewing more generally
whether the rights of the accused have been respected. As Kevin Jon Heller demonstrated in his
masterly analysis of the issue, the “due process thesis,” while “seductive” is also “incorrect.”
35
If
anything, as I have argued elsewhere, the reference to independence and impartiality refers to the
need to ensure that courts are not biased in favor of the accused. The reference to the right to be
tried within a reasonable time is not there to protect the accused from excessive pretrial detention,
but to protect society from a never ending trial that could end up in a sort of permanent judicial
impunity under the guise of protecting the presumption of innocence. As to the reference to due
process it can be understood as meaning that the accused should not be granted more rights than
would be guaranteed as part of the right to a fair trial, or even that the trial should be fair in all
respects, for example in relation to victims. However deplorable the fact that the ICC will not be
able to assert jurisdiction in cases where the rights of the accused are violated, urging
implementation on this account is little more than opportunistic free-riding on complementarity.
D. REQUIRING TOO MUCH IDENTITY WITH THE ROME STATUTE
33 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
34 16 O. Triffterer, Legal and Political Implications of Domestic Ratification and Implementation Processes,
In C. Kress and F. Lattanzi (eds.), The Rome Statute and Domestic Legal Orders.: Volume I: General Aspects and
Constitutional Issues, (2000).
35 17 K. Heller, "The Shadow Side of Complementarity: The Effect of Article 17 of the Rome Statute on
National Due Process" (2006) 17 Criminal Law Forum 255-280.
11
A related pattern is one in which although the issue is of relevance to the discharge of the “willing”
or “able” burden, too much homology is required between domestic law and the Rome Statute. This
is what might be called the “mirror” theory of implementation: the idea that “the local regulatory
framework should reflect as faithfully as possible the actual regime of the ICC itself, and every step
taken from it takes states closer to a finding of inability or unwillingness”.36 As Edoardo Greppi puts
it, “A State wishing to give solid foundations to its right to investigate and to prosecute crimes, and
to avoid being considered in a potential situation of inability, should conform its legal system to the
requirements of the Rome Statute”.37 A good example of this is some demands that are made of
substantive law, beyond the core crimes themselves. For example, the International Centre for
Criminal Law and Policy holds that “States desiring to prosecute criminals in their national courts
under the principle of complementarity should ensure that their national legislation includes all the
forms of individual and inchoate criminal responsibility covered by the Rome Statute”.38 Darryl
Robinson has also argued that “if a State wants to be sure of meeting the complementarity test, then
it would be prudent to review the grounds of responsibility in the Rome Statute, and to ensure that
the national law is at least as broad”.39 For Otto Triffterer, it is “in the interest of justice and the
States, that the latter already have or create (…) regulations in their domestic legal systems, which
are in accordance with the law to be applied by the Court”.40
There is a lot of ambiguity in terms like “conform” or “in accordance” or “consistent with” that is
problematic. The language seems to assume that the general part of international criminal law has
become part of the international consensus about prosecutions of international crimes to the same
degree as the definition of special crimes. The general follows the special, one might say, so that the
general part is now indissociable from the very definition of crimes. The general part of ICL,
however, is a relatively recent creation. Special international criminal law treaties (even though
they may not always have been thought of in this way originally) typically do not, apart from the
genocide convention, prescribe a general part in any detail, being quite happy historically to
delegate this to domestic jurisdiction. The ad hoc international criminal tribunals did without a well
In some cases, states have been
specifically faulted for, for example, allowing ordinary domestic law defenses to remain that may
not be “always consistent with international law”.
36 The “mirror” theory is evident in the language used by several NGOs 5 International Criminal Court:
Manual for the Ratification and the Implementation of the Rome Statute: 3rd edition, (2008).
37 18 E. Greppi, Inability to Investigate and Prosecute under Article 17, In M. Politi and F. Gioia (eds.), The
International Criminal Court and National Jurisdictions, (2008).
38 5 International Criminal Court: Manual for the Ratification and the Implementation of the Rome
Statute: 3rd edition, (2008).
391 D. Robinson, The Rome Statute and its Impact on National Law, In A. Cassese, P. Gaeta and J. R. W. D.
Jones (eds.), The Rome statute of the international criminal court : a commentary, (Oxford ; New York 2002),
pp. 2 v. (cxl, 2018 p.).. Robinson makes a similar point re defenses.
40 16 O. Triffterer, Legal and Political Implications of Domestic Ratification and Implementation Processes,
In C. Kress and F. Lattanzi (eds.), The Rome Statute and Domestic Legal Orders.: Volume I: General Aspects and
Constitutional Issues, (2000). Also 19 O. Bekou and S. Shah, "Realising the Potential of the International
Criminal Court: The African Experience" (2006) 6 Human Rights Law Review 499-544.
12
developed one, and theirs was largely a jurisprudential creation. The general part was elaborated in
the context of the Rome Statute because of fears about the nullum crimen sine lege principle
internationally, rather than necessarily because there was an absolutely major stake for the
international community or because there was only one good “general part” by which to prosecute
international crimes. In fact, this is an area of international criminal law that has been fraught with
tensions, contradictions and evolutions so that the idea that it should now be transposed as such to
domestic systems strikes one as odd. One issue that is almost never raised, in this context (except to
suggest that it is not a problem at all),41
is the risk that this sort of implementation might play
roughshod with delicate and highly defensible domestic balances about the general part.
E. GOING FURTHER THAN THE ICC
In some cases, NGOs seem to be pushing for forms of implementation that go beyond even the
content of the Rome Statute. Implementation is used as a corrective for what are seen as – in
language that sometimes reflects a typical NGO distaste with the limitations of
intergovernmentality - faulty “compromises” in the Rome Statute as a result of “consensus”. As is
well known, the adoption of the Statute led to many compromises that were felt by some to have
been fundamentally misrepresentative of the state of international law, or at least at variance with
the better objectives of international criminal justice. The concern is typically that a particular
provision or set of provisions in the Rome Statute is unsatisfactory and even at variance with
customary international law. As HRW notes, although “some of the Rome Statute definitions reflect
the evolution of international humanitarian and criminal law in the years since the principal
treaties governing these areas were adopted, while others are less progressive and do not meet
recent standards adopted by the international community”.42 The same thing goes with universal
jurisdiction which the Rome Statute, as a result of compromises, “failed” to give the ICC. Similarly,
the FIDH considers that article 33 of the Rome Statute, in failing to mention war crimes, results
from a compromise with the US position at the conference that “does not provide a true reflection
of prior international case law”,43
41 For Triffterer, for example, if after “adaptation” of domestic law to the requirements of the Rome Statute
“the relevant regulations differ from those characterizing the ordinary criminal procedures in the respective
domestic legal system, the latter than (sic) may be adapted too.” (16 O. Triffterer, Legal and Political
Implications of Domestic Ratification and Implementation Processes, In C. Kress and F. Lattanzi (eds.), The
Rome Statute and Domestic Legal Orders.: Volume I: General Aspects and Constitutional Issues, (2000). In other
words, implementation of international law is very much a vehicle for domestic reform along universalizing
lines.
and commends Cambodia for ignoring it. In relation to general
principles of law, the same organization mentions that “gaps and incoherencies remain, even in the
42 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
43 6 Report: International Criminal Court, Implementation of the Rome Statute in Cambodian Law,
(2006).
13
rules stated in the Rome Statute”.44 In other cases, failure to “impos(e) criminal responsibility for
the inchoate offence of conspiracy to commit genocide” is attributed to “an oversight” given the
Genocide convention.45
Far from the corridors of the FAO, therefore, a second battle has been fought to essentially correct
these “glitches” through “progressive” implementing legislation. The “mirror” is this time supposed
to be not so much a faithful as a corrective one, that improves on minor defects in the Court’s image.
“If the international system of justice is to be fully effective,” goes the reasoning, “all States parties
should fill the gap in the Court’s jurisdiction” in relevant ways.
46 Rather than taking the Rome
Statute as a model (it is the source of the problem), NGOs switch to general international law. For
example, when it comes to the failure to incorporate universal jurisdiction in the Rome Statute, or
to principles of superior responsibility that are “not as strict as required by customary international
law”, then “if the international system of justice is to be fully effective, all states parties should” 47
adopt provisions domestically that fill these gaps. The FIDH encourages Cambodian legislators to
“complete or rectify”48
There are some ironies to this process of retrospective “mending”. First, NGOs are telling states
which will often have been involved in the Rome negotiations, will have been privy to and in some
cases ardently wanted the compromises (which did help bring about the ICC), that they should now
implement something different than what they often painstakingly agreed to. Second, and most
importantly, the idea that some definitions in the Rome Statute “do not meet recent standards
adopted by the international community” seems to miss altogether the fact that the Rome Statute…
is the most recent “standard adopted by the international community”, and by a long shot.
unsatisfactory provisions in the Statute.
49
44 6 Ibid.(
It may
be that one wants states to adopt more progressive definitions, but one can hardly claim that this
has anything to do with implementation stricto sensu. What is really being suggested, under cover
of conformity with the Rome Statute, is that implementation becomes something akin to a sort of
“progressive development” that has little to do with implementation stricto sensu.
45 5 International Criminal Court: Manual for the Ratification and the Implementation of the Rome
Statute: 3rd edition, (2008). It was clearly not an oversight, but an attempt to rationalize and harmonize
modes of imputability for all crimes in the Statute.
46 2 The International Criminal Court: Checklist for Effective Implementation, (2000), p. 33. 349
47 2 Ibid.(
48 6 Report: International Criminal Court, Implementation of the Rome Statute in Cambodian Law,
(2006).
49 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
14
III. THE REALITY OF COMPLEMENTARITY50
What is presented under the classic international law language of “implementation,” often turns out
to have more to do with a broad agenda of legal reform. If there is an obligation at all, it is the long
preexisting one of trying international crimes, and that obligation was traditionally satisfied as long
as prosecutions broadly occurred. The truth is that there are few cases in which lack of
implementation alone would make a case receivable before the ICC, and certainly many scenarios
where a case will not be receivable despite lack of implementation. The “stick” with which some
would like to nudge states towards high degrees of implementation is in fact a lot less threatening
than it is made to be.
A. THE PRACTICE OF COMPLEMENTARITY
There is a sense, especially with the passing of years, that some of the more demanding views of
implementation are based on a reading of complementarity that does not fit with reality. It is worth
stressing that the vision of the Court routinely resorting to “hostile” complementarity, i.e.: seeking
to exercise jurisdiction against states actually trying to conduct their own proceedings, although it
might still amount to something in the future, has proved very far from reality. If anything, typical
ICC cases are those in which there will be no investigation whatsoever (regardless of the adequacy
of legislation) and where a state evidences no intention of complying with either their Rome Statute
or general international law obligations to try certain crimes (e.g.: Sudan). In fact, it was always
unlikely politically that states would conduct “mock” proceedings for the purposes of holding off
ICC jurisdiction. If a state is unwilling, it will generally be unwilling all the way. One can hardly
imagine Sudanese authorities giving Al Bashir even a complacent trial. If anything, at the other end
of the spectrum, what we have seen so far is that complementarity is also sidelined because of
states gladly affirming that they are, in effect, unwilling or unable (e.g.: Uganda).51
50 Interestingly, states have resisted this trend. For example, the British Government policy on the question
was described in the following way: “The problem is that the (ICC) Statute does not require universal
jurisdiction, so we do not think that we should go it alone and say that we will do it all if the court will not do
it”. Quoted in 12 D. Turns, Aspects of National Implementation of the Rome Statute: The United Kingdom and
Selected Other States, In P. J. R. Dominic McGoldrick, Eric Donnelly (ed.), The Permanent International
Criminal Court: Legal and Policy Issues, (2004).
51 Incidentally, it may be that some state-parties would consciously fail to fully guard again the risk of being
found “unwilling or unable” by not adopting laws that might better protect them. Although this option is often
not considered in the implementation literature, one could consider that calculated risk taking by states in
this area is not to be excluded and not necessarily shocking. For example, a state might think that it would be
better if it were found “unable” to try an incumbent head of state from a third party, thus allowing the Court
to exercise jurisdiction over it, than for it to create significant tensions with that other state as a result of
asserting jurisdiction. In such a scenario, the involved state party could “blame” prosecutions on the ICC and
safeguard its horizontal relations with the non-party state.
15
This means that the intermediary position, that of “sham” trials, seems rather unlikely to arise
except in the most unusual circumstances. As a result, it is unlikely that the ICC will become
involved in sophisticated qualitative analysis of domestic trials to see if they constitute “genuine”
proceedings or not, and thus deal, among other things, with the finer details of implementation. In
fact, in a context where the prosecutor by necessity has to be hyper-selective, it makes sense that
s/he should concentrate his/her energies on the cases where the Court can makes most of a
difference, rather than aim to uncover the “sham” in trials that at least have the merit of existing. It
therefore seems slightly improbable, in practice, that the Prosecutor would deal with cases where
failure to satisfy the complementarity requirement results simply from a failure to implement the
right legislation.
Of course, it may be that with the passing of years and the decline of the more conspicuous types of
impunity, the ICC will little by little venture into acting as more of a scrutinizer of trials, imposing a
more “qualitative” reading of complementarity. After all, implementation legislation is there for the
long haul and one needs to factor in the possibility of shifting international prosecutorial strategies.
Arguably, the ICC is as much a promoter as an enforcer of complementarity these days, in
accordance with its increasingly cooperative and managerial role, as highlighted by Carsten Stahn
in particular.52
B. THEORETICAL TENSIONS
However given the vast numbers of cases that evidence a dismal level of compliance
with the obligation to prosecute, combined with the ever-present temptation of seizing self-
referrals, prospects for this seem small. In the great majority of cases, moreover, inability or
unwillingness will result from failure to pursue specific cases so that, although lack of
implementation may have a role in this, it will rarely be the decisive factor. Needless to say,
moreover, there will be many cases where states have all the implementing legislation in the world
and yet are clearly not committed to prosecutions.
There are several general theoretical issues at stake in terms of what should be the depth of
implementation to satisfy the complementarity requirement which are as yet unanswered, and
where it is unhelpful to simply raise the prospect of ICC jurisdiction. First, one can discern a
constant tension between two views of complementarity, which the debate over implementation
dramatizes. At one level, complementarity might simply require states to exercise whatever
jurisdiction they have as a result of the operation of their domestic law in a “willing and able”
way;53
52 20 C. Stahn, Complementarity: A Tale of Two Notions, vol. 19, (2008), pp. 87-113.
at another extreme, complementarity might be seen to require states to develop an ability to
exercise certain forms of jurisdiction which they may not have from the start. In the former case,
the state is judged by the standards of its legal system, whatever its deficiencies may be. In the
latter version, states becoming party to the Rome Statute are required to put themselves through a
process of sophisticated legal re-engineering.
53 Paragraph 6 of the Rome Statute Preamble, for example, states that “it is the duty of every State to exercise
its criminal jurisdiction over those responsible for international crimes”. This might be understood to mean
“exercise whatever criminal jurisdiction they have”.
16
The complementarity/implementation debate is often struggling with this tension. For example, it
is quite clear that a state that could not prosecute one of its nationals suspected of genocide
because it had granted that individual an absolute immunity, would be “unwilling” or even “unable”
to prosecute that individual. But one could take this sort of reasoning quite far and say that a state
that fails to condemn an individual for example because it does not prosecute incumbent foreign
heads of state or because it does not allow for a certain form of complicity to lead to liability is also
“unable” or “unwilling”, because it should have implemented laws allowing it to prosecute crimes
“exactly as the ICC would.” Clearly, a compromise is needed between these positions, one that does
not allow states to hide behind the limitations of their jurisdiction when their jurisdiction is clearly
inadequate, yet also one that does not turn into “inability” or “unwillingness” any case where a state
has a form of jurisdiction that is at odds with that of the ICC.
Second, there is a more general problem with the sort of jurisdiction that states should be capable
of asserting over international crimes in terms of international law, to satisfy the requirements of
complementarity. Should states all adopt what at any one time is seen as the “cutting edge” of the
international community’s tools to combat impunity (e.g.: universal jurisdiction, no immunity, large
modes of imputability, narrow defences)? Or should the requirement be merely that states conform
to what can be minimally expected of them? Will the international community be content with
“prosecutions” or does it want “a certain kind of prosecution”? Does the requirement differ for state
parties and non parties? The question is really whether certain features of domestic jurisdiction
have become so central to what is required internationally, that not having them fails the
complementarity test. It is a question also linked to a broader dialectics of unity and diversity, legal
universalism and legal pluralism, on which much of complementarity rests.
Third, a key question is under which of “unwillingness” or “inability” failure to implement
legislation might be subsumed. It should be stressed that nothing in article 17 of the Rome Statute
seems to be particularly oriented towards faulting states for failure to adopt adequate
implementing legislation per se. Part of the inability standard looks inapplicable except in the rarest
of cases (“total or substantial collapse of the judicial system”) and not particularly relevant to the
issue of implementation.54 Lack of implementation might in itself involve “unwillingness” to
prosecute an individual. For example, unwillingness would have a role to play in a case where
failure to implement is the result of a larger culture of inertia that is fundamentally hostile to the
goal of fighting impunity (e.g.: where a state already has foresight that crimes might be committed
and it fails to adopt the right legislation as part of a sophisticated advanced cover-up).55
54 The exception is of course the case where a failure to implement even basic legislation results from the
collapse of the judicial system.
However,
55 In that respect, I am a bit skeptical of Terracino’s suggestion that « unwillingness » must really be
appreciated at the point of the trial, from which issues of implementation are too far removed. It would be too
easy for a state, having utterly failed to implement any legislation, to then say that it was not particularly
unwilling in a given case and that its inability to conduct trials was merely the technical one of not having the
right legislation. Often, impunity will masquerade as technical impossibility, indeed historically it often has.
What will be needed is to understand the conditions in which failure to implement happened, whether it is
the result of a more or less innocent mistake, or whether it conceals a more general intent to sabotage
attempts to make the authors of certain crimes accountable. If the inability in a particular case results from a
broader unwillingness, then it should only be a very short step to finding a case receivable. 21 B.
17
that may be a long route. The idea that a state is “unable” to prosecute a case if its judicial system is
“unavailable” is also particularly interesting. Reading article 17 3. of the Statute, one gets the sense
that “unavailability” is something structurally broad, in the same vein as “collapse” of the judicial
system (e.g.: a post-Rwandan genocide scenario in which most of the judicial profession has been
wiped out). Moreover, unavailability sounds like something in the order of an absence rather than
an “inadequacy,” and there are reasons why the latter term was never a strong contender in Rome.
However, one can construct “unavailability” as meaning not just this sort of logistical, literal
unavailability, but some sort of “legal unavailability” or “unavailability for the purposes of realizing
the purposes of international criminal justice.” Moreover, in many cases unavailability will be a
form of “willed inability,” a failure that was planned in advance, and is no better simply because it
was not intended for a particular case. Under this light, there may be room to more firmly connect
complementarity and implementation. In fact, unavailability, because it arguably requires proving
neither a massive default (i.e.: collapse) or a flagrant ill-faith (i.e.: unwillingness) has emerged as a
strong ground for admissibility in contexts of absent or insufficient implementation.56
IV. SOME ILLUSTRATIONS
Let us, then, look at a few situations where a challenge based on lack of implementation, might be
the criterion. The scenario, to be more specific, is one of states who are willing “politically” and able
“logistically” (in the sense that the judicial system has not collapsed) to exercise jurisdiction but are
unable to “technically” because of lack of adequate legislation. Does this technical “inability” count
as full inability under the “unavailability” scenario? There would seem to be a continuum between
situations in which the domestic judicial system is entirely unable to investigate or prosecute a case
as a result of absent or imperfect implementation, and situations in which it can but does not do so
in a way that is judged satisfying internationally.
A. FAILURE TO EXERCISE JURISDICTION
The total failure to exercise any jurisdiction is perhaps the one where the case for “unavailability” is
at its strongest. A first scenario is the one where the definition of core crimes or any of their
component parts has not been implemented domestically, where the behavior in question is not
even criminalized domestically. According to Bacio Terracino, who wrote one of the most
comprehensive articles on the issue, the failure to adopt legislation criminalizing these domestically
Terracino, "National Implementation of ICC Crimes: Impact on National Jurisdictions and the ICC" (2007) 5
Journal of International Criminal Justice 421. The fact that implementation will hardly ever have a role alone
in a finding of inability or unwillingness, does not mean that it will not be a strong corroborative element in
particular cases.
56 For example, the informal expert paper on the principle of complementarity mentions “lack of substantive
or procedural penal legislation rendering system ‘unavailable’. 22 ICC-OTP, Informal expert paper: the
principle of complementarity in practice, (2003). One could argue that “unavailability” is the closest thing to
the ILC’s old but ultimately abandoned notion of “ineffectiveness”.
18
in any manner, if it prevents a state from prosecuting them, would be a case of unavailability.57
More problematic and troubling issues arise with aspects of states’ jurisdiction which are less
clearly connected to the obvious core of having a legal system that is “available” to prosecute
international crimes. Even though a state party is not obliged to adopt universal jurisdiction by the
Rome Statute or to abolish incumbent head of state immunities, for example, such a state might well
find itself in a situation where it cannot even begin to investigate a case as a result of not having
adapted its domestic law appropriately. In that situation, it will be very difficult for it to mount a
receivability challenge, since it will not even have been able to “put its foot in the door”, as it were,
and secured some sort of jurisdictional foothold. The Rome Statute anticipates that the inability or
unwillingness test is to be applied to “a State which has jurisdiction” over a case that the Court is
also trying to prosecute. Presumably, jurisdiction means just that, so that the total unavailability of
the legal system for the purposes of prosecuting the crime at stake would allow the Court to
exercise its own, less impeded jurisdiction. When it comes to immunities, the question is arguably
more complex since one might argue that the state party does have jurisdiction, except that it is
prevented from using it by a rule of international law that grants immunity to incumbent heads of
state. However, having a “theoretical” jurisdiction one cannot exercise and not having jurisdiction
in the first place fall quite close to each other. In both cases, the judicial system might be considered
“unavailable” to prosecute the crimes involved.
But
this is in a sense the easy case since it covers an area of ICC implementation legislation which
overlaps very substantially with states preexisting obligation to prosecute these crimes and which,
at any rate, goes to the core of what the Court is about. To not be able to prosecute a crime of
genocide because one does not punish genocide domestically or even some of the component acts
of genocide would seem to be quite incompatible with one’s obligations as a party to the ICC.
However, in practice, it is worth noting that the situation of a state that cannot prosecute an
individual because it lacks an adequately broad jurisdiction may not necessarily make it
straightforwardly “unavailable”. Consider the case, for example, where it is not be immediately
apparent to authorities that the lack of universal jurisdiction or the existence of incumbent head of
state immunity were going to be an obstacle to jurisdiction, so that an investigation and even a
prosecution might be launched, in which the prosecutor simply calculated that the judicial system
would eventually find in his favor. Imagine, in other words, a post-ICC Pinochet type trial on the
territory of a state party (but involving the head of a non-state party). Domestic courts might have
jurisdiction enough to at least detain the person, while assessing whether they actually had
jurisdiction, and could probably stave off ICC jurisdiction for a while.
Assuming that a domestic court then decided not to exercise universal jurisdiction or to not ignore
the accused’s immunity and that the ICC then sought to assert jurisdiction, should there be a
difference between failing because one in a sense wants to fail, and failing because one has failed to
make the case, before an independent judiciary, that jurisdiction should ultimately be asserted? In
such a situation it is not obvious that the state concerned could easily be considered unable as a
result of the unavailability of its judicial system, given that it had done everything it could to give
complex issues of international law a fair hearing.
57 21 B. Terracino, "National Implementation of ICC Crimes: Impact on National Jurisdictions and the ICC"
(2007) 5 Journal of International Criminal Justice 421.
19
Indeed, could the state involved argue that “the case has been investigated”, but that it “decided not
to prosecute the person concerned,” (having found that it cannot exercise jurisdiction), rather than
simply being “unable” to as a result of not having a jurisdictional basis to do so?58
The Court would probably still eventually exercise jurisdiction on the basis of the absence of a real
prosecution, or on the basis that limitations on states’ jurisdiction, however legitimate on the
horizontal plane, are not really its problem. Respect for inter-state immunities and should be
distinguished from ordinary “unsuccessful” cases (e.g.: where not enough evidence is assembled to
press charges). But these and other cases like it are worth keeping in mind, to emphasize the
existence of a difference between being straightforwardly “unable or unwilling” and having tried
genuinely but failed to exercise jurisdiction, and being unable to assert jurisdiction as a result of the
operation of international law itself, in an environment where powerful and arguably legitimate
counter-currents at the interstate level still constrain its exercise in some situations.
The challenge is
that the Statute does not tell us with much precision in which situations a decision not to prosecute
will be characteristic of unwillingness or inability. Some decisions not to prosecute will clearly be
more legitimate than others. For example, it would be perfectly legitimate for a domestic
prosecutor to, in good faith, decide not to prosecute someone against whom s/he felt there was no
case. In the case that concerns us, a state might argue to the ICC that respect for interstate
immunities was something worth deferring to, and that inability should not cover situations of
inability that result from the accepted operation of international law between states. For example,
had the House of Lords decided the Pinochet case differently in the ICC era, and concluded that the
UK could not exercise jurisdiction and Pinochet enjoyed immunity at all times (and if that was, at
least arguably, the position of the law at the time), would the UK be “unable”, or would this simply
be a case of failed prosecution? A state could point out that if its respecting immunities of
incumbents on its territory ends up making it “unable” in the eyes of the ICC, then its ability to
guarantee immunities (as arguably required by international law) at all would be fundamentally
curtailed.
B. FAILURE TO SECURE A CONVICTION
A more complex issue arises where the state undeniably has broad jurisdiction to begin with (i.e.:
jurisdiction over the core crimes, no recognition of immunities, possibly universal jurisdiction), but
where it fails to prosecute a case successfully – at least in the sense “successfully” is understood
internationally - because of what might be presented as “gaps” in its substantive law (gaps in
relation to what the ICC could claim).59
A variant of the scenario of non-implementation of the core crimes is one in which a state
nonetheless prosecutes the relevant conduct as “ordinary” crimes, which at least partly cover the
58 Article 17 1. (b).
59 Terracino, 432. 23 B. Broomhall, International justice and the International Criminal Court : between
sovereignty and the rule of law, (Oxford ; New York 2003), pp. x, 215 p.
20
behavior involved. Most commentators seem not to find excessive fault with this, at least as long as
the offenders are tried for correspondingly grave domestic crimes. Presumably the rationale is that
it matters that people be “tried and sentenced for their behavior”, rather than the more “aesthetic”
requirement that they be “tried for what they did according to the international community’s best
characterization of it.” However, this is the one case where this author would find merit in the
contention that insufficient implementation could lead to a real unavailability. In most cases except
perhaps isolated war crimes, judging individuals for “ordinary” domestic crimes will be the
equivalent of denying a crucial contextual and systemic aspect to their acts, which will almost
inevitably reveal an at least partial unwillingness to try them for what they did. This is something
that the ICTY’s statute had foreseen when, as an exception to the non bis in idem rule, it anticipated
a situation where “the act for which he or she was tried was characterized as an ordinary crime”.60
Prosecuting for international rather than ordinary crimes is therefore perhaps the area where
NGOs are on the surest footing when recommending implementation. More controversial from the
point of view of complementarity is any implementation requirement leading to changes in the
nature of the judicial or legal system beyond the incorporation of the core crimes. A number of
demands are made of states in terms of reforming their criminal procedure which seem a long way
away from what complementarity is about. For example Human Rights Watch, insists that “States
should be aware that certain procedural or evidentiary rules that effectively prevent the proper
investigation and prosecution of some crimes, may lead the ICC to determine that the state is
“unwilling” to take action”. The example given is that “some jurisdictions have onerous and
discriminatory requirements relating to the prosecution of crimes of sexual violence, such as a legal
requirement that there be several male eyewitnesses to a rape of a woman. (…) This type of
requirement may not be consistent with an intent to bring the perpetrator to justice.”
Although a similar provision was not adopted at the Rome conference, there is a sense that the
reasoning applicable to prosecutions for ordinary crimes may be the same as when no prosecutions
arise at all: a form of foot-dragging, that lies somewhere between inability and unwillingness.
61 In a
different vein, HRW suggests that “National procedures should also provide for victims and
witnesses or their representatives to make representations to national courts that are considering
an application for interim release by detained persons under article 59(3), and ensuring that the
national courts take those representations into consideration in such decisions”.62 Also relevant
here are all situations where modes of imputability are deemed to be defined “too narrowly” or
defenses “too broadly”, or where sentences are “too” indulgent. As an example of the latter, the
FIDH seeks to amend Cambodian law so that the Cambodian King shall have no right to reduce
sentences in cases of convictions for ICC crimes, and that the National Assembly shall be barred
from adopting amnesties in case of genocide, crimes against humanity or war crimes.63
60 ICTY Statute, article 10 2. (a).
Bruce
Broomhall has argued that “the application of narrower general principles than contemplated by
61 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
62 7 Ibid.(
63 6 Report: International Criminal Court, Implementation of the Rome Statute in Cambodian Law,
(2006).
21
the Statute could make a case admissible before the Court if national law thereby allowed impunity
where the Court would punish”.64
These demands made on implementation clearly represent the qualitative outer-edge of what the
ICC might ever have to consider in admissibility hearings. In such cases, it would surely become
very hard to argue that there was an “unavailability” of the legal system. Clearly complementarity
anticipates a modicum of friction and domestic limitations in the prosecution of international
crimes, and perfection (whatever that might be) is not required. As Darryl Robinson puts it “there is
no obligation to adopt the procedures appearing in the Rome Statute” and “it is not the purpose of
the ICC to impose universal criminal procedure”.
65
The root of the confusion is that the Court might act a bit like a human rights court, supervising a
sort of “right to an effective remedy” of victims, and ensuring that it is not violated substantively,
procedurally, or jurisdictionally. But the ICC’s jurisdictional mechanism is a rougher, less
discriminating one. The ICC’s complementarity scrutiny is quite far removed from a human rights
compliance analysis. The exercise of domestic criminal jurisdiction could not be totally inefficient,
but nor will the Court likely find a state “unwilling or unable” to prosecute someone simply because
domestic prosecutions do not conform with the understanding of how they should unfold
internationally. As the Group of Experts on complementarity put it, “The standard for showing
inability should be a stringent one, as the ICC is not a human rights monitoring body, and its role is
not to ensure perfect procedures and compliance with all international standards”.
The same surely is true of a variety of elements
not central to the ICC’s core mission, as envisaged from the point of view of complementarity.
66
CONCLUSION
In essence, the risk is that complementarity will be used to occasionally implement a liberal
orthodoxy about what international criminal law should be, caricaturing the reality of
complementarity. There is scattered evidence that some NGOs are clearly aware of the fact that
their vision of implementation is not simply about implementation, and involves an attempt to
more generally bring domestic law in line with various international obligations.67
64 23 B. Broomhall, International justice and the International Criminal Court : between sovereignty and the
rule of law, (Oxford ; New York 2003), pp. x, 215 p.
For example
65 1 D. Robinson, The Rome Statute and its Impact on National Law, In A. Cassese, P. Gaeta and J. R. W. D.
Jones (eds.), The Rome statute of the international criminal court : a commentary, (Oxford ; New York
2002), pp. 2 v. (cxl, 2018 p.).
66 22 ICC-OTP, Informal expert paper: the principle of complementarity in practice, (2003).
67 24 J. S. D. K. Piragoff, Commonalities and Differences in the Implementation of the Rome Statute, In M.
Neuner (ed.), National legislation incorporating international crimes: approaches of civil and common law
countries, (2003). (“some national legislation developed to implement the Rome Statute will serve a broader
policy objective, to ensure that national judicial systems can meet all existing obligations under international
law”.)
22
Human Rights Watch emphasizes that “Implementation (…) will also benefit the state by
strengthening its national criminal justice system, bring national law into conformity with
international obligations and up to date with important developments in international law”.68
The process raises issues about the contours of the “implementation industry” more generally, with
its recognizable litany of experts, consultancies and research projects. There is a sense that
complementarity was seized as a “once in a lifetime” opportunity to ensure that states ratify all
major relevant international instruments and reform key aspects of their criminal justice system.
The drive to fully implement the Rome Statute has had all the hallmarks of an ideal donor-project:
manageable, change inducing, concrete and labor intensive. Yet it also tends to rely on a command-
and-control, top-down view of international law. Particularly strong is the rhetoric of the “Rome
Statute incentive”,
In
other words, what is proposed is fundamentally an opportunity for legal reform.
69 and of a “carrot-and-stick mechanism”.70 “Model implementation kits”71
Although very little connection has been drawn in the literature between ICC implementation and
the law reform sector, there are undeniably areas of overlap, and some of the critiques that have
been addressed to the latter could certainly be transferred to the study of international criminal
justice. One should not underestimate the drain on the resources of relatively weak and already
overburdened legislatures, for example in some countries from the Global South, of the sort of
comprehensive package that is now being held up as necessary to adopt. One study, for example,
emphasized “a discernible capacity shortfall in some of the countries manifesting in an over-
stretched and thinly-staffed system, and a lack of sufficient officials with expertise in drafting
legislation or in international criminal co-operation”.
reinforce the sense of implementable upgrades, like a patch for a faulty software, and the
overwhelming emphasis on issues of technique of implementation tends to avoid asking the hard
questions.
72
Although this is typically noted as a weakness that should be remedied, e.g.: through more donor
funded programs, I believe it should also raise issues about the impact of the international
implementing agenda on domestic legal systems. This in a context where there may be much more
68 7 Making the International Criminal Court Work: A Handbook for Implementing the Rome Statute,
(2001).
69 23 B. Broomhall, International justice and the International Criminal Court : between sovereignty and the
rule of law, (Oxford ; New York 2003), pp. x, 215 p.
70 11 J. Kleffner, "The Impact of Complementarity on National Implementation of Substantive International
Criminal Law" (2003) 1 Journal of International Criminal Justice 86-113. ne might remark that states are not
donkeys, and the view of the international community or NGOs as their handlers is more than a little self-
serving.
71 See, e.g.: the SADC Model Enabling Act on Ratification of the Rome Statute of the International Criminal
Court or the Commonwealth Model Law to Implement the Rome Statute of the International Criminal Court.
72 25 M. Du Plessis, "Complementarity and Africa: the promises and problems of international criminal
justice: essay". 19 O. Bekou and S. Shah, "Realising the Potential of the International Criminal Court:
The African Experience" (2006) 6 Human Rights Law Review 499-544.
23
pressing legislation to adopt, indeed even reform legislation (economic, political) that may do more
to prevent genocides than the rather formalist constraints of the Rome statute implementation (not
to mention international criminal law legislation on issues such as terrorism which these countries
are also pressured to implement). It is important to retain an awareness of how the international
implementation agenda risks displacing local legislative priorities. Typically, the campaign to ratify
and implement the Rome statute evidences no such self-doubt.
There are costs to this overloading complementarity. One tactical cost may be the perception by
some states that prosecuting offenders will not suffice and that joining the Court requires a very
substantial overhaul of the criminal system, when it arguably only requires a broad legislative and
political readiness. Although this is clearly not the only factor, punctilious implementation
requirements may be one of the things that make states like the US wary of joining. A more
domestic but no less significant cost may be the creation of tensions within domestic criminal law
(particularly the general part) through the introduction of new and alien concepts that sit oddly
within the state’s legal system. International-domestic transplants are an intriguing concept that
can gain much from the study of comparative law, but it seems important not to create
“international bubbles” within domestic legal systems.73
None of these dangers may seem pressing at present, but there is no mistaking a certain drive that
could have implications in the future. The better view has always been that “states have
considerable leeway in the implementation of the core crimes, however desirable the full
implementation of the core crimes might be politically” -– something which is even more true when
it comes to a number of features linked to the more general operation of the criminal law.
Implementation really only requires at best “equivalent” rather than “identical” standards.
Yet another more systemic danger is that
an excessive drive to homogeneization will reduce the inherent diversity of the world’s legal
traditions, some of which are no doubt a threatened species in an era of globalization. More
broadly, rigid implementation may preempt debates of considerable complexity about issues of
transitional justice (e.g.: amnesties) that should occur domestically, and on which the Rome Statute
is not particularly clear.
74 In fact,
“complementarity does not require that domestic prosecutions are carried out in precise accord
with the legal regime under the ICC Statute,” nor to “change their domestic law to ensure parity
with the Statute in all aspects of domestic criminal law and procedure in order to avail themselves
of the principle of complementarity.”75
73 This might happen, for example, if a particular meaning were given to a domestic legal concept whenever
prosecutions for international crimes occurred, distinct from the meaning attributed to it ordinarily. Or
consider pressure to not impose the death penalty when international crimes are at stake, to align domestic
legal systems with the evolving trend of international criminal justice. If this were done but the death penalty
were not abolished for ordinary crimes, one would be faced with a very iniquitous paradox.
As Darryl Robinson puts it “there is no obligation to adopt
74 26 M. Goldmann, "Implementing the Rome Statute in Europe: From Sovereign Distinction to
Convergence in International Criminal Law?".
75 27 H. Duffy, "National Constitutional Compatibility and the International Criminal Court" (2001) 11
Duke J. Comp. & Int'l L. 5.
24
the procedures appearing in the Rome Statute” and “it is not the purpose of the ICC to impose
universal criminal procedure”.76
Behind the temptation to “over-implement” lies, I suspect, a particular vision of international
criminal justice as a universalizing mechanism, that requires a high degree of uniformity, both in
terms of specifically international (universal jurisdiction, foreign immunities) and domestic
(substantive and procedural criminal law) jurisdictional features. As Otto Triffterer put it in one of
the leading works on implementation:
There is good cause to believe that the jurisprudence of the ICC will have a clarifying and
unifying effect on the international and the domestic level of criminal law. The expectation
concerns not only the definitions of crimes, but also general principles … defences and
procedural rights. (ICC jurisprudence) may … in addition, influence and improve the
respective regulations in domestic legal systems and, thus, contribute not only to the
globalization of criminal justice but also to the rapprochement of the major legal systems of
the world, as well as of the numerous different domestic criminal laws.77
International criminal law is itself the product of a formidable drive to universalization,
harmonization, and standardization. The temptation, having “rationalized” international criminal
law to what some may see as a high degree of perfection, is to think that states should by and large
emulate it. The ICC is supposed, after all, to offer the state of the art in terms of substantive
definitions and procedural due process. But the truth is the ICC is an imperfect institution, meant to
operate in a very specific environment. It is neither the whole of international criminal law, nor in
some cases the best of international criminal law.
78
76 1 D. Robinson, The Rome Statute and its Impact on National Law, In A. Cassese, P. Gaeta and J. R. W. D.
Jones (eds.), The Rome statute of the international criminal court : a commentary, (Oxford ; New York 2002),
pp. 2 v. (cxl, 2018 p.).
In seeking too much homogeneization of
domestic criminal law systems, the ICC might even be seen as sawing the branch on which it is
sitting: one that has formidably benefited from the diversity of sources from which it has been able
to draw, as a vast reservoir of legal diversity.
77 16 O. Triffterer, Legal and Political Implications of Domestic Ratification and Implementation Processes,
In C. Kress and F. Lattanzi (eds.), The Rome Statute and Domestic Legal Orders.: Volume I: General Aspects and
Constitutional Issues, (2000). Also 24 J. S. D. K. Piragoff, Commonalities and Differences in the
Implementation of the Rome Statute, In M. Neuner (ed.), National legislation incorporating international
crimes: approaches of civil and common law countries, (2003). (“The development of national implementing
legislation could facilitate a merger of distinctive national approaches to criminal law relevant to the template
established by the Rome Statute”. These authors note, however, that at the same time “each continues to
reflect characteristics unique to that State’s legal history”).
78 Indeed, one of the ironies is that although the ICC is at the forefront of the fight against impunity, nor is it
(and neither were its predecessors) necessarily the most liberal system of criminal justice. Whether it be
command responsibility, the possibility of life penalties, or of anonymous testimony, there are several areas
where domestic systems can rightly see themselves as more protective of fundamental liberties.
25
What is needed therefore, although this is beyond the scope of this article, is to develop an
understanding of complementarity that sees it not only as a concession to sovereignty, but as
perhaps primarily a device to accommodate diversity. Complementarity tolerates many degrees of
implementation and beyond that, many ways of going about criminal justice, not only because this
diversity exists, but because it has normative value in itself. Although this is something that NGOs
and scholars readily acknowledge when it comes to legal traditions (primarily common and civil
law) and constitutional restraints, the tolerance should extend further to cover such diverse issues
as attitude towards transnational jurisdiction, or the role of criminal and transitional justice. A
broad outline of how international criminal law can be both universal and plural is still largely
missing.
In seeking to take some first steps beyond hegemonic like-mindedness, several ideas come to mind,
taken from regional integration. One would be derived from EU directives, and the idea that their
implementation is more about the result than the means, effectively granting states considerable
latitude as to how they reach a particular result. ICC implementation could thus be characterized as
less about certain forms and more about certain goals, emphasizing a degree of political and social
preparedness about the need to deal with the worst crimes. Another idea that comes to mind is the
theory of the “margin of appreciation”, as developed by the European Court of Human Rights. There
is a “core” to what complementarity and therefore implementation requires no doubt, but most
departures from what international criminal and human rights lawyers might see as the “ideal” way
of making individuals accountable for their crimes, will not make a system “unavailable” for the
purposes of international criminal justice. Only a very significant consensus about the needs of a
particular form of criminal jurisdiction, at least among state parties to the Rome Statute, will justify
considering that a state has failed to fulfill the complementarity test if it has not modified its
domestic laws accordingly.
