Degan - Sources of International Criminal Law

# The Author 2005. Published by Oxford University Press. All rights reserved.
.......................................................................................................................................
On the Sources of
International Criminal Law
Vladimir-Djuro Degan
Abstract
In their efforts to establish a quite original system of procedural and material rules of
international criminal law, by means of the so-called ‘‘judge-made law’’, the two
ad hoc Tribunals for the Former Yugoslavia and Rwanda hold a peculiar approach
to the sources of that law. The most controversial of all is their concept of ‘‘customary
law’’. This paper is an attempt to clarify the meaning and scope of these sources mainly
from some aspects of the respective rules adopted in the 1998 Rome Statute. It is also a
continuance in this author’s research on the sources of public international law.
I. Introduction
The International Criminal Tribunal for the former Yugoslavia (ICTY) was originally estab-
lished by the UN Security Council (SC) Res 827 (1993) of 25 May 1993, with the primary
aim to put an end to the crimes committed in that region, first of all in Bosnia-Herzegovina.
It is very doubtful whether the establishment and functioning of an ad hoc criminal tri-
bunal can fulfil these expectations.1 It is, however, beyond doubt that before and after all
the armed conflicts in the Balkan region ended, the Office of the Prosecutor of the ICTY
marked the locations of the biggest crimes and identified their victims. These tragic facts
are not easy to deny, not even in a judicial procedure.
Much more complicated is the issue of prosecution and punishment of all responsible persons
involved. Under normal circumstances, which, in an atmosphere of hatred against the enemy,
never materialize, it should be expected that the respective States themselves, on the basis of the
principle of territoriality, will identify the direct perpetrators of these horrible misdeeds and their
immediate commanding officers, and prosecute them in a normal criminal procedure according
to their national laws.2 On the basis of such an investigation into the responsibilities ‘‘in

Professor of International Law, Rijeka University; Head of the Adriatic Institute of the Croatian Academy of
Sciences and Arts in Zagreb; Membre de l’Institut de Droit international (email: jz@hazu.hr).
1 Hence, a Swiss journalist, Pierre Hazan, in his remarkable book, La justice face à la guerre, de Nuremberg à La
Haye, Stock, Paris (2000), 260 –261, quoted Judge Antoine Garapon: ‘‘It was not the Nuremberg Tribunal that
liberated the concentration camps of Auschwitz and Mauthausen, but it was done by the Red Army and the
Americans.’’ As is known, the Nuremberg Tribunal was established after the World War II in Europe had ended.
2 The fact is that before and during the process of the dissolution of the former Yugoslavia, in which five new suc-
cessor States appeared, there were no breaks in legislation. During this process, all these States have had criminal
laws in force in which was enshrined the substance of all four international crimes which, in 1993, became
Articles 2–5 of the ICTY Statute. However, there was simply no will to put these provisions into effect.
....................................................................................................................................................................
Chinese Journal of International Law (2005), Vol. 4, No. 1, 45 –83 doi:10.1093/chinesejil/jmi002
46 CJIL (2005)
pyramid’’, it could be possible to identify higher military and political leaders who ordered, or
were not willing to prevent, the crimes in question, and to sue them, too.
But with some rare and insignificant exceptions, such a normal course of events has never hap-
pened in the history of mankind.3 Every party to a conflict is eager to punish the crimes by the
enemy inflicted on its population or soldiers. But their own perpetrators of crimes and their com-
manders are regarded in the public opinion as national heroes, who are, as such, untouchable.
The establishment of international criminal courts and tribunals should be a remedy to
these imperfections in human behavior. These tribunals should sue all suspects for all inter-
national crimes within their jurisdiction, so that none can escape his responsibility and that
all, regardless of their nationality or official position, should be, if convicted, punished
accordingly.
To these noble ends, two ad hoc tribunals have recently been created, as well as the Inter-
national Criminal Court (ICC), in The Hague. Hence, we are now faced with two tracks in inter-
national criminal proceedings. On the one hand, there are the Statutes, the frequently amended
Rules of Procedure and Evidence, and already a rich practice of the ICTY and of the International
Criminal Tribunal for Rwanda (ICTR). On the other hand, there is a much more elaborated text
of the 1998 Rome Statute, as well as the Elements of Crimes, for the ICC, and its very detailed
Rules of Procedure and Evidence. There now only remains to emerge the practice of the ICC in
the application and interpretation of these complex written rules.
One could join these two separate tracks into one larger road. The ICTY and the ICTR
could partly adjust their practices to these newly written rules of international criminal law
since 1998, and consider it as a single system of legal rules and principles.
However, some prominent international lawyers choose to contradict these two different
sets of legal principles and practices. In their writings, they strongly criticize almost all the
aspects in which the Rome Statute and its additional instruments substantially depart
from the existing law and practice. They label the existing practice as the alleged ‘‘customary
law’’ in this field. In addition, in all substantial innovations in the 1998 instruments, they see
the victory of criminal lawyers (together with the United States at the Rome diplomatic
conference) over international lawyers.
The disagreements relate to many fundamental concepts of criminal law, such as the pre-
sumption of innocence of the accused, the scope of the principle nullum crimen sine lege, all
sources of international law laid down in Article 38(1) of the Statute of the International
Court of Justice (ICJ), etc.
Hence, one of the most prominent French legal authors criticizes the exhaustive defi-
nitions of crimes in Articles 6–8 of the Rome Statute, and the subsequent elaboration
and adoption of the Elements of Crimes. He asserts that this was a misapprehension by
3 These few known exceptions relate to the trials by the US courts for war crimes committed by the US armed
forces in the Philippines in 1900 –01, and the few instances of punishment by the Ottoman courts in 1919 –
20 of persons responsible for crimes against Armenians. However, the My Lai ‘‘incident’’, in which 347 Vietna-
mese civilians were massacred, and the process against Lieutenant Calley alone with the pardon granted to him by
President Nixon seem rather to be a mockery of justice. There were many more cases where the commission of
serious crimes was treated as an offence of a disciplinary nature.
On the Sources of International Criminal Law 47
criminal lawyers of the scope of the principle nullum crimen sine lege. In support, he quoted
another author, who ‘‘impeccably demonstrated’’ that the position of the drafters of the
Rome Statute is incorrect. According to the common view of these international lawyers,
the customary law consolidation of the definition of the four ‘‘grand crimes’’ was certainly
sufficient to ensure the respect of the nullum crimen principle. ‘‘Custom is a source of inter-
national law to the same extent as treaties and is just as apt to constitute the indispensable
lex.’’ 4 This appears equally from Article 11(2) of the 1948 Universal Declaration of
Human Rights which confirms that ‘‘international law may be a source of legality for the
purpose of applying the nullum crimen principle . . .’’.5
If I understand this view correctly, international judges should apply this ‘‘customary law’’
lock, stock and barrel part of positive international law. The matter is therefore not of a
lex ferenda created by judges and applied retrospectively to the detriment of the accused.
But that is not exactly what this author has in mind.
Pellet’s additional reasoning opens many new questions of the utmost importance. By
giving way to American pressure, by not trusting the judges to interpret and apply inter-
national law in its present state ‘‘and such as it is evolving’’, the authors of the Rome
Statute have, in his view, limited the chances of making the Court an efficient instrument
in the struggle against the crimes it is supposed to repress, ‘‘the most serious crimes of
concern to the international community as a whole’’. He also argues:
Unfortunately, men’s criminal imagination appears unlimited and, by enclosing the

definition of the crimes in narrow, punctilious formulations, they have forbidden the
judges in advance to suppress future malevolent inventions of the human spirit; all
the more so, and this is undoubtedly the most serious weakness of the Statute,
because, in practice, they have excluded any realistic prospect of amendment.6
Happily enough, the alleged men’s criminal imagination and malevolent inventions of the
human spirit are not the only features of human species as such. Normal human beings are
entirely devoid of that ‘‘gift’’, regardless of their sex, race, colour, descent, language, creed
or national or ethnic origin. We can imagine that only some persons have such a vicious
imagination. But because of their cowardly character, or the limits imposed on them by edu-
cation or religious beliefs, they never dare to display it in their lifetime. However, probably the
narrowest group of men and women reveal their criminal character in situations of armed
conflicts or social turmoil. They commit and force others to commit horrible crimes allegedly
in the name of their nation, or religious group, or their party in the conflict.
The foregoing remarks seem to also reflect the views of many Judges of the ICTY, at least in the
first years of its functioning.7 In prosecuting international crimes, they presumed to act, together
4 Cf. Alain Pellet, referring to Luigi Condorelli, Applicable Law, in: A. Cassese, P. Gaeta and J.R.W.D. Jones (eds),
2 The Rome Statute of the International Criminal Court, A Commentary, Oxford (2002), 1058.
5 Ibid., 1057 –1058.
6 Ibid., 1058 –1059.
7 The ‘‘imagination of future torturers who wish to satisfy their bestial instincts’’, mentioned in the comment of
common Art.3 of the 1949 Conventions, was literally quoted in para.237 of the Judgment by the Trial Chamber
48 CJIL (2005)
with the Prosecutor of the Tribunal, in the name of the international community. However, the
Security Council could not properly authorize them to implement such a far-reaching mandate
because it would simply exceed its own competencies according to the UN Charter. A careful
reading of the text of the Statute that it adopted in 1993 proves that it actually did not do that.
The Judges of the ICTY appropriated in their practice some other prerogatives to them-
selves which no judicial body is expected to exercise. Among them is the large law-making
power, including the creation of new ‘‘customary law’’ in order to repress the unlimited
man’s criminal imagination.
Nevertheless, every judiciary power should be limited and controlled by other powers, by defi-
nition. In normal States, the judiciary, legislative and executive control one another. The Judges
of the ICJ are restrained by the agreement of States by which they place their dispute in its jur-
isdiction. The same applies to inter-State arbitration. For this reason, the Court and the arbitra-
tors are very careful in defining the applicable customary and other law in their judgments. That is
because they try, strictly within the jurisdiction established by them, to convince the parties that
their dispute was decided according to positive law. It is not different with the advisory opinions
by the ICJ, which are rendered also on the basis of the will of States acting in the UN General
Assembly, or the Security Council, or other competent organs which can request them.
It is precisely for the above reasons that the concepts adopted by the ICJ of jus cogens and
of customary law differ from those declared by the Judges of the ICTY. We will discuss this
issue in more detail below.
The foregoing criticism reveals the viewpoints of many Judges of the ICTY on some other
issues as well. They consider that the aim of the international criminal judiciary is, above all,
prosecution of international crimes. The latter is, however, the normal function of the Pro-
secutor in an adversarial procedure under the control of the judiciary. On the contrary, the
function of every judicial body worthy of that name is to do justice according to ( positive)
law.8 As a consequence, there is little room in the above concepts for the presumption of
innocence of the accused, which is the starting point of every fair trial.
against Tihomir Blaškić of 2000, that, in spite of the fact that the Judgment itself stated several times that General
Blaškić did not personally commit any crime for which he was accused. And in a large number of his written
commands cited within it, there is none in which he ordered his subordinates to commit any crimes. And
although Blaškić was not even accused of the crime of genocide, parts of the Nuremberg Judgment of 1946
on persecuting Jews were also quoted in the same Judgment ( para.222). Nonetheless, the Appeals Chamber,
in its Judgment of 2004, reduced his sentence from 45 to 9 years of imprisonment, finding that the Trial
Chamber did not establish beyond reasonable doubt Blaškić’s ordering any of the crimes. He was, however, sen-
tenced for not preventing and punishing his inferior officers who inflicted inhuman and cruel treatment and
hostage-taking of civilians.
8 The full title of that body, ‘‘International Tribunal for the Prosecution of Persons Responsible for Serious Viola-
tions of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991’’,
should not mislead anybody. As the Nuremberg Tribunal stated in its Judgment of 1946 concerning its discre-
tionary power under Art.9 of its Charter to declare a group or organization as criminal: ‘‘This discretion is a judi-
ciary one and does not permit arbitrary action, but should be exercised in accordance with well-established legal
principles, one of the most important of which is that criminal guilt is personal, and the mass punishment should
be avoided.’’ Cf. AJIL (1947), 250 –251. Hence, even if the UN Security Council had, in 1993, in mind the
punishment of all the suspects for crimes committed in the former Yugoslavia, a judicial body cannot exercise
its functions beyond the above-mentioned limits.
Taking this into account, it should finally be observed that the above legal constructions
perfectly fit when the fate of others is at stake. Nevertheless, it is not hard to imagine, in
abstract terms of course, that a legal author or any of the former or present Judges of the
ICTY could himself come into the situation of being indicted by the Prosecutor as a
suspect of horrible crimes in the capacity of a military or political superior, on the basis
of the responsibility under Article 7(3) of its Statute. These accusations can, of course, be
devoid at the outset of any factual or legal grounds.
But suppose that an accused in such a lamentable situation has the right to choose.
Would he, from his personal perspective, like to be submitted to the procedure of the
ICTY and be at the mercy of its Prosecutor and Judges, enjoying large discretionary
powers (even in law-creating), or would he rather opt to be tried by the ICC according to
its Statute and the Elements of Crimes? The author of this paper definitely prefers the
latter procedure.
Concerning the issues discussed so far, there does not seem to be a matter of misunder-
standing between the experts of public international law on the one side, and the experts of
criminal law on the other. More likely, there is a misapprehension by some of the former of
the sources of international criminal law—the topic of the present paper.
II. Sources of international criminal law and Article 38(1) of

the ICJ Statute
The most authoritative conventional provision on the sources of international law in general
is still Article 38(1) of the ICJ Statute. That is because the Statute constitutes an integral part
of the UN Charter and the ICJ is the principal judicial organ of the United Nations. That
provision reads as follows:
The Court, whose function is to decide in accordance with international law such
disputes as are submitted to it, shall apply:
a. international conventions, whether general or particular, establishing rules expressly

recognized by the contesting States;
b. international custom, as evidence of a general practice accepted as law;
c. the general principles of law recognized by civilized nations;
d. subject to the provision of Article 59,9 judicial decisions and the teachings of the most
highly qualified publicists of the various nations, as subsidiary means for the determi-
nation of the rules of law.10
9 Article 59 of the Statute reads: ‘‘The decision of the Court has no binding force except between the parties and in
respect of that particular dispute.’’
10 Paragraph 2 of Art.38, which reads ‘‘This provision shall not prejudice the power of the Court to decide a case ex
aequo et bono, if the parties agree thereto’’, does not reflect sources of international law and can be overlooked in
this paper.
50 CJIL (2005)
It is generally agreed that there is no firm hierarchy between the three main sources of law
indicated in subparagraphs (a) –(c). On the other hand, it is obvious that the sources
indicated in subparagraph (d) are of a subsidiary character in relation to the three main
sources.11
The above sources normally function in the international legal order of sovereign States
and other international persons. Because treaties cannot directly bind States which are not
parties to them, the main source of what is understood as general international law is the
general custom. General principles of law are applied in this order as an autonomous
source only if they are not transformed into general customary law. That transformation
most often happens through codification conventions.
Needless to say that in international criminal law, relations between Judges, the Prosecu-
tor, the accused and his counsel, and the victims of the crime are not the same as the relations
between sovereign States, which enjoy legal equality and possess the autonomy of will. This is
a matter of trials of accused persons, which are not substantially different from criminal trials
before national courts. But one essential difference is that the source of law in international
criminal proceedings is not the national criminal law of any State, but the law emanating
from the sources listed in Article 38(1).
The general principle of all criminal law—nullum crimen nulla poena sine lege—was
already mentioned. Many call it the principle of legality (in its narrow sense).12 The said
principle is an achievement from the 1789 French Declaration of the Rights of Man and
of the Citizen and is embodied in the constitutions of a great many of States as being one
of the guarantees of the rights of individuals.13 Even if, strictly speaking, it is not a peremp-
tory norm of general international law ( jus cogens), it is better to observe it than to
undermine its importance in any criminal proceedings.14
Where we actually take this principle into account, and there are no convincing reasons to
do the contrary, all sources of international law provided in Article 38(1) appear in a
11 This enumeration, although authoritative, is not all-exhaustive any more. In addition to general custom, the inter-
national practice knows particular customary rules which, like treaties, relate to a limited number of international
persons. This paper will, however, deal with the general international customary law alone. In addition, Art.38(1)
does not make any mention of unilateral acts by which a State can assume some legal obligations at its charge, or
can, in some situations, acquire new rights in its profit. See details on all these sources, V.D. Degan, Sources of
International Law, Martinus Nijhoff Publishers, The Hague (1997), 564 pages.
12 It seems to be more appropriate to embrace, under the principle of legality, the sum of general principles of crim-
inal law which should be observed in any criminal proceedings.
13 Antonio Cassese, International Criminal Law, Oxford (2003), 139 –158 has exposed many arguments challen-
ging this principle as allegedly consisting in the doctrine of ‘‘strict legality’’, in contrast to the doctrine of ‘‘sub-
stantive justice’’. His arguments that this ‘‘doctrine’’ (including the principle of non-retroactivity of law) does not
operate in international law in the same way as within States with codified law do not seem to be convincing
enough.
14 In any case, the principles opposed to it—nullum crimen sine poena (that no crime must remain unpunished) and
of ‘‘sound popular feeling’’, being the sources of law for Nazi criminal judges in Germany between 1933 and
1945—were strongly criticized by the Permanent Court of International Justice. See its 1935 Advisory
Opinion on Danzig Legislative Decrees, PCIJ, Series A/B, No.65, 56. Those who consider the former judicial
practice altogether as ‘‘customary law’’, should on the basis of the stare decisis doctrine observe this precedent.
somewhat different light from that in inter-State relations. From this principle proceed the
following requirements of a fair criminal trial:
— Nullum crimen sine lege scripta: the basis of a criminal charge should be either in the
national law of a State, or in a statute of an international criminal court or tribunal. In
both instances, the matter should be subject to a rule of positive law in written form.
This excludes incriminations based exclusively on (unwritten) customary law.15
— Nullum crimen sine lege certa: the elements of crimes must be precisely defined by a
rule. This forbids the criminal judge to resort to analogy. To this end, the Assembly
of States Parties to the Rome Statute of 1998 has adopted the Elements of Crimes, to
be applied by the ICC.
— Nullum crimen sine lege previa: a crime must be forbidden by law at the time of its
commission. Retrospective application of new criminal laws is forbidden, unless
they were more favourable to the accused (lex mitius).
— Nulla poena sine lege: the penalties for specific crimes should also be provided by a
legal rule in advance. It is hard to strictly respect this requirement in international
criminal proceedings. The scale of prison sentences for the crimes within the compe-
tence of international criminal tribunals have so far not been provided in their
Statutes.
Hence, in criminal law, either municipal or international, written sources have the pre-
ference over unwritten ones. This means that in international criminal law, customary
rules cannot have the same importance as in the international legal order of sovereign
States in which a near totality of rules of general international law is of customary character.
There is another substantial difference in this regard. When two or more States agree on
the jurisdiction of the ICJ, or on international arbitration, they expect to obtain in these
procedures the final judgment of their case. To this end, in order to avoid non liquet,
Judges resort to all sources of law provided in Article 38(1). They must find applicable
legal rules to any dispute which States can refer to them.
International criminal judges do not have such a large power. As stated, they are strictly
forbidden to resort to analogy, or to apply new rules of positive criminal law retrospectively
to the accused, or even to create new rules of ‘‘customary law’’ and, by that, cure the
imperfections in public international law.
But this problem of non liquet, which was presented by some internationalists, is not of the
major importance for proper functioning of international criminal tribunals.
One cause for the deficiencies of the ICTY Statute is certainly that the drafters of its
Articles 2–5 did not discern the very fact that most crimes in the former Yugoslavia were
committed in non-international armed conflicts. The process of disintegration of that
Federation lasted some time. It took place in a series of protracted armed conflicts during
15 This prerequisite excludes even more the quasi-legislative function of criminal judges and forbids them to create
new rules which, at the time of the commission of a crime, were not lex lata, and then to apply them on the
accused retrospectively.
52 CJIL (2005)
which new successor States acquired their independence at different dates. Only from these
dates could an originally internal armed conflict transform into an international one.
Hence originates the problem of the application of the grave breaches of the 1949 Geneva
Conventions, as Article 2 relates (like these Conventions themselves, except their common
Article 3) only to international armed conflicts, and Article 3 sparsely defines violations of
the laws and customs of war, which are applicable in all sorts of conflicts. That poses
the major problem of applicable written law to war crimes in non-international armed
conflicts.
On the other hand, poor definitions of the crimes against humanity in Article 5 of the
ICTY Statute, which certainly encompass the practices of ‘‘ethnic cleansing’’ and of enforced
prostitution as particular forms of already existing crimes,16 do not impede the punishment
of any responsible person. If the Tribunal finds an accused guilty of one or more crimes pro-
vided in Article 5(a)– (h) (murder, extermination, deportation, torture, persecution, etc), it
can punish him properly. However, when fixing the penalty, the Judges can take into con-
sideration ‘‘other inhumane acts’’ provided for in subparagraph (i) as aggravating factors. It is
therefore not necessary to define in Article 5(i) new crimes against humanity and apply them
to the accused retrospectively.
Certainly, definitions of all crimes against humanity as set forth in Article 7 of the Rome
Statute seem to be more appropriate and more complete. But it is not a job of judges in
deciding on past crimes to create perfect definitions of crimes which fall into their jurisdic-
tion. It is better for them to stick to the nullum crimen sine lege principle.
All the foregoing suggests a particular order of explanation of the sources of international
criminal law. We shall start with the general principles of law which any criminal judge is
supposed to observe in a fair trial. Then, an explanation of treaties as a source of law for
an international criminal judge will follow and, after that, the scope of custom within the
international criminal law. Following these explanations, we will turn to the subsidiary
sources provided in Article 38(1)(d). We shall finish our presentation with a comment on
Article 21 of the Rome Statute—‘‘Applicable law’’.
III. The general principles of (criminal) law

In the doctrine and practice of public international law, ‘‘the general principles of law recog-
nized by civilized nations’’, as stated in Article 38(1)(c), are understood in two quite different
meanings.
First, the general principles of law can be understood as the basis of any legal relationship.
That is because no legal order can properly function in the absence of such principles as the
rules of positive law. In their opposite sense, in absence of an applicable treaty or cus-
tomary rule, i.e. in case of gaps (lacunae) in what is considered as positive international
law, Article 38(1)(c) is usually interpreted as an option, or even an authorization, given to
international judges to have resort to some legal principles common to national legal
16 See the Report of the Secretary-General pursuant to para.2 of the SC Res 808 (1993) (S/25704), para.48.
systems. These systems are still considered as more developed and more sophisticated than
international law.17
Although analogy is forbidden to any criminal judge in respect of incrimination of human
behaviour, there is still some place for general principles of law in this narrow sense as a sub-
sidiary source of international criminal law, in particular if it is so provided in advance. That
is the case with Article 21(1)(c) of the Rome Statute, to which Article 31(3) of the Statute
refers concerning the grounds for excluding criminal responsibility. Nevertheless, the scope
of application of these principles in this narrow sense is rather exceptional in this discipline.18
It is more important to envisage the general principles of law in the broad sense, as a basis
of any legal order. There, they are not opposed to customary law and treaties but, in this
perspective, all these three main sources form a unity.
In general international law, this is the case with its basic norm of pacta sunt servanda. It is
worth quoting the explanation in this respect given by Sir Gerald Fitzmaurice:
This rule does not require to be accounted for in terms of any other rule. It could
neither not be, nor be other than what it is. It is not dependent on consent, for it
would exist without it. There could not be a rule that pacta sunt non-servanda, or
non sunt servanda, for then the pacta would no longer be pacta. Nor could there
be a rule that pacta sunt interdum servanda et interdum non sunt servanda. The
idea of servanda is inherent and necessary in the term pacta. 19
The same is true with other general principles of law forming part of the law of treaties. They
have been operating from the times immemorial whenever sovereign States agreed on valid
treaties with the intention to respect them. Before the 1969 Vienna Convention, there were
some doubts about how these general principles, mostly derived from the civil law, can apply
to treaties concluded between States. Discussions were, for instance, intense on the scope of
the clausula rebus sic stantibus in international law.
When codifying this part of the law, the International Law Commission relied on State
practice as constituting rules of customary law, or at least on judicial precedents. It
wanted to prove that all its propositions were rules of positive international law. However,
in respect of error, fraud and corruption vitiating the consent of States, there was no such
17 It seems that only in this second narrow sense, and as distinct from treaties and customary law, are these general
principles understood by Bruno Simma and Andreas Paulus: ‘‘Le rôle relatif des différentes sources du droit inter-
national pénal (dont les principes généraux de droit)’’, in: H. Ascensio, E. Decaux and A. Pellet (eds), Droit inter-
national pénal, Paris (2000), 55 –69; see, in particular, 62 –65. On the contrary, Pierre-Marie Dupuy, Normes
internationales pénales et droit impératif ( jus cogens), ibid., 71 –80; see jus cogens in some ‘‘general principles of
criminal law’’, ibid., 73 –75. He therefore envisages these principles in the broader sense.
18 However, the genuine nature of these general principles of law is most transparent in the so-called ‘‘transnational
law’’, which is extremely poor in substantial legal rules. When arbitrators decide disputes about contracts con-
cluded between States and foreign corporations or private banking institutions, there are almost no other
sources of applicable law except the contract itself. Then the arbitrators largely apply ‘‘general principles of
law recognized by civilized nations’’, because the text of the contract would, in their absence, be almost mean-
ingless. See details on that practice, V.D. Degan, Sources (1997), 113 –126.
19 Cf. Sir Gerald Fitzmaurice, Some Problems regarding the Formal Sources of International Law, in: Simbolae
Verzijl, The Hague (1958), 153.
54 CJIL (2005)
practice at all. Still, the Commission proposed the inclusion of these circumstances excluding
the validity of treaties as being a part of positive international law and not lex ferenda or
precepts of natural law. On that ground, there emerged Articles 48– 50 of the 1969
Vienna Convention.20
Today, almost all provisions of the Vienna Convention are considered as rules of custom-
ary international law. They are applicable to all existing or future treaties, even if one or all of
their parties did not ratify or accede to the Convention.
This is very similar with the rationalization of general principles of law as existing rules of
positive international law in codification efforts by the International Law Commission on
State responsibility. If its proposed draft articles become a convention in force, then its
content will also be generally recognized as reflecting customary law in this important domain.
The rules of arbitral procedure were codified by the Hague Conventions for the Pacific
Settlement of International Disputes of 1899 and 1907. The basic principles of international
judicial and arbitral procedure can be detected also in the Statute and Rules of The Hague
Court, in the Rules of the Court of Justice of the European Communities, and in the statutes
of all administrative tribunals of international organizations. If a judicial or arbitral body, in
exercising its functions, ignores one or more of these principles, it will jeopardize the legality
of its judgment and its own character as a judicial organ.21
Because of their continuing to be under the influence of former positivist or normativist
teachings, even if unconsciously, some experts in public international law do not realize
the impact of general principles of law in this broad sense in international criminal law.
Hence their disagreements with experts in criminal law who have no such problems in
the same discipline.
Historically, the general principles of criminal law first appeared in municipal law of some
States, as guarantees against arbitrariness of judges as State organs. Since their very begin-
nings, they form a part of human rights. However, the development of criminal law was
not quite uniform in all States.
From paragraph 39 of the Magna Carta Libertatum, issued by the English King John in
1215, followed the right of every free individual to be judged by his equals. From that royal
guarantee, there appeared criminal trials in which the jury decides whether the accused is
guilty or not, and in which the role of a professional judge is minimal. Later on,
some other pledges of fair trial were extorted from the sovereign also in the form of ‘‘guar-
antees’’. The development in the United States was similar. The first 10 amendments to
the Constitution of 1789 forming the ‘‘Bill of Rights’’ embrace some guarantees in criminal
proceedings, and of liberties of citizens in general.
20 See details in Degan, Sources (1997), 72 –89. It was there also discussed whether these ‘‘everlasting’’ general
principles of law really affect the sovereign will of States.
21 In a case where all parties to an actual dispute provide, in their arbitral agreement, the procedural rules essentially
different from those referred in the above instruments, the matter will not be of a judicial, i.e. arbitral procedure
anymore, but perhaps of a different mode of pacific settlement (mediation or conciliation), or, in some instances,
even of agreement by the parties on otherwise unlawful intervention by third parties into their relations.
The legal development in the continents of Europe and Latin America and in some other
States took a different path. The 1789 French Declaration of the Rights of Man and of the
Citizen was an attempt to codify all rights and freedoms of individuals in a single text. Its first
11 paragraphs provide the most important of what are now understood as the general principles
of criminal law. They are, inter alia: the principle that law must be the same for all, whether it
protects or punishes; the nullum crimen sine lege principle, including the non-retroactivity of
criminal laws; and the principle that limits to human liberties can only be determined by law.
Later on, important legal rules were aggregated concerning the personal criminal respon-
sibility of an individual, the definitions of different forms of perpetration and participation
in a crime, the mental ( psychological) element of a crime (mens rea), grounds for excluding
criminal responsibility, mistakes of fact and mistakes of law.
In these countries with codified law, the sum of detailed rules of this character is provided
in the ‘‘general part’’ of criminal codes. They bind every judge in all criminal proceedings. A
judge cannot ignore them or modify them by creating new legal rules. However, the essential
role in these proceedings is performed by professional judges ( juge d’instruction and judges in
the grand jury). Even if lay assessors form a majority in criminal chambers, they cannot inde-
pendently decide on the guilt of the accused.
It is now important to outline how the procedure of international criminal tribunals devel-
oped since 1945. The London Agreement on the Nuremberg Military Tribunal adopted
mainly the Anglo-Saxon procedure but without any jury of laymen. Since then, all ad hoc
tribunals have consisted of professional judges only.
It should be noted that the Anglo –American adversarial model with a jury has the advan-
tage of releasing the professional judges of some responsibilities and helping them to avoid
the mistakes they can otherwise commit. On the contrary, the correctness of this model
deprived of the jury depends almost exclusively on the professional skill, consciousness,
impartiality and independence of judges. If they are incapable of handling their responsibil-
ities, such a procedure can degenerate into its opposite—the one prevalent in the inquisitions
until the end of the 18th century. Hence the contrast between the quality of the Nuremberg
Judgment of 1946 and that of the Tokyo Judgment of 1948.
A jury would probably release Dražen Erdemović from his responsibility for the crimes he
committed. Perhaps, it would also release General Tihomir Blaškić for most of the crimes for
which the Trial Chamber of the ICTY sentenced him, in 2000, to 45 years of imprisonment.
In these two cases, it probably could not have happened that juries of laymen would create a
miscarriage of justice in respect of the accused.
Nevertheless, for a number of practical reasons, the jury in international criminal proceed-
ings seems unrealistic. But the adversarial model of the procedure without a jury results in
very long and expensive trials, each lasting several years and usually with little effect. For
this very reason, perhaps, the European inquisitorial model, with a juge d’instruction and
trials by chambers of professional judges on the basis of a file (‘‘dossier’’) formed by the inves-
tigating judge, could have been speedier and more efficient. But, in such a case, and in all
other cases in which international criminal tribunals consist of professional judges, the
56 CJIL (2005)
general principles of criminal law should have been codified in advance by the statute of the
tribunal, exactly as it was done in Part 3 of the Rome Statute of 1998.
The Statutes of both ad hoc Tribunals for the former Yugoslavia and Rwanda are poorly
drafted. The subsequent practice of these Tribunals has revealed many important lacunae in
their regulations. They both do not provide the most important general principles of crim-
inal law. However, their common provision (Article 15 of the ICTY Statute and Article 14 of
the ICTR Statute) provides the following:
The judges of the International Tribunal shall adopt rules of procedure and evidence
for the conduct of the pre-trial phase of the proceedings, trials and appeals, the admis-
sion of evidence, the protection of victims and witnesses and other appropriate matters.
The general principles of criminal law are not even mentioned. They probably fall into the
category of ‘‘other appropriate matters’’. It is a usual practice in inter-State arbitrations that
their parties empower the arbitrators, appointed by their agreement, to adopt the procedural
rules before the proceedings. They then just adopt and transcribe the established general
principles of law in that subject matter either from the 1907 Hague Convention, or from
the Model Rules of Arbitral Procedure drafted by the International Law Commission in
1958. There is no disagreement on the content of these basic procedural principles at all.
On the other hand, it did not seem appropriate to confer on Judges of the ad hoc Criminal
Tribunals, who were elected by the UN General Assembly from a list proposed by the Secur-
ity Council, to adopt rules of international criminal procedures in the domains in which,
prior to the 1998 Rome Statute, there were no such codified rules.
However, even in the situation that actually happened, the first concern of conscientious
criminal judges should have been to provide in advance, in the Rules of Procedure and
Evidence, detailed provisions of personal criminal responsibility of the accused, including
those on actus reus and mens rea, different forms of perpetration and participation in a
crime, grounds for excluding criminal responsibility and mistakes of fact or law.
A less favourable option seems to be to provide in these Rules a general clause according to
which, for all issues not regulated by the Statute, the relevant rules of criminal law of the State
in which the crime was committed will apply, provided that these rules are not inconsistent
with the norms of international law. In both instances, international judges should apply the
general principles of criminal law as having already been in written form.
However, the ICTY, from the very beginning, adopted another option which was reflected
in the Judgment of the Trial Chamber in Kupreškić, of 14 January 2000 (IT-95 –16-T):
. . . any time the Statute does not regulate a specific matter, and the Report of the
Secretary-General does not prove to be of assistance in the interpretation of the
Statute, it falls to the Tribunal to draw upon (i) rules of customary international
law or (ii) general principles of international criminal law; or lacking such principles,
(iii) general principles of criminal law common to major legal systems of the world;
or, lacking such principles, (iv) general principles of law consonant with the basic
requirements of international justice. ( para.591)
Here, the matter is, therefore, not of observing the established general principles of criminal
law, but mainly of their ascertainment apparently through a comparative analysis.22 In this
respect, the two ad hoc Tribunals sharply differ from the approach of Rome Statute, the
Elements of Crimes and the Rules of Procedure and Evidence of the ICC.
Now, let us first explain how the above principles were applied in a case. Dražen Erdemo-
vić, according to his confession, took part in the summary execution of unarmed Moslem
civilians after the fall of Srebrenica on 16 July 1995, where he, himself, executed about
60 persons. After his escape from Bosnia, Erdemović was first surrendered to a Serbian
court and then transferred to the Tribunal, and admitted that he did kill the Moslems.
The matter seems to be of the most heinous crime of genocide. However, the Tribunal
admitted that the accused acted under duress (vis absoluta) from his commander, Brano Goj-
ković. Before these executions, he ventured to resist participation in the killing. Later on, he
declared before the Tribunal that ‘‘if I had refused, I would have been killed together with the
victims’’.
In that situation, which was not contested by the prosecution, a jury of laymen would
most probably find Erdemović not guilty. Exactly the same would be done by the ICC in
application of Article 25(3)(a) of the Rome Statute. Erdemović acted as a simple tool or
instrument of his commander who was the actual, though indirect, perpetrator of the geno-
cide. The Court would exclude Erdemović’s responsibility on the basis of Article 31(1)(d) of
the Statute because he acted ‘‘by duress resulting from a threat of imminent death’’.
In his trial before the ICTY, everything was different. The Prosecutor did not indict Erde-
mović for genocide, although it was exactly that crime that was committed, but ‘‘only’’ for
crimes against humanity under Article 5 of its Statute, or, in the alternative, violations of
the laws and customs of war under Article 3. The Trial Chamber, by its judgment of 29
November 1996, sentenced him to 10 years of imprisonment under Article 5 only.23
The Appeals Chamber, in its judgment of 7 October 1997, confirmed that duress was no
defence, but vacated the Trial Chamber Judgment because Erdemović was thought not to
have made an informed plea. It therefore remitted the case to a new Trial Judgment so
that he could make a new plea.
The new Trial Chamber accepted his plea of guilt on violations of laws and customs of war
under Article 3 and, by its judgment of 5 March 1998, sentenced him to five years of
imprisonment.
However, the majority of Judges of the Appeals Chamber in 1997 confirmed the view
from the first sentencing judgment of 1996 that ‘‘duress does not afford a complete
22 Besides, in its practice, the Tribunal frequently makes use of the terms ‘‘customary law’’ and ‘‘general principles of
law’’ as synonyms. That is probably the reflection of its view that it can create itself new ‘‘customary law’’ by its
decisions, like the above quoted in the Kupreškić Judgment. It seems, however, controversial that statements in
judgments by a trial chamber which were subsequently repealed by the Appeals Chamber and all the convicted
released have the force of precedents and fall, as such, within the scope of rules of ‘‘customary law’’. All these issues
will be discussed later on.
23 For a thorough critical analysis of that judgment, see Sienho Yee, Towards an International Law of
Co-progressiveness, Martinus Nijhoff (2004), 115 –161.
58 CJIL (2005)
defence to a soldier charged with a crime against humanity and/or war crime involving the
killing of innocent human beings’’.24
The long individual and separate opinions by each of the five members of the Appeals
Chamber in 1997 reveal how risky the establishment of simple general principles of criminal
law is, if it is presumed that they constitute ‘‘customary rules’’ of international law and are
perhaps different from the respective principles in national legislations, which can also largely
disagree on details. The general principles of criminal law such as duress are not to be proved;
they simply must be applied.
It is, however, most interesting to ask, what happened with Brano Gojković as an indirect
perpetrator of the genocide, and with all other soldiers in the squad who, without resistance,
executed civilians? We do not know the identity of others, but the name of Gojković is not
on the list of accused preferred by the prosecution of the Tribunal. Perhaps he subsequently
died, or there is a secret indictment against him, or the Prosecutor is simply not interested in
this matter anymore. In any case, there is no doubt that this horrible crime against innocent
civilians was committed and, in this light, it is difficult to grasp the purpose which the trial of
Erdemović alone served.
The Tribunal starts from the postulate that the international criminal proceedings are
different from procedures before national courts. It seems hard to support this position.
For sure, international tribunals have within their jurisdiction the prosecution of the most
serious international crimes, while municipal courts have in their competence a large spec-
trum of human misdeeds of very unequal social consequences. But the greater the gravity of a
crime imputed to an accused, the higher the risk of miscarriage of justice. It is exactly for this
reason that codified general principles of criminal law in written form have at least the same
importance for international tribunals consisting of professional judges as for national judi-
cial bodies where professional judges also dominate.
There is no room to analyse here in detail other judgments of the ICTY which involved the
application of the general principles of criminal law which were not fixed in advance in the
Statute. The matter is, first of all, of the presumption of innocence of the accused, but also
of the principle that law must be the same for all, whether it protects or punishes, and of
the principle of ‘‘equality of arms’’. It seems sufficient to make some indications in this respect.
According to the two Statutes, the Prosecutor is invested with very broad powers in pre-trial
investigations. He/she not only has the absolute freedom to decide whether or not to initiate
investigation and against whom, but carries out any judicial scrutiny. Only at the end of inves-
tigations does the reviewing judge admit or dismiss the indictment. But even in that judicial
phase, the suspect and his counsel have no right to assist in the proceedings. For this reason,
the confirmation of an indictment is more often than not a simple formality.25
24 Cf. Judgment by the Appeals Chamber of 7 October 1997, para.19, and in the disposition under (4). Judges
McDonald, Vohrah and Li, in their separate opinions, made, inter alia, an almost complete list of provisions
from national laws on duress. Judges Cassese and Stephen dissented.
25 The UN Security Council has conferred such an extensive power to the Prosecutor probably in the belief that his
or her vigorous action will prevent the commission of further large-scale crimes on both territories of the former
Yugoslavia and Rwanda. That, however, did not happen. In both instances, it probably also wanted to diminish its
To these powers, self-appointed missions were added by both the Prosecutor and the two
Tribunals to act as organs of prosecution in the name of the international community. In per-
forming that common mission, the Judges acquiesced in some practices for making the life of
the prosecution easier. Hence, the strong insistence on the ‘‘equality of arms’’ between the
prosecution and the accused at the stage of trial proceedings only does not seem to be a suffi-
cient guarantee for the respect of the presumption of innocence of the latter throughout the
process. We can here note some practices which, in the view of this author, are not likely to be
fully in accordance with the sum of general principles of criminal law.
Although the two Statutes do not provide the competence of the Tribunals to condemn an
organization as criminal, the Prosecutor is allowed to indict a group of persons for a ‘‘joint
criminal enterprise’’. Unlike Article 25(3)(d) of the Rome Statute, the matter here is not of a
form of contribution to the commission of the principal crime by furthering criminal action
of actual perpetrators, but of a presumed act of co-perpetration in a crime. If the Tribunal
does not strongly insist on the intent and knowledge of each member of such a supposed
group, the crime can be imputed to any executioner of an order issued by his political
superior who was perhaps the ultimate perpetrator. The actus reus and mens rea can then
be attributed to the accused on indirect evidence, or even from factual circumstances.26
The Tribunal admits, in addition, the practice of confirming indictments against a person
at the same time for ‘‘planning, instigation, ordering, committing, or otherwise aiding and
abetting in the planning, preparation, or execution’’ of a crime. This is the full formula from
Article 7(1) of the Statute.27 However, in case the reviewing Judge has, from the outset,
forced the Prosecutor to specify the acts he imputes to the accused, that could only
enhance the prestige of both the Tribunal and its Prosecutor as being impartial. That
could facilitate the co-operation with the Tribunal by governments of all the successor
States of the former Yugoslavia.
Similar to that is the practice followed by the Tribunal without rational restrictions on
indictments of a person for a single act or transaction, for alternatively having been in
breach of Article 3 (on violations of the laws or customs of war) and of Article 5 (crimes
against humanity). This is admitted as a precaution that in case the Prosecutor does not
primary responsibility for maintenance and restoration of international peace under Art.24(1) of the UN Charter.
Instead of undertaking enforcement actions against the violators of the peace by armed forces whenever it was
really necessary, the Security Council settled for the activities of the Prosecutors of these two Tribunals which
it, itself, had established.
26 In this respect, the still pending case of Croatian General Ivan Cermak can be of interest. After the liberation of
the city of Knin, in the operation ‘‘Storm’’, he performed, between 5 August and 15 November 1995, the duty of
military commander charged with the civilian affairs there. He had allegedly no armed forces, or even police,
under his command. He was, inter alia, accused by the prosecution for command responsibility under
Art.7(3) of the Statute, as well as for his participation in a joint criminal enterprise, for committing all crimes
in that larger region. It will be intriguing to analyse the future judgment by the ICTY in case the prosecution
does not prove the responsibility of General Cermak under Art.7(1).
27 These unqualified statements in indictments are abusively interpreted by some groups in Croatia which resist co-
operation with the ICTY. They impute to the Tribunal and its Prosecutor their condemnation of acts of liberation
of the Croatian territory between 1992 and 1995 as allegedly criminal acts. The liberation itself was a perfectly
lawful action, but some crimes against unarmed civilians during it were nevertheless committed which fall within
the jurisdiction of the ICTY.
60 CJIL (2005)
prove in the proceedings the specific requirements of one crime, then the accused could be
punished for another. This is nothing like the equality of arms between the two parties
during the proceedings.
When the matter is of command responsibility under Article 7 of the Statute, the Trial
Chamber has sometimes condemned a person cumulatively for ordering or planning
a crime under paragraph (1) and, at the same time, for not preventing his subordinates to
actually commit it under paragraph (3).28
It is true that, in most cases, the persons who bear responsibility for abhorrent crimes have
been adequately punished by the ICTY. But there are still some borderline cases where the
so-called ‘‘judge-made law’’ in the domain of general principles of criminal law could lead to
unjust condemnations. In their constant wish to create new legal rules, the Judges actually
blurred the threshold of personal responsibility of lower commanding officers for the
deeds of their highest political or military superiors.
The next instance is of ‘‘judge-made law’’ in respect of mental element (mens rea) of the
accused. According to Article 30(1) of the Rome Statute, unless otherwise provided, a person
shall be criminally responsible ‘‘only if the material elements were committed with intent
and knowledge’’. The intent and knowledge of the accused cover dolus directus in the first
and in the second degree.29 However, the so-called dolus eventualis is not provided by the
Statute. Conscious negligence or recklessness is provided only in Article 28 concerning
the responsibility of commanders and other superiors. But unconscious negligence, or dis-
regard of the obligation of due diligence, is not provided in the Rome Statute expressly,
even as a ground of criminal responsibility for military superiors. It is close to the notion
of ‘‘strict liability’’ which, in criminal law of civilized nations, is excluded as a general
principle.30
In what is labelled as ‘‘customary rules’’ in international criminal law, dolus eventualis is
simply called ‘‘recklessness’’ (which is otherwise synonymous with negligence).31 A new
notion of ‘‘culpable negligence’’ is introduced which, in fact, covers the unconscious
28 It seems that the Appeals Chamber has stopped this unlawful practice. Referring to some of its former decisions, it
stated in the Blaškić Judgment of 2004 the following: ‘‘Where both Article 7(1) and Article 7(3) responsibility are
alleged under the same count, and where legal requirements pertaining to both of these heads of responsibility are
met, a Trial Chamber should enter a conviction on the basis of Article 7(1) only, and consider the accused’s
superior position as an aggravating factor in sentencing’’ ( para.91).
29 Dolus directus in the first degree is provided in Art.30(2): ‘‘. . . a person has intent where: (a) in relation to conduct
it means to engage in the conduct; and (b) in relation to a consequence, it means to cause it.’’ Dolus directus in the
second degree is covered by Art.30(2) under (b): ‘‘. . . a person is aware that a consequence will occur in the ordin-
ary course of events’’; and in Art. 30(3): ‘‘. . . the ‘knowledge’ means awareness of the person that a circumstance
exists or a consequence will occur in the ordinary course of events.’’
30 Such a form of responsibility exists in public international law only if it is agreed by a treaty in advance. Criminal
laws of some States provide it for some future traffic or similar delicts of minor importance. For an analysis of all
provisions concerning mens rea, see Albin Eser, Mental Elements: Mistake of Fact and Mistake of Law, in:
A. Cassese, P. Gaeta and J.R.W.D. Jones, The Rome Statute of the International Criminal Court: A
Commentary, Oxford (2002), Vol.I, 890, 902 –903.
31 In American law, in case of ‘‘wanton negligence’’, the intent of the perpetrator is sometimes presumed, even if it is
not proved.
negligence.32 That was an attempt to extend the personal responsibility of accused beyond
reasonable limits.
A paradoxical situation has appeared in the still pending process against Slobodan
Milošević. This process is under careful scrutiny of the world media and critical comments.
Milošević, allegedly holding de facto position of superior authority of Bosnian Serbs, is
indicted, inter alia, for the crime of genocide in Bosnia-Herzegovina, pursuant to Article
7(3) of the ICTY Statute. It would not be a major problem to prove his responsibility for
his recklessness or ‘‘culpable negligence’’ for not preventing this crime. However, the
media expect from the prosecution either direct documents or at least depositions of trust-
worthy witnesses in support of that accusation, which it has not produced so far. Due to the
considerable financial resources it spends and the very broad powers in pre-trial investigation,
this is taken as a failure of the prosecution to do the job.
Hence, in a process against the person which probably bears the greatest political respon-
sibility for the destruction of Yugoslavia and for most crimes that followed, the judge-made
‘‘customary law’’ simply does not function, which was otherwise easily admitted by the Tri-
bunal against persons of much lesser importance and for lesser crimes. This means that the
Tribunal did not favour the prosecution when it did not stick to the universally recognized
standards of mens rea.
It remains to establish what that ‘‘judge-made law’’ qua customary international criminal
law consists of. In case any of its authors or partisans suffered, himself, the ungrounded accu-
sations for a presumed crime, he would intimately opt for the proceedings before the ICC
according to the rules laid down in the Rome Statute.
This, however, does not mean that the Rome Statute itself is free from any criticism. The
main lacuna in its provisions relates to the total absence of a general provision concerning
‘‘the commission by omission’’.33 Again, the omission, as a criminal act, is only provided
in Article 28 on the responsibility of commanders and other superiors. A commander is
responsible for a crime committed by the forces under his effective command and control
‘‘as a result of his or her failure to exercise control properly over such forces’’. For the rest,
there is no general provision on omission which is otherwise found in most municipal
criminal laws.
It seems that this lacuna was deliberate, especially with respect to armed forces sent on
peace-keeping missions in areas affected by conflicts. In the case of crimes of genocide,
such as those committed in Rwanda in 1994 and at Srebrenica in Bosnia in 1995, only
direct perpetrators and their commanders are responsible. Responsibility of representatives
of Member States of the UN Security Council and of commanders of peace-keeping
troops for not averting these crimes is thereby excluded. In all these situations, none of
them had the special intent ‘‘to destroy, in whole or in part, a national, ethnical, racial or
religious group’’. None of them can therefore be indicted for omission under Article 25(3)
32 Instead of a review of the practice by the ICTY, see the explanation in this respect by the first President of the
ICTY, Antonio Cassese, International Criminal Law, Oxford (2003), 168 –175. However, the text of the Judg-
ment in Blaškić by the Trial Chamber of 2000 was the best example of all practical consequences of that doctrine.
33 Such a general rule is provided in Art.86(1) of the Protocol I of 1977 to four Geneva Conventions of 1949.
62 CJIL (2005)
of the Rome Statute. But bearing in mind the huge tragic consequences of such omissions,
this does not seem to be entirely justified.
All the foregoing proves that because of their axiomatic character, the general principles of
criminal law cannot, in this legal branch, be confused with or assimilated into custom as a
source of public international law. The careful codification of these general principles in
Part 3 of the Rome Statute, as well as the submission of the powers of the Prosecutor to the
judicial authority of the ICC, will probably diminish the unwelcome practice indicated above.
IV. Treaties as a source of law in international

criminal proceedings
In the international legal order of sovereign States, treaties are the main source of what is
called ‘‘particular international law’’. National criminal judges in their capacity of State
organs are obliged to apply treaties in force for their respective country. This obligation
relates especially to treaties in the domains of human rights, including the guarantees
of a fair trial, immunities of some foreign persons and international judicial co-operation
including extradition.34
Unlike national courts, international criminal tribunals are not organs of any particular
State and they are not bound to apply treaties. The principle pacta sunt servanda is not appli-
cable there as such, unless a tribunal itself was established by a treaty. In these cases, such as
the London Agreement of 1945 or the Rome Statute of 1998, conventional instruments are
constitutional acts of these bodies, much in the same manner as the UN Charter is for the
United Nations Organization.
In relation to this subject matter, one should distinguish the violation of some treaties as
consisting of particular crimes from the respect of treaties as one of the sources of law to be
applied by international judges.
According to Article 227 of the 1919 Versailles Peace Treaty, the Allied and Associated
Powers publicly accused the former German Emperor William II ‘‘for a supreme offence
against international morality and sanctity of treaties’’. In Article 6 of the London Agree-
ment, it was again provided, among the Crimes Against Peace, waging war ‘‘in violation
of international treaties, agreements and assurances’’. Exactly the same crime under the
same heading was set forth in Article 5 of the Charter of the International Military Tribunal
for the Far East.
Because the prosecution of aggression is not in the competence of any of the current
international tribunals, the violation of treaties does not figure as a specific crime in their
respective statutes.
As a potential source of law for international criminal judges, treaties can be viewed from a
different aspect. The question is whether an accused can be punished for the violation of
34 However, if a State has assumed the obligation by a treaty to prosecute criminal acts, such as, e.g. genocide, or the
obligation concerning non-applicability of the statute of limitations to war crimes and crimes against humanity,
national judges cannot give effect to these legal commitments before the adoption of the necessary national
legislation.
provisions of the conventions codifying humanitarian law or for the breaches of bilateral
agreements which the conflicting parties may have concluded. This problem occurred in
the practice of the ICTY. It is closely related to the transformation of provisions of these con-
ventions concerning their grave breaches in customary international law. Let us concentrate
here primarily on the conventional aspect of this problem.
The Report of the UN Secretary-General (S/25704) which, according to Article 32 of the
1969 Vienna Convention on the Law of Treaties, can be considered as travaux préparatoires
for the purpose of interpretation of the ICTY Statute, was, in this respect, quite clear:
34. In the view of the Secretary-General, the application of the principle nullum
crimen sine lege requires that the international tribunal should apply rules of inter-
national humanitarian law which are beyond any doubt part of customary law so
that the problem of adherence of some but not all States to specific conventions
does not arise. This would appear to be particularly important in the context of
an international tribunal prosecuting persons responsible for serious violations of
international humanitarian law.
35. The part of conventional international humanitarian law which has beyond
doubt become part of international customary law is the law applicable in armed
conflicts as embodied in: the Geneva Conventions of 12 August 1949 for the Protec-
tion of War Victims; the Hague Convention (IV) Respecting the Laws and Customs
of War on Land and Regulations annexed thereto of 18 October 1907; the Conven-
tion on the Prevention and Punishment of the Crime of Genocide of 9 December
1948; and the Charter of the International Military Tribunal of 8 August 1945.
The provisions from the above conventions were the basis for drafting Articles 2– 5 of the
ICTY Statute. Unlike the said 1949 Geneva Conventions, to which practically all the
States of the world have become parties, some important States, such as the United
States, Israel, India, Pakistan and some Arab countries, have never acceded to Protocol I
of 1977 to these Conventions.
It was for this reason that grave breaches of Protocol I were not provided as crimes in
Article 2 of the Statute. This is quite natural because the ICTY has the jurisdiction to prosecute
any person responsible for committing crimes in its jurisdiction in the territory of the former
Yugoslavia since 1991. In case a US citizen is indicted by the Prosecutor for grave breaches of
this Protocol alone, his counsel will be right to claim that the defendant cannot be punished
for this crime. Although the former SFRY and all its successor States were parties to both
Protocols of 1977, a criminal tribunal must respect what is called the ‘‘universal measures
of repression’’. It cannot punish the accused according to their citizenship.
Nevertheless, the drafters of the ICTY Statute have probably not realized the full
importance of the problem of applicable law based on the conventions in non-international
armed conflicts. As already stated, Article 2 is applicable only to the crimes committed
in international armed conflicts, while Article 3, applicable to all sorts of conflicts, was
poorly drafted and it provides an open definition of crimes.
64 CJIL (2005)
In its 1995 Tadić Jurisdiction Decision, the Appeals Chamber decided that ‘‘Article 3 of
the Statute is a general clause covering all violations of humanitarian law not falling
under Article 2 or covered by Articles 4 or 5’’.35 Hence, it construed this poor definition
as an opportunity to amend Article 3 by including other crimes it deems necessary for its
proper functioning. In the view of this author, the Appeals Chamber was wrong to
include, inter alia: ‘‘(iv) violations of agreements binding upon the parties to the conflict,
considered qua treaty law, i.e. agreements which have not turned into customary inter-
national law.’’36
First of all, this newly added crime refutes the necessity of the ‘‘universal measures of
repression’’ by a criminal tribunal, because the extent of the crimes to be imputed to an indi-
vidual depends on the agreements that his party has accepted.
Only a few combatants or their commanders are aware of these commitments ‘‘considered
qua treaty law’’. In case there is proof that the accused himself has participated in the con-
clusion of such an agreement, that fact could be taken as an aggravating factor in his punish-
ment. On the contrary, when there is no direct evidence that the accused did even know of
the existence of such an agreement, he cannot be guilty of its violation.37
For the rest, it is difficult to imagine that a person could be punished only for the violation
of such an agreement without having been found guilty of more serious crimes. That proves
that this amendment is almost of no practical use.
In this light, it seems highly desirable that the statutes of criminal tribunals provide, in
advance, all the crimes within their jurisdiction, exactly as was done in the Rome Statute.
Amendments like that are not in accordance with the principle nullum crimen sine lege.
V. Place of custom in international criminal law

There are legal terms in general use on whose exact meaning there is no general agreement.
Some authors usually resort to some of these terms in order to assert some facts or legal rules
which they are otherwise unable to support by stronger legal arguments. For such purposes,
they sometimes use the terms of equity, natural law, even sources of international law in
general and, as shown above, the general principles of law.38
In the practice of ad hoc tribunals and in the writings of some publicists on international
criminal law, the same frequently occurs with the term ‘‘customary law’’. Their notion of this
term has substantially different features from those that appeared in the long practice of the
two Hague Courts since 1922. We feel it our duty to introduce some necessary clarifications
35 Cf. Tadić Jurisdiction Decision of 1995, para.89.

36 Ibid., para.143.
37 For this reason, the violation of ad hoc agreements of this kind falls into the domain of responsibility of States or
recognized insurgents, for which the ICTY is not competent. This means that the criminal responsibility of indi-
viduals cannot entirely replace the State responsibility here.
38 The author of this paper confronted these misunderstandings by trying to reveal the exact meaning and content of
some of these notions, including, when necessary, their historic developments. See V.D. Degan, L’équité et le
droit international, La Haye (1970), 264 pages; and with respect to other above terms, including that of
natural law, Sources of International Law, The Hague (1997), 564 pages.
here. We shall first briefly explain the meaning of custom in public international law, includ-
ing its two essential elements, and then its place in national and international criminal law.
We shall, of course, not overlook the misunderstandings of customary law created by the
ICTY and its Judges, which have already been partly revealed.
Article 38(1)(b) of the ICJ Statute correctly indicates the two elements of custom.39 A
customary rule consists of a material element: practice, which can be general or common
to two or more international persons in their mutual relations only; and a psychological
element: opinio juris sive necessitatis, i.e. the fact that a uniform, consistent and durable
practice (or ‘‘usage’’) has been ‘‘accepted as law’’ by its participants.
There is not enough space to explain here all the characteristics of these two elements, such
as the authors of the practice, its density, uniformity and frequency, duration and quality, as
well as the opinio juris. 40
The inter-relation between these two elements in the process of creating a new customary
legal rule was best explained in the 1969 Judgment by the ICJ in North Sea Continental Shelf:
. . . for in order to achieve this result, two conditions must be fulfilled. Not only must
the acts concerned amount to a settled practice, but they must also be such, or be
carried out in such a way, as to be evidence of a belief that this practice is rendered
obligatory by the existence of a rule of law requiring it. The need for such a belief,
i.e., the existence of a subjective element, is implicit in the very notion of the opinio
juris sive necessitatis. The States concerned must therefore feel that they are conform-
ing to what amounts to a legal obligation.41
It should be added that the States concerned also begin to believe that there is a reciprocal
duty for all other States to conform to the same legal obligation from the customary rule it
claims to exist. Hence, a claim of existence of the opinio juris is not necessarily equivalent to a
unilateral commitment to or abnegation of rights by such a State.
In the 1986 Judgment in Nicaragua, the Court stressed in addition that: ‘‘The Court
must satisfy itself that the existence of the rule in the opinio juris of States is confirmed by
practice.’’42 As a consequence, somebody’s belief in the existence of a customary rule does
not suffice if there is no confirmation of the opinio juris in the actual practice of States.
It is of particular importance to make a precise note of State practice and the opinio juris in
the domains of international humanitarian law and of law of armed conflicts in general. The
matter is here certainly not of the ‘‘practice’’ in everyday application of respective legal rules,
39 The wording ‘‘international custom, as evidence of a general practice accepted as law’’ was rightly criticized in the
doctrine. Besides its bypassing customary rules of particular international law, it was contended that this wording
put the cart before the horse. It should be understood that the practice as being accepted as law is the evidence
of customary rules, and not vice versa.
40 See an explanation taking into account the case law by the ICJ up to its advisory opinion on the Legality of the
Threat or Use of Nuclear Weapons of 1996, in: V.D. Degan, Sources of International Law, The Hague (1997),
147 – 174. On the basis of the explained materials, the reader can, of course, come to his own conclusions on
various factors in the customary process.
41 ICJ Reports (1969), 44, para.77.
42 ICJ Reports (1986), 98, para.184. See ibid., 99 –100, para.188, etc.
66 CJIL (2005)
such as those concerning diplomatic and consular relations, the law of treaties or the law of
the sea. A ‘‘general practice’’ does not require a worldwide armed conflict in which all States
should apply these rules in their mutual relations. Sufficient proof of the ‘‘practice’’ and
of the opinio juris is general participation by States as parties to the Geneva and other
Conventions on humanitarian law.
However, especially in the domains of humanitarian law and international criminal law,
customary legal rules appear mainly in the practice of sovereign States, coupled with the
opinio juris of their main decision makers. The practice of political organs of international
organizations, such as the UN General Assembly or the Security Council, is easy to assimilate
in the practice of States which take part in their deliberations. But, in any case, the matter
cannot be of the opinio juris of judges or of professors of international law.43 There are a
number of cogent reasons for that.
Only sovereign States keep their regular armed forces under control, which are able to
commit crimes in international armed conflicts in violation of their legal commitments.
In case of internal insurgencies, it is of the utmost importance that the State conducts the
operations against a rebel movement (which usually invokes its right to self-determination),
in accordance with the legal obligations it has assumed for that kind of conflict. In case, in
order to preserve its national and territorial integrity, the State treats rebels with cruelty, the
conflict quickly degenerates into numerous crimes on both sides. Then, retrospective appli-
cation to their perpetrators of new ‘‘customary rules’’, which were invented by judges, will
certainly not dissuade other States and rebels in other regions from repeating the practice
of the most cruel misdeeds.
Hence, because international judges (or professors) have no armed forces under their
command, their own opinio juris as to the existence of some customary rules is not the
opinio juris of any States. And because every legal rule on warfare consists of a compromise
based on ‘‘the desire to diminish the evils of war, as far as military requirements permit’’,44
the ‘‘customary rules’’ in this domain cannot be deduced by judges from ‘‘elementary
considerations of humanity’’ as they conceive them. They will better fulfil their mission
by insisting on the respect of customary rules which form positive international law.
Now we come to the role of international custom as a source of national and international
criminal law. A national criminal judge must observe in his practice peremptory norms of
general international law ( jus cogens) which are all of customary character. These norms
relate mainly to immunities from prosecution of some foreign persons, such as active
Heads of State, other high-ranking State officials and diplomatic agents. He must also
respect the inviolability of some objects, such as foreign diplomatic premises or warships, etc.
43 As stated in the above 1969 Judgment by the ICJ, the opinio juris means a belief (i.e. a conviction) that a practice
is rendered obligatory by the existence of a rule of law requiring it. The opinio juris can, therefore, never be under-
stood as a simple ‘‘legal opinion’’ by a judge or a legal author.
44 That was provided in the preamble to the Fourth Hague Convention Respecting the Laws and Customs of War
on Land of 1907. The said compromise is in the core of all the rules on armed conflicts that States are willing to
accept and observe in their actual practice at a given time. Nevertheless, the history of warfare since 1907 proved
many examples where even this minimum of legal commitments was disavowed.
More important is the question of whether a national judge can directly resort to rules
of customary international law and punish a person for his or her violations on this
ground alone.
It was already stated that the general principle of law nullum crimen sine lege impedes judges
in continental systems of codified law to resort to analogy, or to base the incrimination of some
acts exclusively on customary law. In the best tradition of the dualistic doctrine of relationship
between municipal and international law, even the rules of jus cogens in this domain must first
be transformed into municipal laws in order to be applied by a national criminal judge.
In the Anglo –American law, the situation seems different at the first glance. This problem
will be explained in our next section. It is sufficient to say here that new offences can some-
times be created by judicial precedents on the basis of the stare decisis doctrine. However,
such a ‘‘judge-made law’’ has nothing in common with customary rules of whatever
origin—domestic or international. As a consequence, the Anglo – American judges do not
directly apply in criminal proceedings the existing rules of customary international law;
nor can they create them by their own decisions.
The principle of nullum crimen sine lege is equally valid in international criminal law. As
stated above, international criminal tribunals should apply first of all written legal rules
embracing the definitions of crimes, as provided in their statutes. Resorting to analogy, or
retrospective application of new legal rules to the accused, results in violations of the
principle nullum crimen sine lege in the same way that an unlawful practice by national
courts does.
Customary rules cannot, as a matter of principle, be a direct basis of incrimination by an
international criminal judge. However, as shown in the above-quoted paragraphs 34 and 35
of the 1993 Report of the UN Secretary-General, codified customary rules of general inter-
national law can nevertheless be a model or test when drafting future statutes of international
criminal tribunals. Several requirements, however, must be fulfilled to this end.
(i) As shown above, it must be a matter of general customary norms which, being rules
of positive international law, relate to all States and other international persons in
the world. If an important convention which codified humanitarian law was not
adhered to by an important group of States, in particular by one or several perma-
nent members of the UN Security Council, the ‘‘grave breaches’’ defined in it are
not parts of general international law.
(ii) Such a rule must, at the same time, consist of a peremptory norm of general inter-
national law ( jus cogens), as defined in Article 53 of the 1969 Vienna Convention on
the Law of Treaties.45 There should not be general customary norms belonging to
the jus dispositivum. 46
45 It is provided there that such a norm ‘‘is a norm accepted and recognized by the international community of States
as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent
norm of general international law having the same character’’.
46 Some norms of this kind can be found in the 1961 and 1963 Vienna conventions on diplomatic and consular
relations, or in the 1983 Vienna Convention on Succession of States in respect of State Property, Archives and
Debts, etc.
68 CJIL (2005)
(iii) However, although all international crimes belong to jus cogens, all peremptory
norms of this character do not solely consist of international crimes. According to
the former draft Article 19(2) on the Responsibility of States by the International
Law Commission:
An internationally wrongful act which results from the breach by a State of an inter-
national obligation so essential for the protection of fundamental interests of the
international community that its breach is recognized as a crime by that community
as a whole constitutes an international crime.
The above criterion has two aspects: ‘‘one is the requirement that the obligation brea-
ched shall, by virtue of its content, be essential for the protection of fundamental
interests of the international community; the other, which completes the first and pro-
vides a guarantee that is essential in such a delicate matter, makes the international
community as a whole responsible for judging whether the obligation is essential
and, accordingly, whether its breach is of a ‘criminal’ nature.’’47 Hence, international
crimes cannot be established as such by the opinio juris of one or a few States.
(iv) A further requirement is that such an international crime with all the above attributes
was included in the competence of a particular criminal tribunal by its statute. Hence,
although aggression is generally recognized as an international crime both involving
personal responsibility of its perpetrators and the responsibility of the respective
State, important States have prevented its inclusion in the statutes of the current
ad hoc Tribunals. These Tribunals, as well as the ICC, have not within their jurisdic-
tion the power to prosecute international crimes of piracy on the high seas, slavery
and slave trade, international terrorism, drug-trafficking as a trans-national crime,
etc., although most of them meet all the requirements for international crimes.
Nevertheless, in case of the institution of any new international criminal court or tribunal,
the principle of non-retroactivity will not be disobeyed if the prosecution of past inter-
national crimes that have already been recognized as a part of positive international law is
within their competence. Under such circumstances in respect of these crimes, the principle
nullum crimen sine lege will not be affected.48
As already said, the problem with the law applicable by the ICTY appeared with the fact
that Article 3 of its Statute, which relates to internal and international armed conflicts, was
drafted in sparse terms.49 This insufficient provision on the violations of laws or customs of
47 Cf. ILCYB (1976), Volume II, Part Two, 119, para.61.

48 On this ground, the ICTY is competent to punish the crimes committed in the territory of the former Yugoslavia
since 1991, although it was, itself, established by UN SC Res 827 of 25 May 1993. Nevertheless, Arts 11 and 24
of the Rome Statute exclude the jurisdiction of the ICC in respect of the crimes committed before its entry into
force. See also ICCPR, Art.15(2).
49 The entire text of Art.3 reads as follows: ‘‘The International Tribunal shall have the power to prosecute persons
violating the laws or customs of war. Such violations shall include, but not be limited to:—(a) employment of
poisonous weapons or other weapons calculated to cause unnecessary suffering; (b) wanton destruction of
cities, towns or villages, or devastation not justified by military necessity; (c) attack, or bombardment, by whatever
means, of undefended towns, villages, dwellings, or buildings; (d) seizure of, destruction or wilful damage done to
war repeated some prohibited acts, but not all of them, listed in the Regulations annexed to
the 1907 Hague Convention (IV) Respecting the Laws and Customs of War on the Land.
The Report of the Secretary-General stresses that these provisions were, by 1939, recognized
by all civilized nations and were regarded as being declaratory of the laws or customs of war.50
Article 3 thus provides an open and unfinished definition of crimes. That could be inter-
preted as an invitation for the Tribunal itself to supplement it in one of its Judgments by
enlarging its content to include all other war crimes that may happen, especially in non-
international armed conflicts.
In order to remain within the limits of universally adopted general customary inter-
national law, the Tribunal could amend the text of Article 3 by adding to it some other
crimes as already prescribed in Article 23 of the Hague Regulations, taking into account
the development in the methods and means of warfare and harmful effects of new arms.51
Such violations also include taking hostages, using enemy prisoners of war or civilians as a
shield against attacks, and torture. All these crimes equally affect the civilian population
and combatants and there is no doubt that all of them were customary rules of general inter-
national law in 1993, when the ICTY Statute was adopted.
However, the Tribunal wanted to legislate. In the Tadić Jurisdiction Decision from 1995,
the Appeals Chamber did not enumerate specific crimes that could be committed in non-
international armed conflicts. But, in disrespect of the principle nullum crimen sine lege, it
enclosed under Article 3 the entire body of international instruments with all their
provisions.
Hence, under the Violations of Laws or Customs of War heading, the Appeals Chamber
included: (i) violations of the Hague law on international conflicts (exactly as suggested
above); (ii) infringements of the provisions of the Geneva Conventions other than those classi-
fied as ‘‘grave breaches’’ in the Conventions; (iii) violations of common Article 3 and other
customary rules on internal conflicts; and (iv) violations of agreements binding upon the
parties to the conflict, considered qua treaty law (discussed in the previous paragraph).52
The ICTY found the justification for exceeding its proper judicial function in the travaux
préparatoires concerning the adoption of the SC Res 827 establishing the ICTY. The del-
egates of the United States, the United Kingdom and France shared the view that Article
3 of the Statute included all obligations that flow from the humanitarian law agreements
institutions dedicated to religion, charity and education, the arts and sciences, historic monuments and works of
art and science; (e) plunder of public or private property.’’
50 See para.41 of that Report.
51 The crimes from Art.23 are mainly the following: ‘‘(b) killing or wounding treacherously individuals belonging to
the hostile nation or army; (c) killing or wounding an enemy who, having laid down his arms, or having no longer
means of defence, has surrendered at discretion; (d) declaring that no quarter will be given; (e) employment of
arms, projectiles, or material calculated to cause unnecessary suffering; ( f ) making improper use of a flag of truce,
of the national flag, or of military insignia and uniform of the enemy, as well as the distinctive badges of the
Geneva Convention.’’ Article 23 further provides that: ‘‘A belligerent is likewise forbidden to compel the
nationals of a hostile party to take part in operations of war directed against their own country, even if they
were in the belligerent’s service before the commencement of the war.’’
52 Cf. Tadić Jurisdiction Decision of 1995, 51, para.89.
70 CJIL (2005)
in force on the territory of the former Yugoslavia. The American delegate expressly included
in that law the common Article 3 of the 1949 Geneva Conventions and the 1977 Additional
Protocols to these Conventions.
‘‘Since no delegate contested these declarations, they can be regarded as providing an
authoritative interpretation of Article 3 (of the Statute) to the effect that its scope is much
broader than the enumerated violations of Hague law’’ was the conclusion of the Appeals
Chamber.53 The same practice was followed in subsequent cases.
The above enrichment of the content of Article 3 by the Appeals Chamber gives rise to
many interesting questions. It reveals once again the necessity of a general definition of inter-
national crimes as originally proposed in draft Article 19 on State responsibility which the
International Law Commission, in its new draft articles of 26 July 2001, simply
avoided.54 Besides that, newly added crimes refute the already discussed necessity of the
‘‘universal measures of repression’’ by any criminal tribunal. Let us, however, briefly
comment on some of the above points.
As to point (ii), it should be noted that whoever has read large texts of the 1949 Geneva
Conventions, especially those of the Third and Forth Convention, he cannot but agree with
Jean Pictet that ‘‘violations of certain of the detailed provisions of the Geneva Conventions
might quite obviously be no more than offences of a minor or purely disciplinary nature’’.55
If not the ambassadors of the permanent Member States in the Security Council, at least the
Judges of the ICTY should be aware of this fact. It is therefore difficult to find international
crimes outside the ‘‘grave breaches’’ that were specified in these Conventions as such.
In respect of point (iii), a similar crime—‘‘violations of Article 3 common to the Geneva
Conventions and of Additional Protocol II’’—was provided for the first time in Article 4 of
the ICTR Statute. Both the common Article 3 and the Protocol II do not define their grave
breaches; neither were they provided in the ICTY Statute. The UN Security Council adopted
the ICTR Statute by its Res 955 on 8 November 1994. However, the ICTY is not allowed to
apply this crime by analogy (which is forbidden in criminal law), and especially not against
the accused within its jurisdiction for their acts committed before 8 November 1994.
Finally, in respect of point (iv), it should be added that a general principle of law was
already detected in Roman law, which reads: nemo plus juris in allium transferre potest
53 Cf. Tadić Jurisdiction Decision of 1995, para.88. The authoritative interpretation is only offered in the Report of
the Secretary-General pursuant to para.2 of SC Res 808 (1993), (S/25704) of 27 May 1993, to which the ICTY
Statute was enclosed. Together with the Statute, it was adopted by the Security Council under Chapter VII of the
UN Charter. Statements by representative of Member States in the debate, even if not opposed by others, can
hardly be considered as the travaux préparatoires of the Statute for the purpose of its interpretation. That is
because these representatives cannot by their statements legally bind their country, e.g. in respect of the 1977
Protocol I. On abstract and verbal statements by State representatives as a proof of State practice in customary
process, see Degan, Sources (1997), 160 –161.
54 Instead of the old draft Art.19, the Commission proposed in its new draft Art.26 ‘‘serious breaches of peremptory
norms of general international law’’. Hence, just before the events of 11 October of the same year, in its text on
‘‘Responsibility of States for Internationally Wrongful Acts’’, the Commission unwisely suppressed any mention
of international crimes. In spite of that, we believe that the content of the suppressed article constitutes the law
in force.
55 Cf. Jean S. Pictet, The Geneva Conventions of 12 August 1949, Commentary, I, 370, Geneva (1952).
quam ipse habet. Its equivalent in relations among sovereign States is that no State can impose
legal obligations on another State that it, itself, has dismissed in respect to itself. That relates
to the above-mentioned statement by the delegate of the United States in the Security
Council that the 1977 Additional Protocols should be included in Article 3.
Hence, the Judges of the Appeals Chamber were not successful in their effort to comp-
lement Article 3 of the Statute. Their statements did not satisfy the above-mentioned criteria
for the very existence of international crimes. In respect of the equally open definition of
crimes against humanity in Article 5, it was already stressed that ‘‘other inhumane acts’’
from its paragraph (i) do not need formulation of new crimes and their retrospective
application to the defendant. Such acts, if they are established by the Tribunal, can be
taken as an aggravating factor in punishing their perpetrator.
Unlike the ICTY Statute, Articles 6, 7 and 8 of the Rome Statute were drafted with much
more care. The intention of its drafters, and of all States Parties to the Statute as well, was to
provide very broad and comprehensive definitions of all crimes within the jurisdiction of the
ICC, coupled with the Elements of Crimes. That will exclude in advance the necessity for
the legislative function of its Judges. In criminal matters, legislative and judicial power
mutually exclude one another.
Genocide, in the short text of Article 6, is defined exactly according to Article II of the
1948 Convention, which was reaffirmed in Article 4 of the ICTY Statute and Article 2 of
the ICTR Statute.56 Hence, this definition is deeply rooted in the general customary
international law without any need for subsequent improvements.
Crimes against humanity are more comprehensively enumerated in Article 7(1) of the
Statute than in any previous instrument. Compared with Article 5 of the ICTY Statute
(and Article 3 of the ICTR Statute), this Article comprehends in addition, in its sub-
paragraph (i), ‘‘enforced disappearance of persons’’ and, in subparagraph ( j), the crime of
apartheid, although the Convention on Apartheid of 1973 has never been adhered to by
the States of the former Western block. It is, however, important that in paragraph (2) of
Article 7, the meanings of most of the crimes listed in its paragraph (1) were included.
Article 8, on war crimes, is the largest and the most complex in the entire text of the Rome
Statute. It was a clear intention of its parties to confirm the distinction of war crimes that can be
committed in international armed conflicts from the crimes committed in ‘‘armed conflicts not
of an international character’’.57 The former are set forth in its paragraphs 2(a) and 2(b), and
the latter in paragraphs 2(c) and 2(e). However, in paragraphs 2(d) and 2(f), it repeated that
paragraphs 2(c) and 2(e) do not apply ‘‘to situations of internal disturbances and tensions,
such as riots, isolated and sporadic acts of violence or other acts of a similar nature’’.
56 However, the text of Art.III of the Genocide Convention was not reproduced in Art.6. Acts of perpetration of or
participation in all the crimes within the jurisdiction of the ICC are regulated by Art.25(3) of the Rome Statute.
Nevertheless, direct and public incitement of others to commit genocide constitutes, in its subpara.(e), a separate
crime.
57 Hence, this division has been a part of positive international law at least since the adoption of the 1949 Geneva
Conventions. The views de lege ferenda by some Judges of the ICTY and some other authors on its unfitness from
the aspect of ‘‘elementary considerations of humanity’’ should not influence any judicial decision.
72 CJIL (2005)
By this, Article 8 confirms all sorts of situations from the aspect of applicability of huma-
nitarian law, as previously provided in common Articles 2 and 3 of the 1949 Geneva
Conventions and in Article 1 of the 1977 Protocol II.
Unlike the statutes of ad hoc Criminal Tribunals, the Rome Statute is a conventional instru-
ment. It depends on the sovereign will of any State whether it will become a party to it. From
this aspect, there was a large margin of discretion for its drafters to provide ‘‘law-generating
rules’’ of general international law, i.e. new rules intended to legislate, and to ‘‘improve’’ the
law of armed conflicts and international criminal law. The drafters of the ICTY and ICTR Sta-
tutes did not enjoy the same freedom, and still less the Judges in application of their provisions.
It is, however, to the credit of the drafters of the Rome Statute that in spite of that freedom
to legislate, Article 8 of the Rome Statute sticks to the division between the ‘‘law declaratory
rules’’ of general customary law in this domain and the ‘‘law crystallizing rules’’ in cases
where codification conventions have not yet attained the universal approval.58
Among the rules declaratory of the customary law on international crimes are those con-
cerning war crimes in Article 8(2)(a) and (c). Paragraph 2(a) relates to the well known grave
breaches of the 1949 Geneva Conventions, which are couched in the exactly same terms as in
Article 2 of the ICTY Statute.59
In Article 8(2)(c) of the Rome Statute, serious violations of Article 3 common to the 1949
Geneva Conventions in case of non-international armed conflicts are enumerated. Their
enumeration was obviously inspired by Article 4 of the ICTR Statute. However, it leaves
out acts of terrorism, while pillaging is provided in Article 8(2)(e) of the Rome Statute.
This means that since the entry into force of the Rome Statute, all States Parties consider
these crimes as declaratory of general international law.
On the other hand, Article 8(2)(b) of the Rome Statute contains a list of 26 offences as
constituting ‘‘[o]ther serious violations of the laws and customs applicable in international
armed conflicts, within the established framework of international law’’.
The same reservation appears in Article 8(2)(e) concerning ‘‘[o]ther serious violations
of the laws and customs applicable in armed conflicts not of an international character,
within the established framework of international law’’, which includes no less that
12 sub-paragraphs.
In the view of this author, the above reservation should be understood in the context of the
law of treaties. If a suspect is accused of a crime originally provided in a convention of which
his national State is not a party, he could plead the fact that such a crime cannot be imputed
to him ‘‘within the established framework of international law’’.60
58 The three modalities of customary law resulting from general codification conventions—(a) declaring effect, (b)
crystallizing effect, and (c) constitutive or generating effect—were originally introduced by Eduardo Jiménez de
Arechaga, the former President of the International Court of Justice. See his general course at the Hague Academy
of International Law, International Law in the Past Third of a Century, 159 RCADI (1978), 14 –22. See also,
V.D. Degan, Sources (1997), 201 –215.
59 This means that item (ii) in para.89 of the Tadić Jurisdiction Decision of 1995 on ‘‘infringements of the
provisions of the Geneva Conventions other than those classified as ‘grave breaches’ in the Conventions’’ is mean-
ingless. There are no other genuine international crimes that can be found in these four Conventions.
60 It is nevertheless at the discretion of the Court to attach to this wording any meaning it finds appropriate.
We hope that this reservation is of temporary importance. Once a great majority of States,
including all the permanent members of the UN Security Council, become parties to the Rome
Statute, its Articles 7 and 8 will be considered as a complete codification of customary law on inter-
nationalcrimes,humanitarianlawand the law ofarmed conflicts,all ofwhich form, infact,a unity.
However, until this requirement is fully satisfied, some newly added crimes in Articles 7
and 8 will consist of legal rules obligatory for the ICC only. But they are, on the conventional
basis, already obligatory for all the States Parties to the Rome Statute and their citizens, in
particular if they take part in the enforcement and peace-keeping missions abroad. These
States must, from now on, fully conform the operations of their armed forces with these
provisions. Otherwise, their nationals could fall under the jurisdiction of the ICC for
committing the crimes provided in its Statute.
VI. Judicial practice and doctrine

As stated above, among the sources of international law under Article 38(1) of the ICJ
Statute figure (subject to the provision of its Article 59) ‘‘judicial decisions and the teachings
of the most highly qualified publicists of the various nations, as subsidiary means for the
determination of the rules of law’’.
Because Judges of the ICTY and ICTR frequently refer to former judicial decisions on the pun-
ishment of war crimes as ‘‘customary law’’, for the sake of clarity, it seems important to offer some
explanation in this respect. Otherwise, both notions remain completely obscured. Judicial prac-
tice is the focus of our interest here, while the doctrine will be mentioned only randomly. Nobody
has so far confused customary law with the teachings of the most highly qualified publicists.
The law of England, the United States and other countries which follow that tradition is
developed in the practice of their highest judicial organs. In these States, the doctrine stare
decisis et non quieta movere is valid. Judges are obliged to abide by former judicial precedents
when the same points arise in later litigation. However, this doctrine presupposes a hierarchy
of judicial organs and of their respective decisions. Only final decisions of the highest judicial
court have the power of precedents. We found the best explanation in respect of the legal
scope of precedent in English law:
A judgment or decision of a court, normally recorded in a law report, used as an

authority for reaching the same decision in subsequent cases. In English law,
decisions of the House of Lords are binding upon the Court of Appeals and all
lower courts and are normally followed by the House of Lords itself. Decisions of
the Court of Appeals are binding on all lower courts and, subject to some exceptions,
on the Court of Appeals itself. Decisions of the High Court are binding on inferior
courts, but decisions of inferior courts do not create any binding precedent. A lower
court is not bound by all aspects of a previous decision but only by those parts of the
judgment that constitute the principles of the decision (ratio decidendi) and are not
merely passing comments (obiter dicta) of the judge.61
61 Cf. Elizabeth A. Martin (ed.), A Dictionary of Law, Fourth Edition, Oxford University Press (1997), 348.
74 CJIL (2005)
It is similar with the federal law of the United States and the laws of any of its 50 states. It is
important to stress that in this ‘‘judge-made law’’, there is nothing similar to the creation of
new rules of customary international law, or to the application of the existing ones. Taking all
the circumstances of a case into account, judges form precedents, wishing them to be appli-
cable to most of the subsequent similar cases. By that, they try to adopt the most reasonable
decisions.
On the contrary, in the codified legal systems of Europe, Latin America, China and
other States, judges, in principle, do not create new legal norms, but apply laws passed by
the legislature. In practice, the situation is more complex. However, the transposition of the
stare decisis doctrine is not recognized in these systems as such. The precedents of the
highest courts have a high authority of standards (or examples), which is not good to
neglect. It is normal that a lower court should abide by the decisions of the court of appeals
or of the court of cassation which will finally decide on the appeal of the applicant.a
In the law applied by the ICJ, the stare decisis doctrine is also not formally recognized.
Article 59 of its Statute prevents the doctrine from being applied and confirms the principle
sententia jus facit inter partes. Besides, the will of international judges is not the same as the
wills of sovereign States which are relevant in customary law-creating process. In addition,
according to general international law, the competence of any international court or tribunal
is not compulsory. These bodies can settle international disputes by binding decisions only if
all their parties have agreed on their jurisdiction.
The ICJ and international arbitrators are, in their practice, careful not to exceed their com-
petence and the subject matter of the dispute as agreed by the parties. Particularly the ICJ
spends about three-quarters of its time considering the questions of jurisdiction and of
the admissibility of applications.
In addition, the Court is careful to avoid making comments on all questions not falling
within its jurisdiction or the subject matter of the dispute. It sometimes formulates legal
norms of general character referring to its former decisions, but only within this narrow fra-
mework. It almost never qualifies such a norm as part of jus cogens, although that can be
deduced from its judgment.
In its judgments and advisory opinions, the ICJ normally refers to its former practice and
the practice of its predecessor, the Permanent Court of International Justice (PCIJ). It refers,
however, very infrequently to previous arbitral decisions, which does not seem entirely jus-
tified. In order to corroborate its decisions, it almost never quotes doctrines. It is nevertheless
evident from large texts of individual and dissenting opinions of its judges, which extensively
refer to doctrinal views, that they are not unknown to the Court during its deliberations.
Unlike that, arbitral awards in inter-State disputes extensively refer to the precedents of the
PCIJ and the ICJ, to former arbitral practice, as well as to the viewpoints of the most highly
qualified publicists.
Because The Hague Court and international arbitrators adhere to the above restrictions,
their decisions are carefully reviewed by the doctrine. Some of them are praised and
some criticized.
In respect of the customary process, judicial decisions are often declaratory of already
existing rules of international law. In order to confirm their findings, judges usually show
proof of former State practice and the opinio juris. 62
However, in some cases, judicial decisions have a crystallizing effect. They confirm as
customary rules some existing practices of States which have not yet achieved a degree of
uniformity, and the norms in question have not yet been ‘‘accepted as law’’ (opinio juris)
by the necessary majority of States.63
It is exceptional that the ICJ tried to legislate by asserting the existence of new ‘‘customary
rules’’ without proof of any former State practice and opinio juris. 64
Some judgments and advisory opinions by the Hague Court, especially if they were sup-
ported by convincing legal reasons, enjoy high authority as evidence of existence or non-
existence of some customary rules. Upon their rendering, all States which find that these
statements corroborate their claims in respect to other States refer to them as imposing
rules erga omnes. On other States, with different interests, then falls the heavy burden to
prove the contrary.
The ICTY and ICTR have excelled in their practice of all that is disallowed by the above
restrictions. They do not refer in their decisions to former State practice and the opinio juris
as the only, or even as the main, evidence of customary law. They rather rely on the practice
of former ad hoc military tribunals established by Allied Powers after World War II in occu-
pied Germany and in the Far East, as well as to their own practice.65 They thus call that
former judicial practice, as a whole, ‘‘customary law’’.
However, considering themselves as organs of the international community, these ad hoc
Tribunals do not hesitate to create new ‘‘customary rules’’ deducing them directly from
‘‘elementary considerations of humanity’’, which are presumably their own opinio juris
sive necessitatis. The borderline between lex lata and lex ferenda is almost completely
62 There is still a paradox in that. With the increasing importance of general customary legal rules for the mainten-
ance of the world political order, the Hague Court’s duty to prove State practice and communis opinio juris is
diminishing.
63 Hence, the ICJ had already stated in its Judgment of 24 February 1982 on the Continental Shelf case (Tunisia/
Libya) that the exclusive economic zone ‘‘may be regarded as part of modern international law’’ (ICJ Reports
(1982), 74, para.100). At that time, it was not yet certain that the pending Third UN Law of the Sea Conference
would succeed. The Convention, with its Part V on that zone, was adopted at the Conference on 30 April, and it
was formally signed on 10 December 1982. It entered into force no earlier than 16 November 1994.
64 To our knowledge, it happened with a number of ‘‘equitable principles’’ on maritime delimitations defined in the
1969 judgment in North Sea Continental Shelf. This attempt was abortive. In the cases on maritime delimita-
tions that followed, it became evident that because of their abstract content, these ‘‘equitable principles’’ did not
lead to predictable results. Hence, judicial and arbitral practice returned in later cases to the basic principle of
equidistance and special circumstances, as applicable in delimitations of all maritime areas. For more details,
see V.D. Degan, Equitable Principles in Maritime Delimitations, Le droit international à l’heure de sa codifica-
tion, Etudes en l’honneur de Roberto Ago, Milano, Giuffrè (1987), vol.II, 107 – 137; Sources of International
Law, The Hague (1997), 95 – 99. It is somewhat paradoxical that the same 1969 Judgment by the ICJ formulated
side by side the most precious rules concerning customary process.
65 Even if these military tribunals, which acted in the post-war period, were organs of respective States, there is no
proof that they were willing to prosecute, on the basis of the same principles, the same misdeeds by members of
their own armed forces.
76 CJIL (2005)
blurred. Still, they expect that States themselves will conform their practices in warfare to that
judge-made ‘‘customary law’’ in future international and non-international armed conflicts.
They could perhaps even achieve some desirable results in these efforts of humanizing war,
had not they taken on almost unlimited freedom in this act of law-creating. By that, they
have betrayed their judicial function, without protecting the victims of future armed conflicts
at the same time.
These Tribunals have an ambivalent approach to the relationship between case law and the
customary process. It seems possible to discern two aspects of this problem: the application
of existing case law qua customary rules, and the creation of new legal rules of this kind.
In respect of the first aspect, there is not enough evidence that the ICTY and ICTR recog-
nize in former judicial practice the confirmation of previous State practice and of the com-
munis opinio juris, and that they apply such ‘‘customary law’’ as rules of positive international
law. Consequently, unlike the Anglo –American law, the stare decisis doctrine does not seem
to be recognized by these judicial bodies in all its essential aspects.
Large texts of both the judgments by Trial Chambers and the Appeals Chamber look like
doctoral theses where a mass of documentary material is carefully analysed, such as former
case law, respective conventional instruments, State legislation and doctrinal views. Never-
theless, in this intellectual venture, there is room to pick and choose. The research is not
limited to the final decisions of the Appeals Chamber. All arguments in favour of established
conclusions are welcome, even if they are found in the judgments of Trial Chambers later on
reversed by the Appeals Chamber and the convicts released.
For instance, the famous Yamashita case should not be taken as a precedent by any present
criminal court. The execution of that unfortunate Japanese general was most likely a crime in
itself. That case of miscarriage of justice should be governed by the principle ex injuria jus
non oritur. 66 On the other hand, the Trial Chamber in the Kupreškić case of 14 January
2000 rejected the tu quoque argument as allegedly being ‘‘universally rejected’’, although,
as a specific defence to punishment, it was recognized by the Nuremberg Judgment of
1946 in respect of two German Admirals, Dönitz and Raeder.67
In this light, labelling the former judicial practice altogether as ‘‘customary law’’ seems to
be a distortion of the meaning of this source of public international law.
66 The decision by the US Military Commission on Yamashita was quoted in the Judgment by the Trial Chamber
on Furundžija of 10 December 1998 ( para.168). The Judgment on the same case by the US Supreme Court was
cited in the Judgment by the Trial Chamber on Delalić of 16 November 1998 ( para.338), etc. In that Judgment,
the majority in the Supreme Court refused the habeas corpus in respect to his imprisonment and the death sen-
tence rendered by the Military Commission. However, Justices Murphy and Rutledge (327 US 1 (1946), 26 –81)
noted in their dissenting opinions that the majority had not shown that Yamashita had ‘‘knowledge’’ of gross
breaches perpetrated by his troops, that he had no ‘‘direct connection with the atrocities’’, nor could be found
guilty of a ‘‘negligent failure to discover’’ the atrocities, nor did he, in other words, have ‘‘personal culpability’’.
He spent less than a month in the position of the last supreme commander of Japanese forces in the Philippines. It
was therefore better to forget that example of gross judicial miscarriage, although the text of Art.7(3) of the ICTY
Statute was probably inspired by that case.
67 See Kupreškić, paras 515 –516. Instead of an extensive presentation of this issue here, see an excellent and exten-
sive review of this problem published in this very Journal by Sienho Yee, The tu quoque argument as a defence to
international crimes, prosecution or punishment, 3 Chinese JIL (2004), 87 –133.
The efforts to create new ‘‘customary rules’’ of international criminal law do not fare
better. One of the founding fathers and the first President of the ICTY, Professor
Antonio Cassese, is one of the champions of this ‘‘judge-made law’’. He has exposed in a
recent article copious case law concerning ‘‘elements of both the Statute provisions and
general international law on the definitions of crimes’’.68 Let us cite him with all his
references:
Thus, for instance, ICTY case law has defined, from the viewpoint of international
humanitarian law, the features of armed conflict,69 as well as the conditions on which
one may hold that an international armed conflict has broken out.70 Furthermore, it
has clarified the notion of war crimes, in particular by stating for the first time that
such crimes may also occur in internal armed conflicts,71 the notion of grave
breaches72 and the objective and subjective elements of crimes against humanity.73
Furthermore, it has spelled out the notion of such crimes as torture,74 rape,75 depor-
tation,76 enslavement,77 extermination78 and persecution,79 besides delineating some
important aspects of genocide.80
68 Cf. Antonio Cassese, The ICTY: A Living and Vital Reality, 2 JICJ (2004) 585 –597, 592.
69 See Jurisdiction Appeals Decision, Tadić (IT-94-I/AR72), Appeals Chamber, 2 October 1995, para.70.
70 See Judgment, Tadić (IT-94-A), Appeals Chamber, 15 July 1999, paras 88 –145; Aleksovski (IT-95 –14/I-A),
Appeals Chamber, 24 March 2000, paras 120 –146.
71 See Jurisdiction Appeals Decision, Tadić, 2 October 1995, cit. paras 94 –137.
72 See Jurisdiction Appeals Decision, Tadić, 2 October 1995, cit. para.70; Appeals Judgment, Tadić (IT-94-A),
15 July 1999, paras 80 –87; Kordić and Cerkez, Decision on the Joint Defence Motion to Dismiss the
Amended Indictment for Lack of Jurisdiction Based on the Limited Jurisdiction Reach of Articles 2 and 3
(IT-95–14/2-T), Trial Chamber, 2 March 1999, paras 12 –34; Trial Judgment, Kordić and Cerkez (IT-95 –
142-T), 26 February 2001, paras 65 –160.
73 See Jurisdiction Appeals Decision, Tadić (IT-94-I/AR72), 10 August 1995, paras 141 –142; Tadić (IT-94–1-I),
Trial Chamber, 7 May 1997, para.656; Appeals Judgment, Tadić (IT-94 –1-A), 15 July 1999, paras 271 –272;
Zoran Kupreškić and others (IT-95 –16-T), Trial Chamber, 14 January 2000, paras 547 – 549 and 556; Kunarac
and others (IT-96 –23-T), Trial Chamber, 22 February 2001, paras 413 – 420 and 433; Blaškić (IT-95 –14-T),
Trial Chamber, 3 March 2000, paras 208 – 213.
74 See, e.g. Furundžija (IT-95 –17/1-T), Trial Chamber, 10 December 1998, para.185; Kunarac and others (IT-
96 –23-T), Trial Chamber, 22 February 2001, paras 465 – 497.
75 See, e.g. Furundžija (IT-95–17/1-T), Trial Chamber, 10 December 1998, paras 174 –189; Kunarac and others
(IT-96–23-T), Trial Chamber, 22 February 2001, paras 436 –464.
76 See, e.g. Krstić (IT-98 –33-T), Trial Chamber, 2 August 2001, para.529; Krnojelac (IT-97– 25-T), Trial
Chamber, 15 March 2002, paras 472 –485; Stakić (IT-97 –24-T), Trial Chamber, 31 July 2003, paras 671 –687.
77 See, e.g. Kunarac and others (IT-96–23-T), Trial Chamber, 22 February 2001, paras 515 –541.
78 See Krstić (IT-98 –33-T), Trial Chamber, 2 August 2001, paras 496 –501; Vasiljević (IT-98– 32-T), Trial
Chamber, 29 November 2002, paras 216 –229; Stakić (IT-97 –24-T), Trial Chamber, 31 July 2003, paras
638 – 642.
79 See, e.g. Zoran Kupreškić and others (IT-96 –23-T), Trial Chamber, 14 January 2000, paras 616 –636; Trial
Judgment, Kordić and Cerkez (IT-95 –14/2), 26 February 2001, paras 188 –220.
80 See, e.g. Goran Jelisić (IT-95 –10-I), Trial Chamber, 14 December 1999, paras 59 –108; Krstić (IT-98 –33),
Trial Chamber, 2 August 2001, paras 539 –599; Stakić (IT-97 –24-T), Trial Chamber, 31 July 2003, paras
499 – 561.
78 CJIL (2005)
Let us only take one of these examples. In its judgment in Furundžija of 10 December 1998,
the Trial Chamber made a thorough analysis of torture and rape as international crimes in all
relevant conventional texts, decisions by international tribunals, comparative research of
domestic law, etc. On this basis, it established its own objective elements of these crimes,
as well as the actus reus and mens rea of aiders and abettors.81
Instead of applying the existing law to the accused and determining his sentence on this
basis, the order was, here, as usual, reversed: the Chamber first established that new ‘‘custom-
ary law’’ in extenso, and then it applied it to the accused.82
It must be finally stressed that normative statements in judicial decisions should be con-
sidered only as emerging customary law and not as positive legal rules. They can transform
into genuine customary law subsequently, if a majority of States confirm them in practice
coupled with the communis opinio juris. However, this legislative process is generally inap-
propriate for criminal courts, because of a constant risk of punishing the indicted for
some acts which were not criminal at the time of their commission.
In respect of the new ‘‘customary rules’’ established especially in the Furundžija judgment,
the developments seem to go in the opposite direction. The judicial findings by the ICTY
did not compel parties to the actual conflicts in Chechenya and Afghanistan, or the culprits
at the Guantánamo base and recently in the Abu Ghraib prison in Baghdad, to abide by
them. It seems, therefore, more productive in criminal proceedings to insist on the rigorous
respect for the minimum legal obligations of all the parties in international and internal
armed conflicts than to create new ‘‘customary rules’’.83
However, these embellishments in material law have no harmful effect if the responsibility
of the presumed perpetrator was established concerning the commission of the original crime
which is in the competence of the Tribunal according to its Statute. Much more delicate is
the question of interpretation of command responsibility under Article 7(3) of the ICTY
Statute.84 However, the judgment of the Appeals Chamber in the Blaškić case of 29 July
2004 is, in this respect, encouraging.
81 See paras 134 –256, especially paras 162 and 185 of that Judgment.
82 Hence, in respect of torture, the ‘‘customary law’’ established in that Judgment consisted of its being also inflicted
by omission, which Art.7 of the Rome Statute does not recognize. And although the convict did not himself
commit rape or other sexual assault against the victim, but his culpability was established as his aiding and abet-
ting the actual perpetrator, the ‘‘customary law’’ in that Judgment was nevertheless enriched by defining a sexual
penetration in physical terms. Such a definition, though justified, could fit in a judgment against an actual rapist.
In the light of widespread and numerous acts of rape committed as a part of criminal policy by one party in the
conflicts in the former Yugoslavia, the judges were not patient enough to wait for a genuine crime of this kind.
83 This is, however, not to say that even widespread practices of torture, rape or organized prostitution can lead to
appearance of opposite customary rules by which all this mischief could become lawful. Such a development is
opposed even by rare condemnations by national tribunals of individual perpetrators of these crimes, although
only with symbolic penalties so far. However, in a situation in which States either fail to exercise effective
control over their troops, or they themselves organize concentration camps for suspects who are deprived of judi-
cial guarantees, it seems to be vain to create in judicial practice rules stricter than those proceeding from treaties.
84 For a thorough analysis of all aspects of this problem, see Bing Bing Jia, The Doctrine of Command Responsi-
bility Revisited, 3 Chinese Journal of International Law (2004), 1–42.
It must be finally stated that the opinio juris is definitely not equivalent to simple ‘‘legal
opinions’’ either by judges or by legal experts in their collective works. In both cases,
these ‘‘opinions’’ do not amount to international custom in the meaning of Article
38(1)(b) of the ICJ Statute. Being ‘‘judicial decisions and the teachings of the most
highly qualified publicists of the civilized nations’’, these opinions are but ‘‘subsidiary
means for the determination of the rules of law’’ under subparagraph (d) of that same Article.
VII. A brief comment on Article 21 of the Rome Statute

Instead of an abstract assessment of which system is better than the other, i.e. whether the
‘‘customary law’’ encompassed by the former judicial practice is better than the law set
forth in the Rome Statute, it seems more productive to comment here on the applicable
law as provided in Article 21 of the said Statute.
Unlike Article 38(1) of the ICJ Statute, Article 21 establishes a hierarchy of sources of law
that should be applied by the ICC.85 However, Article 21 does not only provide sources of
the material law on international crimes within the jurisdictions of the ICC, but also relates
to all law that can be applied by it.
In paragraph (1a) of that provision, it is stated that the Court shall apply ‘‘In the first place,
this Statute, Elements of Crimes and its Rules of Procedure and Evidence; . . .’’.
The Rome Statute is the constitutive instrument of the ICC in the same way as the Charter is
for the United Nations Organization. It provides, inter alia, rules on the ‘‘Establishment of the
Court’’ (Part 1); on International co-operation and assistance and on Enforcement (Parts 9
and 10); on the Assembly of States Parties (Part 11); on its Financing (Part 12); etc. The
Court functions as a separate international organization and, like specialized agencies, it is
in relationship with the United Nations. Unlike the ICJ, it is not the organ of the United
Nations. Many additional agreements will be concluded in the implementation of the
Rome Statute.
More important is the fact that for the first time in the history of international criminal
tribunals, the Rome Statute provides, in its Part 3, the General Principles of Criminal Law in
written form and in advance. Among these principles are those on individual criminal
responsibility and on various forms of perpetration of and participation in a crime
(Article 25), on the mens rea (Article 30), on grounds for excluding criminal responsibility
(Article 31), on mistakes of fact or law (Article 32), etc. This is, in fact, what constitutes the
substantial provisions in the ‘‘general part’’ of criminal laws of many States, which judges
must not disregard or interpret at will.
85 The hierarchy was obviously inspired by Art.7 of the Twelfth Hague Convention of 1907 of the International
Court of Prize, which has never entered into force. That former provisions read as follows: ‘‘If a question of
law to be decided is covered by a treaty in force between the belligerent captor and a Power which is itself or
whose subject or citizen is a party to the proceedings, the Court is governed by the provisions in the said
treaty. In the absence of such provisions, the Court shall apply the rules of international law. If no generally recog-
nized rule exists, the Court shall give judgment in accordance with the general principles of justice and equity.’’
80 CJIL (2005)
The Statute contains an extensive enumeration of definitions of all the crimes within the
jurisdiction of the Court. As already stated, they are: Genocide (Article 6), Crimes against
Humanity (Article 7) and War Crimes (Article 8). All these written definitions are sufficient
to ensure that the problem of non liquet will not arise in practice. However, because these defi-
nitions were adopted from codification conventions of humanitarian law, they sometimes miss
the description of actus reus and mens rea of perpetrators or participants in a crime. For that
purpose, the Elements of Crimes are appended to the Statute which, according to Article 9,
‘‘shall assist the Court in the interpretation and application of Articles 6, 7 and 8’’. They
must be consistent with the Statute but can be amended in a simpler procedure that the
Statute itself.
The Statute itself provides detailed procedural provisions. It is also supplemented by the
even more detailed Rules of Procedure and Evidence, which are also adopted as an instru-
ment in the application of the Statute. These Rules and all their subsequent amendments
must be consistent with the Statute. In the event of conflict, the Statute will always
prevail (Article 51).
It seems important to stress that both the Elements of Crimes and the Rules of
Procedure and Evidence were not adopted by Judges, but by a two-thirds majority of
members of the Assembly of States Parties. However, the Prosecutor and Judges, acting by
an absolute majority, may propose amendments to them. Judges adopt, by an absolute
majority, only the Regulations of the Court for its routine functioning, which can be commen-
ted on by States parties. All this has finally brought the legislative function of Judges to an end.
In Article 21(1)(b) of the Statute, it is provided that the Court shall apply, furthermore,
‘‘In the second place, where appropriate, applicable treaties and the principles and rules of
international law, including the established principles of the international law of armed
conflict; . . .’’.
Several supplementary sources of international law are in this hierarchical order provided
here. The applicable treaties (other than the Rome Statute itself ) may, for instance, include
the 1966 International Covenant on Civil and Political Rights, or regional conventions such
as the 1950 European Convention on the Human Rights and Fundamental Freedoms with
protocols attached to it. The provisions of these applicable treaties concerning fair trial and
similar problems that are perhaps not entirely regulated by the Statute and the Rules of
Procedure and Evidence of the ICC can be of special importance.
Under ‘‘principles and rules of international law’’, it should be understood to include cus-
tomary rules of general international law. Probably because of the frequent misuse of these
terms in the past, they were not referred to as such. But this less precise wording has the
advantage that the Court will not need to prove the customary character of ‘‘principles
and rules’’ it wishes to apply, especially not the communis opinio juris.
Established principles of the international law of armed conflicts consist most often of
customary rules confirmed in codification conventions. Hence, they include numerous pro-
visions from the four 1949 Geneva Conventions and the two 1977 Protocols, but not the
definitions of their ‘‘grave breaches’’.
These ‘‘established principles’’ may include, inter alia, rules concerning proper qualifica-
tion of armed conflicts; specific prohibitions of reprisals against protected persons, buildings
or equipment; localities and zones under special protection and on demilitarized zones; con-
ditions to be fulfilled for belligerent occupation; neutrality in a conflict; or rules of the
warfare at sea, when there is room for their application. These rules are most often conven-
tional, but can also be found in the practice of warfare.
All the above sources under paragraph (1)(b) cannot replace, amend or extend either the
general principles of criminal law set forth in the Statute or definitions of crimes within the
jurisdiction of the Court. They are of lower legal force than even the Elements of Crimes and
the Rules on Procedure and Evidence.
Paragraph (1)(c) provides that the Court shall apply furthermore:
Failing that, general principles of law derived by the Court from national laws and
legal systems of the world including, as appropriate, the national laws of States
that would normally exercise jurisdiction over the crime, provided that those
principles are not inconsistent with this Statute and with international law and
internationally recognized norms and standards.
Hence, in the absence of any applicable law under paragraph (1)(a) and (b), the Court can
resort to the general principles of law in their narrow sense, as explained above. National laws
and legal systems of the world (continental, Anglo-Saxon and possibly Islamic) do not
require an analysis of laws of a great many States. It is sufficient to consider a few represen-
tative samples. When appropriate, the Court will take into account the national laws of States
‘‘that would normally exercise jurisdiction over the crime’’, either by the application of the
principle of territoriality, or of active personality, or perhaps according to the law of the State
which arrested the suspect and surrendered him to the Court.
Paragraph (1)(c) stresses that all these derived general principles of law must be consistent
with the Statute and with international law and ‘‘internationally recognized norms and stan-
dards’’ (having probably in mind human rights and fundamental freedoms). The emphasis
is, here, once again, on the rules of positive international law and not of lex ferenda by judges.
In this context, we want to add an explanation on the two kinds of general principles of
law. An obvious example of these principles in their broad sense is Article 31 of the Rome
Statute, which codifies four grounds for excluding criminal responsibility: mental disease or
defect, intoxication, self-defence and duress/necessity. These grounds were, in Article 32,
supplemented by mistakes of fact or law. A great many States have in their national legislation
the same or similar rules.86
86 But this is not all. The 1969 Vienna Convention on the Law of Treaties also codifies in its Arts 48 –52 similar
general principles of law vitiating the consent of the parties that can result in invalidity of a treaty. They are: error,
fraud, corruption and coercion of a representative of a State, and the coercion of a State by the threat or use of
force. The Draft articles on State responsibility provide in all their versions the following circumstances preclud-
ing wrongfulness of an act: consent, self-defence, countermeasures (lawful reprisals), force majeure, distress and
necessity. Hence, all these general principles of law have, in various contexts, the same basis.
82 CJIL (2005)
However, the drafters of the Rome Statute were conscious that by Articles 32 and 33 did
not codify all the possible general principles of law in this respect. For that reason, it is
provided in Article 31(3):
At trial, the Court may consider a ground for excluding criminal responsibility other
than those referred to in paragraph 1 where such a ground is derived from applicable
law as set forth in article 21 . . ..
Here, we come to the domain of the general principles of law in their narrow sense, exactly as
provided in Article 21(2)(c). A comparative analysis of national laws and legal systems
throughout the world indicates the following possible grounds for excluding somebody’s
criminal responsibility: (i) consent by the victim; (ii) the immunity of an active diplomat;
(iii) military necessity; (iv) lawful reprisals; and (v) tu quoque argument. However,
grounds under (i), (iii) and (iv) do not seem to be consistent with the present peremptory
norms of international humanitarian law. The one under (ii) will depend on further clarifica-
tions in case law, while the tu quoque argument as a defence to punishment under (v) in this
precise sense seems only to be justified (hence, not as a defence to crimes, or to
prosecution).87
The short text of Article 21(2) provides that ‘‘The Court may apply principles and rules of
law as interpreted in its previous decisions.’’
This means that the ICC is not bound to follow even its own precedents. The stare decisis
doctrine is therefore not obligatory for it. The previous practice of State military tribunals
and of ad hoc international tribunals is not mentioned in this provision at all. It is, therefore,
obvious that it does not constitute ‘‘customary law’’.
However, on that ground the Court ‘‘may apply’’ its own previous practice, it is most likely
that it will follow the practice of the ICJ in this respect. In order to support its new decisions,
it will not hesitate to refer to its previous judgments, first of all those by its Appeals Chamber.
Nevertheless, when, subject to the circumstances of a new case, it decides otherwise, it will
expose legal arguments for having declined from its former practice.
Finally, Article 21(3) states the following:
3. The application and interpretation of law pursuant to this article must be consist-
ent with internationally recognized human rights, and be without any adverse dis-
tinction founded on grounds such as gender, as defined in article 7, paragraph 3,
age, race, colour, language, religion or belief, political or other opinion, national,
ethnic or social origin, wealth, birth or other status.
The matter is here not of ‘‘super-legality’’.88 The matter is only of confirmation of peremp-
tory norms of general international law in that domain, including the principle of
87 See Sienho Yee, above n.67.

88 This paragraph was thus interpreted and, at the same time, criticized by Alain Pellet, Applicable Law, above n.4,
1079 –1082.
non-discrimination. These norms are obligatory for judges in any national or international
criminal proceedings, even if they were not expressly provided.
In the light of all specific sources of public international law set forth in Article 38(1) of
the ICJ Statute, as applied in its normal practice and in the practice of inter-State arbitrators,
there cannot be raised any serious objections to the text of Article 21 of the Rome Statute.
When sources of international criminal law, as being applied by criminal courts and tribu-
nals, are at stake, hierarchical order in their application seems to be perfectly appropriate.
The rules of the Rome Statute, taken as a whole, will not allow Judges of the ICC to
improvise with the general principles of criminal law, or to manoeuvre with applicable
law on international crimes within its jurisdiction. These, as well as many other guarantees
of legality in the procedure, should encourage the States, which are suspicious with regard to
the previous practice of ad hoc tribunals, to adhere to the Rome Statute.
It seems, in addition, highly desirable to include in the Statute, at the next Review Con-
ference, the definition of the crime of aggression, as well as that of the crime of international
terrorism. There are no more suitable impartial fora than the ICC for combating this scourge
of the international community.

Degan - Sources of International Criminal Law

Uploaded by

Copyright:

Available Formats

Degan - Sources of International Criminal Law

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Degan - Sources of International Criminal Law

Uploaded by

Copyright:

Available Formats

# The Author 2005. Published by Oxford University Press. All rights reserved.

Unfortunately, men’s criminal imagination appears unlimited and, by enclosing the

II. Sources of international criminal law and Article 38(1) of

a. international conventions, whether general or particular, establishing rules expressly

III. The general principles of (criminal) law

IV. Treaties as a source of law in international

V. Place of custom in international criminal law

35 Cf. Tadić Jurisdiction Decision of 1995, para.89.

47 Cf. ILCYB (1976), Volume II, Part Two, 119, para.61.

VI. Judicial practice and doctrine

A judgment or decision of a court, normally recorded in a law report, used as an

VII. A brief comment on Article 21 of the Rome Statute

Paragraph (1)(c) provides that the Court shall apply furthermore:

Finally, Article 21(3) states the following:

87 See Sienho Yee, above n.67.

You might also like