Separate opinion of Judge Bhandari

Document Number
118-20150203-JUD-01-10-EN
Parent Document Number
118-20150203-JUD-01-00-EN
Document File
Bilingual Document File

409

SEPARATE OPINION OF JUDGE BHANDARI

Introduction

1. I have voted with the majority on all three operative clauses of the
present Judgment. However, with respect to the second operative clause, h
i.e., the rejection of Croatia’s principal claim, I wish to qualify ahnd expand

upon the rationales for my vote. In so doing, I shall take the present
opportunity to expound upon certain reservations I continue to harbour
regarding the analysis employed at various points throughout that por -
tion of the Judgment with respect to issues which, in my respectful view,

have received inadequate — or even incorrect — attention.
2. At the outset, I wish to underscore that the principal reason for my
rejection of Croatia’s claim is that the Applicant has failed, in my hconsid-
ered opinion and after having carefully scrutinized the entire evidentiahry

record in these proceedings, to satisfy the minimum standard of credible
evidence required by this Court in its prior jurisprudence (in particulhar
the Bosnia Judgment of 2007 1, which dealt with claims of a highly similar
nature) in relation to the dolus specialis of genocide. In this regard, I take
specific note of Croatia’s near complete inability to substantiate hmost of

the figures it has averred in terms of number of victims as a consequehnce
of the hostilities that occurred in the regions and during the period ath
issue. Moreover, I recall that it is a well -settled principle of law that the
graver the offence alleged, the higher the standard of proof required forh

said offence to be established in a court of law. Consequently, I am not h
“fully convinced” (Judgment, para. 178) that the only inference available
from the evidence on record is that attacks against ethnic Croats on the
territory of Croatia between 1991 and 1995 were perpetrated with the req -

uisite genocidal intent. Thus, although I concur with the majority that hthe
actus reus of genocide has been conclusively satisfied with respect to many
of the localities averred by Croatia, the Applicant’s inability to prhove that
the mens rea of genocide — which, by its very nature, constitutes a charge

“of exceptional gravity” (ibid.) — has been “clearly established” (ibid.)
is necessarily fatal to Croatia’s entire cause of action.

1 Application of the Convention on the Prevention and Punishment of the Cr▯ime of
Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, I.CReports

2007 (I), p. 43 (hereafter the “Bosnia Judgment”).

410

7 CIJ1077.indb 816 18/04/16 08:54 410 application of genochide convention (sep. ohp. bhandari)

3. Indeed, during the oral hearings phase of these proceedings, in responseh

to a question posed by another Member of the Court, Croatia was com -
pelled to concede that many of its written witness statements would haveh
been inadmissible in a domestic Croatian court of law, to which Serbia
responded that such statements would have been likewise inadmissible in hthe

domestic courts of the formerYugoslavia (Judgment, para.195). Moreover,
in response to a question that I posed to the Parties, Croatia maintained
that the Court enjoyed a free hand in determining what weight should be
given to them, based on established Court jurisprudence pertaining to

out-of-court documents (ibid., para. 194). The sum total of these exchanges
is that it stands to reason that a party to proceedings before the Courth can -
not expect to have documents that would be inadmissible before the courths
of its own country, and which bear marked deficiencies when assessed uhsing

the standards applied in this forum, admitted for proof of their contenths;
especiallywhere the matter to be proved is as grave as the crime of genocide.
4. In reaching this conclusion, I share the majority’s sensitivity to “hthe
difficulties of obtaining evidence in the circumstances of th[is] case”h ( ibid.,
para. 198), wherein proof had to be gleaned from a postbellum context

where the juridical infrastructure and other cornerstones of government
and civil society typically relied upon by litigants appearing before thhis
Court have been rendered largely absent or at least severely compromisedh
by years of brutal war, massive displacements of populations and other

seismic socio -political upheavals. Indeed, so Herculean are these obsta -
cles that I must confess to having harboured a fleeting temptation to hrelax
my approach to the methods of proof obtaining before the instant pro -
ceedings, specifically with respect to the documentary evidence adducehd
by Croatia, much of which admittedly lacks the indicia of reliability nohr-

mally demanded of documents presented before a judicial body. How -
ever, the allure of adopting an elastic approach to Croatia’s documenhts
was, to my mind, definitively quelled by the countervailing considerathion
that the crime of genocide, being “an odiousscourge” that is “condemned
3
by the civilized world” , carries with it such grievous moral opprobrium
that a judicial finding as to its existence can only be countenanced uhpon
the most credible and probative evidence. Consequently, despite the
sympathy I have expressed herein regarding the extraordinary evidentiaryh
hurdles faced by the Parties to these proceedings, I ultimately share thhe

majority’s finding “that many of the statements produced by Croahtia are
deficient” (ibid., para. 198), that these deficiencies are irremediable, and
that the remainder of the Applicant’s evidence has failed to conclusively
demonstrate the only conclusion to be drawn from the evidence it has
proffered is that there existed genocidal intent against the targeted group

in question during the time period averred.

2 Convention on the Prevention and Punishment of the Crime of Genocide (“hGenocide
Convention”), Preamble.
3
Ibid.

411

7 CIJ1077.indb 818 18/04/16 08:54 411 application of genochide convention (sep. ohp. bhandari)

5. This premise having been established, I take note of the fact that in

spite of the serious evidentiary deficiencies in Croatia’s case, thhe majority
has elected to assess whether the claims of the Applicant, taken at their
highest, could nevertheless evince genocidal intent . Following this lead,

and notwithstanding my conclusion that Croatia’s charges of genocide
have failed on evidentiary grounds, I intend to profit from the presenht
opportunity in making certain observations and critiques to the analysish

adopted by the majority on the issue of dolus specialis, assuming, argu ‑
endo (as the majority has done), that Croatia’s case may be taken at iths
highest.

6. In brief, it is my respectful view that the Court should have used the
present Judgment to lay down clearer guidelines on three principal issues.
First, I believe the Court could have provided a better and clearer treaht -

ment as to what constitutes genocidal intent. Second, given the proliferha-
tion of international criminal tribunals over the past two decades and thhe
consequent exponential expansion of jurisprudence emanating from these

juridical bodies, I believe the majority has been derelict in not more fhully
canvassing the available authorities to provide clear parameters to disthin-
guish between genocide and the oft closely intertwined offences of extermh - i

nation and/or persecution as a crime against humanity. Third and finalhly,
I believe that the 17 factors advanced by Croatia in support of its conten-
tion that genocide occurred deserved a more comprehensive response than

the majority’s approach of selecting, without any apparent reasoned ehxp-la
nation, five factors deemed “most important” to Croatia’s clahim of geno-
cidal intent (Judgment, para. 413). I believe that a superior treatment of

4 See Judgment, para. 437:

“The Court considers that it is also relevant to compare the size of hthe targeted part
of the protected group with the number of Croat victims, in order to dethermine whether
the JNA and Serb forces availed themselves of opportunities to destroy thhat part of the
group. In this connection, Croatia put forward a figure of 12,500Croat deaths, which
is contested by Serbia. The Court notes that, even assuming that this figure it correc—

an issue on which it will make no ruling — the number of victims alleged by Croatia is
small in relation to the size of the targeted part of the group.” (Ehmphasis added.)
See also, ibid., para. 213:

“Croatia first asserts that, between the end of August and 18November 1991,
Vukovar was besieged and subjected to sustained and indiscriminate shellhing, laying
waste to the city. It alleges that between 1,100 and 1,700 people, 70 per cent of whom
were civilians, were killed during that period.”

See also, ibid., para. 218:

“The Court will first consider the allegations concerning those kilhled during
the siege and capture of Vukovar. The Parties have debated the number of victims,
their status and ethnicity and the circumstances in which they died. The▯ Court need
not resolve all those issues. It observes that, while there is still some uncertainty
surrounding these questions, it is clear that the attack on Vukovar was hnot confined
to military objectives it was also directed at the then predominantly Croat civilian
population (many Serbs having fled the city before or after the fighhting broke out).”
(Emphasis added.)

412

7 CIJ1077.indb 820 18/04/16 08:54 412 application of genochide convention (sep. ohp. bhandari)

these topics would have been commensurate with the Court’s function ahs
not only the principal judicial organ of the UnitedNations but as a “World

Court” from which other international and domestic courts and tribunahls
seek guidance as a legal authority of the highest order.

Genocidal Intent and theh “Substantiality” Critherion

7. For ease of reference, I reproduce the relevant sections of the Geno -
cide Convention in which the substantive provisions of the crime of genoh -
cide are enshrined :

“Article 1: The Contracting Parties confirm that genocide, whether
committed in time of peace or in time of war, is a crime under inter -

national law which they undertake to prevent and to punish.
Article 2 : In the present Convention, genocide means any of the
following acts committed with intent to destroy, in whole or in part,
a national, ethnical, racial or religious group, as such :

(a) Killing members of the group ;
(b) Causing serious bodily or mental harm to members of the group;

(c) Deliberately inflicting on the group conditions of life calculated
to bring about its physical destruction in whole or in part ;
(d) Imposing measures intended to prevent births within the group ;
(e) Forcibly transferring children of the group to another group.”

As the foregoing text illustrates, the chapeau of Article II of the Geno -
cide Convention defines genocide as “any of the following acts commhitted
with intent to destroy, in whole or in part, a national, ethnical, racial or

religious group, as such” (emphasis added). The fact that the Convehntion
expressly envisages situations where a group may be targeted for destruch-
tion “in part” naturally gives rise to the thorny question of whenh exactly
the targeted “part” meets the threshold for genocidal intent. Becahuse the
Convention is silent on this point, in the Bosnia Judgment the Court

relied upon the jurisprudence of the International Criminal Tribunal for
the former Yugoslavia (ICTY) and the International Criminal Tribunal
for Rwanda (ICTR), as well as International Law Commission (ILC)
Commentary, to conclude that “part” of the “group” for the phurpose of

Article II requires5an intent “to destroy at least a substantial part of the
particular group” .

5 Bosnia Judgment, p. 126, para. 198; emphasis added.

413

7 CIJ1077.indb 822 18/04/16 08:54 413 application of genochide convention (sep. ohp. bhandari)

8. While the “substantiality” criterion enunciated in the Bosnia Judg -
ment has been reaffirmed in the instant Judgment (in somewhat modifihed
form, a subject to which I intend to return in short order), this has bheen

done rather tersely and in a way that, in my view, fails to lay down clehar
parameters that would provide guidance to future adjudicative bodies
grappling with this concept. The majority, has also, I fear, neglected to so
much as consider possibly relevant jurisprudential developments emanat -
ing from the ad hoc international criminal tribunals in the intervening

eight years since the issuance of the Bosnia Judgment. Therefore, in the
hopes of elucidating this standard for the sake of posterity, I intend to
revisit the Bosnia formula to see how that test has been applied in practice
by other tribunals in recent years, so as to juxtapose such developmentsh
with how the majority has employed said formula in the instant Judg -

ment.

The Legal Test Enunciatehd
in the Court’s B osnia Judgment of 2007

9. As has been correctly observed in the present Judgment, in the Bosnia

Judgment of 2007, the Court “considered certain issues similar to thohse
before it in the present case” (Judgment, para. 125). On that occasion, the
Court expounded the relevant test for determining what constitutes a “hpart”
of the targeted group for the purpose of analysing genocidal intent as fol -
lows :

“[T]he Court refers to three matters relevant to the determination of
‘part’ of the ‘group’ for the purposes of Article II [of the Genocide

Convention]. In the first place, the intent must be to destroy at least
a substantial part of the particular group. That is demanded by the
very nature of the crime of genocide : since the object and purpose of
the Convention as a whole is to prevent the intentional destruction of
groups, the part targeted must be significant enough to have an impacth
on the group as a whole. That requirement of substantiality is sup ‑

ported by consistent rulings of the ICTY and the International Criminal ▯
Tribunal for Rwanda (ICTR) and by the Commentary of the ILC to
its Articles in the draft Code of Crimes against the Peace and Security h
of Mankind.
Second , the Court observes that it is widely accepted that genocide

may be found to have been committed where the intent is to destroy
the group within a geographically limited area . . . As the ICTY
Appeals Chamber has said . . . the opportunity available to the perpe-
trators is significant. This criterion of opportunity must however be
weighed against the first and essential factor of substantiality . It may

414

7 CIJ1077.indb 824 18/04/16 08:54 414 application of genochide convention (sep. ohp. bhandari)

be that the opportunity available to the alleged perpetrator is so lim -

ited that the substantiality criterion is not met. The Court observes
that the ICTY Trial Chamber has indicated the need for caution, lest
this approach might distort the definition of genocide [. . .]

A third suggested criterion is qualitative rather than quantitative.
The Appeals Chamber in the Krstić case [noted that]

‘. . . In addition to the numeric size of the targeted portion, its
prominence within the group can be a useful consideration. If a
specific part of the group is emblematic of the overall group, or is

essential to its survival, that may support a finding that the part
qualifies as substantial . . .’
Establishing the ‘group’ requirement will not always depend on the

substantiality requirement alone although it is an essential startin‑gpoint.
It follows in the Court’s opinion that the qualitative approach canno▯t
stand alone. The Appeals Chamber in Krstić also expresses that
view.” 6

The Court concluded its remarks by noting that “[t]he above list of
criteria is not exhaustive, but, as just indicated, the substantiality criterion

is critical. They are essentially those stated by the Appeals Chamber in 7
the Krstić case, although the Court does give this first criterion priority.”
Thus, in the Bosnia case the Court fastened a tripartite formula, which it
indicated was open to future expansion and elaboration, for determining h

whether a “part” of a group has been targeted with genocidal intenht ;
according to which the criterion of “substantiality” was pre -eminent in
that calculus.

10. While the Bosnia formula did not draw any bright lines around the
contours of what constitutes genocidal intent toward “a part” of thhe tar-
geted group, it would appear plain from that Judgment and the jurispru -
dence of the international criminal tribunals that a “substantial”h part of

the targeted group need not constitute the majority thereof, and that there
is no numeric threshold for discerning a substantial part of the group.

The Legal Test Enunciatehd in the Present Judgmenht

11. The pertinent analysis of the law on genocidal intent vis -à-vis “a
part” of the targeted group is presented in the present Judgment as fhollows :

“The Court recalls that the destruction of the group ‘in part’ within
the meaning of Article II of the Convention must be assessed by refer ‑

6 Bosnia Judgment, pp. 126-127, paras. 198-200 (internal citations omitted ; emphasis
added).
7
Ibid., p. 127, para. 201; emphasis added.

415

7 CIJ1077.indb 826 18/04/16 08:54 415 application of genochide convention (sep. ohp. bhandari)

ence to a number of criteria . In this regard, it held in 2007 that ‘the
intent must be to destroy at least a substantial part of the particular h
group’ [. . .], and that this is a ‘critical’ criterion. The Court further

noted that ‘it is widely accepted that genocide may be found to have h
been committed where the intent is to destroy the group within a
geographically limited area’ and that, accordingly, ‘[t]he area ofh the
perpetrator’s activity and control are to be considered [. . .]’. Account
must also be taken of the prominence of the allegedly targeted part

within the group as a whole. With respect to this criterion, the Appealsh
Chamber of the ICTY specified in its Judgment rendered in the Krstić
case that ‘[i]f a specific part of the group is emblematic of the overall
group, or is essential to its survival, that may support a finding thaht
the part qualifies as substantial . . .’.

In 2007, the Court held that these factors would have to be assessed
in any particular case. [. . .] It follows that, in evaluating whether the
allegedly targeted part of a protected group is substantial in relation to
the overall group, the Court will take into account the quantitative
element as well as evidence regarding the geographic location and

prominence of the allegedly targeted part of the group.” (Judgment,
para. 142 (internal citations omitted ; emphasis added).)

What I find immediately striking from this slightly rebranded iteratiohn
of the tripartite test promulgated by the Court in the Bosnia Judgment is
that, one fleeting reference to the “critical” nature of the “substantiality”
criterion (now renamed “the quantitative element”) notwithstandihng, the
rigidly hierarchical structure of the Bosnia test, whereby the numerosity

of the targeted population was clearly superordinate to the other, supplhe -
mentary criteria of “opportunity” (now dubbed “the geographic hloca -
tion”) and the “qualitative factor” (now dubbed the “promihnence” of the
targeted group) has been jettisoned in favour of a more equal balancingh
effort. My distinct impression that the stratification inherent in the hBosnia

formula has been mollified by the present Judgment (a jurisprudentialh
evolution I applaud) draws further support from the consistently flexhible
and egalitarian manner in which the Court has applied these three factors
to the facts at bar, wherein I cannot discern any noticeable supremacy
afforded the quantitative element (see, generally, Judgment, para -
graphs 413-441).

12. As I shall undertake to demonstrate at a later juncture in this opin-
ion, I believe that this adapted substantiality test has practical conseh -

quences for the manner in which the majority has applied the assessment h
of genocidal intent in the present Judgment, specifically with respect to
the events occurring in the city of Vukovar and its environs.

416

7 CIJ1077.indb 828 18/04/16 08:54 416 application of genochide convention (sep. ohp. bhandari)

Post- Bosnia Jurisprudence
of the ICTY and ICTR

13. As noted above, in the present Judgment the Court has recalled

and reaffirmed the tripartite formula for genocidal intent enunciated ihn
Bosnia as the approach to be followed in the present case, though not
without a significant restructuring of the normative order of the testh to be
employed. This is naturally consonant with the principle that while no
8
prior judgment of this Court constitutes binding precedent sensu stricto ,
“[i]n general the Court does not choose to depart from previous finhdings,
particularly when similar issues were dealt with in the earlier decisionhs. . .
unless it finds very particular reasons to do so” (Judgment, para.h125). As

I have noted above, in Bosnia the Court explicitly acknowledged the con -
tributions of the ICTY and ICTR in shaping the test that it adopted to
assess genocidal intent vis -à-vis a “part” of a targeted group 9. Conse-
quently — and bearing in mind the nearly eight years that have passed

since the promulgation of this Court’s Bosnia formula — it would seem to
me only natural and appropriate to examine whether the jurisprudence of h
those tribunals in the intervening years reveals any evolution in how the
“substantiality” component of genocidal dolus specialis has been applied

in recent litigious contexts. Such an endeavour is not only consonant with
the Bosnia Judgment’s pronouncement that the criteria enunciated therein
were “not exhaustive” 10and therefore presumably subject to future eluci -
dation, but is also faithful to the present Judgment’s self -admonition that

the Court will “take account, where appropriate, of the decisions of hinte-r
national criminal courts or tribunals, in particular those of the ICTY, has
itdid in 2007, in examining the constituent elements of genocide in the

present case” (ibid., para. 129). In a similar vein, I recall the draft Judg -
ment’s avowal that while it will rely on the Bosnia Judgment “to the
extent necessary for its legal reasoning[, t]his will not . . . preclude it,
where necessary, from elaborating upon this jurisprudence” (ibid.,

para. 125).

14. In my respectful view, because the legal standard for genocidal
intent has a necessarily vague and dynamic character, it was incumbent

upon the Court to fully canvass recent developments in the law to deter -
mine how the Bosnia formula (as restated in the present Judgment) has
been applied in other juridical institutions tasked with applying that test.

8 ICJ Statute, Article 59: “The decision of the Court has no binding force except
between the parties and in respect of that particular case.”
9 Bosnia Judgment, p. 126, para. 198.
10 Ibid., p. 127, para. 201.

417

7 CIJ1077.indb 830 18/04/16 08:54 417 application of genochide convention (sep. ohp. bhandari)

I regret to say that in my estimation the present Judgment has neither
fully nor properly canvassed the current jurisprudential standard of genho -
cidal intent emanating from the ICTY and the ICTR. For this reason, I
shall now conduct a survey of recent trends in the case law of those trihbu -

nals on this subject in an attempt to glean insights as to the present shtate
of the law in this area. As I shall expound hereunder, I take the positihon
that these recent jurisprudential trends would tend to suggest that a pat -
tern of killings such as has been averred in relation to the events thath
11
occurred in Croatia between 1991 and 1995 , and in particular with
respect to the region of Eastern Slavonia and the greater Vukovar area, h
may be more indicative of genocidal intent than the majority has acknowl-
edged.

The t olimiR ICTY Trial
Chamber Judgment

15. On 12 December 2012, the Trial Chamber of the ICTY issued its
judgment in the case of Tolimir (currently under appeal), in which it pro -
vided a comprehensive treatment of the substantiality criterion of geno -

cidal intent. The Trial Chamber recalled that

“[t]he term ‘in whole or in part’, relates to the requirement thhat the
perpetrator intended to destroy at least a substantial part of a pro -
tected group. While there is no numeric threshold of victims required,
the targeted portion must comprise ‘a significant enough [portion] hto

have an impact on the group as a whole’. Although the numerosity of
the targeted portion in absolute terms is relevant to substantiality, this
is not dispositive other relevant factors include the numerosity of the
targeted portion in relation to the group as a whole, the prominence

of the targeted portion and whether the targeted portion of the group
is emblematic of the overall group, or is essential to its survival, as h
well as the area of the perpetrators’ activity, control and reach.”h 12

These observations made repeated reference to the same section of the
analysis contained in the Krstić Judgment of the ICTY Appeals Chamber
that was relied upon by this Court when it adopted its tripartite test fhor
genocidal intent in the Bosnia Judgment.

11For the avoidance of any doubt, I recall that while I have found that Crhoatia’s claim
fails on evidentiary grounds, I am taking the present opportunity, as thhe majority has done,
to assess Croatia’s case taken at its highest.

12Prosecutor v. Tolimir, Trial Judgment, 12 December 2012, para. 749 (internal cita-
tions omitted; emphasis added).

418

7 CIJ1077.indb 832 18/04/16 08:54 418 application of genochide convention (sep. ohp. bhandari)

16. After summarizing these widely accepted elements of the law on
genocidal intent, the Trial Chamber in Tolimir further recalled a passage
from an earlier judgment of the ICTY Trial Chamber in the case of
13
Jelisić , w14ch was cited approvingly in a passage of the Krstić Appeals
Judgment that was referenced favourably by the Court in the Bosnia
Judgment 15. As the Tolimir Trial Chamber recalled :

“The Jelisić Trial Chamber held that as well as consisting of the
desire to exterminate a very large number of members of the group,

genocidal intent may also consist of the desired destruction of a more
limited number of persons selected for the impact that their disappear ‑
ance would have on the survival of the group as such.” 16

The Trial Chamber then made the following further observations about

the Jelisić Trial Judgment:
“The Jelisić Trial Chamber cited the Final Report of the Commis -

sion of Experts formed pursuant to Security Council Resolution 780
which found

‘[i]f essentially the total leadership of a group is targeted, it couhld
also amount to genocide. Such leadership includes political and
administrative leaders, religious leaders, academics and intellec -
tuals, business leaders and others — the totality per se may be a

strong indication of genocide regardless of the actual numbers
killed. A corroborating argument will be the fate of the rest of the
group. The character of the attack on the leadership must be

viewed in the context of the fate or what happened to the rest of the
group. If a group has its leadership exterminated, and at the same
time or in the wake of that, has a relatively large number of the
members of the group killed or subjected to other heinous acts, for

example deported on a large scale or forced to flee, the cluster of
violations ought to be considered in its entirety in order to inter -
pret the provisions of the Convention in a spirit consistent with its
purpose.’. . .

The Commission of Experts Report stated, further, that

‘[s]imilarly, the extermination of a group’s law enforcement and
military personnel may be a significant section of a group in that
it renders the group at large defenceless against other abuses of a
similar or other nature, particularly if the leadership is being eli -

minated as well. Thus the intent to destroy the fabric of a society
through the extermination of its leadership, when accompanied by

13Prosecutor v. Jelisić, Trial Judgment, 14 December 1999, para. 82.
14Prosecutor v. Krstić, Appeals Judgment, 19 April 2004, para. 8 and fn. 10.
15Bosnia Judgment, p. 126, para. 198.
16Prosecutor v. Tolimir, Trial Judgment, 1December 2012, para. 749; citing Pros‑
ecutor v. Jelisić, Trial Judgment, 14 December 1999, para. 82.

419

7 CIJ1077.indb 834 18/04/16 08:54 419 application of genochide convention (sep. ohp. bhandari)

other acts of elimination of a segment of society, can also bedeemed
17
genocide.’”
17. The Trial Chamber then proceeded to apply this more flexible con -

cept of substantiality to the factual circumstances of that case, which h
involved, inter alia, the killing of three prominent members of the Bos -
nian Muslim population of Z epa enclave in Eastern Bosnia and Herze -
govina (“BiH”). As the Trial Chamber recalled, Z epa was a village

situated approximately 20 kilometres from Srebrenica that had a popula -
tion of less than 3,000 inhabitants prior to the war, but which saw its
population swell to as many as 10,000 people by July 1995, as Bosnian
Muslims from other surrounding areas in Eastern BiH sought refuge
from the prevailing hostilities, such that “[d]uring the conflict the popula -
18
tion of Zepa consisted entirely of Bosnian Muslims” .
18. Regarding the three individuals killed, the Trial Chamber made the
following observations:

“The three leaders were Mehmed Hajrić, the Mayor of the muni -
cipality and President of the War Presidency, Colonel Avdo Palić,
Commander of the ABiH Zepa Brigade . . . and Amir Imamović, the

Head of the Civil Protection Unit. They were, therefore, among the
most prominent leaders of the enclave . . . [T]hose responsible for
killing Hajrić, Palić and Imamović targeted them because they were
leading figures in the Zepa enclave at the time that it was populated

by Bosnian Muslims. These killings should not be viewed in isola ‑
tion . . . it is significant to consider the connection between the VRS
operations in Srebrenica and Zepa. The respective attacks and take -
overs of the enclaves were synchronized by the [same] leadership and
included the same forces. The takeover of Zepa enclave followed less

than two weeks after the capture of Srebrenica, during a time in which
the news of the murders of thousands of Bosnian Muslim men was
starting to spread. While the individuals killed were only three in num ‑
ber, in view of the size of Zepa, they constituted the core of its civilian

and military leadership. The mayor, who was also a religious leader,
the military commander and the head of the Civil Protection Unit,
especially during a period of conflict, were key to the survival of a h
small community. Moreover, the killing of Palić, who at this time
enjoyed a special status as the defender of the Bosnian Muslim pop -

ulation of Zepa, had a symbolic purpose for the survival of the Bos -
nian Muslims of Eastern BiH. While the majority accepts that the
Bosnian Serb Forces did not kill the entirety of the Bosnian Muslim
leadership of Z epa . . it does not consider this to be a factor against

17
Prosecutor v. Tolimir, Trial Judgment, 12 December 2012, fn. 3138; emphasis added;
citing Final Report of the Commission of Experts Established pursuant toh Security Council
Resolution 780 (1992), UN doc. S/1994/674, (“Commission of Experts Report”), para. 94.
18
Prosecutor v. Tolimir, Trial Judgment, 12 December 2012, paras. 598-599.

420

7 CIJ1077.indb 836 18/04/16 08:54 420 application of genochide convention (sep. ohp. bhandari)

its determination that the acts of murder against these three men con ‑
stitutes genocide.” 19

The Trial Chamber then proceeded to reach the following further conclu-
sions :

“In accordance with the Jelisić Trial Chamber’s finding in which it
relied on the Commission of Experts Report the Majority also takes
into account the fate of the remaining population of Zepa ; their for‑

cible transfer immediately prior to the killing of these three leaders i▯s a
factor which supports its finding of genocidal intent. To ensure that th▯e
Bosnian Muslim population of this enclave would not be able to recon ‑
stitute itself, it was sufficient inhe case of Zepa to remove its civilian

population, destroy their homes and their mosque, and murder its most
prominent leaders . . . The Majority has no doubt that the murder of
[these three leaders] was a case of deliberate destruction of a limited h
number of persons selected for the impact that their disappearance
20
would have on the survival of the group as such.”

19. I acknowledge that these conclusions — which were subject to a
dissenting opinion and are currently awaiting a judgment from the ICTY
Appeals Chamber — must be treated with a requisite degree of caution.

Such limitations having been duly conceded, in my view the passages
cited above from the Tolimir Trial Judgment nevertheless evince a con -
certed departure from the narrower ambit of the tripartite test adopted hby
this Court in the Bosnia Judgment. Given that the present Judgment has

likewise determined to apply the Bosnia formula in a more flexible man -
ner that places less emphasis on the primacy of the quantitative elementh,
I am both surprised and disheartened by the majority’s refusal to makhe
any mention of the most recent judicial pronouncement of the ICTY on

this highly pertinent and substantively fluid area of law.
20. Specifically, the Tolimir Judgment’s finding of genocide where only
three killings were proven marks a clear and unambiguous departure
from the Bosnia formula’s dogged insistence that the numerosity of the

victims of predicate acts under Article II of the Genocide Convention be
considered a pre -eminent factor in the substantiality equation. Rather,
Tolimir presents a rather striking example of a case where not only were
the three individuals killed low in absolute terms, but against the back -

drop of a homogeneous religious community of approximately 10,000 it
is dubious to suggest that their deaths could constitute a high relative
“numerosity in relation to the group as a whole” 2. Rather, in find-
ing genocidal intent, the Tolimir Trial Chamber placed heavy empha-

sis on the prominence of the targeted population and the fact that the

19Prosecutor v. Tolimir, Judgment, 12 December 2012, paras. 778-780; emphasis
added.
20Ibid., paras. 781-782; emphasis added.
21Ibid., para. 749.

421

7 CIJ1077.indb 838 18/04/16 08:54 421 application of genochide convention (sep. ohp. bhandari)

attackers exercised complete control over the enclave during the period hin
question.

21. Finally, what cannot be overlooked is that apart from three kill -
ings, the gravamen of the atrocities perpetrated at Zepa constituted the
complete forcible transfer of its entire Bosnian Muslim population, a
community of thousands, away from that enclave and into Bosnian -
controlled territory. While Tolimir certainly did not go so far as to pro -

nounce that the “ethnic cleansing” of these thousands of Bosnian Muslims
from Z epa enclave (in conjunction with means taken to ensure their
non-return, such as destruction of homes and places of worship) consti -
tuted genocide per se, it did clearly and unequivocally affirm that this
mass displacement of the civilian population, when combined with the

very limited targeted killing of prominent local leaders, constituted anh
attempt to physically destroy a significant part of the Bosnian Muslim
group of Eastern BiH, by depriving that community of the means of recon ‑
stituting itself within that geographical area. On this final point, it would
appear to be a clear evolution of the position adopted by this Court in hthe

Bosnia Judgment as to what constitutes “physical destruction” of the
group for the purpose of Article II of the Genocide Convention, where it
was held that

“[i]t will be convenient at this point to consider what legal significanc▯e
the expression [‘ethnic cleansing’] may have [under the Genocide Con ‑
vention]. It is in practice used, by reference to a specific region or area,
to mean ‘rendering an area ethnically homogeneous by using force or

intimidation to remove persons of given groups from the area’. . .
It does not appear in the Genocide Convention ; indeed, a proposal
during the drafting of the Convention to include in the definition
‘measures intended to oblige members of a group to abandon their
homes in order to escape the threat of subsequent ill -treatment’

was not accepted . . . It can only be a form of genocide within the
meaning of the Convention, if it corresponds to or falls within one of
the categories of acts prohibited by Article II of the Convention. Nei ‑
ther the intent, as a matter of policy, to render an area ‘ethnically
homogeneous’, nor the operations that may be carried out to implement▯
such policy, can as such be designated as genocide . . . As the ICTY

has observed, while

‘there are obvious similarities between a genocidal policy
and the policy commonly known as ‘ethnic cleansing’. . . [a]
clear distinction must be drawn between physical destruction

and mere dissolution of a group. The expulsion of a group

422

7 CIJ1077.indb 840 18/04/16 08:54 422 application of genochide convention (sep. ohp. bhandari)

or part of a group does not in itself suffice for geno -
cide.’” 22

In the present Judgment, this relationship has been revisited in the folhlow -
ing terms :

“The Court recalls that, in its 2007 Judgment, it stated that

‘[n]either the intent, as a matter of policy, to render an area ‘ehthn-i
cally homogeneous’, nor the operations that may be carried out to
implement such policy, can as such be designated as genocide : the
intent that characterizes genocide is ‘to destroy, in whole or in

part’ a particular group, and deportation or displacement of the
members of a group, even if effected by force, is not necessarily equi ‑
valent to destruction of that group, nor is such destruction an auto ‑
matic consequence of the displacement. [. . .]’

It explained, however, that :

‘[t]his is not to say that acts described as ‘ethnic cleansing’h may
never constitute genocide, if they are such as to be characterized

as, for example, ‘deliberately inflicting on the group conditions ohf
life calculated to bring about its physical destruction in whole or
in part’, contrary to Article II, paragraph (c), of the Convention,
provided such action is carried out with the necessary specific
intent (dolus specialis), that is to say with a view to the destruc -

tion of the group, as distinct from its removal from the region . . .
In other words, whether a particular operation described as ‘eth -
nic cleansing’ amounts to genocide depends on the presence or
absence of acts listed in Article II of the Genocide Convention,
and of the intent to destroy the group as such. In fact, in the

context of the Convention, the term ‘ethnic cleansing’ has no legahl
significance of its own. That said, it is clear that acts of ‘ethnihc
cleansing’ may occur in parallel to acts prohibited by Article II of
the Convention, and may be significant as indicative of the pre -
sence of a specific intent (dolus specialis) inspiring those acts.’”

(Judgment, para. 162 (internal citations omitted ; emphasis
added).)

22. In my respectful view, the ICTY Trial Chamber in Tolimir has
burst open the tight confines of the dictum promulgated in Bosnia and

reaffirmed in the present Judgment. By finding that the confluence hof kil-l
ing three prominent community leaders (which constitute genocidal acts h
as per Article II (a) of the Convention) in parallel to massive acts of

22Bosnia Judgment, pp. 122-123, para. 190 (internal citations omitte; emphasis
added).

423

7 CIJ1077.indb 842 18/04/16 08:54 423 application of genochide convention (sep. ohp. bhandari)

ethnic cleansing (which are non-genocidal atrocities per se; see Judgment,
para. 162) was sufficient to characterize the entire series of events occur ‑
ring at Zepa as possessing genocidal intent, the Trial Chamber clearly went

above and beyond this Court in Bosnia and the present Judgment, if not
in its application of the letter of the applicable law, then clearly in hits
appreciation of the spirit thereof. Stated differently, there is no indication
in Tolimir that the approximately 10,000 denizens of Zepa enclave who
were forcibly removed from the area and prevented from returning were

targeted for physical or biological destruction as envisaged by Article II
of the Convention. Rather, the Trial Chamber found that their permanent
removal from that geographical area (in conjunction with the destruction
of a diminutive core of its civil and military leadership) was enough tho
constitute “physical or biological” destruction under the terms ofh Arti -

cle II of the Convention. This cannot but be described as a clear depar -
ture from the Court’s analysis in Bosnia and certain other judgments
rendered by the ICTY upon which the Court in Bosnia relied. Not only
has the quantitative element that featured so prominently in Bosnia been
eschewed, but the Tolimir Judgment has clearly pushed the boundaries of

what constitutes physical or biological destruction by expressly incorpo -
rating non-fatal geographical concerns. In other words, according to my
reading of Tolimir the Trial Chamber clearly found that genocidal intent
was established not because the approximately 10,000 Bosnian Muslims
of Z epa enclave were targeted for elimination per se, but rather because

they were targeted for elimination from that specific location.

23. Granted, the majority’s reticence to adopt a Tolimir-style approach
may be readily (and defensibly) explained by considerations such as thhe

fact that the case remains under appeal and that the finding of genocihde
at Zepa was linked (although obliquely) to the now widely recognized
genocide that was perpetrated by the same attackers at Srebrenica some
20 kilometres away and mere days beforehand. Nevertheless, in my
respectful view, to ignore Tolimir completely constitutes a failure by the

majority to heed its own undertaking to “take account, where approprih -
ate, of the decisions of international criminal courts or tribunals, in hpar-
ticular those of the ICTY, as it did in 2007, in examining the constituent
elements of genocide in the present case” (Judgment, para. 129). I shall
return to this aspect of the Tolimir precedent when dissecting the present

Judgment’s treatment of Croat victims during the siege of Vukovar andh
its aftermath (see infra).

424

7 CIJ1077.indb 844 18/04/16 08:54 424 application of genochide convention (sep. ohp. bhandari)

The popović ICTY Trial Chamber Judgmenth

24. In the Popović case, the Trial Chamber of the ICTY provided an
analysis on the substantiality component in relation to the killing of shev-
eral thousand Bosnian Muslim men at Srebrenica enclave in Eastern Bos-
nia and Herzegovina in July 1995. It is to be recalled that both the Court

and several Trial and Appeals Chambers of the ICTY have consistently
held that the massacre at Srebrenica constituted genocide. Consequently,h
while the Popović Trial Chamber’s finding of genocidal intent in relation
to the Srebrenica massacre is not in itself a novel jurisprudential devehlop-

ment, in expounding this notion the Trial Chamber made the following
noteworthy remarks:

“The Trial Chamber finds that the Muslims of Eastern Bosnia con-
stitute a substantial component of the entire group, Bosnian Muslims.
As has been found by the Appeals Chamber, although the size of the

Bosnian Muslim population in Srebrenica before its capture . . . was
a small percentage of the overall Muslim population of BiH at the
time, the import of the community is not appreciated solely by its size.h
The Srebrenica enclave was of immense importance to the Bosnian Serb
leadership because : (1) the ethnically Serb state they sought to create

would remain divided and access to Serbia disrupted without Srebren ‑
ica; (2) most of the Muslim inhabitants of the region had, at the
relevant time, sought refuge in the Srebrenica enclave and the elimi -
nation of the enclave would accomplish the goal of eliminating the

Muslim presence in the entire region ; and(3) the enclave’s elimination
despite international assurances of safety would demonstrate to the Bos‑
nian Muslims their defenceless and be ‘emblematic’ of the fate of ▯all
Bosnian Muslims. ” 23

25. In my respectful view, the first and third factors enumerated by the
Trial Chamber in Popović may have warranted consideration when con -
ducting an assessment of genocidal intent in the present Judgment, par -
ticularly with reference to the attack on Vukovar municipality. Regardinhg
the first factor, I note that the Judgment has recalled that :

“Croatia attaches particular importance to the events which took

place in Vukovar and its surrounding area in the autumn of 1991.
According to the Applicant, the JNA and Serb forces killed several
hundred civilians in that multi -ethnic city in Eastern Slavonia,
situated on the border with Serbia and intended to become, under the

23
Prosecutor v. Popović et al., Trial JudgmentJune 2010, para. 865; emphasis
added.

425

7 CIJ1077.indb 846 18/04/16 08:54 425 application of genochide convention (sep. ohp. bhandari)

plans for a ‘Greater Serbia’, the capital of the new Serbian regio▯n of
Slavonia, Baranja and Western Srem. ” (Judgment, para. 212 ; empha-

sis added.)
This averred emblematic significance of Vukovar can be further inferrehd

from the findings of the ICTY Trial Chamber in Mrkšić, as accepted by
the Court in the instant Judgment, which found that during the approxi -
mately three -month siege of Vukovar :

“The duration of the fighting, the gross disparity between the num-
bers of the Serb and Croatian forces engaged in the battle and in the
armament and equipment available to the opposing forces and, above
all, the nature and extent of the devastation brought on Vukovar and

its immediate surroundings by the massive Serb forces over the pro ‑
longed military engagement, demonstrate, in the finding of the Cham ‑
ber, that the Serb attack was also consciously and deliberately directed▯
against the city of Vukovar itself and its hapless civilian population ,

trapped as they were by the Serb military blockade of Vukovar and
its surroundings and forced to seek what shelter they could in the
basements and other underground structures that survived the ongo -
ing bombardments and assaults. What occurred was not, in the find -
ing of the Chamber, merely an armed conflict between a military force h

and an opposing force in the course of which civilians became casu -
alties and some property was damaged. The events, when viewed
overall, disclose an attack by comparatively massive Serb forces, well
armed, equipped, and organized, which slowly and systematically
destroyed a city and its civilian and military occupants to the point

where there was a complete surrender of those that remained . . .
It is in this setting that the Chamber finds that, at the time relevanht
to the Indictment, there was in fact, not only a military operation
against the Croat forces in and around Vukovar, but also a widespread
and systematic attack by the JNA and other Serb forces directed against ▯

the Croat and other non‑Serb civilian population in the wider Vukovar
area . . .”4

In my view, this sustained, ethnically discriminatory attack, aimed in
part at the slow and systematic destruction of the Croat civilian populahce
of Vukovar, provides implicit evidence of its strategic importance in tehrms
of allowing the expansionist policy of “Greater Serbia” to gain a hpivotal

foothold within Croatian territory, and thus heightens the prominence ofh
the Vukovar Croat subgroup when assessing genocidal intent vis -à-vis
that municipality and its environs.

26. Regarding the third criterion enunciated in Popović, I recall that

the evacuation of Vukovar hospital on 20 November 1991, through which
many Croats were interned at nearby concentration camps and subse -

24 Judgment, para. 218; citing Mrkšić, paras. 470 and 472.

426

7 CIJ1077.indb 848 18/04/16 08:54 426 application of genochide convention (sep. ohp. bhandari)

quently killed, severely beaten and/or otherwise subjected to serious fohrms

of physical and psychological abuse, was conducted in violation of the
Zagreb Agreement, which purported to allow for the safe evacuation of
those internally displaced Vukovar Croats who had sought refuge at the
local hospital under the supervision of neutral international monitors. hI

believe that the deliberate and cynical manner in which this internationhal
agreement was violated, to the grave detriment of those who made the
assumption that a widely publicized agreement would guarantee their
safety, allows for the inference that the sorry plight of the victims frhom

Vukovar hospital and those subsequently interned in concentration and
death camps, could certainly, to paraphrase Popović, “demonstrate to the
Vukovar Croats their defencelessness and be emblematic of the fate of alhl
ethnic Croats on Croatian territory”.

The nizeyimana ICTR Trial Chamber Judgmenth

27. In the Nizeyimana case, the accused was convicted at trial for geno-
cide in relation to, inter alia, the killing of Rosaliecanda, a member of
the targeted Tutsi ethnic group and former Queen of Rwanda. In apply -

ing the substantiality criterion of genocidal intent, the Trial Chamber
stressed that

“[t]he fact that this operation targeted one Tutsi in particular in no way
impacts the conclusion that the perpetrators possessed the intent to
destroy at least a substantial part of the Tutsi ethnic group.The Cham-

ber reiterates that this killing must be viewed in the context of the
targeted and systematic killing of Tutsis perpetrated . . . in Butare
[town] around this time. Moreover, the symbolic importance of the kil‑l
ing of Gicanda as a means of identifying the enemy is also relevant.” 25

In that regard, the Trial Chamber noted that it had “no doubt that thhe
murder of Gicanda . . . who was a symbol of the former [Tutsi] monar-

chy, was killed in order to set a striking example that Tutsis, 26 well as
Hutus sympathetic to the plight of the Tutsis, were the enemy ” . The Trial
Chamber further stressed the nexus between this particular attack and thhe
significantly increased violence against Tutsi civilians in Butare town fol -

lowing an incendiary speech by the President of Rwanda on 19 April 1994
in which he exhorted the population to seek out and kill Tutsis.

28. In relation to a separate incident, the Trial Chamber found that the

killing of one Pierre Claver Karenzi, a Tutsi lecturer at a local university

25Prosecutor v. Nizeyimana, Trial Judgment, June 2012, para. 1530; emphasis
added.
26
Ibid., para. 1511; emphasis added.

427

7 CIJ1077.indb 850 18/04/16 08:54 427 application of genochide convention (sep. ohp. bhandari)

who was considered “a prominent figure in Butare” town also constituted

genocide. Again the Trial Chamber found that

“ Karenzi’s murder is also emblematic of the systematic nature in which
Tutsi civilians were identified and killed on an ongoing basis at roadh -
blocks manned by . . . soldiers in Butare town. Consequently, while
this incident only resulted in the killing of one Tutsi, the Chamber has▯

no doubt that the physical perpetrator acted with the spe27fic intent to▯
destroy at least a substantial part of the Tutsi group .”

Once again the Trial Chamber found that this attack was linked to the
broader context of significantly increased targeted killings of the Tuhtsi
ethnic group in Butare town around that time in the wake of the Presi -
dent’s speech.

29. Finally, the Trial Chamber found genocidal intent in relation to
another incident where two Tutsi civilians were killed and another seri -
ously injured at a military roadblock. As the Trial Chamber reasoned,

“ [w]hile these attacks only resulted in the deaths of two Tutsis and the ▯
serious bodily harm of a third, the Chamber has no doubt that the per ‑
petrators acted with the intent to destroy at least a substantial part o▯f

the Tutsi group. These attacks were emblematic of the systematic nature ▯
in which Tutsi civilians were identified and killed on an ongoing basis
at this roadblock and others manned by. . . soldiers in Butare town.” 28

What is particularly noteworthy about this specific finding of genochidal
intent is the Trial Chamber’s determination that the attack on the thhree

Tutsi victims was “emblematic” of the overall group not because of
the individual prominence of the victims within the community (there
was no evidence on record to suggest such a conclusion), but rather
because the attack on them embodied a modus operandi for the system ‑

atic destruction of the Tutsi group in Butare town generally. In other
words, Tutsis at the roadblock were “emblematic” of the overall group
not because of who they were, but rather the manner in which they
were attacked. While indicia of “prominence” through the modus operandi

of the attack may be gleaned from the killings of Gicanda and Karenzi,
the fact that the victims of this third attack did not hold any prominenht
positions within the Tutsi community only further underscores the point.h
Consequently, I find these exemplars from the Nizeyimana Trial

Judgment to signal a clear departure from what was envisaged
by the “qualitative approach” in the Court’s Bosnia Judgment (and
subsequently rebranded as the “prominence” of the targeted group
in the instant Judgment), and believe that the present Judgment’s

analysis would have been enriched by a consideration of this recent,

27Prosecutor v. Nizeyimana, Trial Judgment, 19June 2012, para. 1530; emphasis
added.
28
Ibid., para. 1521; emphasis added.

428

7 CIJ1077.indb 852 18/04/16 08:54 428 application of genochide convention (sep. ohp. bhandari)

pertinent jurisprudential development on the law of genocidal dolus

specialis.

30. Applying the Nizeyimana precedent to the facts at bar, I would
note that the present Judgment recalls that the Mrkšić Trial Chamber
found that the attacks in Eastern Slavonia generally followed a consistent

pattern:

“[T]he system of attack employed by the JNA typically evolved
along the following lines (a) tension, confusion and fear is built up
by a military presence around a village (or bigger community) and
provocative behaviour ; (b) there is then artillery or mortar shelling
for several days, mostly aimed at the Croatian parts of the village ; in

this stage churches are often hit and destroyed (c) in nearly all cases
JNA ultimata are issued to the people of a village demanding the
collection and the delivery to the JNA of all weapons ; village delega-
tions are formed but their consultations with JNA military authorities

do not lead . . . to peaceful arrangements ;.. . (d) at the same time,
or shortly after the attack, Serb paramilitaries enter the village ; what
then follows varied from murder, killing, burning and looting, to dis-
crimination.” 29

The Judgment also recalled that in the Martić case, the ICTY Trial

Chamber made similar findings regarding the pattern of attacks perpe -
trated by Serb forces in Croatia :

“[T]he area or village in question would be shelled, after which
ground units would enter. After the fighting had subsided, acts of
killing and violence would be committed by the forces against the
civilian non -Serb population who had not managed to flee during the
attack. Houses, churches and property would be destroyed in order

to prevent their return and widespread looting would be carried out.
In some instances the police and the TO of the SAO Krajina organ -
ized transport for the non -Serb population in order to remove it from
SAO Krajina territory to locations under Croatian control. More -

over, members of the non -Serb population would be rounded up and
taken away to detention facilities, including in central Knin, and even -
tually exchanged and transported to areas under Croatian control.” 30

After considering, inter alia, these findings from the ICTY, the present
Judgment concludes that :

“The Court likewise notes that there were similarities, in terms of
the modus operandi used, between some of the attacks confirmed to
have taken place. Thus it observes that the JNA and Serb forces

29
30Judgment, para. 414, citing Mrkšić Trial Judgment, para. 43.
Ibid., para. 427.

429

7 CIJ1077.indb 854 18/04/16 08:54 429 application of genochide convention (sep. ohp. bhandari)

would attack and occupy the localities and create a climate of fear
and coercion, by committing a number of acts that constitute the

actus reusof genocide within the meaning of Article II (a) and (b)
of the Convention. Finally, the occupation would end with the forced
expulsion of the Croat population from these localities.
The findings of the Court and those of the ICTY are mutually
consistent, and establish the existence of a pattern of conduct that

consisted, from August 1991, in widespread attacks by the JNA and
Serb forces on localities with Croat populations in various regions of
Croatia, according to a generally similar modusoperandi.” (Judgment,
paras. 415-416.)

31. Bearing this established pattern of conduct throughout various
parts of the territory of Croatia in mind, I would further recall that the

Applicant has presented the siege of Vukovar as representing a paradig -
matic example of the modus operandi outlined above. As counsel for Cro-
atia stated during the oral hearing phase of this case,

“[W]hat happened at Vukovar was repeated again and again across
Eastern Slavonia and across Croatia as a whole in the course of this
conflict. This pattern may have varied from village to village, town to
town and across different regions. But, properly analysed, the ‘pat -

tern’ discloses that there was an intention to ‘destroy’ a parth of the
Croat group in question. The artillery or mortar shelling was wholly
disproportionate and, in places, such as Vukovar, essentially destroyed
the entire city. And the murderous attacks were never intended as part
of the mere expulsion of a part of the Croat group in question.” 31

Consequently, I believe that in view of the Nizeyimana Trial Judgment,
the Judgment’s analysis with respect to substantiality could have beehn

enhanced by considering the modus operandi of the attack on Vukovar,
being a microcosm for the manner in which a much wider conflict was
waged, for the purpose of assessing whether the “prominence” of thhe
Vukovar Croats could factor into the calculus as to whether they were

targeted with genocidal intent.

The H ateGekimana ICTR Trial Chamber Judgmenth

32. In the case of Hategekimana, the accused was convicted, inter alia,
of genocide for the murder of three Tutsi women during an attack on their
home by militia and soldiers. In determining genocidal intent, the Trial

Chamber noted that in addition to the fact that the three women were
singled out because of their ethnicity, the Chamber had received “exten -
sive evidence . . . about the targeting of Tutsi civilians in Butare [prov -
ince] following the speech of interim President Sindikubwabo on
19 April 1994”, which resulted in “many Tutsi civilians being killed in their

31 CR 2014/8, p. 47 (Starmer); emphasis added.

430

7 CIJ1077.indb 856 18/04/16 08:54 430 application of genochide convention (sep. ohp. bhandari)

homes over the course of many days”. The Chamber found that “[g]iven

the scale of the killings and their context, the only reasonable inference is
that the assailants [who killed the women] possessed the intent to destrhoy
in whole or in part a substantial part of the Tutsi group” 3. Once again
we see an example where an attack against a targeted group that resultedh

in a comparatively low absolute and relative number of victims was
nevertheless deemed to possess genocidal intent, due at least in part toh
the modus operandi of the manner in which they were killed.

The m unyakazi ICTR Trial Chamber Judgmenth

33. In the case of Munyakazi, the accused was convicted of genocide
for, inter alia, an attack on a parish that killed between 60 and 100 Tutsi

refugees. The Trial Chamber observed that the attack occurred the day
after a much larger attack on a different parish where approximately
5,000 to 6,000 Tutsis were killed by the same group of perpetrators. Con -
sidering both attacks as a whole, the Trial Chamber found genocidal
33
intent for a substantial part of the Tutsi ethnic group . This finding of
genocide evokes many parallels with the ICTY Trial Chamber’s findinhg
of genocide in relation to Zepa enclave in the Tolimir Judgment, which
also featured an attack on one geographic area where a relatively small h
number of victims were killed (Zepa) but which was closely linked, in

terms of geography, time, and the identity of the perpetrators, to a prehvi-
ous, considerably more sizeable attack (Srebrenica).

Other ICTR Trial Chamber Judghments

34. In keeping with the pattern demonstrated above, since the Court’s

issuance of the Bosnia Judgment in 2007, the ICTR has made findings of
genocide in relation to scenarios where “the quantitative element”h (to use
the nomenclature adopted by the present Judgment) figured far less
prominently in the calculus as to whether the attacks were perpetrated

with genocidal intent than a strict application of the Bosnia formula
would dictate. In this regard, we see that genocidal intent in several
instances was inferred in large part due to the geographic profile of h
the situs of the attack and/or the prominence of the victims (whether

said prominence was measured in terms of personal standing in the

32 Prosecutor v. Hategekimana, Trial Judgment, 6 December 2010, para. 673; emphasis
added.
33
Prosecutor v. Munyakazi, Trial Judgment, 5 July 2010, paras. 496, 499-500.

431

7 CIJ1077.indb 858 18/04/16 08:54 431 application of genochide convention (sep. ohp. bhandari)

community of the victim or the modus operandi of how the attack
unfolded) . In sum, what we see is a clearly more flexible application

of genocidal dolus specialis that would tend to challenge the pre-eminence
afforded the substantiality criterion in the Bosnia case.

35. For the avoidance of any doubt, I wish to underscore my recogni -
tion that there are obviously significant contextual differences between

the crimes prosecuted in relation to Croatia before the ICTY (and, by
extension, the subject -matter presently before the Court) and those relat -
ing to Rwanda before the ICTR, not the least of which is the sheer dis -

parity in scale of the atrocities that occurred during the course of theh
respective conflicts. Hence I recall that Croatia’s case — taken at its high-
est— is that the hostilities that form the backdrop to the present Judg -
ment resulted in 12,500 Croat deaths, whereas conservative estimates of

the carnage in Rwanda posit that at least half a million ethnic Tutsis and
moderate Hutus were killed during the course of the genocide that unfolded
in that country in 1994 — a genocide, which, it should be noted, was the

subject of an express finding of judicial notice by the ICTR Appeals
Chamber 35. Indeed, I recall that the Court stipulated in the present Judg-
ment that it would give particular preference to jurisprudence emanatingh
from the ICTY (see supra, paragraph 13, citing paragraph 129 of the

Judgment), and I understand this perfectly sensible decision to be motih -
vated in large part by the plain fact that the cases before the ICTY invholve
much closer historical, socio -political and legal issues to those presented

in the case at bar than cases appearing before the ICTR.

36. In sum, while I am by no means advocating the wholesale importa -
tion of ICTR case law into the jurisprudence of this Court, my concern

lies with what I find to be essentially the complete disregard of the hmost
prolific judicial body to have interpreted and applied the Genocide
Convention in the course of human history. With the greatest of respect h

to my learned colleagues, failure to so much as consult this ample body hof
jurisprudence, to my mind, constitutes a failure by the Court in its duthy
and its undertaking to keep abreast of the most recent and pertinent
developments in the law of genocide in the present Judgment.

34 See, e.g.,Prosecutor v. Nsengimana, Trial Judgment, 17 November 2009,

paras. 834-836; Prosecutov.Renzaho, Trial Judgment, 14 July 2009, paras. 768-769;
Prosecutor v. Rukundo, Trial Judgment, 27 February 2009, paras. 72, 74, 76; Prosecutor
Nc35mihigo, Trial Judgment, 12 November 2008, paras. 333-336, 346-347, 354, 357.
This landmark decision was delivered by the Appeals Chamber on Prosecutohr’s
Appeal on Judicial Notice, dated 1 6 June 2006, in the trial of Prosv.Karemera,
Ngirumpatse and Nzirorera, ICTR-98-44AR73 (C).

432

7 CIJ1077.indb 860 18/04/16 08:54 432 application of genochide convention (sep. ohp. bhandari)

Conclusion on the Courth’s Treatment of post -B osnia
Jurisprudence of the ICTY ahnd ICTR

37. In view of the observations above, I believe that the jurisprudence
of the ad hoc international criminal tribunals in the years following the
issuance of this Court’s Bosnia Judgment demonstrate a dilution of the
rigidly hierarchical tripartite formula for discerning genocidal intent has

promulgated in that precedent, whereby the numerosity of the targeted
population was clearly designed to serve as the pre -eminent concern in
any such calculus. As I have noted above, while there are traces of a mohl-
lified approach to be found in the Judgment’s gentle refastening ofh the

Bosnia formula into a more egalitarian weighing of the three criteria to be
applied in the present Judgment, in my respectful view it was incumbent h
upon the Court to take the further step of explicitly acknowledging and h
engaging the recent, pertinent jurisprudential developments presented byh
the ICTY and ICTR in this area of law and to incorporate, if and where

appropriate, any evolutions to the Bosnia test that are not only strictly
necessary for the disposition of the merits of the present dispute, but h
which may elucidate the development this legal area has undergone over
the past eight years. In other words, even if it were not appropriate orh
even correct to apply such precedents to the facts at bar, in my considered

opinion it was certainly appropriate to at least consider such key develop -
ments, if only to explain why they ought to be distinguished from the
present case. Such an approach, I suggest, would be wholly commensu -
rate with the Court’s role as a pre-eminent global judicial forum to which
other international dispute resolution mechanisms turn in search of guidh-

ance on such important and arcane points of law. Consequently, I regret h
that the majority has missed a prime opportunity to improve the clarity h
and authority of this area of public international law.

The Majority’s Conclusihons on Genocidal Intenth

in the Present Judgmenth

38. In assessing whether the targeted group was “substantial” for pur -
poses of the chapeau of Article II of the Genocide Convention, the Judg-
ment recalls Croatia’s submission

“that the Croats living in the regions of Eastern Slavonia, Western
Slavonia, Banovina/Banija, Kordun, Lika and Dalmatia [who were tar-

geted for destruction by JNA and affiliated Serb forces] constituted a
substantial part of the protected group , and that the intent to destroy
the protected group ‘in part’, which characterizes genocide as defihned

433

7 CIJ1077.indb 862 18/04/16 08:54 433 application of genochide convention (sep. ohp. bhandari)

in Article II of the Convention, is thus established” (Judgment,
para. 403 ; emphasis added).

The Judgment also recalls that “[i]n its written pleadings, Croatia
defines [the overall protected] group [at issue in its Claim] as the Croat
national or ethnical group on the territory of Croatia, which is not con -

tested by Serbia” (see ibid., para. 205; emphasis added). Relying on offi -
cial census data from 1991 — the year in which the hostilities that are the
subject-matter of the present dispute commenced — adduced by Croatia,
and uncontested by Serbia, the Judgment finds that “the ethnic Croaht

population living in the [identified] regions . . . numbered between 1.7 and
1.8 million [individuals . . . and] constituted slightly less than half of the
ethnic population living in Croatia” (see ibid., para. 406). The Judgment
further concludes “that acts committed by JNA and Serb forces in the
[identified] regions. . . targeted the Croats living in those regions, within

which these armed forces exercised and sought to expand their control”
(ibid). While the majority also found that “as regards the prominence of
that part of the group, the Court notes that Croatia has provided no
information on this point” (ibid.) — a conclusion I do not share, and to
which I shall return presently — “[t]he Court [nevertheless] concludes

from the foregoing that the Croats living in the [identified] regions . . .
constituted a substantial part of the Croat group” (ibid.). Despite my
misgivings about the majority’s pronouncement as to the ostensible lahck
of evidence regarding the prominence of the Croat ethnic group at issue,h
I am in full agreement with the majority’s general conclusion that thhe part

of the ethnic Croat group identified by the Applicant constituted a sub -
stantial part of the overall Croat ethnic group living within the territory
of Croatia during the relevant period.
39. It is to be further recalled that the Judgment concludes that :

“The Court is fully convinced that, in [the] various [identified] loca-l

ities. . . the JNA and Serb forces perpetrated against members of the
protected group acts falling within subparagraphs (a) [killing mem-
bers of the group] and (b) [causing serious bodily or mental harm to
members of the group] of Article IIof the Convention, and that the
actus reus of genocide has been established.” (Ibid., para. 401.)

I am also in complete agreement with the majority on this point. Where Ih
depart from the majority is in the manner of reasoning through which it has

arrived at its conclusion that “[t]he acts constituting the actus reus of geno-
cide within the meaning of Article II (a) and (b) of the Convention were
not committed with the specific intent required for them to be characther-
ized as acts of genocide” (ibid., para. 440). While I again recall that I have
joined the majority in rejecting Croatia’s claim that genocide was com -

mitted against the targeted Croat population on evidentiary grounds, given
that the majority has elected to take Croatia’s case at its highest phrior to
dismissing it, I shall proceed to make certain observations and critiquehs

434

7 CIJ1077.indb 864 18/04/16 08:54 434 application of genochide convention (sep. ohp. bhandari)

regarding its approach to the analysis of genocidal dolus specialis as per-

tains to Croatia’s allegations.

The Geographic Area Consihdered by the Majority whenh
Assessing d olus s pecialis

40. As I have noted, supra, the Judgment characterizes Croatia’s

delimitation of the relevant “part” of the ethnic Croat group as bheing
“the Croats living in the regions of Eastern Slavonia, Western Slavonhia,
Banovina/Banija, Kordun, Lika and Dalmatia” (Judgment, para. 403).
While I agree with this conclusion as pertains to these six geographical

locales, I am also mindful of the fact that the gravamen of Croatia’sh case
focused heavily on the specific region of Eastern Slavonia, and in parhticu -
lar the city of Vukovar and its environs. As counsel for Croatia submitted
during the oral hearing phase of this case :

“Even when judged against the other atrocities detailed by the Appli‑
cant before this Court . . . the events in Vukovar plumbed new depths.

Serbian forces carried out a sustained campaign of shelling ; system-
atic expulsion ; denial of food, water, electricity, sanitation and med -
ical treatment ; bombing; burning; brutal killings and torture which
reduced the city to rubble and destroyed its Croat population. It started

with roadblocks and ended with torture camps and mass execution.
In human terms, the scar will never heal.

The events at Vukovar are significant and they are known around
the world. They deserve to be examined in context, in detail and in
full.”36

This heavy reliance by Croatia on the events at Vukovar throughout this
case is even conceded by Serbia when it acknowledges in its written
Rejoinder that “[t]he most significant episodes in Eastern Slavonia took

place in Vukovar, and these attract the bulk of the discu37ion in the Reply,
as they did in the Memorial and Counter -Memorial” . Indeed, on more
than one occasion it has been expressly recognized by the Court in the
present Judgment that “Croatia has given particular attention” to h“the

events at Vukovar” in pursuing its claims in this case (Judgment,
paras. 429 and 436).
41. Moreover, there is clear precedent from this Court that an analysis
of genocidal intent may be confined to a geographic area notably smaller

than the six expansive regions considered by the present Judgment, even
if the Applicant framed its cause of action with respect to a wider geog▯raph‑

36CR 2014/8, pp. 28-29, paras. 1-2 (Starmer); emphasis added.
37
Rejoinder of Serbia, para. 370; emphasis added.

435

7 CIJ1077.indb 866 18/04/16 08:54 435 application of genochide convention (sep. ohp. bhandari)

ical area. This was of course precisely what occurred in the 2007 Bosnia
Judgment, wherein the Court made a finding of genocide solely with
respect to Srebrenica, a Bosnian Muslim enclave consisting of upwards of
30,000 people where more than 7,000 military-aged Bosnian Muslim men

were systematically rounded up and executed while the remaining popula-
tion of approximately 25,000 Bosnian Muslims — mostly women, chil -
dren and the elderly — were ethnically cleansed from the enclave 3. It is
to be recalled that this isolated finding of genocide was made in spite of the

Applicant Bosnia and Herzegovina’s much broader allegations of genoci▯de,
which included events in the capital city of Sarajevo, as well as acts thhat
occurred at various other municipalities and camps spread across the terh -
ritory of BiH.

42. In view of these considerations, my ensuing remarks shall confine
themselves to the majority’s analysis of genocidal intent regarding thhe
events at Vukovar. While I must reiterate, for the sake of absolute clarity,

that it is not my contention that genocidal intent was established with h
respect to the events occurring on Croatian soil between 1991 and 1995
(including Vukovar), I steadfastly believe that the majority has failed to
fully and properly canvass the events at Vukovar, being as they are the h

cornerstone of Croatia’s case in the instant proceedings, and thus I hintend
to present additional considerations that I believe the majority was remhiss
in failing to consider when determining whether genocide was perpetratedh
against the Vukovar Croats.

The Siege of Vukovar

43. During the oral hearing phase of these proceedings, Croatia cited
uncontested census statistics indicating that in 1991 Vukovar “had a pop-
39
ulation of just over 21,000 Croats [and] 14,500 Serbs” , whereas “[e]ven
after the peaceful reintegration of the region, only 7,500 of the original
21,500 Croat population of Vukovar in 1991 have ever returned to the
city”40. Counsel for Croatia further averred that during the siege of Vuk -

ovar that lasted from August to November 1991, between 1,100 and
1,700 Croats were killed, whereas after the fall of the city and the ensuing
occupation by JNA and Serb forces, an additional 2,000 Croats were
killed41. I recall and share the majority’s conclusion that the Croat popu -

lation in all six geographic regions relied upon by the Applicant constih -
tutes a substantial part of the overall Croat ethnic group within the

38 Bosnia Judgment, p. 155, para. 278, citing Krstić Trial Judgment, para. 1.

39 CR 2014/8, p. 29, para. 7 (Starmer).
40 Ibid., p. 47, para. 85 (Starmer).
41 CR 2014/12, p. 11 (Starmer).

436

7 CIJ1077.indb 868 18/04/16 08:54 436 application of genochide convention (sep. ohp. bhandari)

territory of Croatia. However, I further recall my remarks above that
there is clear precedent for considering a much smaller geographic and

demographic area for the purpose of determining whether that subgroup
constitutes a “substantial” part of the overall group, and hence chonclude
that the Vukovar Croats in and of themselves — in addition to their inclu-
sion in “the substantial” subgroup comprising the six geographic rhegions
as recognized by the Judgment — constituted a substantial part of the

overall ethnic Croat group within the geographical territory of Croatia h
during the relevant period. In this regard, I would recall the Court’hs cha-r
acterization of what constituted a “substantial” part of the targeted group
in question from the following passage of the Bosnia Judgment, in which

certain conclusions of the ICTY Appeals Chamber in the Krstić case were
adopted:

“The Court now turns to the requirement of Article II that there
must be the intent to destroy a protected ‘group’ in whole or in
part . . . the Court recalls the assessment it made earlier in the Judg -
ment of the persuasiveness of the ICTY’s findings of facts and its h

evaluation of them . . . Against that background it turns to the find -
ings in the Krstić case . . . in which the Appeals Chamber endorsed
the findings of the Trial Chamber in the following terms.

‘In this case, having identified the protected group as the natio-
nal group of Bosnian Muslims, the Trial Chamber concluded that
the part . . . targeted was the Bosnian Muslims of Srebrenica, or the
Bosnian Muslims of Eastern Bosnia. This conclusion comports

with the guidelines outlined above. The size of the Bosnian
Muslim population in Srebrenica prior to its capture by the VRS
forces in 1995 amounted to approximately forty thousand people.
This represented not only the Muslim inhabitants of the Srebre -
nica municipality but also many Muslim refugees from the sur -

rounding region. Although this population constituted only a small
percentage of the overall Muslim population of Bosnia and Herze ‑
govina at the time, the importance of the Muslim community of Sre‑
brenica is not captured solely by its size.’

The Court sees no reason to disagree with the concordant findings of
the Trial Chamber and the Appeals Chamber.

The Court concludes that the acts committed at Srebrenica falling
within Article II (a) and (b) of the Convention were committed with
the specific intent to destroy in part the group of the Muslims of Bosni▯a
and Herzegovina as such ; and accordingly that these were acts of
genocide, committed by members of the VRS in and around Srebren -
42
ica from about 13 July 1995.”

42Bosnia Judgment, p. 166, paras. 296-297 (internal citations om; emphasis
added.

437

7 CIJ1077.indb 870 18/04/16 08:54 437 application of genochide convention (sep. ohp. bhandari)

44. Setting aside the different conclusions as to whether genocidal
intent was proven in the Bosnia Judgment versus the present Judgment,
on the issue of how substantiality was assessed, I believe it would have

been entirely appropriate, given, inter alia, the size of the ethnic Croat
population of Vukovar, the Judgment’s recognition that during the siege
and capture of the city the attack was “directed at the then predominhantly
Croat civilian population (many Serbs having fled the city before or hafter
43
the fighting broke out)” — which, to my mind, rendered the city a
de facto ethnic Croat enclave — and finally its emblematic importance to
the ethnic Croat population within Croatia generally (for reasons of mihl-i
tary strategic importance as a key focal point in the expansive strategyh of

“Greater Serbia”, as expounded supra), for the majority to have con -
ducted a specialized analysis of the attack on Vukovar. While I acknowl -
edge that the attack on Vukovar and its aftermath was considered as parth

of an overarching mélange of factors when evaluating whether genocidal
intent existed with respect to the six geographic territories identifihed in
Croatia’s pleadings, in my respectful view such an analysis lacks clahrity
and coherency and would have been improved by an explicit, separate

examination of the events at Vukovar.

45. As I have painstakingly underscored throughout this opinion, it is
my definitive conclusion that Croatia has failed to satisfy the minimuhm
standard of credible evidence required by this Court to allow me to be

“fully convinced” that a finding of genocidal intent vis-à-vis the protected
ethnic Croat group is the only reasonable inference to be drawn from theh
evidentiary record proffered by the Applicant. Indeed, when pressed by a h
Member of the Court during the oral hearing phase, counsel for Croatia

made the critical concession that the number 44 victims it was alleging h
was difficult to ascertain with precision , which I find to epitomize the
many probative shortcomings of the Applicant’s cause of action. This h
position having been reaffirmed, and again following the majority’s elec -

tion to take Croatia’s figures at their highest, I am somewhat puzzhled by
the lack of analysis as to why the averred killing of upwards of 3,000 eth-

43Judgment, para. 218:

“The Court will first consider the allegations concerning those killed during the
siege and capture of Vukovar. The Parties have debated the number of vichtims,
their status and ethnicity and the circumstances in which they died. Theh Court need
not resolve all those issues. It observes that, while there is still somhe uncertainty
surrounding these questions, it is clear that the attack on Vukovar was hnot confined
to military objectives ; it was also directed at the then predominantly Croat civilian
population (many Serbs having fled the city before or after the fightin▯g broke out).”

(Emphasis added.)

44CR 2014/12, pp. 11-12 (Starmer).

438

7 CIJ1077.indb 872 18/04/16 08:54 438 application of genochide convention (sep. ohp. bhandari)

nic Croats in Vukovar out of a pre -war population of 21,500 would not
constitute sufficient physical destruction of the group pursuant to Arthi -
cle II (a) of the Convention to satisfy the “quantitative element” as

adopted by the present Judgment. While there may be good reasons for
such a negative finding, the paucity of analysis conducted by the majohrity
to this end is discouraging.

46. In addition to my misgivings regarding the majority’s application

of the quantitative element regarding the number of Vukovar Croats
allegedly killed during and after the siege of that city, Croatia has prhe -
sented a series of 17 contextual factors which, in its estimation,

“constitute a pattern of conduct from which the only reasonable infe- rh
ence to be drawn is that the Serb leaders were motivated by genocidal
intent . . . [and which], individually or taken together, could lead the

Court to conclude that there was a systematic policy of targeting
Croats with a view to their elimination from the regions concerned”
(Judgment, para. 408).

Consequently, “[a]ll these elements indicate, according to Croatia, the
existence of a pattern of conduct from which the only reasonable infer -
ence is an intent to destroy, in whole or in part, the Croat group” (hibid.,
para. 409). For ease of reference, these factors have been reproduced in

their totality at paragraph 408 of the present Judgment.
47. In the Judgment, the majority has determined

“that of the 17 factors suggested by Croatia to establish the existence
of a pattern of conduct revealing a genocidal intent . . . the most
important are those that concern the scale and allegedly systematic
nature of the attacks, the fact that those attacks are said to have
caused casualties and damage far in excess of what was justified by
military necessity, the specific targeting of Croats and the nature,

extent and degree of the injuries caused to the Croat population (i.e.,h
the third, seventh, eighth, tenth and eleventh factors identified [by h
Croatia])” (ibid., para. 413).

Regrettably, the majority provides no ratio for this critical distinction,
and consequently excludes as “less important”, without any justifihcation,
factors such as “the political doctrine of Serbian expansionism whichh cre -

ated the climate for genocidal policies aimed at destroying the Croat pohp -
ulation living in areas earmarked to become part of ‘Greater Serbia’h
[Croatia’s first factor]”, “the statements of public officiahls, including
demonization of Croats and propaganda on the part of State -controlled
media [Croatia’s second factor]”, “the explicit recognition by the JNA

that paramilitary groups were engaging in genocidal acts [Croatia’s fifth
factor]”; and “the fact that during the occupation, ethnic Croats were
required to identify themselves and their property as such by wearing

439

7 CIJ1077.indb 874 18/04/16 08:54 439 application of genochide convention (sep. ohp. bhandari)

white ribbons tied around their arms and by affixing white cloths to thheir
homes [Croatia’s ninth factor]”, to name a few (see Judgment, parha. 408).

48. While an exhaustive treatment of how these factors may be, con -
trary to the view of the majority, “more important” in decipheringh geno-

cidal intent lies beyond the scope of the present opinion, I must admit I
find myself flummoxed by some of these exclusions. One need only loohk to
readily available historical examples to find scenarios where such fachtors
clearly and unequivocally played a major role in inciting and perpetu -
ating incipient and ongoing genocides. To that end, I would briefly rehcall

the Nazi expansionist political doctrine of Lebensraum (which would
fall neatly under the rubric of Croatia’s first factor) and their ghettho -
ization of marginalized groups through the forced wearing of religiouslyh
denoted attire (e.g., armbands bearing the “Star of David”) for hthe Jews
of occupied Europe (for which one can find many commonalities in Croah -

tia’s ninth factor). To take a more recent historical example, I would note
the undeniable role played by popular media (especially radio) up to ahnd
during the Rwandan genocide in the promotion of a demagogic “Hutu
Power” ideology that sought to vilify and ostracize the Tutsi ethnic h
minority population through the ubiquitous use of the epithet of“inyenzi”

(cockroaches) and other comparable slurs (which aligns with Croatia’hs
second factor). In each of these three examples, the averred acts are nhot,
strictly speaking, genocidal per se in accordance with Article II of the
Genocide Convention, but for the majority to rather summarily dismiss
their potency as precursors to or indicia of genocidal intent is, to my h

mind, both puzzling and troubling. Finally, how “the explicit recognition
of genocidal intent of those carrying out the acts” (Croatia’s fihfth factor)
does not figure prominently into the equation of genocidal intent is shim -
ply beyond me.

The Distinction betweehn Criminal Intent and Mohtive

49. In the present Judgment “the Court notes that in the Mrkšić case,
the ICTY found that the attack on [Vukovar] constituted a response to
the declaration of independence by Croatia, and above all an assertion of
Serbia’s grip on the SFRY” (see para. 429). The Judgment then repro -

duces the following block quotation from Mrkšić in support of this con -
clusion :

“The declaration of Croatia of its independence of the Yugoslav
Federation and the associated social unrest within Croatia was met
with determined military reaction by Serb forces.It was in this pol itical

440

7 CIJ1077.indb 876 18/04/16 08:54 440 application of genochide convention (sep. ohp. bhandari)

scenario that the city and people of Vukovar and those living in close
proximity in the Vukovar municipality became a means of demonstrat ‑
ing to the Croatian people, and those of other Yugoslav Republics, the

harmful consequences of their actions . In the view of the Chamber the
overall effect of the evidence is to demonstrate that the city and civil▯ian
population of and around Vukovar were being punished, and terribly so,
as an example to those who did not accept the Serb ‑ controlled Federal
45
Government in Belgrade.”

As a brief aside, the quoted passage from Mrkšić, which forms part of the
uncontested evidentiary record in this case,is positively laden with explicit
references to the emblematic nature of the Vukovar Croats vis -à-vis the
remainder of the ethnic Croat population, thus only further weakening

the majority’s assertion that Croatia “has provided no informationh” as to
“the prominence of that part of the group” of the ethnic Croat pophula -
tion that it contends was targeted for genocide 46.

50. However, returning to the point under consideration, the majority
relies on the quoted passage from Mrkšić to conclude that

“[i]t follows from the above, and from the fact that numerous Croats h
of Vukovar were evacuated . . . that the existence of intent to physi -
cally destroy the Croat population is not the only reasonable inference h

that can be drawn from the illegal attack on Vukovar” (Judgment,
para. 429).

In my view, this line of reasoning appears to conflate the distinct lehgal
concepts of motive and intent in finding that the “punishment” of the
Vukovar Croats could preclude a finding that they were targeted with

genocidal intent. To this end I would recall the language of the ICTY
Appeals Chamber in the Krnojelac Judgment, which recalled

“its case law in the Jelisićcase which, with regard to the specific intent
required for the crime of genocide, set out ‘the necessity to distinguish
specific intent from motive. The personal motive of the perpetrator of

the crime of genocide may be, for example, to obtain personal eco -
nomic benefits,or political advantage or some form of power. The exist-
ence of a personal motive does not preclude the perpetrator from also
having the specific intent to commit genocide ’.”47

45
46Mrkšić Trial Judgment, para. 471; emphasis added.
Judgment, para. 406. While the Judgment was referring more generally to the ethnic
Croat population in the six geographical areas of Croatia alleged by the Applicant, it
stands to reason that Vukovar, being not only situated within the areas hcontemplated but
constituting the gravamen of Croatia’s case, would constitute at leasht some evidence of the
prominence of at least a part of the targeted group in question.

47Prosecutor v. Krnojelac, Appeals Judgment, 17 September 2003, para. 102; emphasis
added.

441

7 CIJ1077.indb 878 18/04/16 08:54 441 application of genochide convention (sep. ohp. bhandari)

Similar language for this proposition can be found in a number of other h
Judgments pronounced by the ICTY 48and ICTR 49. In view of this dis -

tinction, I find the Judgment’s analysis of the motivation underlyihng the
attack on Vukovar to be problematic, as it fails to account for the posshi-
bility, as clearly stipulated in the aforementioned authorities, that gehno -
cidal intent may exist simultaneously with other, ulterior motives. In this

regard, I would recall the finding in Popović that the massacre at Sre -
brenica enclave was in part motivated by the strategic advantage of uni -
ting a “Greater Serbia”. Never was it suggested that this tacticalh
motivation precluded the attack from possessing genocidal intent. Conse-

quently, I am unpersuaded by the Judgment’s dismissal of genocidal
intent vis -à-vis Vukovar based on the finding that the attack was ani -
mated by political and/or retributive motives, and respectfully but fihrmly

believe that the majority has committed a basic error of law in findinhg
that the existence of a punitive motive for the attack on Vukovar pre -
cludes genocidal intent as “the only reasonable inference that can beh
drawn from the illegal attack” (Judgment, para. 429).

Discretion of the ICTY Proshecutor in Laying a Charghe
of Genocide

51. I recall that in the Bosnia Judgment, the Court determined that :

“[A]s a general proposition the inclusion of charges in an indict -

ment cannot be given [evidentiary] weight. What may however be
significant is the decision of the Prosecutor, either initially or in han
amendment to an indictment, not to include or to exclude a charge
of genocide.” 50

No legal authority whatsoever is cited for the rationale underlying the hdi-s

parate probative weight that the Court decided to afford the ICTY Prose-
cutor’s decision to include or exclude a charge of genocide in an indhictment,

48 See, e.g., Prosecutor v. Blaškić, Appeals Judgment, 29 July 2004, para. 694.

“Mens rea is the mental state or degree of fault which the accused held at the
relevant time. Motive is generally considered as that which causes a perhson to act.
The Appeals Chamber has held that, as far as criminal responsibility is hconcerned,
motive is generally irrelevant in international criminal law . . .”
49
See, e.g., Prosecutor v. Karera, Trial Judgment, 7 December 2007, para. 534. (“The
perpetrator need not be solely motivated by a genocidal intent and having a personal
motive will not preclude such a specific intent.”)

50 Bosnia Judgment, p. 132, para. 217.

442

7 CIJ1077.indb 880 18/04/16 08:54 442 application of genochide convention (sep. ohp. bhandari)

nor does the Bosnia Judgment offer any reasoned explanation for this dis -

tinction. Indeed, in my respectful view such a distinction is unsustainahble as
a matter of basic logical construction, and in contrast to the majority hI find
myself drawn to the poignant submission of counsel for Croatia, who
argued during the oral hearing phase of this case that in accordance withh
the prevailing rules of procedure obtaining at that tribunal,

“[T]he judicial arm of the ICTY will review each indictment, includinhg
the charges that have been included, and has the power to dismiss any
count not supported by the evidence. But the judicial arm has no way
of reviewing the charges that have not been included, or the reasons
for non-inclusion. It would therefore be illogical to afford greater

evidential weight to an unreviewable decision without reasons not to
include a charge, than the reviewable decision to include a charge.” 51

Moreover, I believe that Croatia has raised cogent arguments exposing
the various political, logistical and other constraints that may animateh an

exercise of prosecutorial discretion not to lay a criminal charge, incluhd -
ing : (1) the availability (or lack thereof) of evidence at the onset of pro -
ceedings ; (2) the focus of a criminal prosecution on individual accused,
often in relation to very circumscribed crime sites, rather than the muchh

broader question of State responsibility for genocide encompassing largeh
geographical expanses ; (3)the lack of any obligation falling on the ICTY
Prosecutor to provide reasons for not laying a charge ; (4) the need to
selectively employ the finite resources of that Tribunal, especially ihn view
of the massive institutional constraints imposed by the United Nations

Security Council’s imposition of a “Completion Strategy” mandating the
completion of all the Tribunal’s work by fixed dates ; and (5) the fact that
whereas decisions to include a charge are subject to judicial review, dehci -
sions not to include a charge are not 5.

52. In light of these trenchant insights, and in view of the Court’s pro -
nouncement in Bosnia that the lack of probative value for a decision to
lay a charge of genocide constitutes “a general proposition” rathehr than a

definite rule, in my respectful view the jurisprudence of this Court whould
be fortified by a more expansive treatment of this subject. Alas, givehn the
opportunity to clarify the Court’s position concerning prosecutorial hdis -
cretion in the present Judgment, the majority has apparently elected to h

demur. Instead of a reasoned account that explains the distinction, the
Judgment makes the following pronouncement :

51
52CR 2014/6, p. 39 (Starmer); emphasis in original.
Ibid., pp. 33-42 (Starmer).

443

7 CIJ1077.indb 882 18/04/16 08:54 443 application of genochide convention (sep. ohp. bhandari)

“The fact that the Prosecutor has discretion to bring charges does
not call into question the approach which the Court adopted in its 2007
Judgment . . . The Court did not intend to turn the absence of charges

into decisive proof that there had not been genocide, but took the
view that this factor may be of significance and would be taken into
consideration. In the present case, there is no reason for the Court to
depart from that approach. The persons charged by the Prosecutor

included very senior members of the political and military leadership
of the principal participants in the hostilities which took place in Cro-
atia between 1991 and 1995. The charges brought against them
included, in many cases, allegations about the overall strategy adopted h
by the leadership in question and about the existence of a joint crim -

inal enterprise. In that context, the fact that charges of genocide wereh
not included in any of the indictments is of greater significance thanh
would have been the case had the defendants occupied much lower
positions in the chain of command. In addition, the Court cannot fail
to note that the indictment in the case of the highest ranking defendant▯

of all, former President Milošević, did include charges of genocide in
relation to the conflict in Bosnia and Herzegovina, whereas no such
charges were brought in the part of the indictment concerned with the
hostilities in Croatia.” (Judgment, para. 187 ; emphasis added.)

Not only does this purported defence of the Bosnia distinction skirt the
central issue by failing to provide a single rationale as to why the dechision

to include a charge of genocide in an indictment ought to be given differ -
ential weight than a decision to exclude such a charge, but the example of
the Milošević case relied upon by the Judgment to prove its point in fact
tends to defeat its own position. As that juxtaposition plainly illustrahtes,
if the decision not to charge Milošević with genocide in respect of crimes

committed in respect of Croatia is noteworthy, then surely the same musth
be said of the corollary decision to charge him with genocide in respect of
crimes committed in Bosnia and Herzegovina. To my mind, these are two
sides of the same coin and the draft’s failure to make heads or tailsh of its

quizzical distinction, by invoking a litany of irrelevant considerationsh,
leaves me unmoved.

53. In sum, through its belaboured attempt to justify the distinction
regarding the differential probative value afforded the inclusion or excluh-

sion of charges of genocide in an indictment, which to this day fails toh cite
a single germane legal authority and which poignantly avoids engaging
any of the Applicant Croatia’s arguments, the majority has not, to myh
satisfaction, explained the logically and legally problematic distinctiohn it
first iterated in the Bosnia Judgment and has now reiterated in the present

Judgment. I can only express my regret at this missed opportunity.

444

7 CIJ1077.indb 884 18/04/16 08:54 444 application of genochide convention (sep. ohp. bhandari)

Conclusion

54. For the reasons I have explained at length throughout the course
of this opinion, while I share the majority’s conviction that the Applicant
Croatia has not discharged its evidentiary burden in relation to the sech -
ond operative clause of this Judgment, I have felt compelled to voice myh
many (and at times strenuous) objections to the manner in which the

majority has treated the issue of genocidal intent as regards the claimsh put
forward by Croatia. Given my tepid support for the second operative
clause, which is based primarily on evidentiary concerns, there are manyh
aspects of the reasoning employed by the Judgment en route to the con -

clusion contained in that dispositive paragraph that I would distance
myself from as a jurist. Perhaps most disconcerting is that the foregoinhg
does not constitute an exhaustive exposition of my dissatisfaction with h
the Judgment’s approach to genocidal dolus specialis, but merely a survey
of some of my more salient concerns.

(Signed) Dalveer Bhandari.

445

7 CIJ1077.indb 886 18/04/16 08:54

Bilingual Content

409

SEPARATE OPINION OF JUDGE BHANDARI

Introduction

1. I have voted with the majority on all three operative clauses of the
present Judgment. However, with respect to the second operative clause, h
i.e., the rejection of Croatia’s principal claim, I wish to qualify ahnd expand

upon the rationales for my vote. In so doing, I shall take the present
opportunity to expound upon certain reservations I continue to harbour
regarding the analysis employed at various points throughout that por -
tion of the Judgment with respect to issues which, in my respectful view,

have received inadequate — or even incorrect — attention.
2. At the outset, I wish to underscore that the principal reason for my
rejection of Croatia’s claim is that the Applicant has failed, in my hconsid-
ered opinion and after having carefully scrutinized the entire evidentiahry

record in these proceedings, to satisfy the minimum standard of credible
evidence required by this Court in its prior jurisprudence (in particulhar
the Bosnia Judgment of 2007 1, which dealt with claims of a highly similar
nature) in relation to the dolus specialis of genocide. In this regard, I take
specific note of Croatia’s near complete inability to substantiate hmost of

the figures it has averred in terms of number of victims as a consequehnce
of the hostilities that occurred in the regions and during the period ath
issue. Moreover, I recall that it is a well -settled principle of law that the
graver the offence alleged, the higher the standard of proof required forh

said offence to be established in a court of law. Consequently, I am not h
“fully convinced” (Judgment, para. 178) that the only inference available
from the evidence on record is that attacks against ethnic Croats on the
territory of Croatia between 1991 and 1995 were perpetrated with the req -

uisite genocidal intent. Thus, although I concur with the majority that hthe
actus reus of genocide has been conclusively satisfied with respect to many
of the localities averred by Croatia, the Applicant’s inability to prhove that
the mens rea of genocide — which, by its very nature, constitutes a charge

“of exceptional gravity” (ibid.) — has been “clearly established” (ibid.)
is necessarily fatal to Croatia’s entire cause of action.

1 Application of the Convention on the Prevention and Punishment of the Cr▯ime of
Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, I.CReports

2007 (I), p. 43 (hereafter the “Bosnia Judgment”).

410

7 CIJ1077.indb 816 18/04/16 08:54 409

OPINION INDIVIDUELLE DE M. LE JUGE BHANDARI

[Traduction]

Introduction

1. J’aivoté avec la majorité sur chacun des trois points du dispositif du
présent arrêt. Toutefois, en ce qui concerne le deuxième point — le rejet
de la demande principale de la Croatie —, je souhaite nuancer et explici -
ter les raisons de mon vote. Ce faisant, je saisirai cette occasion d’hexposer
certaines réserves que je continue de nourrir au sujet de l’analyshe à laquelle

il a été procédé en divers points de cette partie de l’arhrêt à propos de que-s
tions qui, à mon humble avis, ont été traitées de manièreh insuffisante,
voire incorrecte.
2. Tout d’abord, je tiens à souligner que la principale raison pour

laquelle j’ai souscrit au rejet de la demande de la Croatie est que lh’Etat
demandeur, après avoir étudié attentivement tous les élémhents de preuve
figurant au dossier, n’a pas satisfait au critère d’établihssement de la preuve
énoncé par la Cour dans sa jurisprudence (en particulier dans l’arrêt
1
de 2007 en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro , qui
mettait en jeu des prétentions très semblables) pour établir lh’élément
moral propre au génocide (dolus specialis). A cet égard, je constate en
particulier que la Croatie a été incapable d’étayer la plupahrt des chiffres
qu’elle avait avancés en ce qui concerne le nombre des victimes dehs hosti -

lités qui ont eu lieu dans les régions et pendant la période enh question. Je
rappelle aussi que, conformément à un principe bien établi en dhroit, plus
l’infraction alléguée est grave, plus le critère d’établissement de la preuve
applicable en justice est élevé. Je ne suis donc pas «pleinement convaincu»

(arrêt, par. 178) que la seule conclusion raisonnable qu’autorisent les élé ‑
ments de preuve figurant au dossier est que les attaques commises contre
des Croates de souche sur le territoire de la Croatie entre 1991 et 1995 ont
été perpétrées dans l’intention génocidaire requise. Ahinsi, même si je
conviens avec la majorité que l’élément matériel (actus reus) du génocide

a été établi de façon concluante pour nombre des localitéhs dont a fait état
la Croatie, l’incapacité de celle-ci à démontrer que l’éhlément moral (mens
rea) de ce crime — qui, par sa nature même, constitue une accusation
«d’une exceptionnelle gravité» (ibid.) — était « clairement avéré» (ibid.)

porte nécessairement un coup fatal à toute sa cause.

1
Application de la convention pour la prévention et la répression d▯u crime ‑e géno
cide (Bosnie‑Herzégovine c. Serbie‑et‑Monténégro), arrêt, C.I.J. Recueil 2007 (I), p. 43
(ci-après arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro).

410

7 CIJ1077.indb 817 18/04/16 08:54 410 application of genochide convention (sep. ohp. bhandari)

3. Indeed, during the oral hearings phase of these proceedings, in responseh

to a question posed by another Member of the Court, Croatia was com -
pelled to concede that many of its written witness statements would haveh
been inadmissible in a domestic Croatian court of law, to which Serbia
responded that such statements would have been likewise inadmissible in hthe

domestic courts of the formerYugoslavia (Judgment, para.195). Moreover,
in response to a question that I posed to the Parties, Croatia maintained
that the Court enjoyed a free hand in determining what weight should be
given to them, based on established Court jurisprudence pertaining to

out-of-court documents (ibid., para. 194). The sum total of these exchanges
is that it stands to reason that a party to proceedings before the Courth can -
not expect to have documents that would be inadmissible before the courths
of its own country, and which bear marked deficiencies when assessed uhsing

the standards applied in this forum, admitted for proof of their contenths;
especiallywhere the matter to be proved is as grave as the crime of genocide.
4. In reaching this conclusion, I share the majority’s sensitivity to “hthe
difficulties of obtaining evidence in the circumstances of th[is] case”h ( ibid.,
para. 198), wherein proof had to be gleaned from a postbellum context

where the juridical infrastructure and other cornerstones of government
and civil society typically relied upon by litigants appearing before thhis
Court have been rendered largely absent or at least severely compromisedh
by years of brutal war, massive displacements of populations and other

seismic socio -political upheavals. Indeed, so Herculean are these obsta -
cles that I must confess to having harboured a fleeting temptation to hrelax
my approach to the methods of proof obtaining before the instant pro -
ceedings, specifically with respect to the documentary evidence adducehd
by Croatia, much of which admittedly lacks the indicia of reliability nohr-

mally demanded of documents presented before a judicial body. How -
ever, the allure of adopting an elastic approach to Croatia’s documenhts
was, to my mind, definitively quelled by the countervailing considerathion
that the crime of genocide, being “an odiousscourge” that is “condemned
3
by the civilized world” , carries with it such grievous moral opprobrium
that a judicial finding as to its existence can only be countenanced uhpon
the most credible and probative evidence. Consequently, despite the
sympathy I have expressed herein regarding the extraordinary evidentiaryh
hurdles faced by the Parties to these proceedings, I ultimately share thhe

majority’s finding “that many of the statements produced by Croahtia are
deficient” (ibid., para. 198), that these deficiencies are irremediable, and
that the remainder of the Applicant’s evidence has failed to conclusively
demonstrate the only conclusion to be drawn from the evidence it has
proffered is that there existed genocidal intent against the targeted group

in question during the time period averred.

2 Convention on the Prevention and Punishment of the Crime of Genocide (“hGenocide
Convention”), Preamble.
3
Ibid.

411

7 CIJ1077.indb 818 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 410

3. De fait, lors des audiences, en réponse à une question posée pahr un

autre membre de la Cour, la Croatie a été contrainte de reconnaîhtre que
nombre des déclarations de témoins qu’elle avait produites aurahient été
irrecevables devant la juridiction interne, ce à quoi la Serbie a réhpondu
que de telles déclarations auraient été également irrecevablhes devant les
juridictions nationales de l’ex -Yougoslavie (arrêt, par. 195). Par ailleurs,

en réponse à une question que j’ai posée aux Parties, la Crohatie a soutenu
que la Cour était libre de déterminer quel poids il fallait leur dhonner, en
fonction de sa jurisprudence établie concernant les documents extrajuhdi -
ciaires (ibid., par. 194). Dans ces conditions, il va de soi qu’une partie à

une procédure engagée devant la Cour ne peut pas s’attendre àh ce que des
documents qui seraient irrecevables devant ses propres tribunaux et qui h
présentent des déficiences marquées au regard des normes applhiquées en
l’espèce soient acceptés comme faisant foi de leur contenu, en particulier
lorsqu’ils se rapportent à un crime aussi grave que le génocideh.

4. En parvenant à cette conclusion, je partage la conscience qu’a la
majorité «des difficultés que pose l’obtention des preuves dans les circonsh -
tances de l’espèce » (ibid., par. 198), étant donné que ces éléments de
preuve ont dû être recueillis dans un contexte d’après guerrhe où l’in -
frastructure judiciaire et d’autres composantes fondamentales de l’hEtat et

de la société civile sur lesquelles s’appuient généralemehnt les justiciables
qui comparaissent devant la présente Cour avaient pour une grande parht
disparu ou, à tout le moins, avaient été gravement endommagéhes par des
années de guerre brutale, des déplacements massifs de populations het

d’autres profonds bouleversements sociopolitiques. Si colossaux sont ces
obstacles, d’ailleurs, que je dois avouer avoir eu la tentation fugache d’as-
souplir la conception que j’avais avant la présente affaire des rèhgles d’ad-
ministration de la preuve, en particulier en ce qui concerne les preuvesh
documentaires produites par la Croatie, dont une grande partie est, il ehst

vrai, dépourvue des marques de fiabilité normalement exigées hdes docu -
ments présentés en justice. Cependant, toute velléité que j’hai pu avoir à cet
égard a été définitivement chassée de mon esprit par l’hidée que le crime de
génocide, parce qu’il constitue « un fléau … odieux »2 qui est « condamné
par le monde civilisé »3, emporte un si terrible opprobre moral que le

constat judiciaire de son existence doit impérativement reposer sur lhes él-é
ments les plus crédibles et probants. En conséquence, tout en prenhant acte
des extraordinaires difficultés qu’ont eues les Parties à réhunir des preuves,
je partage en dernière analyse la conclusion de la majorité, qui ah jugé
«que nombre des déclarations produites par la Croatie sont déficihentes »

(ibid., par. 198), que les vices dont elles sont entachées sont irrémédiablhes
et que le reste des preuves apportées par le demandeur ne suffit pash à
démontrer de façon concluante que la seule conclusion raisonnable qu’au ‑
torisent les éléments de preuve qu’il a produits est qu’il existait une inten -
tion génocidaire à l’égard du groupe visé à l’éphoque considérée.

2
Convention pour la prévention et la répression du crime de génohcide (c«onvention sur
le3génocide»), préambule.
Ibid.

411

7 CIJ1077.indb 819 18/04/16 08:54 411 application of genochide convention (sep. ohp. bhandari)

5. This premise having been established, I take note of the fact that in

spite of the serious evidentiary deficiencies in Croatia’s case, thhe majority
has elected to assess whether the claims of the Applicant, taken at their
highest, could nevertheless evince genocidal intent . Following this lead,

and notwithstanding my conclusion that Croatia’s charges of genocide
have failed on evidentiary grounds, I intend to profit from the presenht
opportunity in making certain observations and critiques to the analysish

adopted by the majority on the issue of dolus specialis, assuming, argu ‑
endo (as the majority has done), that Croatia’s case may be taken at iths
highest.

6. In brief, it is my respectful view that the Court should have used the
present Judgment to lay down clearer guidelines on three principal issues.
First, I believe the Court could have provided a better and clearer treaht -

ment as to what constitutes genocidal intent. Second, given the proliferha-
tion of international criminal tribunals over the past two decades and thhe
consequent exponential expansion of jurisprudence emanating from these

juridical bodies, I believe the majority has been derelict in not more fhully
canvassing the available authorities to provide clear parameters to disthin-
guish between genocide and the oft closely intertwined offences of extermh - i

nation and/or persecution as a crime against humanity. Third and finalhly,
I believe that the 17 factors advanced by Croatia in support of its conten-
tion that genocide occurred deserved a more comprehensive response than

the majority’s approach of selecting, without any apparent reasoned ehxp-la
nation, five factors deemed “most important” to Croatia’s clahim of geno-
cidal intent (Judgment, para. 413). I believe that a superior treatment of

4 See Judgment, para. 437:

“The Court considers that it is also relevant to compare the size of hthe targeted part
of the protected group with the number of Croat victims, in order to dethermine whether
the JNA and Serb forces availed themselves of opportunities to destroy thhat part of the
group. In this connection, Croatia put forward a figure of 12,500Croat deaths, which
is contested by Serbia. The Court notes that, even assuming that this figure it correc—

an issue on which it will make no ruling — the number of victims alleged by Croatia is
small in relation to the size of the targeted part of the group.” (Ehmphasis added.)
See also, ibid., para. 213:

“Croatia first asserts that, between the end of August and 18November 1991,
Vukovar was besieged and subjected to sustained and indiscriminate shellhing, laying
waste to the city. It alleges that between 1,100 and 1,700 people, 70 per cent of whom
were civilians, were killed during that period.”

See also, ibid., para. 218:

“The Court will first consider the allegations concerning those kilhled during
the siege and capture of Vukovar. The Parties have debated the number of victims,
their status and ethnicity and the circumstances in which they died. The▯ Court need
not resolve all those issues. It observes that, while there is still some uncertainty
surrounding these questions, it is clear that the attack on Vukovar was hnot confined
to military objectives it was also directed at the then predominantly Croat civilian
population (many Serbs having fled the city before or after the fighhting broke out).”
(Emphasis added.)

412

7 CIJ1077.indb 820 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 411

5. Ce postulat étant posé, je relève que, malgré les graves déhficiences que
présentaient les éléments de preuve produits par la Croatie, lah majorité a
choisi d’examiner la question de savoir si les allégations du demahndeur, à les
4
supposer avérées, pouvaient néanmoins attester l’intention ghénocidaire . Dans
le même esprit, bien que j’en sois venu à la conclusion que lesh accusations de
génocide de la Croatie n’avaient pas été établies, j’ehntends saisir l’occasion

qui m’est donnée pour formuler certaines observations et critiquesh concernant
l’analyse à laquelle la majorité a procédé au sujet de lah question du dolus
specialis, en partant du principe, pour les besoins de la cause (comme l’a fahit

la majorité), que l’on peut accorder plein crédit aux moyens dhe la Croatie.
6. En bref, j’estime que la Cour aurait dû mettre à profit le présent arrêt
pour énoncer des orientations plus claires sur trois questions princihpales.

Tout d’abord, je crois qu’elle aurait pu traiter mieux et plus clairement la
question de savoir ce qui constitue l’intention génocidaire. Ensuihte, compte
tenu de la multiplication des juridictions pénales internationales auh cours

des deux dernières décennies et, consécutivement, du dévelophpement expo -
nentiel de la jurisprudence en émanant, je crois que la majorité ah manqué
à l’obligation qu’elle avait d’analyser pleinement les sourches à sa disposi-

tion afin d’en tirer des critères clairs permettant de distinguehr le génocide
des infractions souvent étroitement liées que sont l’exterminathion et les
persécutions en tant que crimes contre l’humanité. Enfin, je crois que les

17 critères qu’a invoqués la Croatie pour établir qu’un géhnocide avait eu
lieu méritaient une réponse plus complète que celle de la majorhité qui, sans
explication apparente, a sélectionné cinq critères jugés «hplus importants»

au regard de l’intention génocidaire alléguée par la Croatieh (arrêt, pa 4r1.3).

4
Voir arrêt, par. 437:
«La Cour estime qu’il est également pertinent de comparer la tailleh de la partie visée

du groupe protégé avec le nombre de victimes croates afin de déhterminer si la JNA et
des forces serbes ont saisi les opportunités qui s’offraient à ehlles de détruire ladite partie
du groupe.A cet égard, la Croatie a avancé le chiffre de 12500 morts croates, ce qui est
contesté par la Serbie. La Cour note que, même à supposer que ce chiffre soit correct,
point sur lequel elle ne se prononce pas, le nombre de victimes alléguées par la Croatie est
peu élevé par rapport à la taille de la partie visée du group»e.(Les italiques sont de moi.)

Voir aussi ibid., par. 213:

« La Croatie affirme en premier lieu que, de la fin août au 18 novembre 1991,
Vukovar a été assiégée et bombardée de façon continue het indiscriminée, ce qui a
réduit la ville à l’état de ruines. Elle soutient que de 1100 personnes , dont 70%
étaient des civils, ont été tuées durant cette phase. »

Voir aussi ibid., par. 218:
«La Cour considérera d’abord les allégations concernant les pershonnes tuées au

cours du siège et de la prise de Vukov. es Parties ont débattu des questions du nombre
de ces victimes, de leur statut et ethnicité, ainsi que des c▯ ces dans lesquelles
elles sont mortes. La Cour n’a pas à trancher toutes ces questionElle constate que,
s’il subsiste certaines incertitudes sur celles-ci, il est indéniahble que l’attaque contre
Vukovar ne s’est pas limitée à des objectifs militaires, mais ah aussi visé la population
civile, composée alors en bonne partie de Croates (de nombreux Serbehs ayant fui la
ville avant ou lorsque les combats ont éclaté).» (Les italiques sont de moi.)

412

7 CIJ1077.indb 821 18/04/16 08:54 412 application of genochide convention (sep. ohp. bhandari)

these topics would have been commensurate with the Court’s function ahs
not only the principal judicial organ of the UnitedNations but as a “World

Court” from which other international and domestic courts and tribunahls
seek guidance as a legal authority of the highest order.

Genocidal Intent and theh “Substantiality” Critherion

7. For ease of reference, I reproduce the relevant sections of the Geno -
cide Convention in which the substantive provisions of the crime of genoh -
cide are enshrined :

“Article 1: The Contracting Parties confirm that genocide, whether
committed in time of peace or in time of war, is a crime under inter -

national law which they undertake to prevent and to punish.
Article 2 : In the present Convention, genocide means any of the
following acts committed with intent to destroy, in whole or in part,
a national, ethnical, racial or religious group, as such :

(a) Killing members of the group ;
(b) Causing serious bodily or mental harm to members of the group;

(c) Deliberately inflicting on the group conditions of life calculated
to bring about its physical destruction in whole or in part ;
(d) Imposing measures intended to prevent births within the group ;
(e) Forcibly transferring children of the group to another group.”

As the foregoing text illustrates, the chapeau of Article II of the Geno -
cide Convention defines genocide as “any of the following acts commhitted
with intent to destroy, in whole or in part, a national, ethnical, racial or

religious group, as such” (emphasis added). The fact that the Convehntion
expressly envisages situations where a group may be targeted for destruch-
tion “in part” naturally gives rise to the thorny question of whenh exactly
the targeted “part” meets the threshold for genocidal intent. Becahuse the
Convention is silent on this point, in the Bosnia Judgment the Court

relied upon the jurisprudence of the International Criminal Tribunal for
the former Yugoslavia (ICTY) and the International Criminal Tribunal
for Rwanda (ICTR), as well as International Law Commission (ILC)
Commentary, to conclude that “part” of the “group” for the phurpose of

Article II requires5an intent “to destroy at least a substantial part of the
particular group” .

5 Bosnia Judgment, p. 126, para. 198; emphasis added.

413

7 CIJ1077.indb 822 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 412

Je crois qu’un examen plus approfondi de ces points s’imposait eu égard à
la fonction qu’exerce la Cour en sa qualité d’organe judiciaireh principal

des Nations Unies, mais aussi de « Cour mondiale» chargée de montrer la
voie aux autres juridictions, tant nationales qu’internationales.

L’intention génocidahire et le critère du « caractère substantiehl »

7. Par souci de commodité, je reproduis ci -dessous les articles de la
convention sur le génocide qui énoncent les dispositions de fond rhelatives
à ce crime :

«Article premier: Les Parties contractantes confirment que le géno -
cide, qu’il soit commis en temps de paix ou en temps de guerre, est uhn

crime du droit des gens, qu’elles s’engagent à prévenir et àh punir.
Article 2: Dans la présente Convention, le génocide s’entend de
l’un quelconque des actes ci -après, commis dans l’intention de
détruire, ou tout ou en partie, un groupe national, ethnique, racial
ou religieux, comme tel :

a) Meurtre de membres du groupe ;
b) Atteinte grave à l’intégrité physique ou mentale de membres hdu
groupe;

c) Soumission intentionnelle du groupe à des conditions d’existence
devant entraîner sa destruction physique totale ou partielle ;
d) Mesures visant à entraver les naissances au sein du groupe ;
e) Transfert forcé d’enfants du groupe à un autre groupe. »

Comme on peut le voir, le passage introductif de l’article II de la
convention sur le génocide énonce que le génocide s’entend dhe « l’un quel-
conque des actes ci -après, commis dans l’intention de détruire, ou tout ou

en partie, un groupe national, ethnique, racial ou religieux, comme tel »
(les italiques sont de moi). Le fait que la Convention prévoie exprhessé -
ment des situations où un groupe peut être pris pour cible en vue hde sa
destruction « en partie » soulève naturellement l’épineuse question de
savoir où se situe exactement, s’agissant de la taille de la « partie» visée,

le seuil permettant d’établir l’intention génocidaire. Parceh que la Conven-
tion est muette sur ce point, la Cour s’est appuyée, dans l’arrhêt rendu en
l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, sur la jurisprudence
du Tribunal pénal international pour l’ex -Yougoslavie (TPIY) et du Tri -

bunal pénal international pour le Rwanda (TPIR), ainsi que sur le chom -
mentaire de la Commission du droit international (CDI), pour conclure h
que, appliqué au mot « groupe», à l’article II, l’expression « en partie »
supposait l’intention « de détruire au moins une partie substantielle du
groupe en question » 5.

5 Arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, p. 126, par. 198
(les italiques sont de moi).

413

7 CIJ1077.indb 823 18/04/16 08:54 413 application of genochide convention (sep. ohp. bhandari)

8. While the “substantiality” criterion enunciated in the Bosnia Judg -
ment has been reaffirmed in the instant Judgment (in somewhat modifihed
form, a subject to which I intend to return in short order), this has bheen

done rather tersely and in a way that, in my view, fails to lay down clehar
parameters that would provide guidance to future adjudicative bodies
grappling with this concept. The majority, has also, I fear, neglected to so
much as consider possibly relevant jurisprudential developments emanat -
ing from the ad hoc international criminal tribunals in the intervening

eight years since the issuance of the Bosnia Judgment. Therefore, in the
hopes of elucidating this standard for the sake of posterity, I intend to
revisit the Bosnia formula to see how that test has been applied in practice
by other tribunals in recent years, so as to juxtapose such developmentsh
with how the majority has employed said formula in the instant Judg -

ment.

The Legal Test Enunciatehd
in the Court’s B osnia Judgment of 2007

9. As has been correctly observed in the present Judgment, in the Bosnia

Judgment of 2007, the Court “considered certain issues similar to thohse
before it in the present case” (Judgment, para. 125). On that occasion, the
Court expounded the relevant test for determining what constitutes a “hpart”
of the targeted group for the purpose of analysing genocidal intent as fol -
lows :

“[T]he Court refers to three matters relevant to the determination of
‘part’ of the ‘group’ for the purposes of Article II [of the Genocide

Convention]. In the first place, the intent must be to destroy at least
a substantial part of the particular group. That is demanded by the
very nature of the crime of genocide : since the object and purpose of
the Convention as a whole is to prevent the intentional destruction of
groups, the part targeted must be significant enough to have an impacth
on the group as a whole. That requirement of substantiality is sup ‑

ported by consistent rulings of the ICTY and the International Criminal ▯
Tribunal for Rwanda (ICTR) and by the Commentary of the ILC to
its Articles in the draft Code of Crimes against the Peace and Security h
of Mankind.
Second , the Court observes that it is widely accepted that genocide

may be found to have been committed where the intent is to destroy
the group within a geographically limited area . . . As the ICTY
Appeals Chamber has said . . . the opportunity available to the perpe-
trators is significant. This criterion of opportunity must however be
weighed against the first and essential factor of substantiality . It may

414

7 CIJ1077.indb 824 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 413

8. Si le critère du « caractère substantiel » énoncé en l’affaire Bosnie‑
Herzégovine c. Serbie‑et‑Monténégro a été réaffirmé dans l’arrêt en l’es -

pèce (non sans avoir subi quelques modifications, point sur lequelh j’ai
l’intention de revenir sous peu), il l’a été assez laconiquhement et d’une
manière qui, à mon avis, n’établit pas de paramètres claihrs susceptibles de
donner des orientations aux organes juridictionnels qui se retrouveront h
aux prises avec cette notion dans l’avenir. La majorité s’est aussi, je le

crains, abstenue même de prendre en considération, indépendammehnt de
sa pertinence éventuelle, l’évolution qu’a connue la jurisprhudence éma -
nant des tribunaux pénaux internationaux ad hoc au cours des huit années
écoulées depuis la publication de l’arrêt rendu en l’affaihre en l’affaire

Bosnie‑Herzégovine c. Serbie‑et‑Monténégro. En conséquence, dans l’es -
poir d’élucider cette norme dans l’intérêt de la postéhrité, j’ai l’intention de
revenir sur le critère retenu dans l’arrêt rendu dans cette affahire pour voir
comment il a été appliqué dans la pratique par d’autres tribhunaux au
cours des dernières années, afin de mettre en regard cette évholution et le

raisonnement tenu sur ce point par la majorité en l’espèce.

Le critère énoncé dansh l’arrêt rendu par la Chour en 2007
en l’affaire B osnie‑H eRzéGovine c . seRBie‑et ‑monténéGRo

9. Comme il a été correctement observé dans le présent arrêth, la Cour,
dans son arrêt de 2007 « a examiné … certaines questions comparables à
celles dont elle est saisie en l’espèc» (arrêt, par.125). A cette occasion,
elle avait exposé le critère permettant de déterminer ce qui cohnstitue une

« partie» du groupe visé aux fins d’établir l’intention génohcidaire:

« [L]a Cour relève trois points importants s’agissant de déterminer la
«partie» du «groupe» aux fins de l’article II [de la convention sur le
génocide]. En premier lieu, l’intention doit être de détruire au moins une
partie substantielle du groupe en question. C’est ce qu’exige la nature

même du crime de génocide : l’objet et le but de la Convention dans son
ensemble étant de prévenir la destruction intentionnelle de groupehs, la
partie visée doit être suffisamment importante pour que sa disparhition
ait des effets sur le groupe tout entier. Cette condition relative au carac
tère substantiel de la partie du groupe est corroborée par la j▯ prudence

constante du TPIY et du Tribunal pénal international pour le Rwanda
(TPIR), ainsi par la CDI dans son commentaire des articles du projet
de code des crimes contre la paix et la sécurité de l’humanitéh…
Deuxièmement, la Cour relève qu’il est largement admis qu’il peut
être conclu au génocide lorsque l’intention est de détruire hle groupe

au sein d’une zone géographique précise… Comme la chambre d’hap -
pel du TPIY l’a dit … les possibilités qui s’offrent aux criminels
constituent un élément important. Ce critère des possibilités doit tou ‑
tefois être apprécié au regard du premier facteur, essentiel, à▯ savoir

414

7 CIJ1077.indb 825 18/04/16 08:54 414 application of genochide convention (sep. ohp. bhandari)

be that the opportunity available to the alleged perpetrator is so lim -

ited that the substantiality criterion is not met. The Court observes
that the ICTY Trial Chamber has indicated the need for caution, lest
this approach might distort the definition of genocide [. . .]

A third suggested criterion is qualitative rather than quantitative.
The Appeals Chamber in the Krstić case [noted that]

‘. . . In addition to the numeric size of the targeted portion, its
prominence within the group can be a useful consideration. If a
specific part of the group is emblematic of the overall group, or is

essential to its survival, that may support a finding that the part
qualifies as substantial . . .’
Establishing the ‘group’ requirement will not always depend on the

substantiality requirement alone although it is an essential startin‑gpoint.
It follows in the Court’s opinion that the qualitative approach canno▯t
stand alone. The Appeals Chamber in Krstić also expresses that
view.” 6

The Court concluded its remarks by noting that “[t]he above list of
criteria is not exhaustive, but, as just indicated, the substantiality criterion

is critical. They are essentially those stated by the Appeals Chamber in 7
the Krstić case, although the Court does give this first criterion priority.”
Thus, in the Bosnia case the Court fastened a tripartite formula, which it
indicated was open to future expansion and elaboration, for determining h

whether a “part” of a group has been targeted with genocidal intenht ;
according to which the criterion of “substantiality” was pre -eminent in
that calculus.

10. While the Bosnia formula did not draw any bright lines around the
contours of what constitutes genocidal intent toward “a part” of thhe tar-
geted group, it would appear plain from that Judgment and the jurispru -
dence of the international criminal tribunals that a “substantial”h part of

the targeted group need not constitute the majority thereof, and that there
is no numeric threshold for discerning a substantial part of the group.

The Legal Test Enunciatehd in the Present Judgmenht

11. The pertinent analysis of the law on genocidal intent vis -à-vis “a
part” of the targeted group is presented in the present Judgment as fhollows :

“The Court recalls that the destruction of the group ‘in part’ within
the meaning of Article II of the Convention must be assessed by refer ‑

6 Bosnia Judgment, pp. 126-127, paras. 198-200 (internal citations omitted ; emphasis
added).
7
Ibid., p. 127, para. 201; emphasis added.

415

7 CIJ1077.indb 826 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 414

celui du caractère substantiel. Il se peut que les possibilités s’offrant

au criminel allégué soient si limitées qu’il ne soit pas sathisfait à ce
critère. La Cour relève que la chambre de première instance du hTPIY
a d’ailleurs souligné la nécessité de faire montre de prudenhce pour
éviter que cette approche ne dénature la définition du génhocide…
Un troisième critère proposé est d’ordre qualitatif et non quantita -

tif. Dans l’affaire Krstić, la chambre d’appel du TPIY [a noté que]

«… Il peut être utile de tenir compte non seulement de l’impor -
tance numérique de la fraction du groupe visée, mais aussi de sa
place au sein du groupe tout entier. Si une portion donnée du
groupe est représentative de l’ensemble du groupe, ou essentielle à
sa survie, on peut en conclure qu’elle est substantielle… »

Pour établir s’il est satisfait à la condition relative au «roupe», le
critère du caractère substantiel ne suffit pas toujours, bien qu’▯il soit un
point de départ essentiel. Il s’ensuit, de l’avis de la Cour, q▯ue l’approche

qualitative n’est pas suffisante. La chambre d’appel dans l’affaire
Krstić a exprimé la même idée. »6

La Cour a conclu en faisant observer que « [l]a liste de critères donnée
ci-dessus n’est pas limitative, mais, comme il vient d’être indiqué, le critère
du caractère substantiel est déterminant. Ce sont essentiellement les cri -
tères que la chambre d’appel a exposés dans l’affaire Krstić, bien que la
7
Cour donne priorité au premier. » En conséquence, en l’affaire Bosnie‑
Herzégovine c.Serbie‑et‑Monténégro , la Cour a établi, pour déterminer si
telle «partie» d’un groupe avait été visée dans une intention génochidaire,
une formule tripartite dont elle a indiqué qu’elle était susceptible d’être

enrichie et développée à l’avenir, et dans laquelle elle a dhonné préémi -
nence au « caractère substantiel».
10. Sila formule énoncée dans l’arrêt rendu en l’affaire Bosnie‑Herzégo‑
vine c.Serbie‑et‑Monténégro n’a pas délimité avec précision ce qui constitue
une intention génocidaire à l’égard d’« une partie» du groupe visé, il ressort

clairement de cet arrêt et de la jurisprudence des tribunaux pénaux interna -
tionaux qu’une telle partie peut être «substantielle» sans pour autant const- i
tuer la majorité du groupe et qu’il n’y a pas de seuil numérique à cet égard.

Le critère énoncé dansh le présent arrêt

11. L’analyse du droit en ce qui concerne l’intention génocidaire àh l’égard
d’«une partie» du groupe visé est ainsi exposée dans le présent arrêt:

«La Cour rappelle que la destruction «en partie» du groupe au sens

de l’articleII de la Convention doit être appréciée en fonction de plu ‑

6
Arrêt rendu en l’affairBosnie ‑Herzégovine c. Serbie‑et‑Monténégro, p. 126-127,
pa7. 198-200 (renvois internes omis ; les italiques sont de moi).
Ibid., p. 127, par. 201 (les italiques sont de moi).

415

7 CIJ1077.indb 827 18/04/16 08:54 415 application of genochide convention (sep. ohp. bhandari)

ence to a number of criteria . In this regard, it held in 2007 that ‘the
intent must be to destroy at least a substantial part of the particular h
group’ [. . .], and that this is a ‘critical’ criterion. The Court further

noted that ‘it is widely accepted that genocide may be found to have h
been committed where the intent is to destroy the group within a
geographically limited area’ and that, accordingly, ‘[t]he area ofh the
perpetrator’s activity and control are to be considered [. . .]’. Account
must also be taken of the prominence of the allegedly targeted part

within the group as a whole. With respect to this criterion, the Appealsh
Chamber of the ICTY specified in its Judgment rendered in the Krstić
case that ‘[i]f a specific part of the group is emblematic of the overall
group, or is essential to its survival, that may support a finding thaht
the part qualifies as substantial . . .’.

In 2007, the Court held that these factors would have to be assessed
in any particular case. [. . .] It follows that, in evaluating whether the
allegedly targeted part of a protected group is substantial in relation to
the overall group, the Court will take into account the quantitative
element as well as evidence regarding the geographic location and

prominence of the allegedly targeted part of the group.” (Judgment,
para. 142 (internal citations omitted ; emphasis added).)

What I find immediately striking from this slightly rebranded iteratiohn
of the tripartite test promulgated by the Court in the Bosnia Judgment is
that, one fleeting reference to the “critical” nature of the “substantiality”
criterion (now renamed “the quantitative element”) notwithstandihng, the
rigidly hierarchical structure of the Bosnia test, whereby the numerosity

of the targeted population was clearly superordinate to the other, supplhe -
mentary criteria of “opportunity” (now dubbed “the geographic hloca -
tion”) and the “qualitative factor” (now dubbed the “promihnence” of the
targeted group) has been jettisoned in favour of a more equal balancingh
effort. My distinct impression that the stratification inherent in the hBosnia

formula has been mollified by the present Judgment (a jurisprudentialh
evolution I applaud) draws further support from the consistently flexhible
and egalitarian manner in which the Court has applied these three factors
to the facts at bar, wherein I cannot discern any noticeable supremacy
afforded the quantitative element (see, generally, Judgment, para -
graphs 413-441).

12. As I shall undertake to demonstrate at a later juncture in this opin-
ion, I believe that this adapted substantiality test has practical conseh -

quences for the manner in which the majority has applied the assessment h
of genocidal intent in the present Judgment, specifically with respect to
the events occurring in the city of Vukovar and its environs.

416

7 CIJ1077.indb 828 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 415

sieurs critères. A cet égard, elle a estimé en 2007 que « l’intention doit
être de détruire au moins une partie substantielle du groupe », et qu’il
s’agit d’un critère «déterminant». Elle a également relevé « qu’il est

largement admis qu’il peut être conclu au génocide, lorsque l’hinten -
tion est de détruire le groupe au sein d’une zone géographique hpré -
cise» et que, par conséquent, « [l]a zone dans laquelle l’auteur du
crime exerce son activité et son contrôle doit être prise en cohnsidéra-
tion». Il convient également de prendre en compte la place de la par -

tie du groupe qui serait visée au sein du groupe tout entier. En ce qhui
concerne ce critère, la chambre d’appel du TPIY a précisé dahns l’ar-
rêt rendu en l’affaire Krstić que, « [s]i une portion donnée du groupe
est représentative de l’ensemble du groupe, ou essentielle à sah survie,
on peut en conclure qu’elle est substantielle…

La Cour, en 2007, a estimé qu’il revient au juge d’apprécier ces élé▯ ‑
ments dans chaque espèce. Il en découle que, afin de décider si la partie
qui serait visée était substantielle par rapport à l’ensemble du groupe
protégé, la Cour tiendra compte de l’élément quantitatif ainsi que de la▯
localisation géographique et de la place occupée par cette partie au

sein du groupe. » (Arrêt, par. 142 (renvois internes omis ; les italiques
sont de moi).)

Ce qui me semble frappant de prime abord dans cette itération légèhre -
ment modifiée du critère tripartite établi par la Cour dans lh’arrêt rendu en
l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, c’est que, en dépit
d’une brève référence à la qualité « déterminante» du critère du « carac-
tère substantiel» (renommé « élément quantitatif »), la structure hiérar-

chique rigide du critère adopté en l’affaire Bosnie‑Herzégovine
c. Serbie‑et‑Monténégro, où l’importance numérique de la population
visée l’emportait sans équivoque sur le critère complémenhtaire des « pos-
sibilités (désormais appelé «la localisation géographique») et le «facteur
qualitatif» (devenu la « place occupée » par le groupe visé), a été aban -

donnée au profit d’une démarche mettant plus ou moins ces diffhérends sur
un pied d’égalité. La manière souple et non hiérarchiséhe avec laquelle la
Cour a systématiquement appliqué ces trois facteurs aux faits de l’espèce,
sans accorder, à mon sens, de suprématie notable à l’éléhment quantitatif
(voir, d’une manière générale, arrêt, par. 413-441) esth venue conforter ma
nette impression que la stratification inhérente à la formule éhnoncée dans

l’arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro s’est
effritée à la faveur du présent arrêt (évolution jurispruhdentielle à laquelle
j’applaudis).
12. Comme je vais entreprendre de le démontrer plus loin, je crois que
cette adaptation du critère du caractère substantiel a des conséhquences

pratiques pour la manière dont la majorité a apprécié l’ihntention génoci -
daire dans le présent arrêt, notamment en ce qui concerne les éhvénements
survenus dans la ville de Vukovar et ses environs.

416

7 CIJ1077.indb 829 18/04/16 08:54 416 application of genochide convention (sep. ohp. bhandari)

Post- Bosnia Jurisprudence
of the ICTY and ICTR

13. As noted above, in the present Judgment the Court has recalled

and reaffirmed the tripartite formula for genocidal intent enunciated ihn
Bosnia as the approach to be followed in the present case, though not
without a significant restructuring of the normative order of the testh to be
employed. This is naturally consonant with the principle that while no
8
prior judgment of this Court constitutes binding precedent sensu stricto ,
“[i]n general the Court does not choose to depart from previous finhdings,
particularly when similar issues were dealt with in the earlier decisionhs. . .
unless it finds very particular reasons to do so” (Judgment, para.h125). As

I have noted above, in Bosnia the Court explicitly acknowledged the con -
tributions of the ICTY and ICTR in shaping the test that it adopted to
assess genocidal intent vis -à-vis a “part” of a targeted group 9. Conse-
quently — and bearing in mind the nearly eight years that have passed

since the promulgation of this Court’s Bosnia formula — it would seem to
me only natural and appropriate to examine whether the jurisprudence of h
those tribunals in the intervening years reveals any evolution in how the
“substantiality” component of genocidal dolus specialis has been applied

in recent litigious contexts. Such an endeavour is not only consonant with
the Bosnia Judgment’s pronouncement that the criteria enunciated therein
were “not exhaustive” 10and therefore presumably subject to future eluci -
dation, but is also faithful to the present Judgment’s self -admonition that

the Court will “take account, where appropriate, of the decisions of hinte-r
national criminal courts or tribunals, in particular those of the ICTY, has
itdid in 2007, in examining the constituent elements of genocide in the

present case” (ibid., para. 129). In a similar vein, I recall the draft Judg -
ment’s avowal that while it will rely on the Bosnia Judgment “to the
extent necessary for its legal reasoning[, t]his will not . . . preclude it,
where necessary, from elaborating upon this jurisprudence” (ibid.,

para. 125).

14. In my respectful view, because the legal standard for genocidal
intent has a necessarily vague and dynamic character, it was incumbent

upon the Court to fully canvass recent developments in the law to deter -
mine how the Bosnia formula (as restated in the present Judgment) has
been applied in other juridical institutions tasked with applying that test.

8 ICJ Statute, Article 59: “The decision of the Court has no binding force except
between the parties and in respect of that particular case.”
9 Bosnia Judgment, p. 126, para. 198.
10 Ibid., p. 127, para. 201.

417

7 CIJ1077.indb 830 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 416

La jurisprudence du TPIY eht du TPIR postérieure à l’harrêt rendu
en l’affaire Bosnie ‑H eRzéGovine c . eRBie ‑et‑m onténéGRo

13. Comme je l’ai mentionné plus haut, la Cour a, dans le présent ahrrêt,

rappelé et réaffirmé la formule tripartite énoncée dans hl’arrêt rendu en l’a -f
faire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro pour établir l’intention
génocidaire, indiquant qu’il s’agissait de la démarche à adopter en l’espèce,
non toutefois sans modifier sensiblement l’ordre normatif de la formule à

employer. Cela est naturellement conforme au principe selon lequel, si nul
arrêt antérieur de la Cour ne constitue au sens strict 8un précédent contrai-
gnant, « [e]n règle générale — à moins qu’elle n’estime que des raisons très
particulières doivent la conduire à le faire —, la Cour ne s’écarte … pas de

sa jurisprudence, notamment lorsque des questions comparables à cellehs
qui se posent à elle … ont été examinées dans des décisions antérieures »
(arrêt, par. 125). Comme je l’ai dit plus haut, dans l’arrêt rendu en l’haffaire
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, la Cour a expressément

reconnu les contributions du TPIY et du TPIR à l’élaboration deh la for -
mule qu’elle a adoptée pour apprécier l’intention génocidhaire à l’égard
d’une « partie» d’un groupe visé . Par conséquent et compte tenu de ce
que près de huit années se sont écoulées depuis, il me semble naturel et

opportun d’examiner si la jurisprudence émanée de ces tribunauxh au cours
des années qui ont suivi fait apparaître une évolution dans la hfaçon dont a
été appliqué le critère du « caractère substantiel» dans le contexte de l’élé -
ment moral propre au génocide (dolus specialis) dans les affaires récentes.

Cette démarche est non pas seulement conforme à l’arrêt rendhu en l’affaire
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, où il est dit que la liste des
critères énoncés n’est «pas limitative» 10et est appelée à évoluer, mais elle

est aussi fidèle au conseil que la Cour s’adresse dans le préhsent arrêt lors-
qu’elle écrit qu’elle «rendra en considération, le cas échéant, les décisions
des tribunaux pénaux internationaux, en particulier celles du TPIY,
comme elle l’a fait en 2007, lorsqu’elle examinera en l’espèhce les éléments

constitutifs du génocide » (arrêt, par. 129). Dans la même veine, je rappelle
que la Cour a reconnu dans son projet d’arrêt que, si elle entendahit s’ap -
puyer sur l’arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑
Monténégro «dans la mesure nécessaire au raisonnement juridique[, c]ela

ne l’empêchera[it] cependant pas de compléter cette jurisprudenhce, en tant
que de besoin» ( ibid., par. 125).
14. A mon sens, parce que la norme régissant l’intention génocidairhe a
un caractère nécessairement fluide et dynamique, il incombait àh la Cour

d’examiner de façon approfondie l’évolution récente du drhoit afin de
déterminer comment la formule énoncée dans l’arrêt rendu hen l’affaire
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro (et réaffirmée dans le présent

8 Statut de la Cour internationale de Justice, article 59: « La décision de la Cour n’est
obligatoire que pour les parties en litige et dans le cas qui a étéh décidé.»
9 Arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, p. 126, par. 198.
10 Ibid., p. 127, par. 201.

417

7 CIJ1077.indb 831 18/04/16 08:54 417 application of genochide convention (sep. ohp. bhandari)

I regret to say that in my estimation the present Judgment has neither
fully nor properly canvassed the current jurisprudential standard of genho -
cidal intent emanating from the ICTY and the ICTR. For this reason, I
shall now conduct a survey of recent trends in the case law of those trihbu -

nals on this subject in an attempt to glean insights as to the present shtate
of the law in this area. As I shall expound hereunder, I take the positihon
that these recent jurisprudential trends would tend to suggest that a pat -
tern of killings such as has been averred in relation to the events thath
11
occurred in Croatia between 1991 and 1995 , and in particular with
respect to the region of Eastern Slavonia and the greater Vukovar area, h
may be more indicative of genocidal intent than the majority has acknowl-
edged.

The t olimiR ICTY Trial
Chamber Judgment

15. On 12 December 2012, the Trial Chamber of the ICTY issued its
judgment in the case of Tolimir (currently under appeal), in which it pro -
vided a comprehensive treatment of the substantiality criterion of geno -

cidal intent. The Trial Chamber recalled that

“[t]he term ‘in whole or in part’, relates to the requirement thhat the
perpetrator intended to destroy at least a substantial part of a pro -
tected group. While there is no numeric threshold of victims required,
the targeted portion must comprise ‘a significant enough [portion] hto

have an impact on the group as a whole’. Although the numerosity of
the targeted portion in absolute terms is relevant to substantiality, this
is not dispositive other relevant factors include the numerosity of the
targeted portion in relation to the group as a whole, the prominence

of the targeted portion and whether the targeted portion of the group
is emblematic of the overall group, or is essential to its survival, as h
well as the area of the perpetrators’ activity, control and reach.”h 12

These observations made repeated reference to the same section of the
analysis contained in the Krstić Judgment of the ICTY Appeals Chamber
that was relied upon by this Court when it adopted its tripartite test fhor
genocidal intent in the Bosnia Judgment.

11For the avoidance of any doubt, I recall that while I have found that Crhoatia’s claim
fails on evidentiary grounds, I am taking the present opportunity, as thhe majority has done,
to assess Croatia’s case taken at its highest.

12Prosecutor v. Tolimir, Trial Judgment, 12 December 2012, para. 749 (internal cita-
tions omitted; emphasis added).

418

7 CIJ1077.indb 832 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 417

arrêt) a été appliquée par les autres juridictions compéhtentes. Je regrette

de dire que, selon moi, la Cour n’a dans le présent arrêt ni plheinement ni
dûment examiné la norme jurisprudentielle actuelle émanant du ThPIY et
du TPIR en ce qui concerne l’intention génocidaire. C’est pourquoi je
vais à présent passer en revue les tendances récentes de la jurhisprudence

de ces tribunaux à ce sujet pour tenter de recueillir des indicationsh de
l’état actuel du droit dans ce domaine. Comme je l’exposerai cih -dessous,
j’estime que ces tendances jurisprudentielles récentes donnent àh penser
qu’un schéma de meurtres tel que celui établi en ce qui concernhe les évé -
11
nements survenus en Croatie entre 1991 et 1995 , en particulier s’agis-
sant des régions de Slavonie orientale et de Vukovar, peut être plus
révélateur d’une intention génocidaire que la majorité neh l’a reconnu.

Le jugement rendu en l’haffaire t olimiR
par le TPIY

15. Le 12 décembre 2012, le TPIY a rendu son jugement en l’affaire
Tolimir (actuellement en appel), dans lequel il a livré une analyse exhaush-
tive du critère du caractère substantiel se rapportant à l’intention génoci -
daire. Elle a rappelé que

«[l]’expression en tout ou en partie » se rapporte à la condition vou-
lant que l’auteur ait eu l’intention de détruire au moins une phartie

substantielle du groupe protégé. Même si aucun nombre minimal de
victimes n’est requis, la partie du groupe visée doit être «suffisamment
importante pour que l’ensemble du groupe soit affecté ». Même si le
nombre de personnes visées, considéré dans l’absolu, est pert▯ inent pour

déterminer si la partie du groupe est substantielle, il n’est pas détermi‑
nant; d’autres éléments sont à prendre en compte, notamment l’himp-or
tance numérique de la fraction visée par rapport à la taille duh groupe
dans son ensemble, sa place au sein du groupe, la question de savoir si

elle est « représentative de l’ensemble du groupe, ou essentielle à sa
survie», et la zone dans laquelle les auteurs du crime exercent leur
activité et leur contrôle, ainsi que leur pouvoir d’action.»12

Ces observations comportent de multiples références au passage corres -
pondant du raisonnement exposé dans l’arrêt rendu en l’affaire Krstić par
le TPIY, sur lequel la Cour s’est appuyée quand elle a énoncéh, dans l’arrêt

rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, sa formule
tripartite permettant d’établir l’intention génocidaire.

11 Pour éviter tout malentendu, je rappelle que, même si j’en suish venu à la conclusion
que les prétentions de la Croatie devaient échouer faute de preuve, je saisis la présente
occasion, comme la majorité l’a fait, pour examiner les moyens de hla Croatie en supposant
qu’ils soient avérés.
12 Le Procureur c. Tolimir, jugement du 12 décembre 2012, par. 749 (renvois internes

omis; les italiques sont de moi).

418

7 CIJ1077.indb 833 18/04/16 08:54 418 application of genochide convention (sep. ohp. bhandari)

16. After summarizing these widely accepted elements of the law on
genocidal intent, the Trial Chamber in Tolimir further recalled a passage
from an earlier judgment of the ICTY Trial Chamber in the case of
13
Jelisić , w14ch was cited approvingly in a passage of the Krstić Appeals
Judgment that was referenced favourably by the Court in the Bosnia
Judgment 15. As the Tolimir Trial Chamber recalled :

“The Jelisić Trial Chamber held that as well as consisting of the
desire to exterminate a very large number of members of the group,

genocidal intent may also consist of the desired destruction of a more
limited number of persons selected for the impact that their disappear ‑
ance would have on the survival of the group as such.” 16

The Trial Chamber then made the following further observations about

the Jelisić Trial Judgment:
“The Jelisić Trial Chamber cited the Final Report of the Commis -

sion of Experts formed pursuant to Security Council Resolution 780
which found

‘[i]f essentially the total leadership of a group is targeted, it couhld
also amount to genocide. Such leadership includes political and
administrative leaders, religious leaders, academics and intellec -
tuals, business leaders and others — the totality per se may be a

strong indication of genocide regardless of the actual numbers
killed. A corroborating argument will be the fate of the rest of the
group. The character of the attack on the leadership must be

viewed in the context of the fate or what happened to the rest of the
group. If a group has its leadership exterminated, and at the same
time or in the wake of that, has a relatively large number of the
members of the group killed or subjected to other heinous acts, for

example deported on a large scale or forced to flee, the cluster of
violations ought to be considered in its entirety in order to inter -
pret the provisions of the Convention in a spirit consistent with its
purpose.’. . .

The Commission of Experts Report stated, further, that

‘[s]imilarly, the extermination of a group’s law enforcement and
military personnel may be a significant section of a group in that
it renders the group at large defenceless against other abuses of a
similar or other nature, particularly if the leadership is being eli -

minated as well. Thus the intent to destroy the fabric of a society
through the extermination of its leadership, when accompanied by

13Prosecutor v. Jelisić, Trial Judgment, 14 December 1999, para. 82.
14Prosecutor v. Krstić, Appeals Judgment, 19 April 2004, para. 8 and fn. 10.
15Bosnia Judgment, p. 126, para. 198.
16Prosecutor v. Tolimir, Trial Judgment, 1December 2012, para. 749; citing Pros‑
ecutor v. Jelisić, Trial Judgment, 14 December 1999, para. 82.

419

7 CIJ1077.indb 834 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 418

16. Après avoir résumé ces éléments largement acceptés du hdroit relatif
à l’intention génocidaire, la chambre de première instance ah en outre rap-
pelé en l’affaire Tolimir un passage du jugement antérieur prononcé par le
TPIY en l’affaire Jelisić13, repris dans l’arrêt Krstić 14, et auquel la Cour

a renvoyé en y souscrivant également dans l’arrêt rendu en l’affaire
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro 15:

« La Chambre de première instance saisie de l’affaire Jelisić a
conclu que l’intention génocidaire peut consister à vouloir l’hextermi-
nation d’un nombre très élevé de membres du groupe, et elle hpeut

aussi consister à rechercher la destruction d’un nombre plus limité de
personnes, celles‑ci étant sélectionnées en raison de l’impa▯ct qu’aurait
leur disparition sur la survie du groupe comme tel. » 16

La chambre de première instance a ensuite fait les observations complhé-
mentaires suivantes à propos du jugement prononcé en l’affaire Jelisić :

« La Chambre de première instance saisie de l’affaire Jelisić a cité
le Rapport final de la commission d’experts constituée en applichation

de la résolution 780 du Conseil de sécurité, dans lequel il était conclu
«[qu’i]l peut aussi y avoir génocide lorsque tous les dirigeants
d’un groupe sont visés. Ceux -ci peuvent être des chefs politiques

et administratifs, des chefs religieux, des universitaires et des intel -
lectuels, des industriels, etc.; que tous soient visés en tant que tels
fait naître une forte présomption de génocide, quel que soit le
nombre de ceux qui sont effectivement tués. On peut en outre tirer

argument du sort réservé au reste du groupe. Les attaques contre
les dirigeants doivent être évaluées dans le contexte de ce qui
advient au reste du groupe. Si les dirigeants d’un groupe sont exter ‑
minés et si, en même temps ou peu après, un nombre relativement▯

élevé de membres du groupe sont tués ou soumis à d’autres▯ atrocités,
par exemple expulsés en masse ou forcés de fuir , il faut envisager
les diverses violations dans leur ensemble afin d’interpréter lehs di-s
positions de la Convention dans un esprit conforme à son but… »

[L]e rapport [de la commission d’experts] conclut également

«[que, d]e même, l’extermination du personnel chargé de l’apphlic-a
tion des lois et du personnel militaire peut toucher une importante
section du groupe en ce sens qu’elle met l’ensemble du groupe hors

d’état de se défendre contre des sévices de même ou d’hautre nature,
en particulier si les chefs sont également éliminés. Par conséhquent,
l’intention de détruire le tissu d’une société en en extehrminant les

13Le Procureur c. Jelisić, jugement du 14 décembre 1999, par. 82.
14Le Procureur c. Krstić, arrêt du 19 avril 2004, par. 8, et note de bas de page 10.
15Arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, p. 126, par. 198.
16Le Procureur c. Tolimir, jugement, 12 décembre 2012, par. 749, citant Le Procureur
c. Jelisić, jugement du 14 décembre 1999, par. 82.

419

7 CIJ1077.indb 835 18/04/16 08:54 419 application of genochide convention (sep. ohp. bhandari)

other acts of elimination of a segment of society, can also bedeemed
17
genocide.’”
17. The Trial Chamber then proceeded to apply this more flexible con -

cept of substantiality to the factual circumstances of that case, which h
involved, inter alia, the killing of three prominent members of the Bos -
nian Muslim population of Z epa enclave in Eastern Bosnia and Herze -
govina (“BiH”). As the Trial Chamber recalled, Z epa was a village

situated approximately 20 kilometres from Srebrenica that had a popula -
tion of less than 3,000 inhabitants prior to the war, but which saw its
population swell to as many as 10,000 people by July 1995, as Bosnian
Muslims from other surrounding areas in Eastern BiH sought refuge
from the prevailing hostilities, such that “[d]uring the conflict the popula -
18
tion of Zepa consisted entirely of Bosnian Muslims” .
18. Regarding the three individuals killed, the Trial Chamber made the
following observations:

“The three leaders were Mehmed Hajrić, the Mayor of the muni -
cipality and President of the War Presidency, Colonel Avdo Palić,
Commander of the ABiH Zepa Brigade . . . and Amir Imamović, the

Head of the Civil Protection Unit. They were, therefore, among the
most prominent leaders of the enclave . . . [T]hose responsible for
killing Hajrić, Palić and Imamović targeted them because they were
leading figures in the Zepa enclave at the time that it was populated

by Bosnian Muslims. These killings should not be viewed in isola ‑
tion . . . it is significant to consider the connection between the VRS
operations in Srebrenica and Zepa. The respective attacks and take -
overs of the enclaves were synchronized by the [same] leadership and
included the same forces. The takeover of Zepa enclave followed less

than two weeks after the capture of Srebrenica, during a time in which
the news of the murders of thousands of Bosnian Muslim men was
starting to spread. While the individuals killed were only three in num ‑
ber, in view of the size of Zepa, they constituted the core of its civilian

and military leadership. The mayor, who was also a religious leader,
the military commander and the head of the Civil Protection Unit,
especially during a period of conflict, were key to the survival of a h
small community. Moreover, the killing of Palić, who at this time
enjoyed a special status as the defender of the Bosnian Muslim pop -

ulation of Zepa, had a symbolic purpose for the survival of the Bos -
nian Muslims of Eastern BiH. While the majority accepts that the
Bosnian Serb Forces did not kill the entirety of the Bosnian Muslim
leadership of Z epa . . it does not consider this to be a factor against

17
Prosecutor v. Tolimir, Trial Judgment, 12 December 2012, fn. 3138; emphasis added;
citing Final Report of the Commission of Experts Established pursuant toh Security Council
Resolution 780 (1992), UN doc. S/1994/674, (“Commission of Experts Report”), para. 94.
18
Prosecutor v. Tolimir, Trial Judgment, 12 December 2012, paras. 598-599.

420

7 CIJ1077.indb 836 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 419

chefs peut aussi, si elle s’accompagne d’autres actes visant à éliminer
17
un secteur de la société, être considérée comme un génocide. »»

17. La chambre de première instance a ensuite appliqué cette forme
plus souple du critère du caractère substantiel aux circonstances hfactuelles
de l’affaire, où était en cause, entre autres, le meurtre de trois membres
éminents de la population musulmane de Bosnie dans l’enclave de Zepa

en Bosnie orientale. Comme elle l’a rappelé, le village de Zepa était situé
à une vingtaine de kilomètres de Srebrenica et comptait moins de
3000 habitants avant la guerre, mais sa population avait augmenté pour
atteindre 10 000 personnes en juillet 1995, après que des Musulmans

d’autres régions de Bosnie orientale fuyant les hostilités y avhaient cherché
refuge, de sorte que, « [p]endant le conflit, la population de Zepa était
entièrement composée de Musulmans de Bosnie » 1.
18. En ce qui concerne les trois personnes tuées, la chambre de pre -
mière instance a fait les observations suivantes :

«Les trois dirigeants étaient Mehmed Hajrić, maire de la municipa-

lité et président de la présidence de guerre, le colonel Avdo Palić, com-
mandant de la brigade de Zepa de l’ABiH … et Amir Imamović, chef
de l’unité de la protection civile. Ils figuraient donc parmi lehs princi-
paux dirigeants de l’enclave… [L]es personnes responsables du meurhtre

de Hajrić, de Palić et d’Imamović ont pris ces derniers pourh cibles
parce qu’ils étaient des personnalités de premier plan dans l’henclave de
Zepa, à l’époque où celle -ci était peuplée de Musulmans de Bosnie.
Ces meurtres ne devraient pas être considérés isolément … il importe
d’examiner le lien entre les opérations que la VRS a menées àh Srebre -

nica et à Zepa. Les attaques et la prise des enclaves ont été synchroni-
sées par les [mêmes] dirigeants de la VRS et les mêmes forces yh ont pris
part. La prise de l’enclave de Zepa a eu lieu moins de deux semaines
après la prise de Srebrenica, au moment où la nouvelle du meurtre hde

milliers d’hommes musulmans de Bosnie commençait à se propager.h
Bien que les faits ne concernent que trois personnes, celles‑ci étaien▯ t,
compte tenu de la taille de Zepa, les piliers de la direction civile et mili
taire. Le maire — qui était aussi un chef religieux —, le commandant

militaire et le chef de l’unité de la protection civile, en particulier en
période de conflit, étaient essentiels à la survie d’une petite commu -
nauté. En outre, le meurtre de Palić, qui à cette époque jouhissait d’un
statut spécial en tant que défenseur de la population musulmane deh

Bosnie de Zepa, avait une portée symbolique pour la survie des Musu-l
mans de Bosnie orientale. Même si la majorité admet que les forcesh
serbes de Bosnie n’ont pas tué tous les dirigeants musulmans de

17Le Procureur c. Tolimir, jugement, 12 décembre 2012, note de bas de page 3138 (les
italiques sont de moi); citant le rapport final de la commission d’experts constituée chonfo-r
mément à la résolution 780 (1992) du Conseil de sécurité, Nations Unies, doc. S/1994/674
(«Rapport de la Commission d’experts), par. 94.
18
Le Procureur c. Tolimir, jugement du 12 décembre 2012, par. 598-599.

420

7 CIJ1077.indb 837 18/04/16 08:54 420 application of genochide convention (sep. ohp. bhandari)

its determination that the acts of murder against these three men con ‑
stitutes genocide.” 19

The Trial Chamber then proceeded to reach the following further conclu-
sions :

“In accordance with the Jelisić Trial Chamber’s finding in which it
relied on the Commission of Experts Report the Majority also takes
into account the fate of the remaining population of Zepa ; their for‑

cible transfer immediately prior to the killing of these three leaders i▯s a
factor which supports its finding of genocidal intent. To ensure that th▯e
Bosnian Muslim population of this enclave would not be able to recon ‑
stitute itself, it was sufficient inhe case of Zepa to remove its civilian

population, destroy their homes and their mosque, and murder its most
prominent leaders . . . The Majority has no doubt that the murder of
[these three leaders] was a case of deliberate destruction of a limited h
number of persons selected for the impact that their disappearance
20
would have on the survival of the group as such.”

19. I acknowledge that these conclusions — which were subject to a
dissenting opinion and are currently awaiting a judgment from the ICTY
Appeals Chamber — must be treated with a requisite degree of caution.

Such limitations having been duly conceded, in my view the passages
cited above from the Tolimir Trial Judgment nevertheless evince a con -
certed departure from the narrower ambit of the tripartite test adopted hby
this Court in the Bosnia Judgment. Given that the present Judgment has

likewise determined to apply the Bosnia formula in a more flexible man -
ner that places less emphasis on the primacy of the quantitative elementh,
I am both surprised and disheartened by the majority’s refusal to makhe
any mention of the most recent judicial pronouncement of the ICTY on

this highly pertinent and substantively fluid area of law.
20. Specifically, the Tolimir Judgment’s finding of genocide where only
three killings were proven marks a clear and unambiguous departure
from the Bosnia formula’s dogged insistence that the numerosity of the

victims of predicate acts under Article II of the Genocide Convention be
considered a pre -eminent factor in the substantiality equation. Rather,
Tolimir presents a rather striking example of a case where not only were
the three individuals killed low in absolute terms, but against the back -

drop of a homogeneous religious community of approximately 10,000 it
is dubious to suggest that their deaths could constitute a high relative
“numerosity in relation to the group as a whole” 2. Rather, in find-
ing genocidal intent, the Tolimir Trial Chamber placed heavy empha-

sis on the prominence of the targeted population and the fact that the

19Prosecutor v. Tolimir, Judgment, 12 December 2012, paras. 778-780; emphasis
added.
20Ibid., paras. 781-782; emphasis added.
21Ibid., para. 749.

421

7 CIJ1077.indb 838 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 420

Zepa … elle n’estime pas que cela aille à l’encontre de sa conclusion
19
selon laquelle le meurtre de ces trois hommes constitue un génocide.»

La chambre de première instance est ensuite parvenue aux conclusions h
suivantes:

« En accord avec la conclusion exposée dans le jugement Jelisić —
qui s’appuyait sur le rapport de la commission d’experts —, la majo-
rité tient aussi compte du sort du reste de la population de Z epa; son
transfert forcé juste avant le meurtre des trois dirigeants vient con▯for ‑

ter la conclusion relativement à l’intention génocidaire. Pour ▯s’assurer
que la population musulmane de cette enclave ne puisse pas se recons ‑
tituer, il suffisait dans le cas de Zepa — de chasser les civils, de
détruire leurs maisons et leur mosquée et de tuer les principaux

dirigeants de l’enclave… Il ne fait aucun doute pour la majorité
que le meurtre de [ces trois dirigeants] était un exemple de destruc -
tion délibérée d’un nombre limité de personnes sélectionnéhes en rai-
son de l’impact qu’aurait leur disparition sur la survie du groupe
20
comme tel.»
19. Je reconnais que ces conclusions, qui ont donné lieu à une opinionh

dissidente et qui sont actuellement frappées d’appel au TPIY, doivent être
considérées avec la prudence voulue. Cette réserve étant dûhment concé -
dée, j’estime que les passages précités du jugement Tolimir attestent néan -
moins une prise de distance concertée avec la portée étroite deh la formule

tripartite énoncée par la Cour dans l’arrêt qu’elle a renhdu en l’affaire
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro. Etant donné que, dans le
présent arrêt, la Cour a décidé de donner plus de souplesse hà cette for -
mule en insistant moins sur la primauté de l’élément quantithatif, je suis à

la fois surpris et déçu que la majorité ait refusé de faire hla moindre men -
tion de la plus récente décision rendue par le TPIY dans ce domainhe du
droit hautement pertinent, malgré sa mouvance.
20. En particulier, le jugement Tolimir, où le TPIY a conclu au géno -

cide alors que trois meurtres seulement avaient été prouvés, mahrque une
rupture claire et catégorique avec la formule énoncée en l’ahffaire
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, qui exigeait que l’impor -
tance numérique des victimes d’actes incriminés à l’artichle II de la
convention sur le génocide soit considérée comme un facteur préhéminent

dans l’évaluation du « caractère substantiel». Dans le jugement Tolimir, le
nombre de personnes tuées (trois) était faible non seulement danhs l’absolu,
mais aussi au regard d’une communauté religieuse homogène d’henvi -
ron 10 000 personnes. Il aurait été hasardeux d’avancer que ces morts

puissent avoir une « importance numérique » élevée, ne serait-ce que de
manière relative, «par rapport à la taille du groupe dans son ensemble» 21.

19 Le Procureur c. Tolimir, jugement, 12 décembre 2012, par. 778-780 (les italiques sont
de moi).
20 Ibid., par. 781-782 (les italiques sont de moi).
21
Ibid., par. 749.

421

7 CIJ1077.indb 839 18/04/16 08:54 421 application of genochide convention (sep. ohp. bhandari)

attackers exercised complete control over the enclave during the period hin
question.

21. Finally, what cannot be overlooked is that apart from three kill -
ings, the gravamen of the atrocities perpetrated at Zepa constituted the
complete forcible transfer of its entire Bosnian Muslim population, a
community of thousands, away from that enclave and into Bosnian -
controlled territory. While Tolimir certainly did not go so far as to pro -

nounce that the “ethnic cleansing” of these thousands of Bosnian Muslims
from Z epa enclave (in conjunction with means taken to ensure their
non-return, such as destruction of homes and places of worship) consti -
tuted genocide per se, it did clearly and unequivocally affirm that this
mass displacement of the civilian population, when combined with the

very limited targeted killing of prominent local leaders, constituted anh
attempt to physically destroy a significant part of the Bosnian Muslim
group of Eastern BiH, by depriving that community of the means of recon ‑
stituting itself within that geographical area. On this final point, it would
appear to be a clear evolution of the position adopted by this Court in hthe

Bosnia Judgment as to what constitutes “physical destruction” of the
group for the purpose of Article II of the Genocide Convention, where it
was held that

“[i]t will be convenient at this point to consider what legal significanc▯e
the expression [‘ethnic cleansing’] may have [under the Genocide Con ‑
vention]. It is in practice used, by reference to a specific region or area,
to mean ‘rendering an area ethnically homogeneous by using force or

intimidation to remove persons of given groups from the area’. . .
It does not appear in the Genocide Convention ; indeed, a proposal
during the drafting of the Convention to include in the definition
‘measures intended to oblige members of a group to abandon their
homes in order to escape the threat of subsequent ill -treatment’

was not accepted . . . It can only be a form of genocide within the
meaning of the Convention, if it corresponds to or falls within one of
the categories of acts prohibited by Article II of the Convention. Nei ‑
ther the intent, as a matter of policy, to render an area ‘ethnically
homogeneous’, nor the operations that may be carried out to implement▯
such policy, can as such be designated as genocide . . . As the ICTY

has observed, while

‘there are obvious similarities between a genocidal policy
and the policy commonly known as ‘ethnic cleansing’. . . [a]
clear distinction must be drawn between physical destruction

and mere dissolution of a group. The expulsion of a group

422

7 CIJ1077.indb 840 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 421

Au contraire, en concluant à l’intention génocidaire, la chambrhe de
première instance saisie de l’affaire Tolimir a mis l’accent sur la
population visée et sur l’emprise totale exercée sur l’enclave.
21. Enfin, l’on ne saurait négliger, en dehors de ces trois meurtresh,

l’élément essentiel des atrocités commises à Zepa : le transfert forcé de
toute la population musulmane de Bosnie, soit plusieurs milliers de
personnes, loin de cette enclave et en territoire contrôlé par la hBosnie.
Alors que, dans le jugement Tolimir, le tribunal n’est pas allé jusqu’à dire
que le « nettoyage ethnique » dont ont été la cible ces milliers de

Musulmans de Bosnie chassés de l’enclave de Zepa (associé aux
mesures pour garantir leur non -retour, comme la destruction des habita -
tions et des lieux de culte) était un génocide en soi, il affirme qu’un tel
déplacement de la population civile, en plus du meurtre ciblé de dirigeants
locaux de premier plan, si peu nombreux soient-ils, constituait une tenta-
tive de détruire physiquement une grande partie des Musulmans de Bosnie

de Bosnie orientale, en privant cette communauté des moyens de se recons‑
tituer dans cette zone géographique. Sur ce dernier point, il existe une évo-
lution claire par rapport à la position de la Cour dans l’arrêth
Bosnie‑Herzégovine c. Serbie‑et‑Monténégro au sujet de la « destruction
physique » du groupe pour l’application de l’article II de la convention

sur le génocide ; la Cour s’était exprimée ainsi :
«Il serait utile à ce stade de se pencher sur la portée juridique q▯ue
peut revêtir l’expression «nettoyage ethnique» [au regard de la conven ‑

tion sur le génocide]. Dans la pratique, elle est employée, à propos
d’une région ou d’une zone particulière, avec le sens de «hrendre une
zone ethniquement homogène en utilisant la force ou l’intimidationh
pour faire disparaître de la zone en question des personnes apparte -
nant à des groupes déterminés »… Elle n’apparaît pasdans la conven-

tion sur le génocide ; de fait, lors de la rédaction de la Convention,
une proposition visant à inclure dans la définition les « mesures ten-
dant à mettre les populations dans l’obligation d’abandonner lehurs
foyers afin d’échapper à la menace de mauvais traitements ulthé -
rieurs » fut rejetée… De telles mesures ne sauraient constituer une
forme de génocide au sens de la Convention que si elles correspondent

à l’une des catégories d’actes prohibés par l’article II de la Convention
ou relèvent de l’une de ces catégories. Ni l’intention, sous▯ forme d’une
politique visant à rendre une zone « ethniquement homogène », ni les
opérations qui pourraient être menées pour mettre en œuvre pareille
politique ne peuvent, en tant que telles, être désignées par le terme de

génocide. Ainsi que l’a fait observer le TPIY, si
«[i]l y a donc d’évidentes similitudes entre une politique génochi -
daire et ce qui est communément appelé une politique de « net-

toyage ethnique »… il n’en reste pas moins qu’« [i]l faut faire
clairement le départ entre la destruction physique et la simple dis -

422

7 CIJ1077.indb 841 18/04/16 08:54 422 application of genochide convention (sep. ohp. bhandari)

or part of a group does not in itself suffice for geno -
cide.’” 22

In the present Judgment, this relationship has been revisited in the folhlow -
ing terms :

“The Court recalls that, in its 2007 Judgment, it stated that

‘[n]either the intent, as a matter of policy, to render an area ‘ehthn-i
cally homogeneous’, nor the operations that may be carried out to
implement such policy, can as such be designated as genocide : the
intent that characterizes genocide is ‘to destroy, in whole or in

part’ a particular group, and deportation or displacement of the
members of a group, even if effected by force, is not necessarily equi ‑
valent to destruction of that group, nor is such destruction an auto ‑
matic consequence of the displacement. [. . .]’

It explained, however, that :

‘[t]his is not to say that acts described as ‘ethnic cleansing’h may
never constitute genocide, if they are such as to be characterized

as, for example, ‘deliberately inflicting on the group conditions ohf
life calculated to bring about its physical destruction in whole or
in part’, contrary to Article II, paragraph (c), of the Convention,
provided such action is carried out with the necessary specific
intent (dolus specialis), that is to say with a view to the destruc -

tion of the group, as distinct from its removal from the region . . .
In other words, whether a particular operation described as ‘eth -
nic cleansing’ amounts to genocide depends on the presence or
absence of acts listed in Article II of the Genocide Convention,
and of the intent to destroy the group as such. In fact, in the

context of the Convention, the term ‘ethnic cleansing’ has no legahl
significance of its own. That said, it is clear that acts of ‘ethnihc
cleansing’ may occur in parallel to acts prohibited by Article II of
the Convention, and may be significant as indicative of the pre -
sence of a specific intent (dolus specialis) inspiring those acts.’”

(Judgment, para. 162 (internal citations omitted ; emphasis
added).)

22. In my respectful view, the ICTY Trial Chamber in Tolimir has
burst open the tight confines of the dictum promulgated in Bosnia and

reaffirmed in the present Judgment. By finding that the confluence hof kil-l
ing three prominent community leaders (which constitute genocidal acts h
as per Article II (a) of the Convention) in parallel to massive acts of

22Bosnia Judgment, pp. 122-123, para. 190 (internal citations omitte; emphasis
added).

423

7 CIJ1077.indb 842 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 422

solution d’un groupe. L’expulsion d’un groupe ou d’une partihe
d’un groupe ne saurait à elle seule constituer un génocide ».» 22

Dans le présent arrêt, cette distinction a été revue dans lehs termes sui -
vants :

«La Cour rappelle que, dans son arrêt de 2007, elle a affirmé que

« [n]i l’intention, sous forme d’une politique visant à rendre unhe
zone « ethniquement homogène », ni les opérations qui pourraient
être menées pour mettre en œuvre pareille politique ne peuvent,h en
tant que telles, être désignées par le terme de génocide : l’intention

qui caractérise le génocide vise à « détruire, en tout ou en partie »
un groupe particulier ; ladéportation ou le déplacement de membres
appartenant à un groupe, même par la force, n’équivaut pas n▯écessai ‑
rement à la destruction dudit groupe, et une telle destruction ne
résulte pas non plus automatiquement du déplacement forcé .»

La Cour a néanmoins ajouté la précision suivante :

«Cela ne signifie pas que les actes qui sont décrits comme étant h
du « nettoyage ethnique » ne sauraient jamais constituer un géno -

cide, s’ils sont tels qu’ils peuvent être qualifiés, par ehxemple, de
«[s]oumission intentionnelle du groupe à des conditions d’exis -
tence devant entraîner sa destruction physique totale ou par -
tielle, en violation du litt. c) de l’article II de la Convention, sous
réserve que pareille action soit menée avec l’intention spéchifique

(dolus specialis) nécessaire, c’est-à-dire avec l’intention de détruire
le groupe, et non pas seulement de l’expulser de la région… En
d’autres termes, savoir si une opération particulière présenhtée
comme relevant du « nettoyage ethnique » équivaut ou non à un
génocide dépend de l’existence ou non des actes matériels éhnumé -

rés à l’articleII de la convention sur le génocide et de l’intention
de détruire le groupe comme tel. En réalité, dans le contexte dhe
cette Convention, l’expression « nettoyage ethnique n» revêt, par
elle-même, aucune portée juridique. Cela étant, il est clair que desh
actes de « nettoyage ethnique » peuvent se produire en même

temps que des actes prohibés par l’article II de la Convention, et
permettre de déceler l’existence d’une intention spécifiquhe (dolus
specialis) se trouvant à l’origine des actes en question. »» (Arrêt,
par. 162 (renvois internes omis ; les italiques sont de moi).)

22. A mon avis, la chambre de première instance du TPIY saisie de
l’affaire Tolimir a fait éclater les limites étroites de la règle établie danhs

l’arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro et
réaffirmée dans le présent arrêt. En constatant que le meurhtre de trois
dirigeants importants de la communauté (qui constitue un acte de géhno -

22 Arrêt rendu en l’affairBosnie‑Herzégovine c. Serbie‑et‑Monténégro, p. 122-123,
par. 190 (renvois internes omis ; les italiques sont de moi).

423

7 CIJ1077.indb 843 18/04/16 08:54 423 application of genochide convention (sep. ohp. bhandari)

ethnic cleansing (which are non-genocidal atrocities per se; see Judgment,
para. 162) was sufficient to characterize the entire series of events occur ‑
ring at Zepa as possessing genocidal intent, the Trial Chamber clearly went

above and beyond this Court in Bosnia and the present Judgment, if not
in its application of the letter of the applicable law, then clearly in hits
appreciation of the spirit thereof. Stated differently, there is no indication
in Tolimir that the approximately 10,000 denizens of Zepa enclave who
were forcibly removed from the area and prevented from returning were

targeted for physical or biological destruction as envisaged by Article II
of the Convention. Rather, the Trial Chamber found that their permanent
removal from that geographical area (in conjunction with the destruction
of a diminutive core of its civil and military leadership) was enough tho
constitute “physical or biological” destruction under the terms ofh Arti -

cle II of the Convention. This cannot but be described as a clear depar -
ture from the Court’s analysis in Bosnia and certain other judgments
rendered by the ICTY upon which the Court in Bosnia relied. Not only
has the quantitative element that featured so prominently in Bosnia been
eschewed, but the Tolimir Judgment has clearly pushed the boundaries of

what constitutes physical or biological destruction by expressly incorpo -
rating non-fatal geographical concerns. In other words, according to my
reading of Tolimir the Trial Chamber clearly found that genocidal intent
was established not because the approximately 10,000 Bosnian Muslims
of Z epa enclave were targeted for elimination per se, but rather because

they were targeted for elimination from that specific location.

23. Granted, the majority’s reticence to adopt a Tolimir-style approach
may be readily (and defensibly) explained by considerations such as thhe

fact that the case remains under appeal and that the finding of genocihde
at Zepa was linked (although obliquely) to the now widely recognized
genocide that was perpetrated by the same attackers at Srebrenica some
20 kilometres away and mere days beforehand. Nevertheless, in my
respectful view, to ignore Tolimir completely constitutes a failure by the

majority to heed its own undertaking to “take account, where approprih -
ate, of the decisions of international criminal courts or tribunals, in hpar-
ticular those of the ICTY, as it did in 2007, in examining the constituent
elements of genocide in the present case” (Judgment, para. 129). I shall
return to this aspect of the Tolimir precedent when dissecting the present

Judgment’s treatment of Croat victims during the siege of Vukovar andh
its aftermath (see infra).

424

7 CIJ1077.indb 844 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 423

cide au titre du litt. a) de l’article II de la Convention), commis parallèle-
ment à des actes massifs de nettoyage ethnique (qui sont en eux -mêmes
des atrocités non génocidaires ; voir arrêt, par.162), suffisait pour consi -

dérer que l’ensemble des faits survenus à Zepa procédait d’une intention
génocidaire , la chambre de première instance est allée bien plus loin que la
Cour ne l’a fait dans l’arrêt rendu en l’affaire Bosnie‑Herzégovine
c. Serbie‑et‑Monténégro et dans le présent arrêt, sinon dans son applica -
tion de la lettre du droit applicable, du moins sans aucun doute dans sohn

appréciation de l’esprit de ce droit. En d’autres termes, rien hdans le juge-
ment Tolimir n’indique que les quelque 10 000 habitants de l’enclave de
Zepa, qui ont été déplacés de force de cette région et empêchés de revenir,
ont été pris pour cible de destruction physique ou biologique au shens
de l’article II de la Convention. Mais la chambre de première

instance a constaté que leur déplacement à jamais de cette zone géogra ‑
phique (combinée à la destruction d’un petit noyau de leurs dirigeanhts
civils et militaires) suffisait à constituer la destruction « physique ou bio-
logique » dont il est question à l’article II de la Convention. Cette position
ne peut être qualifiée que de rupture nette avec l’analyse fahite par la

Cour dans l’arrêt qu’elle a rendu en l’affaire Bosnie‑Herzégovine
c. Serbie‑et‑Monténégro et avec certains autres jugements rendus par le
TPIY sur lesquels la Cour s’est fondée dans ledit arrêt. Non sehulement
l’élément quantitatif qui occupait une place si importante dansh ce dernier
arrêt n’y est pas pris en considération, mais le jugement Tolimir a

manifestement repoussé les limites de ce qui constitue la destructionh phy -
sique ou biologique, en intégrant expressément des considératiohns
géographiques non liées au nombre de vies perdues. En d’autres termes,
selon ma lecture du jugement Tolimir , le TPIY a conclu sans équi -
voque que l’intention génocidaire était établie non pas parce que hles

quelque 10 000 Musulmans de Bosnie de l’enclave de Zepa avaient été la
cible d’actes visant leur élimination en tant que telle, mais plutôt
parce qu’ils avaient fait l’objet de mesures visant à les éliminerh de ce lieu
précis.
23. Certes, on peut comprendre aisément que la majorité se soit mon -
trée peu encline (et non sans raison) à adopter la position retehnue dans

l’affaire Tolimir, du fait que le jugement a été porté en appel et que la
conclusion qu’un génocide avait été commis à Zepa était liée (bien
qu’indirectement) au génocide désormais largement reconnu qui havait été
perpétré par les mêmes assaillants à Srebrenica, à une vihngtaine de
kilomètres et quelques jours auparavant seulement. Néanmoins, à mon

avis, en faisant complètement abstraction du jugement Tolimir, la majo -
rité a manqué à son propre engagement de « prendr[e] en considération,
le cas échéant, les décisions des tribunaux pénaux internationahux, en
particulier celles du TPIY, comme elle l’a fait en 2007, lorsqu’elhle exami-
nera en l’espèce les éléments constitutifs du génocide » (arrêt, par. 129). Je

reviendrai sur cet aspect du jugement Tolimir lorsque j’aborderai la façon
dont les victimes croates du siège de Vukovar et de ses retombées hont été
prises en considération dans le présent arrêt (voir infra).

424

7 CIJ1077.indb 845 18/04/16 08:54 424 application of genochide convention (sep. ohp. bhandari)

The popović ICTY Trial Chamber Judgmenth

24. In the Popović case, the Trial Chamber of the ICTY provided an
analysis on the substantiality component in relation to the killing of shev-
eral thousand Bosnian Muslim men at Srebrenica enclave in Eastern Bos-
nia and Herzegovina in July 1995. It is to be recalled that both the Court

and several Trial and Appeals Chambers of the ICTY have consistently
held that the massacre at Srebrenica constituted genocide. Consequently,h
while the Popović Trial Chamber’s finding of genocidal intent in relation
to the Srebrenica massacre is not in itself a novel jurisprudential devehlop-

ment, in expounding this notion the Trial Chamber made the following
noteworthy remarks:

“The Trial Chamber finds that the Muslims of Eastern Bosnia con-
stitute a substantial component of the entire group, Bosnian Muslims.
As has been found by the Appeals Chamber, although the size of the

Bosnian Muslim population in Srebrenica before its capture . . . was
a small percentage of the overall Muslim population of BiH at the
time, the import of the community is not appreciated solely by its size.h
The Srebrenica enclave was of immense importance to the Bosnian Serb
leadership because : (1) the ethnically Serb state they sought to create

would remain divided and access to Serbia disrupted without Srebren ‑
ica; (2) most of the Muslim inhabitants of the region had, at the
relevant time, sought refuge in the Srebrenica enclave and the elimi -
nation of the enclave would accomplish the goal of eliminating the

Muslim presence in the entire region ; and(3) the enclave’s elimination
despite international assurances of safety would demonstrate to the Bos‑
nian Muslims their defenceless and be ‘emblematic’ of the fate of ▯all
Bosnian Muslims. ” 23

25. In my respectful view, the first and third factors enumerated by the
Trial Chamber in Popović may have warranted consideration when con -
ducting an assessment of genocidal intent in the present Judgment, par -
ticularly with reference to the attack on Vukovar municipality. Regardinhg
the first factor, I note that the Judgment has recalled that :

“Croatia attaches particular importance to the events which took

place in Vukovar and its surrounding area in the autumn of 1991.
According to the Applicant, the JNA and Serb forces killed several
hundred civilians in that multi -ethnic city in Eastern Slavonia,
situated on the border with Serbia and intended to become, under the

23
Prosecutor v. Popović et al., Trial JudgmentJune 2010, para. 865; emphasis
added.

425

7 CIJ1077.indb 846 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 424

Le jugement rendu en l’haffaire popović par le TPIY

24. En l’affaire Popović, la chambre de première instance du TPIY a
procédé à une analyse du critère du caractère substantielh dans le contexte
du meurtre de plusieurs milliers d’hommes musulmans de Bosnie commis h
en juillet 1995 dans l’enclave de Srebrenica en Bosnie orientale. Il convient
de rappeler que tant la Cour que plusieurs chambres de première instahnce

et la chambre d’appel du TPIY ont systématiquement conclu que le mhas -
sacre de Srebrenica constituait un génocide. En conséquence, si leh fait que
la chambre de première instance saisie de l’affaire Popović ait conclu à
l’existence de l’intention génocidaire en ce qui concerne le mahssacre de

Srebrenica n’a rien d’un revirement jurisprudentiel, ladite chambrhe a, ce
faisant, fait les observations suivantes, qui méritent qu’on s’hy arrête :

«La Chambre de première instance conclut que les Musulmans de
Bosnie orientale représentent une partie substantielle du groupe des h
Musulmans de Bosnie dans son ensemble. Comme l’a conclu la
Chambre d’appel, si, avant la prise de Srebrenica … le nombre des
habitants musulmans était peu élevé par rapport à la populathion

musulmane totale de la Bosnie-Herzégovine à l’époque des faits, il ne
faut pas se méprendre sur l’importance de la communauté musul -
mane de Srebrenica. L’enclave de Srebrenica revêtait une importance
stratégique capitale pour les dirigeants serbes de Bosnie, car 1), sans
Srebrenica, l’Etat ethniquement serbe qu’ils souhaitaient créer▯ serait

resté divisé et coupé de la Serbie ; 2)la plupart des habitants musul -
mans de la région ayant trouvé à l’époque refuge dans l’henclave de
Srebrenica, l’élimination de cette enclave aurait permis de débarras -
ser la région tout entière de sa population musulmane ; et 3) l’élimi‑
nation de l’enclave, malgré les assurances données par la commu▯nauté

internationale, devait faire prendre conscience à tous les Musulmans ▯de
Bosnie de leur impuissance, et être « représentati[ve]» du sort de l’en ‑
semble des Musulmans de Bosnie. » 23

25. A mon avis, les premier et troisième points énumérés par la h
chambre de première instance dans l’affaire Popović auraient dû être pris
en considération, dans le présent arrêt, lors de l’appréchiation de l’inten -

tion génocidaire, notamment en ce qui concerne l’attaque de la munhicipa -
lité de Vukovar. S’agissant du premier point, je constate qu’ilh a été rappelé
ce qui suit dans l’arrêt:

«La Croatie attache une importance particulière aux événements
qui se sont déroulés à Vukovar et dans ses environs à l’ahutomne
1991. Selon elle, la JNA et des forces serbes auraient tué plusieurs h
centaines de civils dans cette ville à population mixte de Slavonie

orientale, située à la frontière avec la Serbie, qui avait vocation, dans ▯

23 Le Procureur c. Popović et consorts, jugement du 10 juin 2010, par. 865 (les italiques
sont de moi).

425

7 CIJ1077.indb 847 18/04/16 08:54 425 application of genochide convention (sep. ohp. bhandari)

plans for a ‘Greater Serbia’, the capital of the new Serbian regio▯n of
Slavonia, Baranja and Western Srem. ” (Judgment, para. 212 ; empha-

sis added.)
This averred emblematic significance of Vukovar can be further inferrehd

from the findings of the ICTY Trial Chamber in Mrkšić, as accepted by
the Court in the instant Judgment, which found that during the approxi -
mately three -month siege of Vukovar :

“The duration of the fighting, the gross disparity between the num-
bers of the Serb and Croatian forces engaged in the battle and in the
armament and equipment available to the opposing forces and, above
all, the nature and extent of the devastation brought on Vukovar and

its immediate surroundings by the massive Serb forces over the pro ‑
longed military engagement, demonstrate, in the finding of the Cham ‑
ber, that the Serb attack was also consciously and deliberately directed▯
against the city of Vukovar itself and its hapless civilian population ,

trapped as they were by the Serb military blockade of Vukovar and
its surroundings and forced to seek what shelter they could in the
basements and other underground structures that survived the ongo -
ing bombardments and assaults. What occurred was not, in the find -
ing of the Chamber, merely an armed conflict between a military force h

and an opposing force in the course of which civilians became casu -
alties and some property was damaged. The events, when viewed
overall, disclose an attack by comparatively massive Serb forces, well
armed, equipped, and organized, which slowly and systematically
destroyed a city and its civilian and military occupants to the point

where there was a complete surrender of those that remained . . .
It is in this setting that the Chamber finds that, at the time relevanht
to the Indictment, there was in fact, not only a military operation
against the Croat forces in and around Vukovar, but also a widespread
and systematic attack by the JNA and other Serb forces directed against ▯

the Croat and other non‑Serb civilian population in the wider Vukovar
area . . .”4

In my view, this sustained, ethnically discriminatory attack, aimed in
part at the slow and systematic destruction of the Croat civilian populahce
of Vukovar, provides implicit evidence of its strategic importance in tehrms
of allowing the expansionist policy of “Greater Serbia” to gain a hpivotal

foothold within Croatian territory, and thus heightens the prominence ofh
the Vukovar Croat subgroup when assessing genocidal intent vis -à-vis
that municipality and its environs.

26. Regarding the third criterion enunciated in Popović, I recall that

the evacuation of Vukovar hospital on 20 November 1991, through which
many Croats were interned at nearby concentration camps and subse -

24 Judgment, para. 218; citing Mrkšić, paras. 470 and 472.

426

7 CIJ1077.indb 848 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 425

le cadre du projet de la « Grande Serbie », à devenir la capitale de la
nouvelle région serbe de Slavonie, Baranja et Srem occidental.» (Arrêt,

par. 212 ; les italiques sont de moi.)
L’importance symbolique apparente de Vukovar ressort également desh

conclusions de la chambre de première instance du TPIY dans l’affaihre
Mrkšić, reprises dans l’arrêt en l’espèce par la Cour, relativemenht au siège
de Vukovar, qui a duré trois mois environ :

« [L]a durée des combats, la très forte disparité tant numérique que
matérielle des forces engagées dans la bataille et, surtout, la nature et
l’étendue des destructions imputables aux forces serbes à Vukov▯ar et

aux abords immédiats de la ville pendant cet engagement militaire pro▯
longé démontrent que l’attaque serbe était consciemment et d▯élibéré‑
ment dirigée contre la ville et sa malheureuse population civile, prise au
piège par le siège de Vukovar et des alentours par les forces serbes et
contrainte de se réfugier dans les caves et autres constructions sou -

terraines qui avaient résisté aux bombardements et aux assauts.
Selon la Chambre, il ne s’agissait pas d’un simple conflit arméh entre
une force militaire et des forces adverses qui aurait fait des victimes
civiles et causé certains dommages matériels. Une vue d’ensemblhe des
événements révèle l’existence d’une attaque par les forces serbes

numériquement bien supérieures, bien armées, bien équipéehs et bien
organisées, qui ont lentement et systématiquement détruit une ville et
ses occupants civils et militaires jusqu’à la reddition complète des der -
niers survivants.

La Chambre en conclut que, à l’époque des faits, il existait nohn
seulement une opération militaire menée contre les forces croates
présentes à Vukovar et aux environs, mais aussi une attaque générali ‑
sée et systématique dirigée par la JNA et d’autres forces se▯rbes contre

la population civile croate et d’autres civils non serbes dans le secteur
de Vukovar.» 24

A mon avis, cette attaque prolongée et discriminatoire sur le plan ethh-
nique, qui visait en partie à la destruction lente et systématiqueh de la
population civile croate de Vukovar, apporte une preuve implicite de
l’importance stratégique de celle-ci, en ce qu’elle permettait hà la politique

expansionniste de la « Grande Serbie» de prendre pied en territoire croate,
ce qui accroît l’importance qu’il convient d’accorder au souhs-groupe des
Croates de Vukovar au moment d’apprécier l’intention génocidhaire à
l’égard de cette municipalité et de ses environs.
26. En ce qui concerne le troisième point énoncé dans le jugement

Popović, je rappelle que l’évacuation de l’hôpital de Vukovar le
20 novembre 1991, à l’issue de laquelle de nombreux Croates ont été intehr -

24 Jugement Mrkšić, par. 470 et 472, cité au paragraphe 218 de l’arrêt.

426

7 CIJ1077.indb 849 18/04/16 08:54 426 application of genochide convention (sep. ohp. bhandari)

quently killed, severely beaten and/or otherwise subjected to serious fohrms

of physical and psychological abuse, was conducted in violation of the
Zagreb Agreement, which purported to allow for the safe evacuation of
those internally displaced Vukovar Croats who had sought refuge at the
local hospital under the supervision of neutral international monitors. hI

believe that the deliberate and cynical manner in which this internationhal
agreement was violated, to the grave detriment of those who made the
assumption that a widely publicized agreement would guarantee their
safety, allows for the inference that the sorry plight of the victims frhom

Vukovar hospital and those subsequently interned in concentration and
death camps, could certainly, to paraphrase Popović, “demonstrate to the
Vukovar Croats their defencelessness and be emblematic of the fate of alhl
ethnic Croats on Croatian territory”.

The nizeyimana ICTR Trial Chamber Judgmenth

27. In the Nizeyimana case, the accused was convicted at trial for geno-
cide in relation to, inter alia, the killing of Rosaliecanda, a member of
the targeted Tutsi ethnic group and former Queen of Rwanda. In apply -

ing the substantiality criterion of genocidal intent, the Trial Chamber
stressed that

“[t]he fact that this operation targeted one Tutsi in particular in no way
impacts the conclusion that the perpetrators possessed the intent to
destroy at least a substantial part of the Tutsi ethnic group.The Cham-

ber reiterates that this killing must be viewed in the context of the
targeted and systematic killing of Tutsis perpetrated . . . in Butare
[town] around this time. Moreover, the symbolic importance of the kil‑l
ing of Gicanda as a means of identifying the enemy is also relevant.” 25

In that regard, the Trial Chamber noted that it had “no doubt that thhe
murder of Gicanda . . . who was a symbol of the former [Tutsi] monar-

chy, was killed in order to set a striking example that Tutsis, 26 well as
Hutus sympathetic to the plight of the Tutsis, were the enemy ” . The Trial
Chamber further stressed the nexus between this particular attack and thhe
significantly increased violence against Tutsi civilians in Butare town fol -

lowing an incendiary speech by the President of Rwanda on 19 April 1994
in which he exhorted the population to seek out and kill Tutsis.

28. In relation to a separate incident, the Trial Chamber found that the

killing of one Pierre Claver Karenzi, a Tutsi lecturer at a local university

25Prosecutor v. Nizeyimana, Trial Judgment, June 2012, para. 1530; emphasis
added.
26
Ibid., para. 1511; emphasis added.

427

7 CIJ1077.indb 850 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 426

nés dans des camps de concentration à proximité, puis tués, roués de coups

et/ou soumis à d’autres formes graves de violences physiques et psycholo -
giques, a été effectuée en violation de l’accord de Zagreb, qhui était censé
permettre l’évacuation en toute sécurité des Croates de Vukohvar déplacés
qui avaient cherché refuge à l’hôpital local sous la supervihsion d’observa-

teurs internationaux neutres. Je crois que la façon délibéréhe et cynique
dont cet accord international a été violé, au grave détriment de ceux qui
avaient pu supposer qu’un accord largement médiatisé garantiraiht leur
sécurité, permet de conclure que le destin tragique des victimes dhe l’hôpital

de Vukovar et de ceux qui ont été internés ultérieurement dahns des camps
de concentration et d’extermination pouvait certainement, pour paraphhra -
ser le jugement Popović, «faire prendre conscience aux Croates de Vukovar
de leur impuissance et être représentati[ve] du sort de l’ensemhble des

Croates de souche sur le territoire croate ».

Le jugement rendu en l’haffaire nizeyimana par le TPIR

27. Dans l’affaire Nizeyimana, l’accusé a été reconnu coupable lors de
son procès pour génocide à raison, notamment, du meurtre de Rosalieh
Gicanda, membre du groupe ethnique tutsi visé et ancienne reine du

Rwanda. Lorsqu’elle a appliqué le critère du caractère substhantiel par
rapport à l’intention génocidaire, la chambre de première instance a souligné
ce qui suit

«Le fait que cette opération visait un Tutsi en particulier n’a aucune
incidence sur la conclusion que les auteurs avaient l’intention de dé▯truire
au moins une partie substantielle du groupe ethnique tutsi. La Chambre

rappelle que ce meurtre doit être vu dans le contexte du massacre
ciblé et systématique des Tutsis perpétré à cette époqhue … [à] Butare.
De plus, l’importance symbolique du meurtre de Gicanda en tant que
moyen d’identifier l’ennemi est également pertinente. » 25

A ce sujet, la chambre de première instance a fait observer qu’ellhe
n’avait « aucun doute … que … Gicanda, … symbole de l’ancienne

monarchie [tutsie] …, a[vait] été tuée pour servir d’exemple frappant et
montrer que l’ennemi, c’était les Tutsis et les Hutus perçus▯ comme sensibles
au sort des Tutsis » 26. Elle a aussi souligné le lien qui existait entre cet
attentat et l’augmentation notable des actes de violence commis contrhe

des civils tutsis à Butare après un discours incendiaire prononcéh le
19 avril 1994 par le président rwandais, qui avait exhorté la population àh
traquer et à tuer les Tutsis.
28. La chambre de première instance a conclu que le meurtre de

Pierre Claver Karenzi, professeur tutsi dans une université locale et consi -

25 Le Procureur c.Nizeyimana, jugement du 19 juin 2012, par. 1530 (les italiques
sont de moi).
26
Ibid., par. 1511 (les italiques sont de moi).

427

7 CIJ1077.indb 851 18/04/16 08:54 427 application of genochide convention (sep. ohp. bhandari)

who was considered “a prominent figure in Butare” town also constituted

genocide. Again the Trial Chamber found that

“ Karenzi’s murder is also emblematic of the systematic nature in which
Tutsi civilians were identified and killed on an ongoing basis at roadh -
blocks manned by . . . soldiers in Butare town. Consequently, while
this incident only resulted in the killing of one Tutsi, the Chamber has▯

no doubt that the physical perpetrator acted with the spe27fic intent to▯
destroy at least a substantial part of the Tutsi group .”

Once again the Trial Chamber found that this attack was linked to the
broader context of significantly increased targeted killings of the Tuhtsi
ethnic group in Butare town around that time in the wake of the Presi -
dent’s speech.

29. Finally, the Trial Chamber found genocidal intent in relation to
another incident where two Tutsi civilians were killed and another seri -
ously injured at a military roadblock. As the Trial Chamber reasoned,

“ [w]hile these attacks only resulted in the deaths of two Tutsis and the ▯
serious bodily harm of a third, the Chamber has no doubt that the per ‑
petrators acted with the intent to destroy at least a substantial part o▯f

the Tutsi group. These attacks were emblematic of the systematic nature ▯
in which Tutsi civilians were identified and killed on an ongoing basis
at this roadblock and others manned by. . . soldiers in Butare town.” 28

What is particularly noteworthy about this specific finding of genochidal
intent is the Trial Chamber’s determination that the attack on the thhree

Tutsi victims was “emblematic” of the overall group not because of
the individual prominence of the victims within the community (there
was no evidence on record to suggest such a conclusion), but rather
because the attack on them embodied a modus operandi for the system ‑

atic destruction of the Tutsi group in Butare town generally. In other
words, Tutsis at the roadblock were “emblematic” of the overall group
not because of who they were, but rather the manner in which they
were attacked. While indicia of “prominence” through the modus operandi

of the attack may be gleaned from the killings of Gicanda and Karenzi,
the fact that the victims of this third attack did not hold any prominenht
positions within the Tutsi community only further underscores the point.h
Consequently, I find these exemplars from the Nizeyimana Trial

Judgment to signal a clear departure from what was envisaged
by the “qualitative approach” in the Court’s Bosnia Judgment (and
subsequently rebranded as the “prominence” of the targeted group
in the instant Judgment), and believe that the present Judgment’s

analysis would have been enriched by a consideration of this recent,

27Prosecutor v. Nizeyimana, Trial Judgment, 19June 2012, para. 1530; emphasis
added.
28
Ibid., para. 1521; emphasis added.

428

7 CIJ1077.indb 852 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 427

déré comme «une personnalité en vue à Butare », constituait aussi un acte

génocidaire. Là encore, la chambre de première instance a conclhu que
« le meurtre de Karenzi illustre, lui aussi , la manière systématique

dont les civils tutsis étaient identifiés et tués aux barragehs contrôlés
par des militaires … à Butare. Aussi, même s’il n’y a eu, cette fois‑là,
qu’un seul Tutsi qui a été tué, il ne fait aucun doute pour ▯la Chambre
que [l’]auteur [du meurtre] était animé de l’intention spé▯cifique de
27
détruire au moins une partie substantielle du groupe ethnique tutsi. »
Une fois de plus, la chambre de première instance a conclu que cette hattaque

s’inscrivait dans le contexte plus général de l’augmentationh appréciable du
nombre de meurtres dirigés contre des membres du groupe ethnique tutshi
dans la ville de Butare à cette époque, à la suite du discours hdu président.
29. Enfin, la chambre de première instance a conclu à l’existenceh de l’in -
tention génocidaire à raison d’un autre épisode au cours duqhuel deux civils

tutsis ont été tués et un troisième, grièvement blesséh, à un barrage militaire :

«Ces attaques s’étaient soldées par la mort de deux Tutsis seule▯ment
et l’atteinte grave portée à l’intégrité physique d’un troisième, mais la
Chambre n’a aucun doute que les auteurs étaient animés de l’▯intention
de détruire au moins une partie substantielle du groupe ethnique tutsi.
Ces attaques attestaient la nature systématique de la manière dont▯ les

civils tutsis ne cessaient d’être identifiés et tués à ce barrage et à
d’autres tenus par les militaires … à Butare. »28

Cette décision concluant à l’existence de l’intention génhocidaire est rema-r
quable en ce que la chambre de première instance a considéré que l’at -
taque contre les trois victimes tutsies « attestait» que l’ensemble du groupe
était visé non pas à cause de l’importance que les victimes avaient en tant

qu’individus au sein de la communauté (pareille conclusion n’étant étayée
par aucune élément de preuve), mais plutôt parce que cette attaque incar -
nait un mode opératoire visant la destruction systématique du groupe tutsi
dans la ville de Butare en général. En d’autres termes, les Tutsis tombés au

barrage routier symbolisaient l’ensemble du groupe en raison non pas hde
leur personnalité, mais plutôt de la manière dont ils avaient été attaqués. Si,
dans les meurtres de Gicanda et de Karenzi, l’on peut considérer qhue la
«notoriété» des victimes a joué un rôle dans le schéma de l’attaqueh, le fait
que les victimes de la troisième attaque n’occupaient pas de posithion

importante au sein de la communauté tutsie ne fait en revanche qu’hap -
puyer encore cette thèse. J’estime donc que ces extraits du jugemehnt rendu
en l’affaire Nizeyimana marquent une rupture nette avec ce qui avait été
envisagé sous l’angle « qualitatif» dans l’arrêt rendu par la Cour en l’af -

faire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro (devenu dans le présent
arrêt la « place occupée » par le groupe visé), et je crois que l’analyse à

27
Le Procureur c. Nizeyimana, jugement du 19 juin 2012, par. 1530 (les italiques sont
de 28i).
Ibid., par. 1521 (les italiques sont de moi).

428

7 CIJ1077.indb 853 18/04/16 08:54 428 application of genochide convention (sep. ohp. bhandari)

pertinent jurisprudential development on the law of genocidal dolus

specialis.

30. Applying the Nizeyimana precedent to the facts at bar, I would
note that the present Judgment recalls that the Mrkšić Trial Chamber
found that the attacks in Eastern Slavonia generally followed a consistent

pattern:

“[T]he system of attack employed by the JNA typically evolved
along the following lines (a) tension, confusion and fear is built up
by a military presence around a village (or bigger community) and
provocative behaviour ; (b) there is then artillery or mortar shelling
for several days, mostly aimed at the Croatian parts of the village ; in

this stage churches are often hit and destroyed (c) in nearly all cases
JNA ultimata are issued to the people of a village demanding the
collection and the delivery to the JNA of all weapons ; village delega-
tions are formed but their consultations with JNA military authorities

do not lead . . . to peaceful arrangements ;.. . (d) at the same time,
or shortly after the attack, Serb paramilitaries enter the village ; what
then follows varied from murder, killing, burning and looting, to dis-
crimination.” 29

The Judgment also recalled that in the Martić case, the ICTY Trial

Chamber made similar findings regarding the pattern of attacks perpe -
trated by Serb forces in Croatia :

“[T]he area or village in question would be shelled, after which
ground units would enter. After the fighting had subsided, acts of
killing and violence would be committed by the forces against the
civilian non -Serb population who had not managed to flee during the
attack. Houses, churches and property would be destroyed in order

to prevent their return and widespread looting would be carried out.
In some instances the police and the TO of the SAO Krajina organ -
ized transport for the non -Serb population in order to remove it from
SAO Krajina territory to locations under Croatian control. More -

over, members of the non -Serb population would be rounded up and
taken away to detention facilities, including in central Knin, and even -
tually exchanged and transported to areas under Croatian control.” 30

After considering, inter alia, these findings from the ICTY, the present
Judgment concludes that :

“The Court likewise notes that there were similarities, in terms of
the modus operandi used, between some of the attacks confirmed to
have taken place. Thus it observes that the JNA and Serb forces

29
30Judgment, para. 414, citing Mrkšić Trial Judgment, para. 43.
Ibid., para. 427.

429

7 CIJ1077.indb 854 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 428

laquelle il a été procédé dans le présent arrêt auraith bénéficié de l’examen

de cette évolution récente et pertinente de la jurisprudence concehrnant le
droit relatif au dolus specialis du génocide.
30. Appliquant le principe du jugementNizeyimana aux faits de l’espèce,
je tiens à souligner qu’il est rappelé dans le présent arrêt que la chambre de
première instance a conclu en l’affaire Mrkšić que les attaques en Slavonie

orientale s’étaient en général déroulées selon le mêhme scénario :

«[L]es attaques de la JNA se déroulaient en général selon le schhéma
suivant: «a) elle attisait les tensions et semait la confusion et la peur
par une présence militaire aux alentours du village (ou d’une com -
munauté plus grande) et par des provocations ; b) elle tirait ensuite,
plusieurs jours durant, à l’artillerie ou au mortier le plus souvehnt sur

les parties croates du village ; c’est à ce stade que, souvent, les églises
étaient touchées et détruites ;c) dans presque tous les cas, la JNA
lançait un ultimatum aux habitants, leur enjoignant de rassembler et h
de remettre leurs armes ; les villages constituaient des délégations,

mais les négociations avec les autorités militaires de la JNA n’hont
abouti à aucun accord de paix ; d) pendant ou juste après l’attaque,
des paramilitaires serbes entraient dans le village, assassinant ou
tuant les habitants, incendiant et pillant leurs biens, pour des raisonsh
discriminatoires ou non. » 29

Il est également rappelé dans l’arrêt que, dans l’affaire Martić, la chambre

de première instance du TPIY est parvenue à des conclusions similahires
au sujet de la série d’attaques perpétrées par les forces sehrbes en Croatie:

«[D]es unités de l’armée de terre entraient dans le secteur ou lhe
village en question à la suite d’un bombardement. Une fois que les
combats avaient cessé, les assaillants tuaient ou maltraitaient les
civils non serbes qui n’avaient pas réussi à fuir pendant l’hattaque. Ils
détruisaient les maisons, les églises et d’autres bâtiments hpour empê -

cher le retour des non -Serbes, se livrant en même temps à un pillage
systématique. Dans certains cas, la police et la TO de la SAO de
Krajina ont organisé le transport de la population non serbe vers desh
localités sous contrôle croate. En outre, les non -Serbes étaient pris

dans des rafles et incarcérés, notamment dans le centre de déhtention
de Knin ville, en vue d’être échangés et transportés versh des régions
sous contrôle croate. » 30

Après avoir examiné, entre autres, ces constatations du TPIY, la Cour en
vient dans le présent arrêt à la conclusion suivante :

«La Cour constate également que, parmi les attaques dont l’exis -
tence a pu être établie, certaines présentaient des similaritéhs quant au
mode opératoire utilisé. Elle note ainsi que la JNA et des forces h

29
30Jugement Mrkšić, par. 43, cité au paragraphe 414 de l’arrêt.
Ibid., par. 427.

429

7 CIJ1077.indb 855 18/04/16 08:54 429 application of genochide convention (sep. ohp. bhandari)

would attack and occupy the localities and create a climate of fear
and coercion, by committing a number of acts that constitute the

actus reusof genocide within the meaning of Article II (a) and (b)
of the Convention. Finally, the occupation would end with the forced
expulsion of the Croat population from these localities.
The findings of the Court and those of the ICTY are mutually
consistent, and establish the existence of a pattern of conduct that

consisted, from August 1991, in widespread attacks by the JNA and
Serb forces on localities with Croat populations in various regions of
Croatia, according to a generally similar modusoperandi.” (Judgment,
paras. 415-416.)

31. Bearing this established pattern of conduct throughout various
parts of the territory of Croatia in mind, I would further recall that the

Applicant has presented the siege of Vukovar as representing a paradig -
matic example of the modus operandi outlined above. As counsel for Cro-
atia stated during the oral hearing phase of this case,

“[W]hat happened at Vukovar was repeated again and again across
Eastern Slavonia and across Croatia as a whole in the course of this
conflict. This pattern may have varied from village to village, town to
town and across different regions. But, properly analysed, the ‘pat -

tern’ discloses that there was an intention to ‘destroy’ a parth of the
Croat group in question. The artillery or mortar shelling was wholly
disproportionate and, in places, such as Vukovar, essentially destroyed
the entire city. And the murderous attacks were never intended as part
of the mere expulsion of a part of the Croat group in question.” 31

Consequently, I believe that in view of the Nizeyimana Trial Judgment,
the Judgment’s analysis with respect to substantiality could have beehn

enhanced by considering the modus operandi of the attack on Vukovar,
being a microcosm for the manner in which a much wider conflict was
waged, for the purpose of assessing whether the “prominence” of thhe
Vukovar Croats could factor into the calculus as to whether they were

targeted with genocidal intent.

The H ateGekimana ICTR Trial Chamber Judgmenth

32. In the case of Hategekimana, the accused was convicted, inter alia,
of genocide for the murder of three Tutsi women during an attack on their
home by militia and soldiers. In determining genocidal intent, the Trial

Chamber noted that in addition to the fact that the three women were
singled out because of their ethnicity, the Chamber had received “exten -
sive evidence . . . about the targeting of Tutsi civilians in Butare [prov -
ince] following the speech of interim President Sindikubwabo on
19 April 1994”, which resulted in “many Tutsi civilians being killed in their

31 CR 2014/8, p. 47 (Starmer); emphasis added.

430

7 CIJ1077.indb 856 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 429

serbes attaquaient les localités, les occupaient et imposaient un clih -
mat de coercition et de peur, en commettant un certain nombre

d’actes constitutifs de l’élément matériel du génocideh, au sens des
litt a) et b) de l’article II de la Convention. Enfin, l’occupation se
soldait par l’expulsion forcée de la population croate de ces locahlités.
Les conclusions de la Cour et du TPIY sont concordantes et per -
mettent d’établir l’existence d’une ligne de conduite constihtuée d’at-

taques généralisées par la JNA et des forces serbes de localitéhs
peuplées de Croates dans différentes régions de Croatie, selon uhn
schéma généralement similaire, à compter d’août 1991. » (Arrêt,
par. 415-416.)

31. Gardant à l’esprit cette ligne de conduite établie dans diversehs par-
ties du territoire de la Croatie, il importe de rappeler que le demandeur a

présenté le siège de Vukovar comme constituant un exemple paradhigma-
tique du mode opératoire décrit plus haut. Ainsi, le conseil de lah Croatie
a tenu les propos ci-après lors des audiences :

«[C]e qui est arrivé à Vukovar s’est reproduit à maintes repr▯ises en
Slavonie orientale et dans toute la Croatie au cours du conflit . Les
méthodes ont peut -être varié selon les villages, les villes, les régions,
mais, examinées de près, elles font toutes apparaître l’intehntion de

« détruire» une partie du groupe croate en question. Les pilonnages
à l’artillerie ou au mortier étaient totalement disproportionnéhs et, à
Vukovar par exemple, ont pratiquement réduit la ville en cendres.
Ces attaques meurtrières ne visaient pas simplement à expulser une
partie du groupe croate en question. » 31

En conséquence, je crois que, compte tenu du jugement Nizeyimana,
l’analyse à laquelle il a été procédé dans l’arrêht au sujet du caractère sub-

stantiel aurait pu être étoffée si l’on avait pris en considéhration le schéma
de l’attaque de Vukovar, en tant que microcosme d’un conflit beahucoup
plus large, pour évaluer si la « place occupée» par les Croates de Vukovar
pouvait entrer en ligne de compte dans l’examen de la question de savhoir

s’ils avaient été pris pour cible dans une intention génocidhaire.

Le jugement rendu en l’haffaire HateGekimana par le TPIR

32. Dans l’affaire Hategekimana, l’accusé a été reconnu coupable, entre
autres, de génocide, pour le meurtre de trois femmes tutsies lors de l’at ‑
taque de leur maison par des miliciens et des soldats. Dans sa recherche de

l’intention génocidaire, la chambre de première instance a obsehrvé que,
outre le fait que les trois femmes avaient été choisies en raison hde leur
appartenance ethnique, elle avait reçu «d’abondants témoignages … [selon
lesquels] les civils tutsis [avaient] été visés dans la [provinhce] de Butare,

plus particulièrement après le discours du président par intéhrim, Sin -

31 CR 2014/8, p. 47 (Starmer) (les italiques sont de moi).

430

7 CIJ1077.indb 857 18/04/16 08:54 430 application of genochide convention (sep. ohp. bhandari)

homes over the course of many days”. The Chamber found that “[g]iven

the scale of the killings and their context, the only reasonable inference is
that the assailants [who killed the women] possessed the intent to destrhoy
in whole or in part a substantial part of the Tutsi group” 3. Once again
we see an example where an attack against a targeted group that resultedh

in a comparatively low absolute and relative number of victims was
nevertheless deemed to possess genocidal intent, due at least in part toh
the modus operandi of the manner in which they were killed.

The m unyakazi ICTR Trial Chamber Judgmenth

33. In the case of Munyakazi, the accused was convicted of genocide
for, inter alia, an attack on a parish that killed between 60 and 100 Tutsi

refugees. The Trial Chamber observed that the attack occurred the day
after a much larger attack on a different parish where approximately
5,000 to 6,000 Tutsis were killed by the same group of perpetrators. Con -
sidering both attacks as a whole, the Trial Chamber found genocidal
33
intent for a substantial part of the Tutsi ethnic group . This finding of
genocide evokes many parallels with the ICTY Trial Chamber’s findinhg
of genocide in relation to Zepa enclave in the Tolimir Judgment, which
also featured an attack on one geographic area where a relatively small h
number of victims were killed (Zepa) but which was closely linked, in

terms of geography, time, and the identity of the perpetrators, to a prehvi-
ous, considerably more sizeable attack (Srebrenica).

Other ICTR Trial Chamber Judghments

34. In keeping with the pattern demonstrated above, since the Court’s

issuance of the Bosnia Judgment in 2007, the ICTR has made findings of
genocide in relation to scenarios where “the quantitative element”h (to use
the nomenclature adopted by the present Judgment) figured far less
prominently in the calculus as to whether the attacks were perpetrated

with genocidal intent than a strict application of the Bosnia formula
would dictate. In this regard, we see that genocidal intent in several
instances was inferred in large part due to the geographic profile of h
the situs of the attack and/or the prominence of the victims (whether

said prominence was measured in terms of personal standing in the

32 Prosecutor v. Hategekimana, Trial Judgment, 6 December 2010, para. 673; emphasis
added.
33
Prosecutor v. Munyakazi, Trial Judgment, 5 July 2010, paras. 496, 499-500.

431

7 CIJ1077.indb 858 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 430

dikubwabo, le 19 avril 1994», si bien que « de nombreux … Tutsis avaient

été attaqués chez eux [et] tués en grand nombre pendant plusieurs jours ».
Elle a jugé que, « [v]u l’ampleur des massacres et le contexte dans lequel ils
[avaient] été perpétrés, la seule conclusion raisonnable quih s’impose est
que les assaillants [qui ont tué ces femmes] étaient animés de hl’intention
32
de détruire, en tout ou en partie, une part substantielle du groupe thutsi » .
Il s’agit là d’un autre cas où l’on a estimé, en raisohn, au moins en partie,
du mode opératoire utilisé, qu’une intention génocidaire éhtait à l’œuvre
dans une attaque contre un groupe ciblé qui s’est pourtant soldéhe par un

nombre de victimes assez faible, qu’il soit pris dans l’absolu ou en tant
que valeur relative.

Le jugement rendu en l’haffaire m unyakazi par le TPIR

33. Dans l’affaire Munyakazi, l’accusé a été reconnu coupable de géno-
cide à raison, notamment, d’une attaque dirigée contre une parohisse au

cours de laquelle ont péri 60 à 100 réfugiés tutsis. La chambre de première
instance a relevé que cette attaque s’était produite le lendemahin d’une
attaque de bien plus grande ampleur contre une autre paroisse, au cours
de laquelle le même groupe d’assaillants avait tué de 5000 à 6000 Tutsis.

Considérant que ces deux attaques devaient être prises ensemble, ehlle a
conclu à l’existence d’une intention génocidaire visant une hpartie substan -
tielle du groupe ethnique tutsi 33. Cette conclusion présente de nombreux
parallèles avec celle à laquelle le TPIY est parvenu dans le jugemhent Toli‑
mir au sujet de l’enclave de Zepa, où avait été menée une attaque qui allait

faire un nombre relativement faible de victimes, mais qui était éthroitement
liée, sur les plans géographique et temporel, et quant à l’ihdentité des
auteurs, à une attaque antérieure beaucoup plus importante (Srebrhenica).

Les jugements rendus phar le TPIR dans d’autres ahffaires

34. Conformément à la tendance exposée ci -dessus, depuis que la Cour

a rendu en 2007 son arrêt en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑
Monténégro, le TPIR a conclu à la perpétration d’un génocide dans
des contextes où l’« élément quantitatif » (pour reprendre la terminologie
adoptée dans le présent arrêt) tenait dans la démarche visahnt à déterminer

si les attaques avaient été lancées dans une intention génochidaire, une
place nettement moins importante que ce qu’aurait exigé une applichation
stricte de la formule établie dans l’arrêt rendu en l’affaireh Bosnie‑
Herzégovine c. Serbie‑et‑Monténégro. A ce sujet, on constate que, dans

plusieurs affaires, l’intention génocidaire a été inféréhe en grande partie en

32 Le Procureur c. Hategekimana, jugement du 6 décembre 2010, par. 680 (les italiques
sont de moi).
33
Le Procureur c. Munyakazi, jugement du 5 juillet 2010, par. 496, 499-500.

431

7 CIJ1077.indb 859 18/04/16 08:54 431 application of genochide convention (sep. ohp. bhandari)

community of the victim or the modus operandi of how the attack
unfolded) . In sum, what we see is a clearly more flexible application

of genocidal dolus specialis that would tend to challenge the pre-eminence
afforded the substantiality criterion in the Bosnia case.

35. For the avoidance of any doubt, I wish to underscore my recogni -
tion that there are obviously significant contextual differences between

the crimes prosecuted in relation to Croatia before the ICTY (and, by
extension, the subject -matter presently before the Court) and those relat -
ing to Rwanda before the ICTR, not the least of which is the sheer dis -

parity in scale of the atrocities that occurred during the course of theh
respective conflicts. Hence I recall that Croatia’s case — taken at its high-
est— is that the hostilities that form the backdrop to the present Judg -
ment resulted in 12,500 Croat deaths, whereas conservative estimates of

the carnage in Rwanda posit that at least half a million ethnic Tutsis and
moderate Hutus were killed during the course of the genocide that unfolded
in that country in 1994 — a genocide, which, it should be noted, was the

subject of an express finding of judicial notice by the ICTR Appeals
Chamber 35. Indeed, I recall that the Court stipulated in the present Judg-
ment that it would give particular preference to jurisprudence emanatingh
from the ICTY (see supra, paragraph 13, citing paragraph 129 of the

Judgment), and I understand this perfectly sensible decision to be motih -
vated in large part by the plain fact that the cases before the ICTY invholve
much closer historical, socio -political and legal issues to those presented

in the case at bar than cases appearing before the ICTR.

36. In sum, while I am by no means advocating the wholesale importa -
tion of ICTR case law into the jurisprudence of this Court, my concern

lies with what I find to be essentially the complete disregard of the hmost
prolific judicial body to have interpreted and applied the Genocide
Convention in the course of human history. With the greatest of respect h

to my learned colleagues, failure to so much as consult this ample body hof
jurisprudence, to my mind, constitutes a failure by the Court in its duthy
and its undertaking to keep abreast of the most recent and pertinent
developments in the law of genocide in the present Judgment.

34 See, e.g.,Prosecutor v. Nsengimana, Trial Judgment, 17 November 2009,

paras. 834-836; Prosecutov.Renzaho, Trial Judgment, 14 July 2009, paras. 768-769;
Prosecutor v. Rukundo, Trial Judgment, 27 February 2009, paras. 72, 74, 76; Prosecutor
Nc35mihigo, Trial Judgment, 12 November 2008, paras. 333-336, 346-347, 354, 357.
This landmark decision was delivered by the Appeals Chamber on Prosecutohr’s
Appeal on Judicial Notice, dated 1 6 June 2006, in the trial of Prosv.Karemera,
Ngirumpatse and Nzirorera, ICTR-98-44AR73 (C).

432

7 CIJ1077.indb 860 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 431

fonction de la géographie du lieu de l’attaque et/ou de la place ohccupée
par les victimes (selon que cette place a été évaluée en fohnction de la situa -

tion de la victime au sein de la c34munauté ou du mode opératoire hselon
lequel s’est déroulée l’attaque) . En somme, ce que nous voyons est une
application à l’évidence plus souple de la formule permettant dh’établir
l’intention génocidaire qui tendrait à mettre en question la prhééminence

accordée au critère du caractère substantiel que l’on trouve dans l’arrêt
rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro.
35. Pour éviter tout malentendu, je tiens à souligner que je suis
conscient des différences contextuelles majeures qui existent entre les

crimes commis en Croatie et poursuivis devant le TPIY (et, par extensiohn,
le différend porté devant la Cour) et ceux qu’a connus le Rwandha et qui
ont été jugés par le TPIR, ne serait-ce que du point de vue de hla disparité

de l’échelle des atrocités commises au cours de chacun de ces chonflits.
C’est pourquoi je tiens à rappeler que, à en croire le chiffre ahvancé par la
Croatie — à le supposer avéré —, les hostilités qui forment la toile de
fond du présent arrêt ont fait 12 500 morts chez les Croates, alors que,

selon les estimations prudentes du carnage au Rwanda, au moins un
demi‑million de Tutsis et de Hutus modérés ont été tués au cours du géno -
cide qui a eu lieu dans ce pays en 1994, génocide, qui, il convient de le

souligner, a fait l’objet d’un constat judiciaire explicite de la part de la
chambre d’appel du TPIR 35. En effet, je rappelle que la Cour a précisé
dans le présent arrêt qu’elle accorderait la priorité à lha jurisprudence ém- a
nant du TPIY (voir ci-dessus, le paragraphe 13, citant le paragraphe 129

de l’arrêt), et je crois comprendre que cette décision parfaithement raison-
nable est motivée en grande partie par le simple fait que les affairesh por -
tées devant le TPIY avaient trait à des questions historiques, sochiopolitiques

et juridiques beaucoup plus proches de celles soulevées en l’espèhce que de
celles dont le TPIR a eu à connaître.
36. En résumé, si je ne préconise en aucun cas l’importation en bloc de
la jurisprudence du TPIR dans celle de la Cour, je suis préoccupé hpar ce

qui me semble être sur le fond une méconnaissance complète de lh’organe
judiciaire qui s’est le plus employé à interpréter et à ahppliquer la conven -
tion sur le génocide dans l’histoire de l’humanité. Avec tout le respect que

je dois à mes éminents collègues, le fait de n’avoir mêmeh pas consulté
cette ample jurisprudence constitue, selon moi, un manquement par la
Cour à son devoir et à son engagement de tenir compte dans le préhsent
arrêt de l’évolution la plus récente et la plus pertinente qhu’ait connue le

droit relatif au génocide.

34Voir, par exemple, Le Procureur c. Nsengimana, jugement du 17 novembre 2009,

par. 834-836; Le Procureur c. Renzaho, jugement du 14 juillet 2009, par. 768-769; Le
Procureur c.ukundo , jugement du 27février 200par.72, 74 et6; Le Procureur c.cha‑
mih35o, jugement du 12 novembre 2008, par. 333-336, 346-347, 354 et 357.
Cette décision historique a été rendue par la chambre d’appehl à la suite de l’appel
interlocutoire interjeté par le procureur contre la décision relathive au constat judiciaire,
en date du 16juin2006, en l’affaire Le Procureur c.emera, Ngirumpatse etNzirorera,
ICTR-98-44AR73 (C).

432

7 CIJ1077.indb 861 18/04/16 08:54 432 application of genochide convention (sep. ohp. bhandari)

Conclusion on the Courth’s Treatment of post -B osnia
Jurisprudence of the ICTY ahnd ICTR

37. In view of the observations above, I believe that the jurisprudence
of the ad hoc international criminal tribunals in the years following the
issuance of this Court’s Bosnia Judgment demonstrate a dilution of the
rigidly hierarchical tripartite formula for discerning genocidal intent has

promulgated in that precedent, whereby the numerosity of the targeted
population was clearly designed to serve as the pre -eminent concern in
any such calculus. As I have noted above, while there are traces of a mohl-
lified approach to be found in the Judgment’s gentle refastening ofh the

Bosnia formula into a more egalitarian weighing of the three criteria to be
applied in the present Judgment, in my respectful view it was incumbent h
upon the Court to take the further step of explicitly acknowledging and h
engaging the recent, pertinent jurisprudential developments presented byh
the ICTY and ICTR in this area of law and to incorporate, if and where

appropriate, any evolutions to the Bosnia test that are not only strictly
necessary for the disposition of the merits of the present dispute, but h
which may elucidate the development this legal area has undergone over
the past eight years. In other words, even if it were not appropriate orh
even correct to apply such precedents to the facts at bar, in my considered

opinion it was certainly appropriate to at least consider such key develop -
ments, if only to explain why they ought to be distinguished from the
present case. Such an approach, I suggest, would be wholly commensu -
rate with the Court’s role as a pre-eminent global judicial forum to which
other international dispute resolution mechanisms turn in search of guidh-

ance on such important and arcane points of law. Consequently, I regret h
that the majority has missed a prime opportunity to improve the clarity h
and authority of this area of public international law.

The Majority’s Conclusihons on Genocidal Intenth

in the Present Judgmenth

38. In assessing whether the targeted group was “substantial” for pur -
poses of the chapeau of Article II of the Genocide Convention, the Judg-
ment recalls Croatia’s submission

“that the Croats living in the regions of Eastern Slavonia, Western
Slavonia, Banovina/Banija, Kordun, Lika and Dalmatia [who were tar-

geted for destruction by JNA and affiliated Serb forces] constituted a
substantial part of the protected group , and that the intent to destroy
the protected group ‘in part’, which characterizes genocide as defihned

433

7 CIJ1077.indb 862 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 432

Conclusion sur le poihds accordé par la Cour dhe la jurisprudence

du TPIY et du TPIR postérieurhe à l’arrêt rendu en l’haffaire
Bosnie ‑H eRzéGovine c . eRBie et ‑m onténéGRo

37. Compte tenu des observations ci -dessus, je crois que la jurispru-

dence émanée des tribunaux pénaux internationaux adhoc dans les années
qui ont suivi le prononcé par la Cour de son arrêt en l’affaire h Bosnie‑
Herzégovine c. Serbie‑et‑Monténégro met au jour une dilution de la for -
mule tripartite fortement hiérarchisée qu’elle a énoncée dans cet arrêt pour
discerner l’intention génocidaire et dans laquelle l’importanceh numérique

de la population visée était clairement destinée à être lha préoccupation p-ré
dominante. Comme je l’ai mentionné plus haut, même si l’on pheut distin-
guer dans le présent arrêt des signes d’assouplissement dans lah légère
reformulation de la règle élaborée dans l’arrêt rendu en hl’affaire Bosnie‑

Herzégovine c.Serbie‑et‑Monténégro, se traduisant par une mise en balance
plus libre des trois critères à appliquer, il incombait selon moi hà la Cour de
faire un pas de plus en prenant expressément acte de la récente évolution,
exprimée dans la jurisprudence du TPIY et du TPIR, qu’a connue ce

domaine du droit au cours des huit dernières années et d’intéhgrer aux cri-
tères énoncés en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, en
tant que de besoin, tout changement qui serait non seulement nécessaihre
pour trancher quant au fond le présent différend, mais aussi suscephtible de

mettre en lumière cette évolution. En d’autres termes, même hs’il n’était pas
opportun ou même juste d’appliquer ces précédents aux faits de l’espèce, à
mon avis, il était certainement souhaitable d’en examiner au moins les
principaux éléments nouveaux, ne serait-ce que pour expliquer pourhquoi la

présente affaire devait en être distinguée. Une telle démarche, me semble-
t-il, serait pleinement conforme au rôle qui revient à la Cour en hsa qualité
de juridiction mondiale suprême, vers laquelle les autres organes intherna-
tionaux de règlement des différends se tournent lorsqu’ils cherchhent des

orientations sur des points de droit aussi importants et obscurs. Je reghrette
donc que la majorité ait manqué une excellente occasion de clarifiher et de
renforcer ce domaine du droit international public.

Les conclusions de la mhajorité sur l’intenthion génocidaire
dans le présent arrêth

38. Pour déterminer si le groupe visé était «substantiel» aux termes du

passage introductif de l’article II de la convention sur le génocide, il est
rappelé dans l’arrêt que la Croatie a allégué

«que les Croates habitant dans les régions de Slavonie orientale, de Sla‑
vonie occidentale, de Banovina/Banija, de Kordun, de Lika et de Dalma‑
tie [pris pour cible de la campagne de destruction menée par la JNA et
les forces serbes affiliées]stituaient une partie substantielle du groupe

protégé. L’intention de détruire « en partie » le groupe protégé, qui

433

7 CIJ1077.indb 863 18/04/16 08:54 433 application of genochide convention (sep. ohp. bhandari)

in Article II of the Convention, is thus established” (Judgment,
para. 403 ; emphasis added).

The Judgment also recalls that “[i]n its written pleadings, Croatia
defines [the overall protected] group [at issue in its Claim] as the Croat
national or ethnical group on the territory of Croatia, which is not con -

tested by Serbia” (see ibid., para. 205; emphasis added). Relying on offi -
cial census data from 1991 — the year in which the hostilities that are the
subject-matter of the present dispute commenced — adduced by Croatia,
and uncontested by Serbia, the Judgment finds that “the ethnic Croaht

population living in the [identified] regions . . . numbered between 1.7 and
1.8 million [individuals . . . and] constituted slightly less than half of the
ethnic population living in Croatia” (see ibid., para. 406). The Judgment
further concludes “that acts committed by JNA and Serb forces in the
[identified] regions. . . targeted the Croats living in those regions, within

which these armed forces exercised and sought to expand their control”
(ibid). While the majority also found that “as regards the prominence of
that part of the group, the Court notes that Croatia has provided no
information on this point” (ibid.) — a conclusion I do not share, and to
which I shall return presently — “[t]he Court [nevertheless] concludes

from the foregoing that the Croats living in the [identified] regions . . .
constituted a substantial part of the Croat group” (ibid.). Despite my
misgivings about the majority’s pronouncement as to the ostensible lahck
of evidence regarding the prominence of the Croat ethnic group at issue,h
I am in full agreement with the majority’s general conclusion that thhe part

of the ethnic Croat group identified by the Applicant constituted a sub -
stantial part of the overall Croat ethnic group living within the territory
of Croatia during the relevant period.
39. It is to be further recalled that the Judgment concludes that :

“The Court is fully convinced that, in [the] various [identified] loca-l

ities. . . the JNA and Serb forces perpetrated against members of the
protected group acts falling within subparagraphs (a) [killing mem-
bers of the group] and (b) [causing serious bodily or mental harm to
members of the group] of Article IIof the Convention, and that the
actus reus of genocide has been established.” (Ibid., para. 401.)

I am also in complete agreement with the majority on this point. Where Ih
depart from the majority is in the manner of reasoning through which it has

arrived at its conclusion that “[t]he acts constituting the actus reus of geno-
cide within the meaning of Article II (a) and (b) of the Convention were
not committed with the specific intent required for them to be characther-
ized as acts of genocide” (ibid., para. 440). While I again recall that I have
joined the majority in rejecting Croatia’s claim that genocide was com -

mitted against the targeted Croat population on evidentiary grounds, given
that the majority has elected to take Croatia’s case at its highest phrior to
dismissing it, I shall proceed to make certain observations and critiquehs

434

7 CIJ1077.indb 864 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 433

caractérise le génocide tel que défini à l’article II de la Convention,
serait ainsi établie. (Arrêt, par. 403 ; les italiques sont de moi.)

On lit également dans l’arrêt que, « [d]ans ses écritures, la Croatie défi -
nit [l’ensemble du] groupe [protégé dont il est question dans sha demande]
comme le groupe national ou ethnique croate se trouvant sur le territoire de
la Croatie, ce qui n’est pas contesté par la Serbie » (voir ibid., par. 205; les

italiques sont de moi). Se fondant sur les données tirées du dernhier recen -
sement officiel réalisé en 1991 — année où les hostilités qui font l’objet du
présent différend ont commencé —, fournies par la Croatie et non contes-
tées par la Serbie, la Cour constate que « les Croates de souche vivant
dans les régions [considérées] étaient … entre 1,7 et 1,8 million

[et]… représentaient un peu moins de la moitié des Croates de souche
vivant en Croatie » (voir ibid., par. 406). Elle conclut ensuite « que les
actes commis par la JNA et des forces serbes dans les régions [considhé -
rées] visaient les Croates habitant ces régions, dans lesquelles ches forces
armées exerçaient et cherchaient à étendre leur contrôle » (ibid.) . Alors

que la majorité constate aussi que, « concernant la place occupée par la
partie du groupe, la Croatie n’a pas fourni d’information sur ce point »
(ibid.), conclusion que je ne partage pas et sur laquelle je reviendrai à
présent, « [l]a Cour déduit [néanmoins] de ce qui précède que les Croathes
habitant dans les régions [considérées] constituaient une partihe substan -

tielle du groupe des Croates » (ibid.). Malgré mes réserves au sujet de la
position majoritaire quant à l’absence manifeste de preuves, concehrnant
la place occupée par les habitants des régions en question au seinh du
groupe des Croates, je souscris pleinement à la conclusion générale tenant
pour substantielle cette partie de l’ensemble du groupe des Croates dhe

souche vivant sur le territoire de la Croatie pendant la période conshidérée.
39. Il convient de rappeler en outre que dans l’arrêt, la Cour énonhce la
conclusion suivante:

« La Cour est pleinement convaincue que la JNA et des forces
serbes ont commis dans [les différentes] localités [citées] àh l’encontre
de membres du groupe protégé des actes relevant des litt. a) et b) de
l’article IIde la Convention et que l’élément matériel du génocide
(actus reus) est constitué. » (Ibid., par. 401.)

Je suis aussi entièrement d’accord avec la majorité sur ce poinht. Je m’écarte

d’elle, toutefois, en ce qui concerne le raisonnement à l’issue duquel elle est
parvenue à la conclusion que « [l]es actes constituant l’élément matériel du
génocide, au sens des litt. a) et b) de l’article II de la Convention, n’ont
pas été commis dans l’intention spécifique requise pour êhtre qualifiés
d’actes de génocide » (ibid., par. 440). Tout en rappelant une nouvelle fois

que je me suis associé à la majorité pour rejeter les prétenhtions de la Croa -
tie concernant la perpétration d’un génocide contre la populatihon croate
visée en raison du manque de preuves , étant donné que la majorité avait
choisi d’accorder plein crédit à l’argumentation de la Croatie avant de la

434

7 CIJ1077.indb 865 18/04/16 08:54 434 application of genochide convention (sep. ohp. bhandari)

regarding its approach to the analysis of genocidal dolus specialis as per-

tains to Croatia’s allegations.

The Geographic Area Consihdered by the Majority whenh
Assessing d olus s pecialis

40. As I have noted, supra, the Judgment characterizes Croatia’s

delimitation of the relevant “part” of the ethnic Croat group as bheing
“the Croats living in the regions of Eastern Slavonia, Western Slavonhia,
Banovina/Banija, Kordun, Lika and Dalmatia” (Judgment, para. 403).
While I agree with this conclusion as pertains to these six geographical

locales, I am also mindful of the fact that the gravamen of Croatia’sh case
focused heavily on the specific region of Eastern Slavonia, and in parhticu -
lar the city of Vukovar and its environs. As counsel for Croatia submitted
during the oral hearing phase of this case :

“Even when judged against the other atrocities detailed by the Appli‑
cant before this Court . . . the events in Vukovar plumbed new depths.

Serbian forces carried out a sustained campaign of shelling ; system-
atic expulsion ; denial of food, water, electricity, sanitation and med -
ical treatment ; bombing; burning; brutal killings and torture which
reduced the city to rubble and destroyed its Croat population. It started

with roadblocks and ended with torture camps and mass execution.
In human terms, the scar will never heal.

The events at Vukovar are significant and they are known around
the world. They deserve to be examined in context, in detail and in
full.”36

This heavy reliance by Croatia on the events at Vukovar throughout this
case is even conceded by Serbia when it acknowledges in its written
Rejoinder that “[t]he most significant episodes in Eastern Slavonia took

place in Vukovar, and these attract the bulk of the discu37ion in the Reply,
as they did in the Memorial and Counter -Memorial” . Indeed, on more
than one occasion it has been expressly recognized by the Court in the
present Judgment that “Croatia has given particular attention” to h“the

events at Vukovar” in pursuing its claims in this case (Judgment,
paras. 429 and 436).
41. Moreover, there is clear precedent from this Court that an analysis
of genocidal intent may be confined to a geographic area notably smaller

than the six expansive regions considered by the present Judgment, even
if the Applicant framed its cause of action with respect to a wider geog▯raph‑

36CR 2014/8, pp. 28-29, paras. 1-2 (Starmer); emphasis added.
37
Rejoinder of Serbia, para. 370; emphasis added.

435

7 CIJ1077.indb 866 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 434

rejeter, je poursuivrai en formulant certaines observations et critiquesh

concernant la démarche qu’elle a adoptée dans l’analyse du dolus specialis
du génocide dont la Croatie fait état dans ses allégations.

La zone géographique chonsidérée par la majohrité

lors de la recherche duh dolus specialis

40. Comme je l’ai mentionné plus haut, d’après l’arrêt, lah Croatie défi-
nit la «partie» visée du groupe des Croates de souche comme étant « les

Croates habitant dans les régions de Slavonie orientale, de Slavonie hocci-
dentale, de Banovina/Banija, de Kordun, de Lika et de Dalmatie » (arrêt,
par. 403). Même si je suis d’accord avec cette conclusion pour ce qui a
trait à ces six lieux géographiques, j’ai aussi conscience du fait que l’essen-
tiel de la demande de la Croatie portait sur la région de Slavonie orhientale

et en particulier la ville de Vukovar et ses environs. Le conseil de la hCroa -
tie a ainsi fait valoir ce qui suit au cours des audiences :

« Même à l’aune des atrocités décrites à la Cour par le ▯demandeur …
celles commises à Vukovar ont atteint des sommets inimaginables. Les
forces serbes ont mené une campagne soutenue de bombardements,
d’expulsions systématiques, de privation de nourriture, d’eau, d’élec -

tricité, d’hygiène et de soins médicaux, d’incendies, de htueries et
d’actes de torture qui ont réduit la ville en cendres et détruit la popula ‑
tion croate. Cette campagne a commencé par l’installation de bar -
rages routiers et s’est terminée dans les camps de torture et par hles

exécutions de masse. Du point de vue humain, les plaies ne pourront
jamais se refermer.
Les événements à Vukovar sont d’une importance considérabhle et
le monde entier en a eu connaissance. Ils méritent d’être examihnés
dans leur contexte, en détail et dans leur totalité. »36

Même la Serbie s’incline devant cette invocation insistante par lah Croatie,

tout au long de l’affaire, des événements qui se sont produits àh Vukovar,
reconnaissant dans sa duplique que « [l]es événements les plus marquants
du conflit en Slavonie orientale se sont déroulés à Vukovar et [qu’]ils for-
ment la majeure partie des développements dans la réplique, comme hdans
le mémoire et le contre -mémoire »37. D’ailleurs, à plus d’une occasion, la

Cour a reconnu expressément dans le présent arrêt que « la Croatie a[vait]
prêté une attention particulière » au « cas de Vukovar » lorsqu’elle a fait
valoir ses moyens en l’espèce (arrêt, par. 429 et 436).
41. En outre, il est bien établi dans la jurisprudence de la Cour que

l’analyse de l’intention génocidaire peut être restreinte àh une zone géogra -
phique nettement moins étendue que les six vastes régions prises ehn cons-i
dération par le présent arrêt, même si la demande fait état d’une région

36
37 CR 2014/8, p. 28-29, par. 1-2 (Starmer) (les italiques sont de moi).
Duplique de la Serbie, par. 370 (les italiques sont de moi).

435

7 CIJ1077.indb 867 18/04/16 08:54 435 application of genochide convention (sep. ohp. bhandari)

ical area. This was of course precisely what occurred in the 2007 Bosnia
Judgment, wherein the Court made a finding of genocide solely with
respect to Srebrenica, a Bosnian Muslim enclave consisting of upwards of
30,000 people where more than 7,000 military-aged Bosnian Muslim men

were systematically rounded up and executed while the remaining popula-
tion of approximately 25,000 Bosnian Muslims — mostly women, chil -
dren and the elderly — were ethnically cleansed from the enclave 3. It is
to be recalled that this isolated finding of genocide was made in spite of the

Applicant Bosnia and Herzegovina’s much broader allegations of genoci▯de,
which included events in the capital city of Sarajevo, as well as acts thhat
occurred at various other municipalities and camps spread across the terh -
ritory of BiH.

42. In view of these considerations, my ensuing remarks shall confine
themselves to the majority’s analysis of genocidal intent regarding thhe
events at Vukovar. While I must reiterate, for the sake of absolute clarity,

that it is not my contention that genocidal intent was established with h
respect to the events occurring on Croatian soil between 1991 and 1995
(including Vukovar), I steadfastly believe that the majority has failed to
fully and properly canvass the events at Vukovar, being as they are the h

cornerstone of Croatia’s case in the instant proceedings, and thus I hintend
to present additional considerations that I believe the majority was remhiss
in failing to consider when determining whether genocide was perpetratedh
against the Vukovar Croats.

The Siege of Vukovar

43. During the oral hearing phase of these proceedings, Croatia cited
uncontested census statistics indicating that in 1991 Vukovar “had a pop-
39
ulation of just over 21,000 Croats [and] 14,500 Serbs” , whereas “[e]ven
after the peaceful reintegration of the region, only 7,500 of the original
21,500 Croat population of Vukovar in 1991 have ever returned to the
city”40. Counsel for Croatia further averred that during the siege of Vuk -

ovar that lasted from August to November 1991, between 1,100 and
1,700 Croats were killed, whereas after the fall of the city and the ensuing
occupation by JNA and Serb forces, an additional 2,000 Croats were
killed41. I recall and share the majority’s conclusion that the Croat popu -

lation in all six geographic regions relied upon by the Applicant constih -
tutes a substantial part of the overall Croat ethnic group within the

38 Bosnia Judgment, p. 155, para. 278, citing Krstić Trial Judgment, para. 1.

39 CR 2014/8, p. 29, para. 7 (Starmer).
40 Ibid., p. 47, para. 85 (Starmer).
41 CR 2014/12, p. 11 (Starmer).

436

7 CIJ1077.indb 868 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 435

géographique plus vaste. C’est précisément ce que la Cour a fait en 2007
dans l’arrêt qu’elle a rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et
Monténégro, où elle n’a conclu à la commission d’un génocide que pohur
la région de Srebrenica, enclave comptant plus de 30 000 Musulmans de

Bosnie, dont plus de 7000 hommes en âge de porter les armes avaient été
systématiquement raflés et exécutés, alors que le reste, shoit environ
25 000 personnes — pour la plupart des femmes, enfants et personnes
âgées —, était victime du nettoyage ethnique de l’enclave 38. Il convient de

rappeler que cette décision concluant à la perpétration d’un génocide ▯géo ‑
graphiquement limité a été rendue alors que les allégations ▯du demandeur
avaient une portée bien plus large , incluant des faits survenus à Sarajevo,
la capitale, ainsi que des actes commis dans divers autres camps et munih-

cipalités un peu partout sur le territoire de la Bosnie -Herzégovine.
42. Compte tenu de ces considérations, je limiterai mes observations
suivantes à l’analyse que la majorité a faite de l’intentionh génocidaire en
ce qui concerne les événements de Vukovar. Je tiens à réaffihrmer, pour

que ma position soit parfaitement claire, que je ne soutiens pas que l’hin -
tention génocidaire a été établie pour ce qui est des événements qui se
sont produits sur le sol croate entre 1991 et 1995 (y compris à Vukovar).
Je suis cependant absolument convaincu que la majorité n’a pas dûhment

et correctement examiné les événements de Vukovar, alors qu’hils consti -
tuent l’essentiel de l’argumentation de la Croatie dans la préshente espèce.
J’exposerai donc d’autres éléments que la majorité a, à mon avis, eu tort
de ne pas prendre en considération pour déterminer si un génocihde avait

été perpétré contre les Croates de Vukovar.

Le siège de Vukovar

43. Pendant les audiences, la Croatie a cité des statistiques de recense -
ment, non contestées, indiquant que, en 1991, Vukovar comptait « un peu
39
plus de 21 000 Croates [et] 14 500 Serbes » , alors que, « [m]ême après le
retour de la paix dans la région, seuls 7500 des 21 500 Croates qui vivaient à
Vukovar en 1991 y [étaient] revenus» 40. Le conseil de la Croatie a en outre
affirmé que, pendant le siège de Vukovar, qui a duré d’aoûht à novem1b 9r9e1,

de 1100 à1700 Croates avaient été tués, tandis que, après la chute de la vhille,
l’occupation ultérieure par la JNA et les forces serbes avait faith 2000 vic-
times croates supplémentaires 41. Je rappelle et partage la conclusion de la
majorité, qui a constaté que la population croate qui vivait dans hles

six régions géographiques dont a fait état le demandeur constituaith une p -ar
tie substantielle de l’ensemble du groupe des Croates de souche se trhouvant

38 Arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, p. 155, par. 278,
citant le jugement Krstić, par. 1.
39 CR 2014/8, p. 29, par. 7 (Starmer).
40 Ibid., p. 47, par. 85 (Starmer).
41 CR 2014/12, p. 11 (Starmer).

436

7 CIJ1077.indb 869 18/04/16 08:54 436 application of genochide convention (sep. ohp. bhandari)

territory of Croatia. However, I further recall my remarks above that
there is clear precedent for considering a much smaller geographic and

demographic area for the purpose of determining whether that subgroup
constitutes a “substantial” part of the overall group, and hence chonclude
that the Vukovar Croats in and of themselves — in addition to their inclu-
sion in “the substantial” subgroup comprising the six geographic rhegions
as recognized by the Judgment — constituted a substantial part of the

overall ethnic Croat group within the geographical territory of Croatia h
during the relevant period. In this regard, I would recall the Court’hs cha-r
acterization of what constituted a “substantial” part of the targeted group
in question from the following passage of the Bosnia Judgment, in which

certain conclusions of the ICTY Appeals Chamber in the Krstić case were
adopted:

“The Court now turns to the requirement of Article II that there
must be the intent to destroy a protected ‘group’ in whole or in
part . . . the Court recalls the assessment it made earlier in the Judg -
ment of the persuasiveness of the ICTY’s findings of facts and its h

evaluation of them . . . Against that background it turns to the find -
ings in the Krstić case . . . in which the Appeals Chamber endorsed
the findings of the Trial Chamber in the following terms.

‘In this case, having identified the protected group as the natio-
nal group of Bosnian Muslims, the Trial Chamber concluded that
the part . . . targeted was the Bosnian Muslims of Srebrenica, or the
Bosnian Muslims of Eastern Bosnia. This conclusion comports

with the guidelines outlined above. The size of the Bosnian
Muslim population in Srebrenica prior to its capture by the VRS
forces in 1995 amounted to approximately forty thousand people.
This represented not only the Muslim inhabitants of the Srebre -
nica municipality but also many Muslim refugees from the sur -

rounding region. Although this population constituted only a small
percentage of the overall Muslim population of Bosnia and Herze ‑
govina at the time, the importance of the Muslim community of Sre‑
brenica is not captured solely by its size.’

The Court sees no reason to disagree with the concordant findings of
the Trial Chamber and the Appeals Chamber.

The Court concludes that the acts committed at Srebrenica falling
within Article II (a) and (b) of the Convention were committed with
the specific intent to destroy in part the group of the Muslims of Bosni▯a
and Herzegovina as such ; and accordingly that these were acts of
genocide, committed by members of the VRS in and around Srebren -
42
ica from about 13 July 1995.”

42Bosnia Judgment, p. 166, paras. 296-297 (internal citations om; emphasis
added.

437

7 CIJ1077.indb 870 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 436

sur le territoire de la Croatie. Je rappelle aussi néanmoins que, comhme je l’ai
fait observer plus haut, il est parfaitement envisageable de prendre en hconsi-

dération une zone géographique et démographique nettement plus hrestreinte
dans le but de déterminer si le sous -groupe constitue une partie « substan-
tielle de l’ensemble, pour conclure que les Croates de Vukovar en et par
eux‑mêmes, en plus de faire partie du sous -groupe « substantiel» habitant
dans les six régions géographiques reconnues dans l’arrêt, chonstituaient une

partie substantielle de l’ensemble des Croates de souche se trouvant hsur le
territoire de la Croatie pendant la période considérée. A ce suhjet, je souhaite
rappeler comment la Cour a qualifié ce qui constitue une partie « substan -
tielle» du groupe visé en question dans le passage c -i essous de l’arrêt qu’elle

a rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, où elle a
repris certaines conclusions énoncées par la chambre d’appel duh TPIY dans
l’affaireKrstić:

«La Cour abordera maintenant le critère de l’article II selon lequel
doit exister l’intention de détruire en tout ou en partie un « groupe»
protégé … [Elle] rappelle … ce qu’elle a dit plus haut au sujet du
caractère convaincant des conclusions du TPIY concernant les faits

et de l’appréciation qu’il en a donné. Gardant cela à l’hesprit, elle en
vient à présent aux conclusions formulées dans l’affaire Krstić… dans
laquelle la Chambre d’appel a fait siennes les conclusions de la
Chambre de première instance dans les termes suivants :

« En l’espèce, après avoir identifié le groupe protégéh comme
étant le groupe national des Musulmans de Bosnie, la Chambre de
première instance a conclu que la partie du groupe visée … était
celle des Musulmans de Srebrenica, ou des Musulmans de Bosnie

orientale. Cette conclusion est dans le droit fil des lignes direc -
trices esquissées plus haut. Avant la prise de la ville par les forcehs
de la VRS en 1995, Srebrenica comptait environ 40 000 Musul-
mans de Bosnie. Etaient compris dans ce chiffre non seulement les
habitants musulmans de la municipalité de Srebrenica, mais aussi

de nombreux réfugiés musulmans de la région. Si, par rapport à la
population musulmane totale de la Bosnie‑Herzégovine à l’épo▯que
des faits, ce nombre est peu élevé, il ne faut pas se méprendre▯ sur
l’importance de la communauté musulmane de Srebrenica. »

La Cour ne voit pas de raison de s’écarter des conclusions concor -
dantes de la Chambre de première instance et de la Chambre d’appelh.

La Cour conclut que les actes relevant des litt. a) et b) de l’ar -
ticleII de la Convention commis à Srebrenica l’ont été avec l’inten ‑
tion spécifique de détruire en partie le groupe des Musulmans de
Bosnie‑Herzégovine comme tel ; et que, en conséquence, ces actes
étaient des actes de génocide, commis par des membres de la VRS àh
42
Srebrenica et à proximité à partir du 13 juillet 1995.»

42 Arrêt rendu en l’affaire Bosnie‑Herzégovine c.Serbie‑et‑Monténégro, p.166,
par. 296-297 (renvois internes omis ; les italiques sont de moi).

437

7 CIJ1077.indb 871 18/04/16 08:54 437 application of genochide convention (sep. ohp. bhandari)

44. Setting aside the different conclusions as to whether genocidal
intent was proven in the Bosnia Judgment versus the present Judgment,
on the issue of how substantiality was assessed, I believe it would have

been entirely appropriate, given, inter alia, the size of the ethnic Croat
population of Vukovar, the Judgment’s recognition that during the siege
and capture of the city the attack was “directed at the then predominhantly
Croat civilian population (many Serbs having fled the city before or hafter
43
the fighting broke out)” — which, to my mind, rendered the city a
de facto ethnic Croat enclave — and finally its emblematic importance to
the ethnic Croat population within Croatia generally (for reasons of mihl-i
tary strategic importance as a key focal point in the expansive strategyh of

“Greater Serbia”, as expounded supra), for the majority to have con -
ducted a specialized analysis of the attack on Vukovar. While I acknowl -
edge that the attack on Vukovar and its aftermath was considered as parth

of an overarching mélange of factors when evaluating whether genocidal
intent existed with respect to the six geographic territories identifihed in
Croatia’s pleadings, in my respectful view such an analysis lacks clahrity
and coherency and would have been improved by an explicit, separate

examination of the events at Vukovar.

45. As I have painstakingly underscored throughout this opinion, it is
my definitive conclusion that Croatia has failed to satisfy the minimuhm
standard of credible evidence required by this Court to allow me to be

“fully convinced” that a finding of genocidal intent vis-à-vis the protected
ethnic Croat group is the only reasonable inference to be drawn from theh
evidentiary record proffered by the Applicant. Indeed, when pressed by a h
Member of the Court during the oral hearing phase, counsel for Croatia

made the critical concession that the number 44 victims it was alleging h
was difficult to ascertain with precision , which I find to epitomize the
many probative shortcomings of the Applicant’s cause of action. This h
position having been reaffirmed, and again following the majority’s elec -

tion to take Croatia’s figures at their highest, I am somewhat puzzhled by
the lack of analysis as to why the averred killing of upwards of 3,000 eth-

43Judgment, para. 218:

“The Court will first consider the allegations concerning those killed during the
siege and capture of Vukovar. The Parties have debated the number of vichtims,
their status and ethnicity and the circumstances in which they died. Theh Court need
not resolve all those issues. It observes that, while there is still somhe uncertainty
surrounding these questions, it is clear that the attack on Vukovar was hnot confined
to military objectives ; it was also directed at the then predominantly Croat civilian
population (many Serbs having fled the city before or after the fightin▯g broke out).”

(Emphasis added.)

44CR 2014/12, pp. 11-12 (Starmer).

438

7 CIJ1077.indb 872 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 437

44. Indépendamment des conclusions différentes auxquelles a donné
lieu la question de savoir si l’intention génocidaire avait étéh prouvée dans
l’arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro par

rapport au présent arrêt, s’agissant de la manière dont le caractère sub -
stantiel a été apprécié, je crois qu’il aurait été hparfaitement justifié que la
majorité procède à une analyse spécifique de l’attaque hde Vukovar,
compte tenu notamment de la taille de la population croate qui s’y trhou-

vait, de la reconnaissance dans l’arrêt que, pendant le siège et la prise de
la ville, l’attaque «a … visé la population civile, composée alors en bonne
partie de Croates (de nombreux Serbes ayant fui la ville avant ou lorsque
les combats ont éclaté) » 43 — ce qui, à mon avis, a fait de la ville une

enclave croate de fait — et enfin de l’importance symbolique de ce groupe
pour la population croate de l’ensemble de la Croatie (pour des raisons de
stratégie militaire dans l’optique de la mise en œuvre du projeht expansion -

niste de « Grande Serbie », comme il est exposé plus haut). Même si je
reconnais que l’attaque de Vukovar et ses retombées ont été hprises en
considération parmi un ensemble de facteurs lors de l’examen de lah ques-
tion de savoir si une intention génocidaire existait à l’égahrd des six terri -

toires dont il est fait état dans les pièces de procédure de lah Croatie, à
mon avis cette analyse manque de clarté et de cohérence, et aurait gagné
à ce que les événements de Vukovar fassent l’objet d’un ehxamen exprès et
distinct.

45. Comme je me suis efforcé de le souligner tout au long du présent
exposé, ma conclusion globale est que la Croatie n’a pas réussih à remplir
le critère d’établissement de la preuve imposé par la Cour phour que je

puisse être «pleinement convaincu » que l’existence d’une intention géno -
cidaire à l’égard du groupe croate protégé est la seule chonclusion raison-
nable à tirer des éléments de preuve présentés par le demhandeur. En effet,
sur l’insistance d’un membre de la Cour pendant les audiences, le conseil

de la Croatie est allé jusqu’à concéder que le nombre de vichtimes44u’avan -
çait le demandeur était difficile à vérifier avec précision , ce qui me
semble caractéristique des nombreuses déficiences qui entachent les
moyens de preuve du demandeur. Cette position ayant été réaffihrmée, et

tout en me ralliant là aussi au choix qu’a fait la majorité d’haccorder plein
crédit aux chiffres produits par la Croatie, je suis quelque peu perplhexe

43Arrêt, par. 218:

«La Cour considérera d’abord les allégations concernant les pershonnes tuées
au cours du siège et de la prise de Vukovar.s Parties ont débattu des questions
du nombre de ces victimes, de leur statut et ethnicité, ainsi que des circonstances
dans lesquelles elles sont mortes. La Cour n’a pas à trancher toutes ces questions.
Elle constate que, s’isubsiste certaines incertitudes sur celles -ci, il est indéniable
que l’attaque contre Vukovar ne s’est pas limitée à des objehctifs militaires, mais a
aussi visé la population civile, composée alors en bonne partie de▯ Croates (de nombreux

Serbes ayant fui la ville avant ou lorsque les combats ont éclaté). » (Les italiques sont
de moi.)
44CR 2014/12, p. 11-12 (Starmer).

438

7 CIJ1077.indb 873 18/04/16 08:54 438 application of genochide convention (sep. ohp. bhandari)

nic Croats in Vukovar out of a pre -war population of 21,500 would not
constitute sufficient physical destruction of the group pursuant to Arthi -
cle II (a) of the Convention to satisfy the “quantitative element” as

adopted by the present Judgment. While there may be good reasons for
such a negative finding, the paucity of analysis conducted by the majohrity
to this end is discouraging.

46. In addition to my misgivings regarding the majority’s application

of the quantitative element regarding the number of Vukovar Croats
allegedly killed during and after the siege of that city, Croatia has prhe -
sented a series of 17 contextual factors which, in its estimation,

“constitute a pattern of conduct from which the only reasonable infe- rh
ence to be drawn is that the Serb leaders were motivated by genocidal
intent . . . [and which], individually or taken together, could lead the

Court to conclude that there was a systematic policy of targeting
Croats with a view to their elimination from the regions concerned”
(Judgment, para. 408).

Consequently, “[a]ll these elements indicate, according to Croatia, the
existence of a pattern of conduct from which the only reasonable infer -
ence is an intent to destroy, in whole or in part, the Croat group” (hibid.,
para. 409). For ease of reference, these factors have been reproduced in

their totality at paragraph 408 of the present Judgment.
47. In the Judgment, the majority has determined

“that of the 17 factors suggested by Croatia to establish the existence
of a pattern of conduct revealing a genocidal intent . . . the most
important are those that concern the scale and allegedly systematic
nature of the attacks, the fact that those attacks are said to have
caused casualties and damage far in excess of what was justified by
military necessity, the specific targeting of Croats and the nature,

extent and degree of the injuries caused to the Croat population (i.e.,h
the third, seventh, eighth, tenth and eleventh factors identified [by h
Croatia])” (ibid., para. 413).

Regrettably, the majority provides no ratio for this critical distinction,
and consequently excludes as “less important”, without any justifihcation,
factors such as “the political doctrine of Serbian expansionism whichh cre -

ated the climate for genocidal policies aimed at destroying the Croat pohp -
ulation living in areas earmarked to become part of ‘Greater Serbia’h
[Croatia’s first factor]”, “the statements of public officiahls, including
demonization of Croats and propaganda on the part of State -controlled
media [Croatia’s second factor]”, “the explicit recognition by the JNA

that paramilitary groups were engaging in genocidal acts [Croatia’s fifth
factor]”; and “the fact that during the occupation, ethnic Croats were
required to identify themselves and their property as such by wearing

439

7 CIJ1077.indb 874 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 438

devant l’absence d’analyse des raisons pour lesquelles le meurtre hsupposé
de plus de 3000 Croates à Vukovar, sur une population de 21 500 Croates
avant guerre, ne constituerait pas, au sens du litt. a) de l’article II de la
Convention, une destruction physique suffisante pour satisfaire au critère

de l’« élément quantitatif » retenu dans le présent arrêt. Pour fondée que
puisse être cette constatation négative, l’indigence de l’anhalyse à laquelle
la majorité a procédé pour y parvenir est décourageante.
46. En sus de mes réserves au sujet de l’appréciation par la majorihté de
l’élément quantitatif en ce qui concerne le nombre de Croates dhe Vukovar

qui auraient été tués pendant et après le siège de cette ville, je rappelle que
la Croatie a présenté une série de 17 critères qui, selon elle, permettent
d’établir

«une ligne de conduite dont la seule déduction raisonnable serait que h
les responsables serbes étaient animés d’une intention génochidaire …
[et] qui, pris individuellement ou ensemble, pourraient … conduire la
Cour à conclure qu’il existait une politique systématique consihstant à
prendre les Croates pour cible en vue de les éliminer des régions h

concernées » (arrêt, par.408).
En conséquence, « [l]’ensemble de ces éléments dénoterait, selon la Croa -

tie, l’existence d’une ligne de conduite dont la seule déductiohn raisonnable
serait l’intention de détruire, en tout ou en partie, le groupe dehs Croates
(ibid., par. 409). Par souci de commodité, ces critères ont été reproduihts
dans leur totalité au paragraphe 408 du présent arrêt.
47. Dans son arrêt, la majorité a conclu

«que, parmi les 17 critères proposés par la Croatie pour établir l’exis-
tence d’une ligne de conduite traduisant une intention génocidaire,
les plus importants sont ceux qui ont trait à l’ampleur et au carahctère

systématique des attaques, au fait que ces attaques auraient fait biehn
plus de victimes et de dégâts que ce qui était nécessaire d’hun point de
vue militaire, au fait que les Croates étaient spécifiquement prhis pour
cible et à la nature, à la gravité et à l’étendue des hlésions infligées à la
population croate (c’est -à-dire les troisième, septième, huitième,
dixième et onzième critères énumérés [par la Croatie])h » ( ibid.,

par. 413).
Malheureusement, la majorité ne dit pas pourquoi elle opère cette distinc -

tion capitale, et exclut en conséquence, sans aucune justification,h des cri-
tères qu’elle qualifie de « moins importants », tels que « la doctrine
politique de l’expansionnisme serbe, qui a créé les conditions hpropices à la
mise en œuvre de politiques génocidaires visant à détruire lha population
croate dans les zones appelées à faire partie de la « Grande Serbie » [pre-
mier critère de la Croatie]», «les déclarations de personnalités publiques,

notamment la diabolisation des Croates et la propagande par les médias
sous contrôle de l’Etat [deuxième critère de la Croatie] », « la reconnais-
sance expresse, par la JNA, de ce que des groupes paramilitaires se
livraient à des actes génocidaires [cinquième critère de la hCroatie]» et « le

439

7 CIJ1077.indb 875 18/04/16 08:54 439 application of genochide convention (sep. ohp. bhandari)

white ribbons tied around their arms and by affixing white cloths to thheir
homes [Croatia’s ninth factor]”, to name a few (see Judgment, parha. 408).

48. While an exhaustive treatment of how these factors may be, con -
trary to the view of the majority, “more important” in decipheringh geno-

cidal intent lies beyond the scope of the present opinion, I must admit I
find myself flummoxed by some of these exclusions. One need only loohk to
readily available historical examples to find scenarios where such fachtors
clearly and unequivocally played a major role in inciting and perpetu -
ating incipient and ongoing genocides. To that end, I would briefly rehcall

the Nazi expansionist political doctrine of Lebensraum (which would
fall neatly under the rubric of Croatia’s first factor) and their ghettho -
ization of marginalized groups through the forced wearing of religiouslyh
denoted attire (e.g., armbands bearing the “Star of David”) for hthe Jews
of occupied Europe (for which one can find many commonalities in Croah -

tia’s ninth factor). To take a more recent historical example, I would note
the undeniable role played by popular media (especially radio) up to ahnd
during the Rwandan genocide in the promotion of a demagogic “Hutu
Power” ideology that sought to vilify and ostracize the Tutsi ethnic h
minority population through the ubiquitous use of the epithet of“inyenzi”

(cockroaches) and other comparable slurs (which aligns with Croatia’hs
second factor). In each of these three examples, the averred acts are nhot,
strictly speaking, genocidal per se in accordance with Article II of the
Genocide Convention, but for the majority to rather summarily dismiss
their potency as precursors to or indicia of genocidal intent is, to my h

mind, both puzzling and troubling. Finally, how “the explicit recognition
of genocidal intent of those carrying out the acts” (Croatia’s fihfth factor)
does not figure prominently into the equation of genocidal intent is shim -
ply beyond me.

The Distinction betweehn Criminal Intent and Mohtive

49. In the present Judgment “the Court notes that in the Mrkšić case,
the ICTY found that the attack on [Vukovar] constituted a response to
the declaration of independence by Croatia, and above all an assertion of
Serbia’s grip on the SFRY” (see para. 429). The Judgment then repro -

duces the following block quotation from Mrkšić in support of this con -
clusion :

“The declaration of Croatia of its independence of the Yugoslav
Federation and the associated social unrest within Croatia was met
with determined military reaction by Serb forces.It was in this pol itical

440

7 CIJ1077.indb 876 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 439

fait que, sous l’occupation, les membres du groupe ethnique croate éhtaient
tenus de s’identifier comme tels, de même que leurs biens, en pohrtant un
ruban blanc autour du bras et en attachant un drap blanc à leurs habihta-

tions [neuvième critère de la Croatie] », pour n’en citer que quelques -uns
(voir arrêt, par. 408).
48. Il serait hors de propos de procéder ici à un examen exhaustif de hla
façon dont ces critères peuvent, contrairement à ce qu’a dith la majorité,

être « plus importants » pour discerner l’intention génocidaire, mais je
dois admettre que je suis déconcerté par certaines de ces exclusiohns. L’his -
toire regorge d’exemples montrant que ces critères ont clairement het sans
équivoque joué un rôle majeur dans l’incitation à commetthre le génocide
et dans la perpétuation de génocides naissants ou en cours. Ainsi,h je rap-

pellerai brièvement la doctrine politique expansionniste du Lebensraum
du régime nazi (qui s’inscrirait parfaitement sous le premier crihtère de la
Croatie) et sa ghettoïsation de groupes marginalisés (qui préhsente de
nombreux points communs avec le neuvième critère de la Croatie), hfor -
çant les juifs de l’Europe occupée à porter des signes à hconnotation reli -

gieuse (par exemple, des brassards marqués de l’étoile de David). Pour
prendre un exemple historique plus récent, pensons au rôle incontehstable
qu’ont joué les médias populaires (surtout la radio), avant eht pendant le
génocide rwandais, dans la promotion d’une idéologie démagogique du
« pouvoir Hutu » acharnée à vilipender et à ostraciser la minorité tutsihe

par l’emploi omniprésent de l’épithète «inyenzi» (cafards) et d’autres
insultes comparables (ce qui coïncide avec le deuxième critèreh de la Croa -
tie). Dans aucun de ces trois exemples, les actes supposés ne sont, hà pro -
prement parler, génocidaires en eux -mêmes au sens de l’article II de la
convention sur le génocide, mais le fait que la majorité ait écharté assez

sommairement l’importance qu’ils pouvaient revêtir en tant que précur-
seurs ou indices de l’intention génocidaire est pour moi aussi éhnigmatique
que troublant. Enfin, je ne m’explique tout simplement pas que «h la recon-
naissance expresse de l’intention génocidaire des auteurs des actehs » (cin-

quième critère de la Croatie) ait pu être mise de côté dhans la recherche de
l’intention génocidaire.

La distinction entre lh’intention et le mobihle criminels

49. Dans le présent arrêt « la Cour note que, dans l’affaire Mrkšić …,
le TPIY a constaté que l’attaque contre [Vukovar] constituait une hréponse
à la proclamation d’indépendance de la Croatie, et surtout une haffirma -
tion de la mainmise de la Serbie sur la RFSY » (voir par. 429). L’arrêt

reproduit ensuite à l’appui de cette conclusion la citation suivanhte tirée du
jugement Mrkšić :

«Les forces serbes ont riposté militairement avec détermination àh
la proclamation par la Croatie de son indépendance et aux troubles
sociaux qui s’en sont ensuivis sur son territoire. C’est dans ce contexte

440

7 CIJ1077.indb 877 18/04/16 08:54 440 application of genochide convention (sep. ohp. bhandari)

scenario that the city and people of Vukovar and those living in close
proximity in the Vukovar municipality became a means of demonstrat ‑
ing to the Croatian people, and those of other Yugoslav Republics, the

harmful consequences of their actions . In the view of the Chamber the
overall effect of the evidence is to demonstrate that the city and civil▯ian
population of and around Vukovar were being punished, and terribly so,
as an example to those who did not accept the Serb ‑ controlled Federal
45
Government in Belgrade.”

As a brief aside, the quoted passage from Mrkšić, which forms part of the
uncontested evidentiary record in this case,is positively laden with explicit
references to the emblematic nature of the Vukovar Croats vis -à-vis the
remainder of the ethnic Croat population, thus only further weakening

the majority’s assertion that Croatia “has provided no informationh” as to
“the prominence of that part of the group” of the ethnic Croat pophula -
tion that it contends was targeted for genocide 46.

50. However, returning to the point under consideration, the majority
relies on the quoted passage from Mrkšić to conclude that

“[i]t follows from the above, and from the fact that numerous Croats h
of Vukovar were evacuated . . . that the existence of intent to physi -
cally destroy the Croat population is not the only reasonable inference h

that can be drawn from the illegal attack on Vukovar” (Judgment,
para. 429).

In my view, this line of reasoning appears to conflate the distinct lehgal
concepts of motive and intent in finding that the “punishment” of the
Vukovar Croats could preclude a finding that they were targeted with

genocidal intent. To this end I would recall the language of the ICTY
Appeals Chamber in the Krnojelac Judgment, which recalled

“its case law in the Jelisićcase which, with regard to the specific intent
required for the crime of genocide, set out ‘the necessity to distinguish
specific intent from motive. The personal motive of the perpetrator of

the crime of genocide may be, for example, to obtain personal eco -
nomic benefits,or political advantage or some form of power. The exist-
ence of a personal motive does not preclude the perpetrator from also
having the specific intent to commit genocide ’.”47

45
46Mrkšić Trial Judgment, para. 471; emphasis added.
Judgment, para. 406. While the Judgment was referring more generally to the ethnic
Croat population in the six geographical areas of Croatia alleged by the Applicant, it
stands to reason that Vukovar, being not only situated within the areas hcontemplated but
constituting the gravamen of Croatia’s case, would constitute at leasht some evidence of the
prominence of at least a part of the targeted group in question.

47Prosecutor v. Krnojelac, Appeals Judgment, 17 September 2003, para. 102; emphasis
added.

441

7 CIJ1077.indb 878 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 440

politique que la ville de Vukovar, ses habitants et ceux des environs
immédiats de la municipalité de Vukovar ont servi d’exemple pou▯r
montrer aux Croates et aux autres Républiques yougoslaves à quelle▯s

conséquences fâcheuses ils s’exposaient par leurs actions. De l’avis de
la Chambre, les éléments de preuve montrent dans l’ensemble que la
punition terrible infligée à Vukovar et à la population civile ▯de la ville
et des environs avait valeur d’exemple pour ceux qui n’acceptaient▯ pas
45
le gouvernement fédéral de Belgrade contrôlé par les Serbes. »

Si je puis me permettre un bref aparté, le passage cité du jugemenht Mrkšić,
qui fait partie des éléments de preuve incontestés dans la préhsente affaire,
déborde littéralement de références expresses au caractère représentatif
des Croates de Vukovar par rapport au reste de la population croate, ce h

qui ne fait qu’affaiblir encore l’affirmation de la majorité sehlon laquelle la
Croatie «n’a pas fourni d’information» concernant «la place occupée par
la partie du groupe » au sein de la population croate qu’elle affirme avoir
été la cible d’une intention génocidaire 46.

50. Quoi qu’il en soit, pour revenir au point considéré, la majorithé
s’appuie sur le passage précité du jugement Mrkšić pour conclure que

«[i]l découle de ce qui précède, ainsi que du fait que de nombrehux
Croates de Vukovar ont été évacués …, que l’intention de détruire
physiquement la population croate n’est pas la seule conclusion rai -

sonnable que l’on puisse déduire de l’attaque illégale de Vuhkovar »
(arrêt, par. 429).

A mon avis, ce raisonnement confond les notions juridiques distinctes
du mobile et de l’intention lorsqu’il en est conclu que la « punition» des
Croates de Vukovar pourrait empêcher de conclure que ceux -ci étaient

visés par une intention génocidaire. A cette fin, je citerai la hchambre d’ap -
pel du TPIY, qui a rappelé ce qui suit dans l’arrêt Krnojelac :

« sa jurisprudence dans l’affaire Jelisić, selon laquelle s’agissant de
l’intention spécifique requise pour le crime de génocide «h il est néces-
saire de distinguer entre l’intention spécifique et le mobile. Le mobile

personnel du génocidaire peut, par exemple, être la perspective d’un
profit économique personnel, d’avantages politiques ou d’une certaine
forme de pouvoir. L’existence d’un mobile personnel n’empêche pas
que l’auteur soit également animé de l’intention spécifiq▯ue de perpétrer
47
un génocide ».»

45
46Jugement Mrkšić, par. 471 (les italiques sont de moi).
Arrêt, par. 406. Alors que l’arrêt renvoyait plus généralement à la populahtion croate
de souche vivant dans les six régions géographiques de la Croatie hmentionnées par le
demandeur, il va de soi que Vukovar, qui non seulement était située dans l’une des régions
considérées mais constituait l’élément fondamental de la hdemande de la Croatie, tendrait à
établir, au moins dans une certaine mesure, l’importance d’au moins une partie du groupe
en question.
47Le Procureur c. Krnojelac, arrêt du 17 septembre 2003, par. 102 (les italiques sont
de moi).

441

7 CIJ1077.indb 879 18/04/16 08:54 441 application of genochide convention (sep. ohp. bhandari)

Similar language for this proposition can be found in a number of other h
Judgments pronounced by the ICTY 48and ICTR 49. In view of this dis -

tinction, I find the Judgment’s analysis of the motivation underlyihng the
attack on Vukovar to be problematic, as it fails to account for the posshi-
bility, as clearly stipulated in the aforementioned authorities, that gehno -
cidal intent may exist simultaneously with other, ulterior motives. In this

regard, I would recall the finding in Popović that the massacre at Sre -
brenica enclave was in part motivated by the strategic advantage of uni -
ting a “Greater Serbia”. Never was it suggested that this tacticalh
motivation precluded the attack from possessing genocidal intent. Conse-

quently, I am unpersuaded by the Judgment’s dismissal of genocidal
intent vis -à-vis Vukovar based on the finding that the attack was ani -
mated by political and/or retributive motives, and respectfully but fihrmly

believe that the majority has committed a basic error of law in findinhg
that the existence of a punitive motive for the attack on Vukovar pre -
cludes genocidal intent as “the only reasonable inference that can beh
drawn from the illegal attack” (Judgment, para. 429).

Discretion of the ICTY Proshecutor in Laying a Charghe
of Genocide

51. I recall that in the Bosnia Judgment, the Court determined that :

“[A]s a general proposition the inclusion of charges in an indict -

ment cannot be given [evidentiary] weight. What may however be
significant is the decision of the Prosecutor, either initially or in han
amendment to an indictment, not to include or to exclude a charge
of genocide.” 50

No legal authority whatsoever is cited for the rationale underlying the hdi-s

parate probative weight that the Court decided to afford the ICTY Prose-
cutor’s decision to include or exclude a charge of genocide in an indhictment,

48 See, e.g., Prosecutor v. Blaškić, Appeals Judgment, 29 July 2004, para. 694.

“Mens rea is the mental state or degree of fault which the accused held at the
relevant time. Motive is generally considered as that which causes a perhson to act.
The Appeals Chamber has held that, as far as criminal responsibility is hconcerned,
motive is generally irrelevant in international criminal law . . .”
49
See, e.g., Prosecutor v. Karera, Trial Judgment, 7 December 2007, para. 534. (“The
perpetrator need not be solely motivated by a genocidal intent and having a personal
motive will not preclude such a specific intent.”)

50 Bosnia Judgment, p. 132, para. 217.

442

7 CIJ1077.indb 880 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 441

On retrouve ce principe exprimé dans des termes similaires dans plusiheurs
autres jugements et arrêts prononcés par le TPIY 48 et le TPIR . Compte

tenu de cette distinction, j’estime que l’analyse que la Cour a faite dans
l’arrêt de la motivation sous -tendant l’attaque de Vukovar pose pro -
blème, car elle ne tient pas compte de la possibilité, clairement hmention -
née dans les sources précitées, que l’intention génocidaire puisse exister

simultanément avec d’autres mobiles sous‑jacents. A ce sujet, je rappellerai
qu’il a été conclu dans l’affaire Popović que le massacre perpétré dans
l’enclave de Srebrenica avait été en partie motivé par l’hintérêt stratégique
de constituer la « Grande Serbie ». Il n’a jamais été envisagé que cette

motivation tactique ait pu empêcher les auteurs de l’attaque de nohurrir
une intention génocidaire. En conséquence, je ne puis souscrire àh la posi -
tion majoritaire concluant à l’absence d’intention génocidaihre en ce qui

concerne Vukovar, sous prétexte que les auteurs de l’attaque éthaient ani-
més par des mobiles à caractère politique et/ou punitif, et je hsuis persuadé,
avec tout le respect que je lui dois, que la majorité a commis une erhreur de
droit fondamentale en décidant que l’existence d’une motivation punitive

dans l’attaque de Vukovar empêchait d’inférer que l’intenhtion génocidaire
était « la seule conclusion raisonnable que l’on puisse déduire de l’athtaque
illégale» (arrêt, par. 429).

Le pouvoir discrétionhnaire du procureur du hTPIY à l’égard
du chef de génocide

51. Je rappelle que, dans son arrêt rendu en l’affaire Bosnie‑Herzégo ‑
vine c. Serbie‑et‑Monténégro, la Cour a conclu que

«l’on ne saurait, en règle générale, accorder de poids au faiht que tel

ou tel chef figure dans un acte d’accusation. Ce qui, en revanche, h
peut être important, c’est la décision prise par le procureur, hd’emblée
ou par modification de l’acte d’accusation, de ne pas inclure ouh de
retirer le chef de génocide. »50

Aucune source, quelle qu’elle soit, n’est citée pour justifiehr la différence de

valeur probante que la Cour a décidé d’accorder à la décihsion du procu -
reur du TPIY de retenir ou non le chef de génocide dans un acte

48 Voir, par exemple, Le Procureur c. Blaškić, arrêt du 29 juillet 2004, par. 694:

«L’élément moral est le dol ou degré de faute au moment des fhaits, alors que
l’on considère que le mobile est généralement ce qui pousse hune personne à agir. La
Chambre d’appel a estimé que, pour ce qui est de la responsabilité pénale, le mobile
était généralement indifférent en droit pénal internationahl…»
49
Voir, par exemple, Le Procureur c. Karera, jugement du 7 décembre 2007, par. 534:
«L’auteur des crimes ne doit pas être mû uniquement par une intehntion génocidaire et
l’existence d’un mobileersonnel n’empêche pas que l’auteur soit également animéh de cette
intention spécifique.»
50 Arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro, p. 132, par. 217.

442

7 CIJ1077.indb 881 18/04/16 08:54 442 application of genochide convention (sep. ohp. bhandari)

nor does the Bosnia Judgment offer any reasoned explanation for this dis -

tinction. Indeed, in my respectful view such a distinction is unsustainahble as
a matter of basic logical construction, and in contrast to the majority hI find
myself drawn to the poignant submission of counsel for Croatia, who
argued during the oral hearing phase of this case that in accordance withh
the prevailing rules of procedure obtaining at that tribunal,

“[T]he judicial arm of the ICTY will review each indictment, includinhg
the charges that have been included, and has the power to dismiss any
count not supported by the evidence. But the judicial arm has no way
of reviewing the charges that have not been included, or the reasons
for non-inclusion. It would therefore be illogical to afford greater

evidential weight to an unreviewable decision without reasons not to
include a charge, than the reviewable decision to include a charge.” 51

Moreover, I believe that Croatia has raised cogent arguments exposing
the various political, logistical and other constraints that may animateh an

exercise of prosecutorial discretion not to lay a criminal charge, incluhd -
ing : (1) the availability (or lack thereof) of evidence at the onset of pro -
ceedings ; (2) the focus of a criminal prosecution on individual accused,
often in relation to very circumscribed crime sites, rather than the muchh

broader question of State responsibility for genocide encompassing largeh
geographical expanses ; (3)the lack of any obligation falling on the ICTY
Prosecutor to provide reasons for not laying a charge ; (4) the need to
selectively employ the finite resources of that Tribunal, especially ihn view
of the massive institutional constraints imposed by the United Nations

Security Council’s imposition of a “Completion Strategy” mandating the
completion of all the Tribunal’s work by fixed dates ; and (5) the fact that
whereas decisions to include a charge are subject to judicial review, dehci -
sions not to include a charge are not 5.

52. In light of these trenchant insights, and in view of the Court’s pro -
nouncement in Bosnia that the lack of probative value for a decision to
lay a charge of genocide constitutes “a general proposition” rathehr than a

definite rule, in my respectful view the jurisprudence of this Court whould
be fortified by a more expansive treatment of this subject. Alas, givehn the
opportunity to clarify the Court’s position concerning prosecutorial hdis -
cretion in the present Judgment, the majority has apparently elected to h

demur. Instead of a reasoned account that explains the distinction, the
Judgment makes the following pronouncement :

51
52CR 2014/6, p. 39 (Starmer); emphasis in original.
Ibid., pp. 33-42 (Starmer).

443

7 CIJ1077.indb 882 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 442

d’accusation, et l’arrêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑ ‑ et

Monténégro n’offre pas davantage d’explication pour justifier cette dis -
tinction. J’estime que celle-ci est d’ailleurs indéfendable du hpoint de vue
de la logique élémentaire et, contrairement à la majorité, je ne suis pas
insensible à l’argument particulièrement parlant du conseil de la Croatie,
qui a fait valoir au cours des audiences que, selon le règlement de procé -

dure du Tribunal,

«l’organe juridictionnel du TPIY procède à l’examen de chaqueh acte
d’accusation, ainsi que des inculpations qui ont effectivement été
retenues, et dispose du pouvoir de rejeter celles qui ne seraient pas
étayées par les éléments de preuve. Il n’a au contraire ahucun moyen
d’examiner les inculpations qui ont été écartées ni les raisons pour

lesquelles elles l’ont été. Il serait par conséquent illogiqhue d’accorder
une valeur probante plus grande à la décision, insusceptible de
contrôle et non motivée, d’écarter une accusation qu’à celle, suscep -
tible de contrôle, de retenir une inculpation. »51

En outre, j’estime que la Croatie a fait valoir des arguments convainhcants
en exposant les diverses contraintes politiques, logistiques et autres phou-

vant conduire le procureur à exercer son pouvoir discrétionnaire de ne
pas retenir un chef d’accusation, comme 1) la disponibilité (ou l’absence)
d’éléments de preuve à l’ouverture de la procédure ; 2)le fait que la pro -
cédure pénale soit ciblée sur la responsabilité pénale inhdividuelle et géné-

ralement sur des lieux de crimes très circonscrits, et non sur la quehstion
bien plus générale de la responsabilité de l’Etat à l’hégard d’un génocide de
portée géographique plus vaste ; 3)le fait que le procureur ne soit nulle -
ment tenu de justifier sa décision de ne pas retenir tel ou tel chehf ; 4) la
nécessité d’user avec parcimonie des ressources limitées du Tribunal,

compte tenu en particulier des contraintes institutionnelles considérhables
découlant de la « stratégie d’achèvement des travaux » imposée par le
Conseil de sécurité de l’Organisation des Nations Unies, qui établit un
échéancier pour l’achèvement de tous les travaux du Tribunalh ; et 5)le fait

que, contrairement à la décision de retenir tel ou tel chef d’ahc52sation,
celle de l’écarter n’est assujettie à aucun contrôle judihciaire .
52. A la lumière de ces considérations décisives, et étant donné que la
Cour a dit dans l’arrêt qu’elle a rendu en l’affaire Bosnie‑Herzégovine
c. Serbie‑et‑Monténégro que l’absence de valeur probante de la décision

de retenir le chef de génocide constituait une « règle générale» et non une
règle absolue, il me semble que la jurisprudence de la Cour gagnerait à ce
que la question soit traitée de manière plus approfondie. Hélash, alors
qu’elle avait la possibilité de clarifier dans le présent arrhêt la position de

la Cour concernant le pouvoir discrétionnaire du procureur, la majorihté a
apparemment choisi de s’abstenir. Au lieu de motiver la distinction opé -
rée, elle s’est prononcée dans les termes suivants :

51
52CR 2014/6, p. 39 (Starmer) (les italiques sont dans l’original).
Ibid., p. 33-42 (Starmer).

443

7 CIJ1077.indb 883 18/04/16 08:54 443 application of genochide convention (sep. ohp. bhandari)

“The fact that the Prosecutor has discretion to bring charges does
not call into question the approach which the Court adopted in its 2007
Judgment . . . The Court did not intend to turn the absence of charges

into decisive proof that there had not been genocide, but took the
view that this factor may be of significance and would be taken into
consideration. In the present case, there is no reason for the Court to
depart from that approach. The persons charged by the Prosecutor

included very senior members of the political and military leadership
of the principal participants in the hostilities which took place in Cro-
atia between 1991 and 1995. The charges brought against them
included, in many cases, allegations about the overall strategy adopted h
by the leadership in question and about the existence of a joint crim -

inal enterprise. In that context, the fact that charges of genocide wereh
not included in any of the indictments is of greater significance thanh
would have been the case had the defendants occupied much lower
positions in the chain of command. In addition, the Court cannot fail
to note that the indictment in the case of the highest ranking defendant▯

of all, former President Milošević, did include charges of genocide in
relation to the conflict in Bosnia and Herzegovina, whereas no such
charges were brought in the part of the indictment concerned with the
hostilities in Croatia.” (Judgment, para. 187 ; emphasis added.)

Not only does this purported defence of the Bosnia distinction skirt the
central issue by failing to provide a single rationale as to why the dechision

to include a charge of genocide in an indictment ought to be given differ -
ential weight than a decision to exclude such a charge, but the example of
the Milošević case relied upon by the Judgment to prove its point in fact
tends to defeat its own position. As that juxtaposition plainly illustrahtes,
if the decision not to charge Milošević with genocide in respect of crimes

committed in respect of Croatia is noteworthy, then surely the same musth
be said of the corollary decision to charge him with genocide in respect of
crimes committed in Bosnia and Herzegovina. To my mind, these are two
sides of the same coin and the draft’s failure to make heads or tailsh of its

quizzical distinction, by invoking a litany of irrelevant considerationsh,
leaves me unmoved.

53. In sum, through its belaboured attempt to justify the distinction
regarding the differential probative value afforded the inclusion or excluh-

sion of charges of genocide in an indictment, which to this day fails toh cite
a single germane legal authority and which poignantly avoids engaging
any of the Applicant Croatia’s arguments, the majority has not, to myh
satisfaction, explained the logically and legally problematic distinctiohn it
first iterated in the Bosnia Judgment and has now reiterated in the present

Judgment. I can only express my regret at this missed opportunity.

444

7 CIJ1077.indb 884 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 443

«Le fait que le procureur dispose d’un pouvoir discrétionnaire de
poursuite ne remet pas en cause l’approche que la Cour a adoptée d▯ans
son arrêt de 2007 . En effet, elle n’a pas entendu faire de l’absence de

poursuite une preuve décisive de l’inexistence du génocide, maihs elle
a estimé qu’il pouvait s’agir d’un élément important àh prendre en
considération. En la présente affaire, il n’y a pas de raisons qui
devraient conduire la Cour à s’écarter de cette approche. Parmi les

personnes inculpées par le procureur figuraient de très hauts rehspon-
sables politiques et militaires des principales parties prenantes aux
hostilités qui s’étaient déroulées en Croatie entre 1991 et 1995. Dans
nombre de cas, les accusations portées à leur encontre se rappor -
taient à la stratégie globale qu’ils avaient mise en œuvre ainsi qu’à

l’existence d’une entreprise criminelle commune. Dans ce contexte,h
l’absence systématique du chef de génocide dans les actes d’haccusa -
tion les concernant revêt davantage d’importance que cela n’aurhait
été le cas s’ils avaient occupé des positions inférieuresh dans la chaîne
de commandement. Par ailleurs, la Cour ne peut manquer de relever

que, dans l’acte d’accusation dressé à l’encontre de l’▯accusé le plus haut
placé, l’ancien président Milošević, le chef de génocide avait bien été
retenu en ce qui concerne le conflit en Bosnie‑Herzégovine, alors qu’▯il
était absent dans la partie se rapportant aux hostilités dont la C▯roatie
avait été le théâtre. » (Arrêt, par.87 ; les italiques sont de moi.)

Non seulement cette prétendue défense de la distinction opéréhe dans l’ar-
rêt rendu en l’affaire Bosnie‑Herzégovine c. Serbie‑et‑Monténégro élude la

question centrale en omettant de donner une seule raison pour laquelle ihl
convient d’accorder à la décision de retenir un chef de génocide dans un
acte d’accusation un poids différent de celui accordé à celleh d’exclure un
tel chef, mais l’exemple de l’affaire Milošević cité dans l’arrêt pour étayer
cette affirmation tend en réalité à démontrer le contraire.h Comme il res -

sort clairement de la confrontation des deux situations, si la décision de ne
pas inculper Milošević de génocide pour les crimes commis à l’égard de la
Croatie mérite de retenir l’attention, alors il faut certainement hen dire
autant de la décision corollaire de l’inculper de génocide pour les crimes

commis en Bosnie -Herzégovine. Selon moi, il s’agit des deux faces d’une
même médaille et, faute d’explication quant à cette distincthion curieuse, la
litanie de considérations non pertinentes invoquées me laisse scephtique.
53. En résumé, dans sa tentative laborieuse de justifier la distincthion
opérée quant à la valeur probante différente devant être ahccordée à la

décision de retenir le chef de génocide et à celle de l’éhcarter dans un acte
d’accusation, sans invoquer la moindre source pertinente et en évihtant
résolument de répondre aux arguments de la Croatie, la majoritéh n’a pas
expliqué d’une manière assez convaincante à mes yeux cette dhistinction,
tout aussi singulière en droit que d’un point de vue logique, qu’helle a

d’abord opérée dans l’arrêt rendu en l’affaire Bosnie‑Herzégovine c.Serbie‑
et‑Monténégro puis reprise dans le présent arrêt. Je ne peux que faire part
de mon regret devant cette occasion manquée.

444

7 CIJ1077.indb 885 18/04/16 08:54 444 application of genochide convention (sep. ohp. bhandari)

Conclusion

54. For the reasons I have explained at length throughout the course
of this opinion, while I share the majority’s conviction that the Applicant
Croatia has not discharged its evidentiary burden in relation to the sech -
ond operative clause of this Judgment, I have felt compelled to voice myh
many (and at times strenuous) objections to the manner in which the

majority has treated the issue of genocidal intent as regards the claimsh put
forward by Croatia. Given my tepid support for the second operative
clause, which is based primarily on evidentiary concerns, there are manyh
aspects of the reasoning employed by the Judgment en route to the con -

clusion contained in that dispositive paragraph that I would distance
myself from as a jurist. Perhaps most disconcerting is that the foregoinhg
does not constitute an exhaustive exposition of my dissatisfaction with h
the Judgment’s approach to genocidal dolus specialis, but merely a survey
of some of my more salient concerns.

(Signed) Dalveer Bhandari.

445

7 CIJ1077.indb 886 18/04/16 08:54 application de convehntion génocide (op. inhd. bhandari) 444

Conclusion

54. Pour les motifs que j’ai expliqués dans le détail ci-dessus, alhors que
je partage la conviction de la majorité selon laquelle le demandeur nhe s’est
pas acquitté de la charge de la preuve qui lui incombait en ce qui concerne
le deuxième point du dispositif du présent arrêt, je me suis sehnti obligé
d’exprimer mes nombreuses (et parfois vives) objections à la manhière dont

la majorité a traité la question de l’intention génocidaire heu égard aux
arguments avancés par la Croatie. Compte tenu du peu d’enthousiasmhe
que j’éprouve envers ce même point du dispositif, lequel est fohndé princi-
palement sur des considérations relevant de l’administration de lah preuve,

je préfère me tenir à distance, en tant que juriste, de nombreuhx aspects du
raisonnement appliqué dans l’arrêt jusqu’à aboutir à lha conclusion y figu-
rant. Plus troublant encore est peut-être le fait que ce qui précède consti -
tue non pas un exposé exhaustif de l’insatisfaction que suscite enh moi
l’analyse du dolus specialis propre au génocide, mais simplement un

aperçu de certaines de mes principales préoccupations.

(Signé) D alveer Bhandari.

445

7 CIJ1077.indb 887 18/04/16 08:54

Document file FR
Document Long Title

Separate opinion of Judge Bhandari

Links