DECLARATION OF JUDGE SHAHABUDDEEN

I agree with the judgment but, as I had made a reservation in a previous case concerning the subject of international armed conflict, I state below my views on some aspects of that matter as it relates to this case.

The main question concerns the test to be applied where the impugned state is alleged to have acted through a foreign military entity, as distinguished from direct military intervention. Paragraph 27 of a joint opinion of Judges Vohrah and Nieto-Navia spoke of "overall control"; that was in Aleksovski, decided by a Trial Chamber on 25 June 1999. A test similarly named appeared in the judgment rendered, shortly afterwards, by the Appeals Chamber in Tadic on 15 July 1999. It was in respect of this test that I entered the reservation above-mentioned. The present Trial Chamber applies this test. Institutionally, I agree with the Trial Chamber in following the judgment of the Appeals Chamber on the point. Individually, however, I have a different approach and would like to preserve it, although it leads to the same conclusion as that produced in this case by the overall control test.

A prefatory consideration is that, as it was said by Judge Gros, "a change in terminology does not suffice to avoid a problem".¹ The problem of determining whether a state is in armed conflict with another state through a foreign military entity is not resolved by asking whether there is "overall control" unless this term is understood by reference to a criterion which enables the court to decide whether such control exists in the particular circumstances of any case.

It seems to me that this criterion is to be found by asking what is the degree of control which is effective in any set of circumstances to enable the impugned state to use force against the other state through the intermediary of the foreign military entity concerned. It is only that standard which can help to determine whether "overall control" for one purpose is "overall control" for another. For example, what is overall control for the purpose of committing breaches of international humanitarian law need not be overall control for the purpose of using force short of committing such breaches, as is possible. It was in respect of the former purpose (which is not germane here) that Nicaragua spoke of the need for specific instructions.² I do not read that case as intended to say that, as a matter of law, proof of specific instructions is necessarily required where the question is whether a state was using force through a foreign military entity short of committing breaches of international humanitarian law, which is all that the prosecution has to prove in support of its proposition that an "armed conflict" had arisen between two states within the meaning of Article 2, first paragraph, of the Fourth Geneva Convention.

In all cases, the juridically meaningful criterion is whether there is effective control. Whether there is such control has to be considered in the light of the particular purpose in view. This means that it may be necessary to distinguish between the criterion and evidence of satisfaction of the criterion. In some cases , proof of satisfaction of the criterion may conceivably require evidence of specific instructions; in others it may not. But, whatever may be the evidence required to satisfy it, the criterion remains that of effectiveness. There is no way of eliminating the concept; it is both constant and ineradicable. Any needed flexibility is provided by its operation.

Thus, a criterion of universal applicability displaces a number of tests (apparent in the Tadic appeal judgment) to determine what is but a single issue, namely , whether the acts of others (individuals or groups, organised or not) are acts of the impugned state – whether a neighbouring state or a distant state. In the result, it may be doubted whether analysis is promoted by adding the word "overall " to the word "control" unless recourse is had to a criterion of effectiveness to determine whether there is overall control in any circumstances. A test is not by itself sufficient and is not therefore adequate if, to apply it in particular circumstances, it is necessary to have recourse to another test of more fundamental applicability.

As mentioned above, as a matter of institutional correctness, I agree with the Trial Chamber in following the Appeals Chamber as to the test to be applied. In my respectful individual opinion, however, the effective control test to be extracted from the judgment of the International Court of Justice in Nicaragua is sound . The defence was right in contending for that test, but I cannot see that , in the circumstances of this case, it produces effects which differ from those resulting from the overall control test.

A second question concerns the argument of the defence that the ICRC-sponsored agreement of 22 May 1992 showed that the parties, together with the ICRC, regarded the conflict as internal, and not as international.³ The argument is based on the fact that, under the agreement, the parties committed themselves to abide by the substantive rules of internal armed conflict set out in common Article 3 of the Geneva Conventions and in addition agreed, on the strength of the third paragraph of that article, to apply certain provisions of the Conventions concerning international armed conflict.

It appears to me that the proposition that the ICRC must have held the view that the conflict was internal is not the same as the proposition that the ICRC must have held that it was only internal. The Appeals Chamber itself recognised that the same armed conflict may have both internal and international aspects.⁴ As the Appeals Chamber put it:

"Taken together, the agreements reached between the various parties to the conflict in the former Yugoslavia bear out the proposition that, when the Security Council adopted the Statute of the International Tribunal in 1993, it did so with reference to situations that the parties themselves considered at different times and places as either internal or international armed conflicts, or as a mixed internal-international conflict". ⁵

It was therefore open to the parties to agree on the regulation of their conflict in so far as it was internal without thereby excluding the possibility that it also had international aspects. An admission that it had these latter aspects was not likely to be made so long as any intervening external state was in a position to exercise influence on key parties to the agreement; if there had to be such an admission , there might have been no agreement. Likewise, it was competent for the ICRC to encourage the parties to regulate the conflict in so far as it was internal without thereby necessarily taking the position that the conflict was only internal .

Further, the position was an evolving one. The agreement was made on 22 May 1992 . Even if the armed conflict was exclusively internal on that date, it did not follow that it remained so. From paragraph 26 of their opinion in Aleksovski , it appears that the majority in that case accepted that the conflict became international but that they considered that this happened after the end of the period covered by the indictment in that case. Their view of the time (seemingly, April 1993) with effect from which the conflict became international does not of course bind this Trial Chamber.

A third question concerns a view that there could have been no international armed conflict in this case because, if there were, that would lead to the absurdity of holding that persons having the nationality of Bosnia and Herzegovina who were held by the HVO were protected persons on the ground that the HVO was acting for a foreign state⁶ whose nationality the victims did not have, but that such persons were not protected persons if held by the ABiH since they would have the nationality of the state for which the latter was acting. In the result, atrocities would be punishable as grave breaches in the former case, but not in the latter.

The difficulty can be overcome by argument to the effect that, on one ground or another, in a particular situation Bosnian Croat victims of the ABiH did not in fact have the nationality of Bosnia and Herzegovina for purposes of the Fourth Geneva Convention, recourse being had to the broad objects of the relevant provisions. But it is right to recognise that those possibilities have a limit beyond which the victims indisputably have the same nationality as that of the state in whose hands they find themselves. What then, when that limit is reached?

The hypothesis given above in relation to victims of the HVO assumes a common nationality of the victims and those by whom they were directly victimised, the former being nevertheless protected persons on the ground that the latter were acting as agents of a state whose nationality the victims did not have. That agency element is missing in a case in which citizens of a state are victimised by the armed forces of the same state who are acting only for that state and not for another. In such a situation , the victims have the nationality of the party in whose hands they find themselves and are not protected persons.

The resulting difference between the two situations is obvious. But the lack of symmetry is superficial and does not attract an absurdity argument. In one case, the captors would have been acting as agents of a foreign state; not so in the other . The law itself is symmetrical. If a party puts itself within the reach of its sanctions for a reason which does not apply to another, there can be no complaint on the ground of inequality in the operation of the law. In the result, there is no such absurdity as to justify rejection of the view that there was an international armed conflict in this case.

A fourth and last question concerns a submission by the defence that, in accordance with the Tadic Interlocutory Appeal on Jurisdiction decision rendered by the Appeals Chamber on 2 October 1995 and the judgment given by the Trial Chamber in Aleksovski on 25 June 1999, the mere intervention of foreign troops in a conflict between local forces does not convert the conflict from a local one into an international one. This holding, said the defence, denies the judgment rendered in Celebici on 16 November 1998 and the review decision given in the Rajic Indictment case on 13 September 1996, when the opposite position was taken . ⁷

I understand the submission to contemplate this situation: There is an armed conflict between a secessionist group and the government of the state. A foreign state intervenes militarily in support of the secessionist group and is resisted by the local state . The external military intervention clearly constitutes an armed conflict between states for the purpose of making the Fourth Geneva Convention applicable. But does the internal conflict itself become an armed conflict between states?

The answer is in the affirmative if the foreign state assumes control over the secessionist group such that the use of force by the secessionist group becomes a use of force by the foreign state against the local state, thereby giving rise to an armed conflict between states within the meaning of Article 2, first paragraph , of the Fourth Geneva Convention.

Where that is so, a remaining question is whether the local armed conflict can retain any part of its original internal character. It seems to me that an affirmative answer is indicated both by principle and by authority. As to principle, it is difficult to see why an ongoing internal armed conflict should suddenly and necessarily lose that character altogether because of foreign intervention. The circumstance that a secessionist group is using force under the control of a foreign state does not necessarily mean that it cannot be also using the same force on its own account . As to authority, words used in the Tadic Interlocutory Appeal on Jurisdiction decision suggest the possibility of dual characterisation. For example, paragraph 72 of that decision spoke of the armed conflict in that case being "characterised as both internal and international" or as having "some combination" of internal and international aspects.

Thus, I think the defence is right to the extent that foreign intervention does not of necessity deprive an internal armed conflict of its internal character altogether – it may, but it may not. However, I do not think it is necessary to pursue the matter further. The conclusion of the Trial Chamber, with which I respectfully agree , is that, in the particular circumstances of the case, Croatia was always in overall control of the HVO; that conclusion is unqualified. No room is left for a finding that the armed conflict was in part internal or to explore the legal implications of such a finding.

Done in both English and French, the English text being authoritative.

Mohamed Shahabuddeen

Dated this third day of March 2000
At The Hague
The Netherland

1 - Gulf of Maine, I.C.J. Reports 1984, p. 363, para. 6, dissenting opinion.
2 - Nicaragua v United States of America, I.C.J. Reports 1986, p. 14.
3 - For supporting argument, see Tadic Interlocutory Appeal on Jurisdiction, (1994-1995)1 ICTY JR, p. 433, para. 73, and para. 16 of the joint opinion in Aleksovski, IT-95-14/1-T of 25 June 1999.
4 - Tadic Interlocutory Appeal on Jurisdiction, (1994-1995), 1 ICTY JR 357, p. 433, para. 72; and see ibid., p. 441, para. 77.
5 - Ibid., p. 435, para. 73.
6 - See Tadic Interlocutory Appeal on Jurisdiction, p. 439, para. 76; and see and compare Prosecutor v. Delalic, IT-96-21-T, 16 November 1998, paras. 245-266.
7 - T.25248-9, Mr Nobilo.