The Gotovina Omnishambles

Friday’s judgment in Gotovina and Markac by the ICTY Appeals Chamber (summary; judgment), in which it by 3 votes to 2 reversed a unanimous Trial Chamber and acquitted the defendants, is a disaster at almost every level. I say this not as an aggrieved Serb lamenting the selectiveness of international justice and its failure to punish crimes against his own people – I have long since developed antibodies to all forms of nationalism, including the very virulent type thriving on self-victimization, and I have no personal axe to grind here. I say this rather as an international (human rights) lawyer who has always thought of the ICTY as an indispensable, if imperfect, instrument of justice for the atrocities of the Yugoslav conflicts. That said, how and why then is the Gotovina appeals judgment so bad? To my mind, the problem is not with the acquittal as such – even though as far as public opinion in the former Yugoslavia is concerned the bottom line is all that mattered. Rather, the problem is with the process, the reasoning, the appearances, and the broader repercussions that all these will have.

First, with regard to process: as the dissents by Judges Agius and Pocar correctly point out, the majority make a complete mess of the appellate standards for review. Readers will recall that in the common law-inspired procedure of the ICTY the main task of the Appeals Chamber is to correct errors of law made by the Trial Chamber. The Trial Chamber is owed deference with regards to its findings of fact, which are not to be disturbed lightly on appeal, but only if no reasonable trier of fact could have made the relevant finding on the strength of the record. In short, unlike in most continental systems, the appellate process should not amount a retrial, a de novo examination of the entire case. This ensures both procedural economy and the integrity of the exhaustive fact-finding process in the trial court.

While the majority endorses these standards as they are set out in the ICTY’s long-established jurisprudence, it does not actually follow them – to the extent that its approach to standards of review is actually even discernible, as I will now explain. The whole case ultimately turned around the Trial Chamber’s unfortunate finding that in assessing the shelling by the Croatian artillery of the four Serb towns in the separatist Serb entity in Croatia, chief of them Knin, any shell that fell further than 200 meters from a legitimate military target in the towns should be presumptively considered as evidence of an unlawful indiscriminate attack. The Appeals Chamber was actually unanimous that this rigid standard was not supported by the evidence in the trial record and was not given adequate reasons for by the Trial Chamber.

So far so good. But what the majority then does with this finding turns appellate review on its head. The majority does not explain whether the Trial Chamber’s error with regard to the 200 meter standard itself was an error of law or an error of fact. If it was the former, the majority would have had to articulate a new, proper legal standard for testing the facts established in the record, upon whose application we could know whether the shelling of Knin and the other towns was indiscriminate or not. If it was the latter, the majority would have had to pay due deference to the totality of the factual findings made by the Trial Chamber and should only have disturbed them if no reasonable trier of fact could have found that the shelling was indiscriminate on the basis of all of the evidence in the record.

But this is not what the majority of the Appeals Chamber in fact did. It rather treated the Trial Chamber’s error with regard to the 200 meter standard as a fatal flaw that unraveled the entire trial judgment, and proceeded with a rather bizarre form of de novo review of the facts on the apparent basis of the Trial Chamber’s failure to provide adequate reasoning as a legal error. It then went on to examine each item of evidence in clinical isolation, determined that essentially all other evidence depended on the core issue of the 200 meter standard, even though the Trial Chamber never assigned that standard such importance in the trial judgment itself, and proceeded to discard one by one almost all of the Trial Chamber’s principal factual findings (the majority does most of this work in paras. 64-67 of the judgment).

Thus, in the majority’s view the Trial Chamber’s error with regard to the 200 meter standard made it impossible to establish beyond a reasonable doubt whether the shelling of Knin and the other Serb towns was indiscriminate or not. In effect, the majority thought there was no standard that could supplant the erroneous 200 meter standard. The consequence of this agnosticism was the majority holding that no reasonable trier of fact could have found that Knin was shelled indiscriminately, even though, as Judge Agius pointed out in his dissent, it was undisputed that over 900 artillery projectiles fell on Knin in the course of one and half days in the absence of any resistance from the town itself (Agius dissent, para. 18). The majority similarly overturned the Trial Chamber’s finding that the distribution of projectile impacts in Knin could not be explained by the Croatian army engaging moving targets of opportunity. In this respect, as Judge Agius correctly observed (Agius dissent, para. 32), the majority’s reasoning becomes simply contradictory:

With respect to the one police car that was hit in Knin, the Majority assumes that HV [Croatian Army] artillery weaponry could be so accurate as to obtain a direct hit, but with regard to all of the military targets which had been pre-established with proper co-ordinates, the Majority effectively gives the HV the benefit of the doubt ad infinitum. I would be enlightened by an explanation from the Majority as to how, if the HV could be so accurate with regard to a moving object, it could miss military targets by hundreds of metres?

(Note the tone of the dissent, to which I will turn later).

The majority then went on to give the 200 meter standard a key role in its assessment of the existence of a joint criminal enterprise on the part of the Croatian leadership to ethnically cleanse the Croatian Serbs. In its view, the Trial Chamber considered unlawful shelling to have been an indispensable component of this enterprise, as this was the primary way of forcibly displacing civilians. Accordingly, as there was no way to establish beyond a reasonable doubt that the shelling was unlawful, there was also no way for a reasonable trier of fact to establish that a JCE even existed. In doing so, the Appeals Chamber essentially imputed to the Trial Chamber views that it had never adopted, and overturned yet one more factual finding by using de novo review of individual items of evidence taken in isolation, which it found lacking in the absence of the context of unlawful shelling. These included the transcripts of the meetings of the Croatian leadership at Brioni, evidence of numerous individual crimes on the ground perpetrated by Croatian forces, inflammatory speeches by Croatian president Tudjman, and the subsequent implementation of discriminatory measures designed to prevent the return of Serbs to Croatia.

With the JCE gone, the convictions had to be vacated. In a rather remarkable turn of irony, the majority then refused to enter convictions on alternate grounds (such as aiding and abetting), which were not pleaded in the original indictment or considered by the Trial Chamber, finding that this would effectively amount to a retrial and would exceed the proper scope of appellate review – which is precisely what they did in the remainder of the judgment.

Of course, there is nothing wrong with de novo appellate review as such, if we forget for a moment the majority’s assessing of individual items of evidence in isolation rather than in their totality. The problem here is one of institutional design. Not only did the majority of the Appeals Chamber think that they better appreciated the facts of the case than the Trial Chamber, but they did so from the absolutely passive position of not remedying in any way the (alleged) deficiencies in the evidentiary process at trial, which is what de novo review in a continental system would do. If, for instance, the evidence collected at trial really made it impossible to conclusively establish whether the shelling was indiscriminate, a court exercising proper de novo review could have reopened evidentiary proceedings and called new witnesses and experts. In the way it went about the matter, however, the majority combined the worst of both worlds.

So much about the law. What makes matters worse is that the majority’s reasoning is lacking in all practical wisdom. It not only overturns the Trial Chamber’s factual findings lightly, it does so casually, in a couple of paragraphs for issues where the trial judgment ran at hundreds of pages, and with apparent disregard for the consequences.  Perhaps the individualized guilt of Gotovina and Markac really wasn’t there. Perhaps they should have been acquitted because they were not members of the JCE or their contribution to it was not a substantial one, or convicted on different grounds and given a lesser sentence. Without having sat in court for the many months of trial and without looking in detail at the whole record one simply cannot be sure, and those who say they are nonetheless are more likely than not victims of their own prejudice. But even if Gotovina and Markac were to be acquitted, was it truly necessary for the majority to disregard the considered views of the Trial Chamber on so many issues, and especially with regard to the existence of the JCE?

This is not only where the majority’s reasoning is especially problematic, but where it is most consequential. Note that people in Croatia and Serbia didn’t really care about the two generals as individuals; what they did care about was about how the Tribunal characterized the systemic nature of the crimes (or not), and it is here that the Appeal Chamber’s decision is at its most dramatic. From a unanimous Trial Chamber declaring that the highest ranks of the Croatian leadership, including President Tudjman, formed a joint criminal enterprise with the purpose of ethnically cleansing Serbs from Croatia, to a divided, 3 to 2 decision by the Appeals Chamber that no reasonable trier of fact could have found that JCE to exist on the evidence heard by the Trial Chamber. Not only is this outcome hard to rationally explain to non-specialists, it only serves to harden the conflicting nationalist narratives in Croatia and Serbia.

In Croatia, the appeals judgment is conclusive evidence that the war they fought with the Serbs was not only defensive and just, but also pure and unsullied. Yes, there were some crimes perpetrated against the Serbs, but this was done by a few bad apples, to an extent understandably inspired by revenge, not the Croat leadership and state apparatus. Rather than being ethnically cleansed, the Serbs left Croatia willingly, only because their own leaders told them to do so in order to make the Croats look bad in the eyes of the world. In Serbia, the judgment only confirms the perpetual victim narrative – the ICTY and the international community never really cared about crimes against Serbs, and the Tribunal has shown itself to be nothing more than a political court. Its credibility is now not just in tatters – it is nil. In assessing these two narratives, if that is even possible, the Croatian, victorious brand of nationalism is even more poisonous and harder to cure. To the Serbs, whose leadership were objectively undoubtedly the greatest villains of the wars of the 1990s, these wars ultimately brought nothing but misery and defeat. To the Croats, however, the war was a foundational event in the creation of their modern state, with all its accompanying paraphernalia, so that even those on the more liberal side of the political spectrum, like President Ivo Josipovic, have to pay it due obeisance, and have indeed done so in the wake of the generals’ acquittal.

The entrenchment of these narratives means that in Croatia in particular, but also in Serbia, there will be no appetite for any further attempts for reconciliation or the search for a common, if possibly objective, truth. Each will continue to live in their own bubble, their own little alternate reality. Civil society attempts to establish a regional truth and reconciliation commission (REKOM), whose success was unlikely to begin with, will now probably falter no matter how laudable they substantively are. Voices dissenting from the two official narratives will continue to be marginalized, and ever more so.

And finally, it is at the same time both fascinating and deeply disturbing to observe how the gloating Croats and crying Serbs (see this BBC report for some flavor) could so easily have found their respective roles reversed, if only one judge of the Appeals Chamber had changed his mind. While the Croatian nationalists will always say that a majority is a majority and that is that, dismissing the dissents completely, Serbs will also equally be able to point to the 5 international judges (3 judges of the unanimous Trial Chamber and 2 judges of the minority on appeals) who looked at the same record and voted to convict. Add to this the sheer vehemence of the disagreement between the majority and the minority, which is probably unprecedented in the ICTY’s jurisprudence. I personally cannot recall a judgment in which dissenting judges so bluntly stated that the majority was misinterpreting, ignoring and dismissing the findings of the Trial Chamber (Judge Agius), characterized the majority’s reasoning as ‘grotesque’ (Judge Pocar), suggested that its reasoning had other than purely legal motives (Pocar dissent, para. 30), and characterized the judgment as ‘contradict[ing] any sense of justice’ (ibid, para. 39). It is indeed a collective failure of the ICTY as an institution, and a professional failure of all those involved in the case, regardless of their good intentions, to have had a decision as important as this one ultimately decided by 3 votes to 2, and in such poor form.