Guest Post: The Complexity of International Trials Is Necessary

by Stuart Ford

[Stuart Ford is an Assistant Professor at The John Marshall Law School.]

International criminal trials are extremely complex. The average trial at the International Criminal Tribunal for the former Yugoslavia (ICTY) takes 176 trial days and involves more than 120 witnesses and 2,000 exhibits. See here at table 2. In comparison, the average criminal trial in the United States takes less than one day, and even the average murder trial takes only three or four days. Id. at 53-55. As a result, there is a widespread belief that international criminal trials are too complex, and international tribunals have come under enormous pressure to reduce that complexity. See here at Part I.

The ICTY, for example, made a number of changes to the Rules of Evidence and Procedure that were intended to reduce trial complexity. See, for example, here. Professors Langer and Doherty found that those changes failed to reduce the complexity of the ICTY’s trials, but why? The answer is important because if we understood what drove the complexity of international trials, perhaps we could find ways to reduce their complexity (and the associated cost) without undermining the purposes of international criminal justice.

My latest project attempts to answer that question by taking the complexity data I collected for my earlier work on the efficiency of international criminal courts and using it to build a model of trial complexity. Trial complexity is the response variable in the model, while the explanatory variables were based on a number of hypotheses about what might cause trial complexity. The hypotheses are summarized below:

Hypotheses
H1 Complexity increases as the number of accused tried together increases
H2 Complexity increases as the seniority of the accused in the political and military hierarchy increases
H3 Complexity increases as the total number of counts in the indictment increases
H4 Complexity increases as the number of crime sites in the indictment increases
H5 Complexity increases if the accused are charged with genocide
H6 Complexity increases is the accused are charged as members of a joint criminal enterprise (JCE)
H7 Complexity increases if the accused are charged under a theory of superior responsibility
H8 Complexity decreases if the accused are charged as a direct perpetrators

The association between the hypotheses and trial complexity was then tested using a multiple regression model. The results of the regression are presented below:

Model Results
Variable Effect Size Significant
Total Accused   0.077 Yes
Seniority   0.065 Yes
Total Counts   0.0061 Yes
Crime Sites   0.0006 No
Genocide   0.066 No
JCE   0.048 No
Superior Responsibility -0.088 No
Direct Perpetrator -0.20 Yes

The results suggest that international tribunals will have a very hard time reducing the complexity of their trials. First of all, the number of crime sites is not significant in the model. Thus simply permitting the judge to impose limits on the number of crime sites in the indictment will probably not be successful. The number of counts in the indictment is statistically significant, but the magnitude of the effect is very small, indicating that any reduction in the overall complexity from imposing limits on the number of counts would also be small. Neither the mode of liability used to prosecute the accused or the legal qualification of the charge had a statistically significant effect on trial complexity either.

In contrast, two factors are both statistically significant and have a large impact on the resulting trial complexity. The most important factor is the accused’s seniority within their respective military or political hierarchies. The complexity that resulted from increasing the accused’s seniority by one level was approximately equivalent to adding an additional ten counts to the indictment. Direct perpetration also had a large impact on overall complexity. Accused who were alleged to be direct perpetrators of violence (i.e., they carried it out themselves), had significantly shorter trials.

The figure below shows the relative contributions of the various factors to the complexity of the median ICTY case. The seniority of the accused and whether the accused was a direct perpetrator account for the majority of the resulting trial complexity.

sford

So, what does this all mean? The results suggest that the key factor driving the complexity of international criminal trials is the geographic and organizational distance of the accused from the crimes they are alleged to be responsible for. Indirect perpetration, where the accused is alleged to be legally responsible for crimes that were physically carried out by others, is a hallmark of international criminal trials. Such individuals tend to be distant, both organizationally and geographically, from the violence that results. (In the model, this distance is captured by the seniority and direct perpetration variables.) As a result, proving that they are criminally responsible for the acts of the direct perpetrators is very difficult and accounts for the bulk of the trial’s complexity.

This has implications for the future of international trials. For at least the last ten years, international tribunals have sought to reduce trial complexity by tinkering with the rules of procedure and evidence. This is unlikely to ever be successful because changes in the procedure cannot change the accused’s seniority or whether that person is a direct perpetrator.

There are some ways that trial complexity could be reduced, but they all come with fairly serious drawbacks. For example, courts could significantly reduce complexity by trying only low-level direct perpetrators. For policy reasons, however, international courts have been encouraged to focus on the most senior leaders. See, for example, here at 71-74. The result is very complex and expensive trials.

Making international criminal law a strict liability regime would also probably reduce complexity significantly by reducing the difficulty of linking accused to crimes from which they are organizationally and geographically distant. The cost, however, would be too high. Strict liability crimes are only appropriate when the violation is not associated with strong moral condemnation and the penalties are small. Serious violations of international criminal law, however, involve both stiff penalties and strong moral condemnation. Importing strict liability into international criminal law would be extremely undesirable, even if it did dramatically reduce trial complexity.

Another possibility would be to embrace symbolic charging at international tribunals. My calculations (see here at 38-42) indicate that using symbolic charging rather than representative charging would have reduced trial complexity at the ICTY by, at best, about a quarter. At the same time, it would have made it significantly harder for the court to achieve its goals by limiting its ability to inform the historical record, promote post-conflict reconciliation, and help victims find closure. These are important goals of international criminal justice and they are probably not worth compromising for a relatively modest reduction in trial complexity.

Finally, an increased use of plea bargaining might be another way to reduce overall trial complexity by simply avoiding the need to have some trials. Unfortunately, it is unlikely to have a significant impact because the cases involving the most senior leaders are the cases least likely to be resolved through a plea bargain and simultaneously the largest source of trial complexity. Prosecutors, for instance, are probably reluctant to enter into a plea deal with the individuals they believe masterminded the crimes. At the same time, senior accused are more likely to see their prosecutions in political terms and thus less likely to accept a plea deal. Indeed, the majority of plea bargains at the ICTY were accepted by low to mid-level accused.

The last ten years have seen most international tribunals focus their efforts on the most senior leaders, almost none of whom are direct perpetrators of violence.   The unsurprising result is trials of enormous complexity. Moreover, this complexity is largely out of the hands of individual judges and prosecutors. It arises from the policy decision to focus on senior leaders, and the model suggests it cannot be meaningfully changed by tinkering with the rules of procedure and evidence. The cost and complexity of international criminal trials is a necessary consequence of that policy decision.

http://opiniojuris.org/2015/05/03/guest-post-the-complexity-of-international-trials-is-necessary/

7 Responses

  1. Thanks for a very interesting post . Of course there is a very complicated moral and philosophical problem with the system :

    Everywhere , systems would work no less for avoiding a situation were an innocent person , is sent to prison , over , conviction of guilty perpetrator , for this is the meaning of justice , and just system . We need to invest more sometimes , for avoiding such wrongdoing , over the declared purpose of the system :

    Conviction of felons .

    You are right , probably ” remote perpetrator ” demands more resources , why ?? it has always been so :

    Dealing with the mental element of the crime , and that one of the accused , is far greater complicated , and doesn’t necessarily matter , whether the accused has such or such seniority , it is the occupation with mental state , wich typically can complicate thinks.

    But I can come up with a suggestion if you’d considere it :

    As you know , every system has that hierarchy of : clerkship , registry , pre trial , trial . in cases which are particularly complicated , it would make it easier , if the weight or work load would be shifted towards administrative discretion and preparation for final discretion , over higher posts or links of final ones ( senior judges ) .I think worth a try or thinking .Thanks

  2. Thanks Professor Ford — I am going to check out your paper and download it soon. I wonder if you compared trials in national jurisdictions dealing with similar level crimes?

  3. Non-Liquet: I have looked at the complexity of domestic trials. That is in an earlier paper on the relative efficiency of international criminal trials compared to domestic trials. A link to that paper is embedded in the main text above.

    I have not tried to build a model of what causes the complexity of domestic trials. It might be possible, but I suspect that it would be hard to find enough domestic prosecutions of similar crimes (with all the key documents in English – so I can read and code them) to have a large enough sample to produce any meaningful results.

  4. I have a couple additional hypothesis for you to test if you are interested:

    1) The amount of time that has passed since the activity being tried ended.
    2) The length of time the activity being tried persisted.

    The second is only imperfectly captured by the number of counts test, although not likely I could imagine a regime that both began and ended with abuses but was relatively free of them during some middle period (if for no other reason than that the opposition was so crushed during that middle period as to be irrelevant).

    Although neither of these would really provide much assistance in reducing the complexity of such trials. (other than perhaps ditching the historic record making function and only trying the most recent counts of a long running pattern of abuse from long ago). But I suspect that would not be a palatable option to those performing such trials.

  5. Soronel: Your first hypothesis seems plausible. I assume the mechanism would be something like: the more time has passed the more key witnesses die and key evidence is destroyed thus making proof of responsibility harder and the trial more complex. I haven’t coded this information, although I probably could for the ICTY.

    As to your second hypotheses – why do you think the length of time the activity persisted would effect complexity? I can’t see any obvious mechanism for this relationship.

  6. My second idea was mostly related to the first,in that if the activity occurred over a long period (especially where the people involved held power) they would be in a position to make evidence about earlier crimes disappear. More recent crimes on the other hand (again especially if those crimes were committed during a failed attempt to retain power) are more likely to have readily obtainable evidence.

    Anyway, that was my basic thinking when I wrote out that test.

  7. Excellent post, thank you Stuart!

Trackbacks and Pingbacks

  1. There are no trackbacks or pingbacks associated with this post at this time.