Sep 26, 2016

The STL Contempt Cases in Context: An Empirical Review of International Practice

It is no secret that the available body of international criminal jurisprudence is still developing and expanding. The International Criminal Court (ICC) began hearing cases only a decade ago, and the Special Tribunal for Lebanon (STL) has not yet reached a verdict in its seven years of operation. The STL has nevertheless attracted considerable attention for its controversial indictment of four legal persons—two individuals and two corporations—for contempt of court. Though the first case against Ms. Karma Al-Khayat and Al-Jadeed S.A.L. ultimately resulted in an acquittal, the Contempt Judge recently convicted Ibrahim Al-Amin and Al-Akhbar Beirut S.A.L. He issued a fine of €20,000 and €6,000 respectively, but did not impose a prison sentence.

But what do these fines really mean in the broader context of “international” contempt cases? I decided to take an empirical approach to answering this question in light of the fact no systematic study of such contempt cases has been undertaken. In fact, it appears that data is rarely used to answer questions that are relevant to international criminal law (ICL) unless they fit within the broader context of international relations theory. Lawyers and judges must be informed by the facts just as much as political scientists do. This is especially true when it comes to sentencing in international contempt cases, as this represents relatively underdeveloped territory for ICL. As the title suggests, this piece will provide a rough overview of the state of contempt cases at international tribunals. Though the data that was collected is limited, it allows us to evaluate the STL’s decisions in the two contempt cases


I attempted to generate the most exhaustive record of contempt cases that have been litigated in an international criminal tribunal. The list is comprised of proceedings from the International Criminal Tribunal for the Former Yugoslavia (ICTY), the International Criminal Tribunal for Rwanda (ICTR), the Special Court for Sierra Leone (SCSL), the “Subsequent Nuremberg Trials,” (SNT), the ICC, and the STL. No cases from the International Military Tribunal (IMT), International Military Tribunal for the Far East, the British War Crimes Courts in Hong Kong, the Special Panels of the Dili District Court, or the Extraordinary Chambers in the Courts of Cambodia could be found. It is possible that some contempt cases have been handled confidentially, but these naturally cannot be included in this study.

Any set of proceedings regarded by the tribunal in question as a contempt case was considered in this analysis. A trial must have taken place and a verdict must have been issued. Thus, an interim order to detain an unruly witness or defendant would not qualify.

I then recorded the case, the court, the year the final judgment was issued, the defendant’s sentence in months (if applicable), the number of months that individual actually served in prison, the fine imposed in Euros (if applicable), whether the defendant was acquitted, and whether the defendant’s sentence was suspended or served concurrently with a separate (non-contempt) sentence.

Additionally, joined cases were separated where sentences differed in order to make a more accurate record of prison terms and fines that were imposed. For example, the case against Al-Amin and Al-Akhbar were separated in this dataset because each received a separate fine. On the other hand, the case against Al-Khayat and Al-Jadeed was left joined because each was ultimately acquitted.

Observations and Findings

Before passing judgment on the sentence imposed by Judge Lettieri, it is important to understand that contempt of court has not been adjudicated as frequently as “substantive” ICL cases. The data shows that only 42 cases have been completed, 71% of which have resulted in convictions. It bears noting that indictments have been issued against multiple individuals in four additional cases but two of the defendants remain at large, two are awaiting extradition, and five are awaiting judgment. The following graph illustrates the frequency of formal contempt proceedings being brought against defendants at all international criminal tribunals from 2000 to 2016, including three cases from 1947.

The graph demonstrates that apart from the first three cases at the SNT, contempt of court was first prosecuted at the turn of the century. The first such case, against Avramović and Simić, resulted in an acquittal for both defendants. Nevertheless, the number of contempt cases generally continued to grow as existing tribunals became more established and new ones came into existence. It is possible that different courts were more prone to try individuals for contempt but as the following graph illustrates, the ICTY—which was established before nearly every other tribunal in this sample—accounts for the vast majority of cases. This may be explained by the sheer number of defendants at the ICTY compared to other tribunals, but would not account for why the ICTR (which also processed a large number of defendants) produced so few contempt cases.

In any event, the positive correlation between the number of contempt cases and time appears somewhat robust (r=0.45, p-value<0.01; excluding SNT cases). Furthermore, this graph and measure of correlation under-represents the number of individuals tried for contempt by international tribunals. Every tribunal except the SNT joined multiple defendants into a single case in several instances. Most of these cases, including the ICC trial against Bemba and four others, concluded after 2013. This correlation would likely be even stronger if every defendant were included individually in this dataset.

Prison Sentences

The number of proceedings is, of course, only half the story. Comparing the sentences imposed by Judge Lettieri to the sentences handed down by other tribunals will provide a clearer picture of whether the STL is in line with current international practice. First of all, Judge Lettieri’s decision not to sentence Al-Amin to prison should be viewed in light of the fact that 83% of those convicted of contempt at an international tribunal were handed a prison term. The mean sentence doled out to these individuals totaled to 8.32 months, which is slightly inflated by the three separate contempt cases in which Vojislav Šešelj was sentenced by the ICTY to 15, 18, and 24 months in prison respectively.

Residual international tribunals such as the MICT (for the ICTY and ICTR) and the RSCSL (for the SCSL) tend to consider applications for early release once the offender has served two-thirds (67%) of his sentence.[1] In light of this practice and other factors (such as providing credit for time served and past human rights abuses), many defendants convicted of contempt have not served their full sentences. On average, an individual sentenced to prison for contempt will have served 5.61 months. Though this figure is almost exactly equal to two-thirds of the average sentence, a year-by-year analysis shows some variation (see the following graph). Moreover, if any sentences in which the defendant served no time in prison following their conviction (on account of credit for time served or other factors) are excluded, along with those from the SNT, then the average time served rises to 9.75 months. Thus in cases where the defendant was not confined to excessive time in prison pending trial, international tribunals tend to require nearly full completion of the sentence.

If anything, this data suggests that the average prison sentence being delivered by international tribunals in contempt cases is increasing and the average time served is increasing accordingly. Nevertheless, a look at the ratio of the average time served to the average sentence imposed reveals that no discernible trend exists with regards to the portion of the sentences that is actually served. Only time will tell whether the duration of prison terms for contempt sentences will align with the two-thirds standard that has been introduced by the MICT in later years.

Lastly, it bears mentioning that 46% of the prison sentences that were imposed were served to 0-60% completion, excluding pre-trial detention. Of these, 42% resulted in no additional prison time. Only 15% of the prison terms fell within the range of 61-71% completion (roughly corresponding to two-thirds), half of which were terminated at exactly 2/3 of the original sentence. Finally, 38% of those sentenced to prison were incarcerated for 72% or more of the length of their original sentence. Of these, 90% served their full sentence.


Perhaps most relevant to evaluating the sentence imposed on Al-Amin and Al-Akhbar is, of course, a review of the history of contempt fines at international tribunals. However, there is little data to draw from because only 7 separate fines have been issued for such cases of contempt of court (including the separate fines handed down to Al-Amin and Al-Akhbar). This monetary penalty averages to about 12,216.75. Excluding the STL case, the mean drops to 11,903.45.


For starters, the fact that Judge Lettieri merely imposed a fine on Al-Amin may suggest that his sentence was relatively lenient in light of the fact that the vast majority of those found in contempt received a prison sentence. But how do Al-Amin’s actions compare to the crimes committed by those who were sent to jail? Vojislav Šešelj was handed the three most severe sentences of all the contempt cases (15, 18, and 24 months) for precisely similar acts as Al-Amin and Al-Akhbar: disclosing confidential witness information and refusing to remove such information. To be sure, Šešelj was already detained pending the completion of a separate trial at the ICTY, perhaps encouraging the Tribunal to impose a harsher sentence. Moreover, Šešelj was particularly blatant about his contempt of the Court. But this disparity does call into question whether Al-Amin should have received a prison sentence in light of the fact that he too displayed open disregard for the Tribunal’s authority.

Other crimes resulting in prison terms were decidedly less grave than those committed by the defendants at the STL. For instance, Milan Tupajić was sentenced to two months in prison for refusing to comply with a subpoena ordering him to testify. In another case, Kosta Bulatović was sentenced to four months’ imprisonment for refusing to complete his testimony before the ICTY. Bulatović made this decision in light of the Trial Chamber’s ruling to continue the proceedings despite the accused’s absence from the courtroom on account of his ill health. The significance of any contempt case should not be downplayed, but these examples cast doubt on the appropriateness of Judge Lettieri’s sentence in the case against Al-Amin and Al-Akhbar.

Though Judge Lettieri may have been lenient in declining to issue a prison sentence, the fine he imposed on Al-Amin was comparatively strict. It lies well above the mean fine imposed by other international tribunals, and is exactly equivalent to the maximum amount ever levied in a case of contempt. Both tribunals—the ICTY and the STL—allow for a maximum fine of €100,000 according to their respective Rules of Procedure and Evidence. Thus, as far as the fine itself is concerned, Al-Amin received a harsh sentence.

The same may not be said of the Al-Akhbar news corporation. The fine of €6,000 is well below the international standard. Considering that Judge Lettieri labeled the acts committed by Al-Amin and Al-Akhbar as “particularly egregious” in the written reasons for his sentencing judgment, the €6,000 would appear relatively light. Moreover, a corporation cannot be sentenced to prison meaning the Contempt Judge could only ever issue a fine to Al-Akhbar. This dilutes the severity of the sentence further still.

Nevertheless, the sentence sought by the Amicus Prosecutor against Al-Amin and Al-Akhbar bears no connection to international practice whatsoever. As we reported in our previous STL Blog post, the Amicus Prosecutor sought a prison sentence of 24 months and a fine of €75,000 for Al-Amin, as well as a fine of €112,700 for Al-Akhbar. This length of prison sentence was only matched in the third case against Vojislav Šešelj. Though Al-Amin did demonstrate repeated indifference to the STL’s protection orders, the prison term sought by the Amicus appears unduly harsh for a first-time offender. It goes without saying that the fines sought by the Amicus for both defendants far exceed the international norm. With respect to Al-Akhbar, the Amicus Prosecutor would likely argue that the company is a repeat offender. In their view, each day the confidential information remained published constitutes a separate offense. Judge Lettieri handily dismissed this claims in the written reasons for his sentencing judgment.

Data Limitations

Any conclusions drawn from this data must be met with caution. At the outset, the strength of this analysis is weakened by the small population size. It is possible that this dataset only represents a sample of the total population of contempt cases but even so, there are simply too few cases from which we may establish any patterns or trends with certainty. Instead, this data provides a snapshot of what international practice has looked like for the past sixteen years, with the addition of the 1947 cases. Nevertheless, it is debatable whether the SNT proceedings should qualify for this dataset. Unlike the IMT, the SNT was staffed by American judges and prosecutors, thus undermining the “international” character of the tribunal. Moreover, singular cases that are nearly 70 years old are unlikely to be instructive with regards to current practice or future trends.

However, it could also be said that international contempt cases in general are too “singular” to allow for systematic analysis. Activities that have qualified as contempt of court in an international setting range from ignoring a subpoena to bribing a witness to releasing confidential witness information. The facts have often been highly particular to the case, and the gravity of the crime relative to these facts. The general inferences that may be drawn from this data would ignore these important details.

While the nature of the data itself presents its own challenges, the decisions made when constructing the dataset offer benefits and drawbacks. For instance, the decision to subtract credit for time served from the measure of time actually served deflates the latter figure. One may argue that the time was still served, but the sentence was simply completed (in whole or in part) before the final verdict. I justify this decision by pointing to the interplay between the brevity of a typical contempt sentence and the lethargy of the international justice apparatus. Many of those convicted of contempt were already detained for long periods of time in relation to other criminal cases. Including this time in the record of actual time served would often convert this figure into a measure of how long the detention/trial process lasted. This is an interesting issue, but for distinct reasons. Nevertheless, it may be useful to recalculate the measure of actual time served by including credit for time served in the final sum. Thus, only early releases would account for any difference between the sentence and the measure of actual time served.

I included three exceptions to this rule. I coded Vojislav Šešelj as having fully served each of his three sentences in light of the fact that he was incarcerated for over a decade but later acquitted of the separate charges filed against him. Keeping to the aforementioned coding guideline would have eliminated the three most significant sentences in the dataset. There are legitimate grounds for criticizing this form of cherry-picking, but I will simply reiterate that this piece is merely intended to provide a convenient overview of the topic. It does not rise to the standard of a formal study.

Lastly, the decision to separate joined cases where the sentences differ naturally inflates the number of entries in the dataset. However, I wanted to capture the most accurate range of sentences as possible in this dataset. Admittedly, restricting this separation to cases with differing sentences effectively eliminates one 18 month sentence and one €15,000 fine. However, separating all cases would massively inflate the number of cases even though most of these joined proceedings resulted in acquittals. Perhaps these two instances were appropriate opportunities for selective cherry-picking as described above. If included, the average prison sentence would increase to 8.7 months and the average fine would increase to €12,564.65.


Though it may be too soon to tell exactly what direction sentencing patterns in international contempt cases are taking, the cases that we have at our disposal paint a telling picture. International criminal tribunals appear to favor criminal sentences over fines for those convicted. However, it is unclear whether these tribunals are more or less likely to grant early release to those convicted of contempt. The data certainly shows that the Contempt Judge at the STL is not operating in a vacuum—his decision to impose a fine rather than a prison sentence does not follow a clear trend in international practice. Nevertheless, it is difficult to interpret the appropriateness of the fines issued to Al-Amin and Al-Akhbar given how few fines have been issued by other tribunals for contempt of court. What is certain is that the saga of the STL contempt cases will likely serve as important precedent not only for the prosecution of legal persons under international criminal law, but also sentencing practices that may have a profound impact on a defendant’s human rights.

[1] For more data and analysis of post-conviction issues at international tribunals, see Holá, Barbora and Joris van Wijk. (2014). Life after Conviction at International Criminal Tribunals: An Empirical Overview. Journal of International Criminal Justice, 12(1), 109-132.