The nature of international criminal trial practice is integral to the perception of the legitimacy of international criminal justice. However, our understanding of what transpires within the trial chambers of international courts and tribunals (ICTs) comes primarily from the reports of judges, lawyers, and stakeholders within the system. This article argues that, while the vast body of international criminal justice scholarship barely draws on socio-legal research, empirical work can contribute to a more objective understanding of international criminal trial practice. It examines prevailing academic approaches to the study of international trial practice as a backdrop to the assessment of data from one of the most expansive empirical studies of international trial practice, undertaken during the second mandate (1999–2003) of the International Criminal Tribunal for Rwanda (ICTR). The findings illustrate significant variations in how judges in different Trial Chambers chose to exercise discretion, revealing the co-existence of two distinct modalities of practice in ‘proactive’ and ‘reactive’ Trial Chambers. Quantitative and qualitative data allow for an assessment of the efficiency of these modalities, revealing the critical role of the performance of the judge in the trial process. It is argued that these findings highlight the potential for further socio-legal research to motivate ‘light-touch reform’ within the international criminal justice system.

International criminal tribunals try defendants for horrific acts: genocide, war crimes, and crimes against humanity. At sentencing, however, evidence often arises of what I will call defendants’ ‘good deeds’ – humanitarian behaviour by the defendants towards those on the other side of the conflict that is conscientious relative to the culture in which the defendants are operating. This article examines the treatment of good deeds in the sentencing practices of the International Criminal Tribunal for the former Yugoslavia and International Criminal Tribunal for Rwanda. I show that the tribunals’ approaches are both undertheorized and internally inconsistent. I argue that the tribunals should draw upon the goals that underlie international criminal law in developing a coherent approach to considering good deeds for sentencing purposes.