A dead body is found. The first thing law enforcement or health practitioners seek to do is to find out what happened: did the person die of natural causes of was it an unnatural death? A suicide? An accident? A homicide? Police may want to start an investigation to clarify this matter. Typically, recordings of homicides start with an unnatural, suspicious death. Such suspicious deaths may be followed by an investigation, clearance, arrest, prosecution, conviction, and ultimately, imprisonment of the perpetrator. However, at each of these steps, homicide cases can ‘flow out’ of the judicial system.

To begin with, the police may decide not to proceed with investigating a case. In other cases a suspect is known to the police, but cannot be arrested. Suspects may have fled the country. Sometimes, they die as a result of suicide before being arrested or get justifiably killed by the police during a confrontation. Also, after an arrest, as police hands a suspect over to the public prosecution office, homicide cases may flow out of the system. For example, when it is decided not to prosecute the suspect for a homicide, but for another crime. Or, when it is expected that court conviction will fail, because evidence is too limited. Moreover, the age or perpetrators (too young or too old) or their mental or physical health conditions may be reasons for discretionary dismissal. Some suspects will not be convicted because their alleged involvement cannot be proven beyond doubt.

Roughly, two separate schools have studied factors relating to homicide clearance. One school holds that the police and political factors, such as workload or organisational policies, play a decisive role in the (un)cleared status of a crime. The other school considers extra-legal factors, such as the likeability of victims and of offenders, as key factors that determine homicide clearance. When studying the homicide flow, one may question whether some victims and perpetrators receive ‘more law’ than others? Do some types of homicide get investigated, followed up on, and eventually solved more often than other types? And, if so, why?

Trying to empirically study the flow through the system is quite a challenge. Data needed to complete the entire process are scattered and registration at each level is often incomplete. Preliminary findings, presented at a recent UNODC conference on Statistics, Governance, Security and Justice reveal large international variations in terms of the amount of (data) loss at each of these steps, as well as in terms of the length of time it takes for homicide cases to flow through the system. To further advance our theoretical understanding of the homicide flow, future research in our research group Physical Violence and Public Order, in collaboration with international partners, seeks to provide insight into arrest-, impunity- and conviction rates and explanations for such variations.