Genocide is a complex problem and predicting and responding to it are usually constrained by several factors including geography, the difficulty of information gathering, international legality, and political will (within the state in question and outside of it). Mapping and analysing the factors leading up to genocide is undoubtedly an ambitious goal, but the Sentinel Project is equipping itself to achieve this.

A variety of tools are already available which can inform our development process. The Violent Risk Appraisal Guide (VRAG) is a clinical tool designed to determine the likelihood of violent recidivism in individual high-risk offenders. Like genocide, the likelihood of an offender re-offending after they have been released back into the community has been problematic and controversial matter. Clinical assessments in violence research which were previously the mainstay of dealing with offenders have been found to be increasingly ineffective. This is due mainly to the fact that clinical assessments rely on the interpretive and subjective evaluation of a practitioner guided by professional literature, and as such tend to be unreliable. This is the reason why the VRAG employs an actuarial model which makes predictions of a particular outcome (violent recidivism) based on objectively measured variables such as age and sex. Variables in VRAG are selected based on their distinctive contribution to the outcome and weights for each variable are then computed.

Numerous models have been developed to assess the risk level of genocide taking place in a given country. Once genocide has begun, other tools such as Gregory Stanton’s Eight Stages of Genocide model have shown that genocides generally follow predictable patterns. With this in mind, the Sentinel Project research team has begun collecting the risk factors and variables identified by various scholars. Unlike violent recidivism as monitored by VRAG, genocides are rarely discrete or isolated events. They take place across various territories, sometimes concurrently or sequentially and are always systematically carried out by groups. Compared to individual violent crime, genocide is more fluid in terms of both likelihood and occurrence.

Another tool that may help to inspire increased organization in genocide detection and response is the Violent Crime Linkage Analysis System (ViClass) database. ViCLAS is a national police database used for tracking both offenders and the offenses they commit. It is also designed as an automated case linkage system for the purpose of capturing, collating, and comparing violent crimes through analysis of victimology, offender/suspect descriptions, modus operandi, and forensic and behavioural data. The ViCLAS system has been successful at tracking predatory/mobile offenders, since it has served as a unifying information store across jurisdictions, and has increased and strengthened cooperative law enforcement endeavours.

In light of the challenges of early warning and cooperative response to genocide, the Sentinel Project has moved forward with the understanding that every potentially genocidal situation will be unique in its pattern of variables. This has informed our situation-of-concern (SOC) format. Additionally, our early warning system (EWS) architecture bears some similarity to VRAG and ViCLAS but will be continually fine-tuned as our research analysts gather more information and the technology team develops new ways of managing it. The Sentinel Project continues to embark this journey expecting to make full use of the tools at its disposal to fulfill its mandate. Stay tuned for future developments!

Troy Powell
Research Analyst at the Sentinel Project.