I just had a particularly interesting meeting at the UN with several well-placed and highly experienced colleagues. The topic of conversation, unsurprisingly, was conflict early warning and conflict prevention. Academics have long drummed up the various albeit few “successes” of early warning, so it was interesting that my UN colleagues cited Ghana, Guyana and Sierra Leone as their own recent success stories. Each intervention involved substantial prevention-related programs/projects, such as “social cohesion programs,” some one to three years prior to scheduled elections.
Equally interesting were the comments made in relation to Kenya and Georgia. In the case of the former, one senior colleague mentioned that,
Our own early warning ‘systems’… or rather analyses, mislead us… they suggested that the most conflict prone places would be in the north of the country, so we focused our preventive, training efforts there to reduce the likelihood of escalating ethnic tensions… this was back in March 2007. What we didn’t realize or expect, was that the Rift Valley would become so volatile, let alone the coastal region of Kenya.
In the case of Georgia, another senior colleague commented on the fact that,
We knew full well what was about to happen, we had our teams in the field, reporting on the increasingly dicy situation several months ago. In fact, we were fully expecting the situation to escalate in August. The problem, again, was not early warning.
When I pressed my colleague further on how exactly they knew, i.e., whether they were using specific and/or sophisticated methodologies for their conflict monitoring and analysis, the answer was no. Situational awareness, fact finding, in-country missions, sharing of information between agencies/contacts in Georgia and regular meetings to discuss the situation was in effect what constituted their conflict early warning system.
The conclusion I take from this meeting is not that early warning is not important, but that “good enough” analysis is more important than sophisticated approaches to conflict early warning and forecasting.