Read other letters about this article
As usual, you raise an important point of the kind that's generally lost in the din ... much appreciated.
The comment (William Timberman's?) about the Irish made me think about all those Noraid jars you used to see in Irish pubs in the US, which all inevitably ended up in the IRA's hands.
The World Bank (and Yale) had a project a while back about the political economy of civil war and persistent conflict. The starting point for it was a large-n statistical study that tested different variables' influence on the persistence of war (which were followed up later by case studies). None were monocausal; some factors contributed robustly, others not so much. Some of them were what you'd expect, others not.
One of the surprises had to do with 'diaspora' communities ... the larger the diaspora, the more likely the war was to persist. One of the conclusions came back to that Noraid jar (or its correlaries); the diaspora communities inevitably sent money back to the combatant parties, which kept things going long after exhaustion of resources would have driven them to the negotiating table.