Yes, I didn't intend my comment to mean there should be no ceasefires (or that there always should be ceasefires as a tool for peace for that matter). I think the article has the seed of some interesting data for analysis, but isn't at a strong conclusion yet. What might really be interesting to get to is if one can see in data if certain pre-conditions to ceasefires are associated with an increased likelihood to get to longer term peace - or just that they're not as employed as time to regroup-for-the-next-attack - though it's hard to see in data if those waves would have occurred anyway (there are flareups and lulls in conflict even without cease fires). On the flip side, in a very grim way, could more attacks, more intensely bring the conflict to an end more quickly?