When you cite a social-scientific study, it usually helps if the study is relevant to the issue at hand

...which is a long-winded way of saying that Stephen Walt seems to be a bit offbase here. He cites an unpublished working paper by Alexander Downes which concludes that "foreign-imposed regime change" tends to "ignite" civil wars. That may well be, but it's of limited relevance to Libya because in Libya there already is a civil war. Kindred Winecoff beat me to this point, but I figure it's worth repeating.