Over at Political Violence at a Glance Andy Kydd describes the situation in Syria this way:
We are now past at least 70,000 dead in the Syrian conflict and mass atrocities are being committed on an almost daily basis.
That reminds me that I find it irritating that prognosticators are rarely held to account when they make a case about what they expect to happen and it does not happen. In July of last year dissidents in Syria pulled off an unexpected bombing in Damascus that led me to argue that the Assad regime was likely to face an onslaught of defection and fall apart:
if I had to place a wager, I would bet that more and more of them are going to conclude that Assad will lose. Let the mass shirking begin, and I suspect it will be only days or at most weeks before we see the key act of insubordination that will collapse Syria’s current regime (though see Erica Chenoweth’s view to the contrary here).
Uh, yeah, well, it has been pretty clear for some time that I was wrong and that Chenoweth was right. I was correct that there would be defections, but very wrong about the ability of the Assad regime to persist. I have been meaning to publicly point this out, and Kydd’s post got me to do so.
In September of last year I had the opportunity to present a very slowly developing project of mine at Yale’s Program on Order, Cooperation and Violence. During the visit Stathis Kalyvas and I got to yacking about the situation in Syria, and it turns out that he was as surprised as I at the joint level and duration of dissent given the state’s coercion, and the ability of the state to persist given the level of dissident mobilization. It was nice to know I am not the only one to find the course of events in Syria puzzling.
@WilHMoo
Will Opines 
Will,
Is this possibly because the real axis of conflict isn’t for/against Assad, but Alawite/Sunni? If that’s true, then Alawites (and those who have tied their fortunes to them) could see the survival of the regime as lashed to their own survival – making defection less likely, because defection = death. I don’t know if this is what’s going on (although some recent signs suggest it may be), but it does explain why we’re not seeing the level of defection a reasonable person (like you) would expect.
Bill, thanks for the comment. My understanding is that Alawites are less than 15% of the population (i.e., not much larger than the European populations of Rhodesia and South Africa). So while I suspect Assad can count on very high Alawite support for the reason you note, I have always assumed that the Assad family rule has never been about keeping the Alawite population solid. But, dem’s fightin words, and further discussion quickly devolves into the sort of thing I complained about last month in my “Did Charles Tilly Labor in Vain?” post. So, I guess my reaction is: that probably is a small part of the story, but not where things are interesting.