One of the frustrations with being a political scientist is that those that we study often have multiple strategies to deal with their problems, so we often cannot provide great predictions. To say that politicians want to keep their positions is a useful starting point, but we need to know their constraints before making guesses.
The example du jour is, of course, Libya. Qaddafi considered the problem of coup prevention since he came to power via the military, and provided an interesting but not entirely unique "solution": create a weak, divided armed forces so that any attempt to seize power would fall apart. Of course, the tradeoff is that it would be a lousy tool for repressing an uprising. But he was more or less prepared for that with his mercenaries. As I have been repeating for the past month (yes, it has only been a month of Mideast protests, more or less), the big question is whether the guys with the guys will shoot. Qaddafi, nutjob that he is, figured out that he can be more confident about the guys shooting the citizens if they are not from the neighborhood. The mercs seem not just willing but eager to shoot, in sharp contrast to the Egyptians.
I wonder if the Saudis are taken notes. They cannot farm out their repression to Palestinian guest workers, can they?
Anyhow, each coup prevention strategy has tradeoffs. Mubarak sought to ensure the loyalty of his armed forces by imitating the Chinese and enriching the officers through their own businesses. The problem then becomes that the military may see the dictator as a threat to their income. Qaddafi chooses mercs and a weak military, but this means civil war or something like it when enough folks get together that mercs cannot crush easily. Others have multiple, overlapping security institutions challenging each other. Iraq had this with the Special Republican Guard, the ordinary Republican Guard, the army and various other actors. Very expensive, especially when you have a real military threat as you have probably impoverished the majority of the forces confronting the external threat. The Saudis? I don't know enough, but money does have its limits, as any one who controls the oil and cash spigots can pay the guys with the guns.
What does all this mean? That the field of civil-military relations (along with social movements, repression, and foreign aid) is going to be back in fashion. This area has not gone completely out of fashion, but the focus was elsewhere. And in this field, the focus was more on how do stable democracies control their militaries and whether there are crises or divides--like the US military being far more conservative than the society. The issue of coups receded after being the focal point for so long. Recent events indicate that we need to figure out some answers to the classic questions: when will the military shoot at the civilians? when will authoritarian leaders lose control? And, of course, when will the military step back after seizing power and intervening? As hard as it is to get a coup rolling or to get the military to support the citizens, getting them back into the barracks is hardly a trivial question. We still don't know what is going to happen in Egypt and Tunisia, not to mention Libya.
All we do know is that social scientists probably should not be speaking on behalf of the dictators.