What does Libya's uprising mean, in terms of showing which regimes are vulnerable to revolution, and which are not?
It's been an article of faith in some circles (usually among "realists" of the right, but not always) that if a regime is ruthless enough, then it doesn't have to worry about being overthrown. It's all very well for us misty-eyed human rightsers to get euphoric about the downfall of Hosni Mubarak, but cold hard realists should realize that this "people power" thing is only good against genteel dictatorships. "Why are the more oppressive governments of Syria, Iran, and Libya not subject to the same degree of popular unrest that is said to be surely spreading to Jordan or the Gulf?" asked Victor Davis Hanson in National Review Online. "Is it because for all the authoritarianism of a Mubarak or a Zine El Abidine Ben Ali, there was never the threat of a genocidal Hama, or thousands perishing on the proscription lists under a Khomeini, or international assassinations of dissidents in the Libyan manner?" A week after the article was published, the rebel flag was flying over Benghazi.
The perception that only hesitant dictators fall has fed a doctrine in the United States favoring support for "traditional" authoritarian regimes, the kind that leave a bit of space for dissent, because their totalitarian (that is to say, Communist) counterparts were both crueler (maybe true) and harder to dislodge (questionable after 1989, particuarly as Ceausescu's Romanian regime was both unreconstructedly Stalinist and completely ready to use force). Variations on this doctrine persisted into the 1990s and 00s in the Middle East, when it was used to justify the support of pro-American despots threatened by Islamists, and also to justify the overthrow of Saddam, because it did not seem likely that his people were going to do it for him.
Some of those who dismiss the effectiveness of crowd-led revolutions may be imagining that you appeal to the better angels of the other side's characters. Instead, what you often have is a rather deadly battle of wills: the protesters march to a place where their mere presence challenges the legitimacy of the regime: a public square to start, a palace if need-be, a security headquarters. If there are enough people willing to join the march, then the regime faces a choice: let them occupy it, use non-lethal force, or shoot early. Letting them occupy it runs the risk of emboldened crowds drawing in more and more protesters, and making the country ungovernable. Using non-lethal force runs the risk of handing the crowds a victory, such as when the Egyptian crowds defeated Central Security on Qasr al-Nil bridge and began the overnight occupation of Tahrir. Using lethal force runs the risk of a mutiny - possibly at the top ranks of command, as happened in Tunisia, or at the very bottom ranks, as happened in St Petersburg during the February Revolution, or in the middle, as seems to have happened in Libya. You may kill protesters, but often they will come back more emboldened then ever. Being in a crowd has an extraordinary psychological effect, and sometimes the deaths of one's comrades makes one more willing to die oneself. Soldiers on the other hand may find their will to keep killing eroded: the real downfall of the tsar, for example, started with a mutiny in a unit which had killed 50 people the day before, and was gripped by remorse.
The solution for a repressive regime is to cultivate a handful of extremely loyal troops, who will kill without compunction. But this is easier said than done -- it is difficult to find people who lack social ties with anyone who might be in a protesting crowd. Libya had a number of regime-defense units, plus mercenaries of African descent recruited in the south and abroad, whom it could throw into the conflict. But their numbers were not sufficient. In Benghazi, it appears that they needed to be backed up by the regular army, and then the regular army turned on them. (Mixing army and security was also one of Ceausescu's mistakes) In Tripoli, Qaddafi appears to have massed enough supporters to prevent protesters from seizing key installations -- but they cannot completely suppress the protests, and appear to be moving in small groups and shooting to disperse gatherings, rather than completely sealing down the streets.
But some regimes are obviously capable of acting ruthlessly. What about Hama, or Khomeini's lists, or Saddam's crackdown on Shia uprisings in 1991 and 1999? Not all of these were crowd-based uprisings, but they do point to some factors which allow a regime to build a force that is both large enough and cohesive enough to kill when ordered. It makes a difference when an uprising is a potential threat to a ruling minority (the Muslim Brother uprising in Hama, for example, or Shia uprisings in Iraq). It makes a difference when you have a Revolutionary or Red Guard that is a product of a real revolution -- that itself has accomplished the amazing feat of overthrowing a regime, and is fiercely loyal to that memory, and does not want to allow the millenarian ideals for which comrades died to be "betrayed". It makes a difference when protesters and rebels are linked to a real external enemy, as were Iraqi rebels to Iran. And it make a difference too when potential protesters are worn down or demoralized by past reverses, or do not have sufficient contact with each other or with the outside world, or are ambivalent because of an external enemy, and insufficient numbers take to the streets to force the regime to make a shoot-or-surrender decision. Maybe a regime will not be vulnerable at all until a major setback -- like the demonstration effect of a neighbor which suffers a revolution, or a lost war -- shatters its facade of permanence.
However, one does not know the ability of a regime to crush protesters until it is put to the test -- and that goes for regimes who have successfully crushed uprisings in the past, as the sense of identity of both praetorians and protesters can change dramatically from one generation to the next. Libya however indicates that ruthlessness alone is no guarantee that a regime can hold power.