Bryan Caplan asks, a propos of events in Egypt, why some revolutions end up making things better while others make things worse. His answer (which he admits is unconvincing) is that revolutions make things better when they are against totalitarian regimes and worse when they are against authoritarian regimes, because “the point of totalitarian regimes is to give people less freedom than the median voter wants, but the point of authoritarian regimes is often to give people more freedom than the median voter – or at least the median man of violence – wants.”
I don’t think that works. Marcos wasn’t a totalitarian, for example, and neither was Milosevic. When I consider which revolutions turned out badly and which turned out well, the thing that really jumps out at me is the degree to which the revolution in question was achieved by peaceful as opposed to violent means. There are exceptions, of course, but for the most part violent revolutions have tended to end badly (very often making things even worse than before), whereas largely non-violent revolutions have tended to make things better. Violent revolutions end up being led by violent men, and once in charge they have a tendency to turn their talents on others. Whereas the leaders of non-violent revolutions tend to be better at democratic politics (and if they aren’t they don’t try to hold onto power by killing their opponents).
Something to think about.