The most encouraging generalization in international relations is what is known as "the democratic peace": no two democracies have ever gone to war with each other. This is-- said to be-- an empirical fact. One of the evidences of its truth that I find most convincing is the lameness of the objections that are occasionally raised. For example, one Foreign Affairs article I read pointed out that if you define democracy "by the standards of the time," the 17th-century war between England and Holland would be an exception to the democratic peace. But 17th-century England and Holland were not democracies "by the standards of the time," they were just not democracies. Their hereditary monarchs were not just figureheads but had considerable power, the franchise was restricted to a small fraction of the population, and protection of civil liberties fell far short of modern freedoms of press and assembly.
The democratic peace combined with some other trends, e.g. (a) the growing number of democracies in the world, (b) the pattern that democracy never falls in countries above certain income levels (as of ten years ago I think the figure was $7,000 GDP per capita), and (c) the tendency of the world to get wealthier over time, points towards a pretty optimistic prognosis for the future, with the whole world being wealthy, democratic, and at peace. This is not utopianism. There will still-- at least, so we must guess from observing today's democracies-- be crime, economic inequality, disease, adultery and divorce and broken hearts, envy and boredom and ennui, back-breaking toil, class stratification, vapid people, rude people, etc. There might also be civil wars: democracy is, unfortunately, no guarantee against that. Just no international war. Which is still pretty good.
But this optimistic prognosis is based on an empirical pattern which could fail at any time. So far, the democratic peace is an iron law of history, without a single exception, yet it might not hold tomorrow. Induction is pretty persuasive yet ultimately is absolute proof of nothing. So for those who regard the democratic peace as the most hopeful promise-- in temporal terms, anyway-- for the future of man, every new war provokes nervous analysis. Is this the first one? Is this a war between democracies? Does the democratic peace hold?
Is the Russia-Georgia war an exception to the democratic peace? No, because while Georgia is a democracy, Russia is not. In the 1990s, when Yeltsin and Zyuganov fought a close election in 1996, Russia was on the margins of democracy; one might call it a "proto-democracy." Since then, Putin has run roughshod over the Russian constitution, enacting a "reform" that made provincial governors appointed from the center, shutting down newspapers, using the judicial process-- Russia does not, unfortunately, have a tradition of judicial independence-- to break up the best-practice oil firm Yukos in order to prevent the emergence of an honest business class that would be less vulnerable to demands for bribes, shutting down independent media, etc. In the last election, Kasyanov, one of Putin's prime ministers, wanted to run against Medvedev and was disqualified, precisely because he had a chance of winning. Russia at this point is not a democracy, as most Russians recognize (though, oddly perhaps, it is still to some extent administered under a democratic constitution). A democracy invaded by a non-democracy is not an exception to the democratic peace.
In a sense, then, one might answer the question, "Why did Russia invade Georgia?" by saying "Because Russia is not a democracy." In logical language this is a necessary but not a sufficient cause: if Russia were a democracy, it would not have invaded Georgia, though of course a non-democratic Russia might still have acted differently. The democratic peace per se is unilluminating in the sense that it does not explain why. Natan Sharansky, in Bush's favorite book, The Case for Democracy, gives a reason why. Leaders in both democracies and non-democracies want power. In a democracy, the way to keep power is to serve the interests of the people. War is never in the interests of the people, so democratic leaders never want war. Of course, sometimes aggression by autocracies makes war inevitable. Sometimes democracies don't fight when they ought to fight. But two democracies can always keep the peace with each other. By contrast, in unfree countries-- democracy and freedom are distinct in principle but closely related in practice-- the government needs external enemies, real or imagined, to promote internal cohesion, marginalize its critics, and justify repression. It wants its people to live in state of paranoia and mental war, and sometimes it needs or wants real wars in order to maintain this mentality.
Sharansky's theory fits contemporary Russia pretty well. From my perusal of the Russian media, it's clear that there is a widespread belief that they are under attack, that America is-- somehow-- waging war against them, that Georgia is the "aggressor" and Saakashvili is an American stooge engaging in deliberate provocation. Such attitudes could not endure long in the context of public discourse with a free press. For example, the Moscow Times and Novaya Gazeta, two Russian publications with limited circulations, have already pointed out that the war was planned in advance by Russia and that the ethnic cleansing/genocide charges which have been widely believed by the Russian people are not based in fact. If such information were allowed to be shown on major TV stations the Putin regime might fall, and would at any rate be unlikely to rally support for new wars, e.g. in Ukraine.
Only when democracy establishes itself in Russia can the West expect lasting peace.