The twentieth century saw a great advance of democracy: not at all even, of course; it was always a question of two steps forward, one step back. But the long-term trend is unmistakable. I believe that this trend will continue in the twenty-first century. This is not because I am a simple-minded optimist (although some of my friends accuse me of this) or because I believe in the inevitability of progress. It is rather because democracy is becoming the only regime that can claim legitimacy over the long term. Human history is full of earlier, hierarchical regimes that were stable because they reposed on some belief in an ordered cosmos, where a ruler might draw his mandate from heaven; on the divine right of some dynasty; or on the natural superiority of some class or tribe. But the progress of what we call modernity, in its different forms, undermines all these traditional ideas and eventually makes them incredible. In the end, the only basis for stable legitimacy we can fall back on turns out to be that we should obey our rulers because in doing so we are really obeying ourselves.

This is not to say that democracies are stable. They may fail to deliver the goods and become discredited, they may fail to hold together in face of tensions between classes or ethic groups, and they may fall apart in a host of ways. But in our day, the regimes that displace them, often through coups, turn out to have even less legitimacy once the first flush of the novelty wears off. And they lack this because they have no good grounds to hold our allegiance over time. They have to claim superefficacy as their excuse for overturning democratic institutions, and this claim cannot fail to look unconvincing over the long term. In other words, democracies may not be stable, but other regimes are even less so; and they end up having to sustain themselves by more and more naked force, like Saddam Hussein in Iraq. That is why putschist generals in our day usually begin by assuring people that they will revert to elected government once they've "cleared up the mess" (like Musharaf in Pakistan).

But this greater geographical scope of democracy makes it harder to understand. Or, rather, it makes us aware how little we have understood it. What are the conditions that make for stable democracies? Why does democracy take root here and not there? (For instance, why in India and not in Pakistan, when both achieved independence at the same time from the same Raj.) In the postwar period, books such as Lipset's Political Man tried to answer these questions.¹ Some of their intuitions were good, but others now seem wide of the mark.²

It may well be that the attempt to answer totally general questions—such as, what are the conditions of democracy?—is misguided. General questions assume that there is some recognizable political culture of democracy and a set of economic and social conditions that enable this. In fact, it would seem more sensible to start from another basic assumption: that there are cultures of democracy, in the plural. Just as we have (many of us) stopped talking about modernity and speak now of multiple modernities, so we will have to recognize different democratic forms.

If nothing else has, the existence of the Indian Republic ought to drive this point home. This is the world's biggest democracy. But more than that, in some respects it is showing signs of health where some older democratic regimes show signs of decline. In the North Atlantic world, we are all concerned by the downward trend in voter participation in elections. This is most noticeable in the United States, but it is happening across the North Atlantic world. This is indeed one of the paradoxes of democratic life today—that as this form spreads over most of the globe, it is showing signs of running out of breath in its original heartland.