As a rule, however, optimists gained the upper hand in both Cold War and post-Cold War contexts. After all, superior ideas did do in the Soviet Union, didn't they? Old arguments that Confucian or caudillo political cultures made democracy impossible in East Asia and Latin America were proved wrong, weren't they? During the Clinton era, at least until the 1997 Asia crisis, it seemed self-evident that democracy and market economics advanced and reinforced one another without any obvious contradiction or downside. Doubters warned that democracy and market economics were revolutionary ideas which always caused havoc in societies unused to them, but even as evidence mounted that this was correct, such was the Zeitgeist that a defensiveness about this claim came as naturally as the claim itself. It is largely thanks to how the Cold War ended and the general trajectory of the past dozen years that democracy optimists have carried the day in the third context: that of post-September 11, 2001.

Three Differences

The stakes over the arguments about democracy seemed high during the Cold War, and nearly as high in the decade that followed. But to many Americans, those stakes seem small compared to what stands before us today. In the Cold War we faced an immense Soviet nuclear weapons arsenal; but the Soviet government could be reasoned with and deterred. Now we face suicidal fanatics who, with nothing earthly to lose, may be essentially undeterrable, and who may nonetheless devise ways to kill us as effectively as Soviet H-bombs would have been. That prospect is what recommends a strategy of pre-emption, and the imperative of bringing democracy to countries that incubate terrorists or provide support and sanctuary for their work. God help us, for this will be much harder to achieve than defeating Soviet communism.

It is, of course, perilous to generalize about the large and diverse number of Muslim political cultures, and even about the smaller and somewhat less diverse number of Arab ones. They are not all the same, and it is hard, too, to separate out the theological, historical and, what for want of a better word, the anthropological strands of these cultures (for Islam is a venerable civilization, not just a religion, its historical equivalent in Western history being not Christianity but Christendom). That said, there are few electoral democracies in the Muslim world (Turkey's is the most mature), and absolutely none in the Arab world. The risks of generalizing aside, can this really be a coincidence?

It isn't a coincidence. Arab societies lack certain dispositional prerequisites for democracy; let us mention just three: the belief that the proximate source of political authority is intrinsic to the society; a concept of majority rule; and the acceptance of all citizens' essential equality before the law. Without the first, the idea of pluralism-and of a "loyal opposition"-cannot exist. Without the second, a polity can be neither free nor liberal. Without the third, the idea of elections as a means to form a government is incomprehensible. These attitudes are second-nature to Americans; indeed, they help define what America is. So smoothly do these attitudes flow into our consciousness that we often assume that they are also second-nature to others. It is not so, however.

There are only two ways to conceive of the operable source of political authority: either it is intrinsic-"of the people, by the people, for the people"-or it is extrinsic (coming from God, or from some accepted imperial source outside the society in question). The first of these conceptions is the newer in human history, and for it we have Copernicus to thank. As John Donne wrote in "To the Praise of the Dead, and the Anatomie" (1611):

And new Philosophy calls all in doubt,
The Element of fire is quite put out;
The Sun is lost, and th'earth, and no mans wit
Can well direct him where to look for it. . . .
'Tis all in peeces, all cohaerence gone;
All just supply, and all Relation:
Prince, Subject, Father, Sonne, are things forgot.

It was not long after that the Glorious Revolution ensued, and that Thomas Hobbes and his successors took the final measure of the divine right of kings, dispatching it as they launched forth the era of modern social contract politics.

Not so within Islamic civilization, which has never recognized any intrinsic source of political authority. Islam is a radically monadic religion of divine revelation, and Islamic political culture has developed over more than 1,300 years wholly true to that principle. Divine, extrinsic authority cannot be disputed, so there is no logic to political pluralism as a permanent or ideal condition. Tolerance for any other set of social and political first principles amounts to heresy; tolerance of other private religious beliefs is conceived as virtuous forbearance, not as a recognition that truth might really be in dispute. A Saudi professor of Islamic law thus explained tolerance to a visiting journalist in these terms: "Well, of course I hate you because you are a Christian, but that doesn't mean I want to kill you."

A particular concept of political leadership flows from these predicates. A leader is someone who enunciates and spreads God's law. It follows that since there is only one God and He has only one law, there should only be one political structure (the caliphate) and one leader of it. Accountability is not democratic in the Western procedural sense, but organic in a religious communal will. Government is legitimate when it accords with a priori religious truth. This is why, as Bernard Lewis relates, the first Muslim jurists to observe the House of Commons at work at the end of the 18th century felt sorry for the British when they learned that the purpose of that large, noisy assemblage of "common" people was to make laws. The English, they later explained to their readers, "had not accepted a divine law and so were reduced to the expedient of making their own laws."

This model of leadership is still relevant to the more secular context of our own day. The explicit content of a priori truth may have changed (or not, as in Saudi Arabia), but the form remains the same. As William Brown, a former U.S. diplomat and Arabist, sums it up:

According to the liberal democratic norms of the West, political institutions are dedicated to enacting the wishes of a tolerant majority. In the Middle East the purpose of political institutions is to facilitate the constant unfolding or revelation of a popular consensus. . . . The Arab perceives a single community of faith and language that contrasts sharply with our emphasis on competing but mutually adjusting political factions. In the West, politics has a flavor of controlled conflict that the Arab regards as destructive to community.

Which brings us directly to the matter of majority rule.

If truth is intrinsic to a society and men are fallible, then political life must amount to trial-and-error attempts to find the best way to govern. If no one can invoke the authority of unquestioned truth, it follows that the majority should decide which path to follow, and which leaders to trust, until it comes time to take stock, vote again, and perhaps try another approach. We regard this as simple common sense, but most Arabs do not-and the reason is not obscure.

For thousands of years, most people-certainly most Middle Easterners-lived in villages of several dozen to perhaps a few hundred people whose organizing principle was usually that of the clan or tribe. They also lived, more often than not, in an insecure world where the dangers posed by other tribes, and the prospect of natural disaster and epidemic disease, were very real. This put a huge premium on preventing serious rifts within village society. While the variety of decision-making procedures in traditional societies is manifest in the literature of cultural anthropology, they all boil down to governance though consensus-building. Leadership was usually centralized and hereditary, but that did not necessarily make it despotic. A leader engaged in open-ended negotiation with the dominant, usually elder, males representing the main branches of the clan; problems were talked out, compromises and understandings reached, and in return all swore personal loyalty to the leader (bay'a in Arabic-a still current and meaningful term). This methodology of governance was absorbed into and sanctified by Islam, wherein a leader comes to his position through a consensus of elders (ijma) and remains in power through the acquiescence of the community (umma). A leader who breaks faith, it is assumed, will be rejected by the umma and replaced by a new standard bearer who will reconstitute a social consensus.

Now consider in this light the idea that a contender who wins 54 percent of the vote in an election should be given 100 percent of the power, while the person who wins 46 percent should end up with none. This strikes people used to consensus decision-making as not only illogical and unfair, but dangerous-an open invitation to civil strife should society come under pressure. This is why, by the way, when Hosni Mubarak or Bashir Asad wins 95 percent of the vote in an election-which we usually interpret as an empty act of egomaniacal perversity-it does not strike a typical Egyptian or Syrian as odd. It also helps to explain the result of a recent UN Development Program report on the Arab world, which shows that the Arab region "has the lowest value of all regions in the world for voice and accountability"-UN-speak for political participation and democracy.