World View:Jean-Jacques Rousseau identified a fundamental paradox of democracy in his Social Contract published in 1762. "For an emerging people to be capable of appreciating the sound maxims of politics and to follow the fundamental rules of statecraft", he wrote, "the effect would have to become the cause, writes Paul Gillespie
"The social spirit which ought to be the work of that institution would have to preside over the institution itself. And men would be, prior to the advent of laws, what they ought to become by means of laws."
His remarks remain extraordinarily apposite as another wave of democratisation is advocated by the Bush administration. It argues that such change in the Islamic world, rogue states and China will enhance the security of the United States, since tyranny creates violence and democracies coexist peacefully.
European policies have the same objectives, but adopt rather different means. Both of them feed into an international rhetoric in which democracy is universally accepted even if in practice it is far less often achieved.
Rousseau's paradox asks the chicken-
and-egg question about the relationship between demos (the body of citizens) and democracy. Which comes first? Or is that the wrong question, since they are mutually constitutive?
It all points to an obvious but underappreciated fact about democracies. They are contained within territorial boundaries which were themselves nearly all established by non-democratic means. War, violence, coercion, imperialism, colonialism and ethnic cleansing have established most of Europe's and the world's borders.
Since 1989, 8,000 miles of new borders have been added in central and eastern Europe, while most European states have different borders now than 100 years ago.
This matters, because borders are considered by many to be integral to human behaviour. A product of the desire for order, control, protection, they are boundaries of sameness and difference, markers of us and them.
Thus Nicholas Sarkozy has criticised the European Union's enlargement, saying the French and Dutch No votes "were in part provoked by people's hostility to a Europe without borders. I regret that European leaders have not taken account of this. I believe it is necessary not to proceed with further enlargement as long as new institutions have not been adopted".
Political identification, he argues, (along with most French leaders and commentators) requires borders and is defined by them - reflecting Rousseau's influence on their republicanism.
In western Europe democracy, welfare and economic planning grew remarkably within state boundaries in the postwar period. That is why most voters and many leaders still believe transboundary extensions of these functions must themselves be securely bordered.
If they are to carry popular approval they must be matched, on this argument, by careful attention to creating a cultural infrastructure of education, language, party and interest-group organisation, and media, to ensure transboundary communication and deliberation.
This relationship between institutions, borders and popular sovereignty becomes topical and germane as soon as another wave of democratisation is proposed. Two well-known US political scientists, Edward Mansfield and Jack Snyder, criticise the Bush administration's assumption that there will be an easy transition between promoting forced-pace democracy at gunpoint, containing violence and peaceful coexistence.
In the current National Interest journal they draw on their wide-ranging comparative study published last year, Electing to Fight: Why Emerging Democracies Go to War.
On the contrary, they argue, where political institutions are lacking, and particularly a coherent state grounded in a consensus about which citizens will exercise self-determination, unfettered electoral politics often gives rise to nationalism and violence at home and abroad.
"Absent these preconditions", they write, "democracy is deformed, and transitions toward democracy revert to autocracy or generate chaos. Pushing countries too soon into competitive electoral politics not only risks stoking war, sectarianism and terrorism, but it also makes the future consolidation of democracy more difficult."
They look at Iraq, where much of the political argument revolves around the question of who will form the nation on which democracy is to be based. Nationalist rhetoric is geared to ethnic and religious groups, whereas secular and catch-all leaders are put at a huge disadvantage.
The earlier elections come in a transition from authoritarian or totalitarian rule the worse this problem becomes, since ethnic mobilisation captures the political space and determines the shape of future competition. Bosnia and Serbia are recent examples.
Such internal nationalism can easily be externalised, as we saw in Armenia and Azerbaijan, Ethiopia, Kashmir, Rwanda and Burundi over the last 15 years. Statistical studies show that countries with weak institutions undergoing incomplete democratic transitions are more than four times as likely to become involved in international wars than other states, while incomplete democracies are more likely to experience civil wars than either pure autocracies or fully consolidated democracies.
These lessons are highly topical, given the various scenarios now under consideration for bringing democracy to north Africa, the Middle East - and China.
In Iraq, Iran, Palestine, Egypt and North African states Islamic movements may be the immediate beneficiaries of forced-pace democracy. Imposing democracy in quasi-imperial fashion without a commitment to "nation-building" understood as institutional development can breed war, not peace.
Iran is a good example of arrested and deadlocked democratisation, with untold regional and world consequences unfolding this week. It is interesting to see the Chinese communist elite beginning to discuss these issues more openly, arguing forcefully that their careful sequencing of the transition there under one-party control is the best way to hold the country together and develop the necessary institutional infrastructure.
But they, too, resort to nationalism the more the old communist ideology is left behind. The contrast with democratic India is increasingly drawn.
The political lessons to be drawn from all this are plain. Democracy cannot be imposed without a huge commitment to build the civil society on which it needs to be based. Such commitment is rare - although the US involvement in postwar Europe comes to mind.
Mansfield and Snyder advocate a long game of promoting democracy by inducement, involving close US-EU co-operation. EU enlargement is a good example of how to go about it, pace Sarkozy. Sequencing long-term organisational capacity over immediate elections is a better way to ensure peaceful transitions.
But Rousseau would agree that reform, revolution and regime change are best left to domestic constituents, not imposed from without.