Ever since Plato, political theorists have played with the idea of abandoning existing arrangements and starting again with a blank slate. And ever since Aristotle, their sober-minded critics have reminded them of the intractability of human affairs and the inescapability of something looking very much like politics as they know it. Indeed, exasperated historians and political practitioners from Polybius to Machiavelli and on have always thought that impractical philosophers should keep quiet: because their utopias never existed and never could exist, they could teach no lessons.

And yet. Given the incapacity of the US political system to deliver sensible policies and attend to the well-being of anyone other than the donors who bankroll the politicians who look after the donors who bankroll the politicians, it¡¯s hard not to wonder whether something utterly different might work better - something that was recognisably democratic but did not require the services of professional politicians. Congress currently enjoys the approval of no more than 15 per cent of the American public, and it¡¯s hard to imagine Founding Fathers such as John Adams, Thomas Jefferson, James Madison and Alexander Hamilton being happy with any of the three branches of government today.

Unrepresentative politicians were once thought to be less ideologically motivated than many of their voters, more apt to compromise and more far-sighted. This now seems the reverse of the truth

A long time ago, conservative intellectual William Buckley said he¡¯d rather be governed by the first 2,000 names in the Boston telephone directory than by the 2,000 members of the Harvard faculty, and the obvious question is whether we might not do better to select members of Congress - and the UK Parliament, too - by some sort of lottery rather than by an electoral process so easily corrupted by money. You¡¯d have to make sure that the sample was drawn properly to avoid the risk of getting 400 or 500 very strange people - just sampling a population of 40 million British adults, let alone 200 million Americans, would produce such an outcome often enough to be disastrous - but a properly drawn sample would provide a representative body. It would also reflect the population whose welfare it was supposed to consider in ways that a collection of professional politicians does not.

ÖйúAƬ

Sixty years ago, the unrepresentative character of politicians was praised by observers because it was thought that they were less ideologically motivated than many of their voters, more apt to compromise, more far- sighted and disinterested. In the US at least, this now seems the reverse of the truth. Things aren¡¯t helped by the role of rich, ideologically motivated donors who can destroy a politician¡¯s career by financing a challenger more to their taste in a primary election where few people other than the most ideologically committed bother to vote. So, on the face of it, bypassing the whole business by replacing professional politicians with amateurs picked by some form of lottery would get money out of politics (and at least some of the posturing as well).

I¡¯m tempted. My students never are. Their great objection is that politics requires expertise: professionals have it but amateurs don¡¯t. I¡¯ve never thought much of that argument. Like all of us, politicians know quite a lot about some things - they are often real experts on one thing - and on almost everything else are complete innocents. In the US, what they spend most of their time doing is raising money for the next campaign and doing favours for their constituents - sometimes in the geographical sense, but more usually in the financial sense. The philosopher John Dewey dismissed politicians of all parties as ¡°bag carriers for business¡±, and there¡¯s a large element of truth in that. It¡¯s not the whole truth, because some of what politicians do - albeit mostly local politicians - is to help their less well-off and less well-connected constituents deal with an unresponsive welfare system; and that¡¯s a valuable service, unlike the insertion of loopholes into legislation that cripple attempts to improve regulation of the pharmaceutical or financial industries.

ÖйúAƬ

It¡¯s when you get into the details of trying to bring back Athenian democracy in a modern setting that problems emerge. Size isn¡¯t a fatal obstacle, although the literature insists that it is: if you accept a sample of the population as a legitimate stand-in for the demos, you achieve a version of the political equality that Athenians aimed at two and a half thousand years ago. If you want more participation, you could recruit amateur legislators by lottery down to local levels; we pick juries by lottery, and juries are not unlike legislatures. Indeed, you could do a lot worse than recruit citizens¡¯ committees for assessing housing schemes, transport proposals and a great deal else. The work of James Fishkin, professor of communication at Stanford University, and others suggests that the results would be perfectly sensible.

There are perhaps two insoluble puzzles here. The first is that people would be unwilling to serve; the evidence of American towns with a town meeting system of local direct democracy is that, outside quite a small circle, almost nobody shows up to the meetings. People chosen at random to serve in Congress or Parliament would be no keener on doing their legislative duty than those chosen for jury service. The other puzzle is how to generate proposals for rational public policy in the first place. It looks like a choice between Scylla and Charybdis - either benign and public-spirited bureaucrats who could too easily bamboozle the ignorant citizens they must persuade to endorse their policies, or pressure groups, demagogues and the recreation of political parties - and of politicians beholden to their donors - all over again. We may not live in the best of all possible worlds, but it may be harder to replace than it looks at first sight.