Unpredictable, that's what we are

"Never predict anything, especially the future," Sam Goldwyn once said. Wise advice for publications that trade on dignity and gravitas - not least The Economist, which imprudently lets its hair down each December and publishes its forecast for the following year.

But just how accurate were The Economist's forecasts for the world in 2001? For all the great minds behind it, this annual volume has not scored much better this year than Old Moore's Almanack, that rather less rigorous 404-year-old astrological guide.

The editor of The Economist's 2001 preview, former Kensington Tory MP Dudley Fishburn, began his keynote foreword with something of an own goal: "2001 will be a year in which the world becomes a richer and sharply more decent place."

He predicted correctly that China would join the World Trade Organisation (it happened quietly earlier this month), but incorrectly guessed that third-generation mobile phones were "a certainty" for the next 12 months. Outside Japan, they are still some way off.

Fishburn more perceptively warned of the dangers of "the growing multitude of angry, unemployed young men" in Mediterranean Arab countries. "Add an irascible Israel into this cauldron of testosterone and Europe's southern border will surely be the world's most unsettled place," he wrote. Right, and somehow not right. Fishburn's piece on Britain in 2001 similarly displayed how lame the most thoughtful prediction can become in the face of events: he posited above all a sharp increase in employment. What a difference an "un" makes to the growing numbers now being fired.

Elsewhere, The Economist rather rashly predicted an 80 per cent

turnout in the General Election that Anthony King, professor of government at Essex University, said would take place on 3 May. It was actually less than 60 per cent on 7 June, but the poor chap could hardly have known about the foot-and-mouth effect before it happened - damn those events, again.

King also forecast that the Conservatives would gain seats in the election. What a difference, in this case, an "s" makes; they gained just one.

Elsewhere there is a correct prediction that Jack Straw would go some way towards legalising cannabis, a correct (but not difficult) prophecy that "Arabs and Israelis will struggle to reach a comprehensive peace agreement",

an almost spot-on augury that the United States might come under terrorist attack by non-conventional weapons such as trucks, and a not-at-all-bad prophecy that the worst country to live in would be not Angola or Iraq or Sierra Leone, but Afghanistan, which is friendless, with America especially not liking it "because Osama bin Laden lives there".

As for Old Moore's record for 2001: actually, it has proved rather good. In its Fishburn-style keynote, all the more remarkable for having ostensibly been written by the long-dead Dr Moore himself, the almanack conjectures that the planetary positions "promote emotional confrontations, even on the world stage", especially in early September.

Furthermore, the foreword says: "We should expect to see greater activity among terrorist groups. The two countries most likely to be affected are the United States and Russia." Spooky.

It doesn't quite name the day, but its October 2001 preview says: "Reli-gious crusades and battles for civil freedoms may pepper this month with world flashpoints. Clashes between Hindus, Jews, Muslims and Christians are likely. World attention will be focused on the American midwest, Afghanistan, Pakistan and Indonesia." Not bad.

Oddly, Old Moore's for next year omits any retrospective reference to 11 September.

Could it be that it embarrassingly went to press before that date?