Probably, they're all wrong

When some people want to settle a dispute, they reach for a lawyer. Others look for the nearest car-park and slug it out. Two distinguished scientists have just decided to sort out their differences in the classic style of a couple of barflies: by making a wager. As one might expect, the focus of the wager is a bit more high-minded than who's going to win the World Cup. The biologist Professor Lewis Wolpert of University College, London, insists that DNA contains everything needed to predict how living organisms develop. In contrast, the former Cambridge University biologist Dr Rupert Sheldrake believes genes are only part of the story. In a series of controversial books, he has argued that there are also "morphic fields" at work, which influence development.

For years these two academics have stuck firmly to their convictions, resulting in some entertaining spats at public lectures. Now the two have finally agreed on something. Dr Sheldrake will accept Prof Wolpert's argument if, by May 1, 2029, every detail of at least one organism is correctly predicted solely by knowing its DNA. Not only that, but Dr Sheldrake will present his adversary with a case of fine port, Quinta do Vesuvio 2005, which by 2029 should have reached perfect maturity.

By that time, both parties to the wager will be pretty mature too: Dr Sheldrake will be pushing 86, while Prof Wolpert will be nearly 100. Yet the potential futility of the wager is one thing; why should two perfectly rational academics take part in what many would regard as an irrational way of settling a scientific dispute? They are certainly in distinguished company. In 1870, the British naturalist Alfred Russel, co-founder with Darwin of the theory of evolution, took on a £500 wager - worth more than £45,000 (DH 270,000) today - to settle a dispute with someone who insisted the Earth is flat.

Russel must have thought he was on to a winner here, as conclusive evidence for the roundness of the Earth has been in existence for millennia. Mariners in ancient Greece noted how ships vanish over the horizon as they sail away, their hulls first being carried out of sight by the curvature of the Earth, followed by their sails and finally their mast-tops. Russel decided he could win the wager by showing the same effect on a stretch of the old Bedford Canal, north of London. Using a telescope to observe two markers held at the same height 5km apart, an independent referee agreed that the further of the two markers appeared lower - consistent with the level of the water curving away into the distance. Russel then made a dreadful discovery: his crackpot adversary simply wouldn't accept the result. Legal action didn't help: the man went bankrupt, leaving Russel to pick up his bill.

Even so, as a means for dealing with irksome adversaries, making wagers remains popular. In 1980, the gloomy predictions of environmentalists prompted the American economist Julian Simon to wager that the price of industrial metals would fall in real terms over the following decade, thus contradicting the view that all natural resources are running out. The famously pessimistic biologist Paul Ehrlich took on the wager, selected a set of metals to monitor - and watched as their price tumbled over the following decade by an average of almost 40 per cent, forcing him to pay up.

Not surprisingly, perhaps, environmental issues have since become a rich source of scientific and technological wagers, on everything from the disappearance of all Arctic ice to the emergence of fuel cell vehicles. Yet the most prolific wager-monger of all works in the most hard-headed of sciences: for more than 30 years Professor Stephen Hawking has laid a series of bets with fellow physicists about the correctness or otherwise of his cosmic theories.

His success rate is less than impressive. In 1974, he wagered an American theorist that a deep-space object called Cygnus X-1 would not prove to harbour a black hole - on which he is regarded as the world expert. Astronomers proved him wrong a few years later. In 1991, Prof Hawking made another wager with the same theorist, this time that it would never be possible to see inside a black hole. Six years later, he had been forced to concede on this as well. Undaunted, he went on to make yet another bet about the properties of black holes; by 2004, he had lost that as well.

Prof Hawking currently has a wager on physicists failing to find a long-sought subatomic particle called the Higgs Vector Boson. Regarded as an essential ingredient of any theory describing all the forces of nature, the Higgs particle is regarded as the key to explaining why objects have mass. Prof Hawking is not impressed, and has a bet of $100 with an American physicist that it won't turn up. As a three-time loser, Prof Hawking looks to be someone in the grip of a bad habit. Yet his persistence reflects a profound truth about the process of wagering, discovered in the 1920s.

At the time, mathematicians were wrestling with the problem of how to capture the fuzzy concept of belief. In particular, they were searching for the rules for changing belief in the light of fresh evidence: what could serve to measure the strength of "belief"? They found the answer in the mathematics of wagers, better known as probability theory. It turns out that the strength of a belief can be captured by probabilities ranging between zero and one, while the impact of fresh evidence follows the laws of probabilities.

Mathematicians found something else too: that using any other way of capturing belief can lead to situations where you can lose a wager even if you're right. Despite appearances, the eminent Professor Hawking is thus on rock-solid logical ground in using wagers to settle his disputes. His only problem is that what he believes often turns out to be wrong. Robert Matthews is Visiting Reader in Science at Aston University, Birmingham, England