to notice a mistaken assumption. A PhD student can prove a Nobel prize-winner wrong.
If at some future date new observations conflict with what we think we know today, scientists will – after considerable soul-searching, some stubborn conservatism, and a lot of heated argument – revise their theories to resolve the difficulties. This does not imply that they are merely making everything up as they go along: each successive refinement has to fit more and more observations. The absence of complete certainty may seem a weakness, but it is why science has been so successful. The truth of a statement about the universe does not depend on how strongly you believe it.
Sometimes an entire area of science can become trapped in a massive conceptual error. A classic instance is ‘phlogiston’. The underlying scientific problem was to explain the changes that occur in materials when they burn. Wood, for instance, gives off smoke and flame, and turns into ash. This led to the theory that wood emits a substance, phlogiston, when it burns, and that fire is made from phlogiston.
Volume 2 of the first edition of Encyclopaedia Britannica, dated 1771, says: ‘Inflammable bodies … really contain the element fire as a constituent principle … To this substance … chemists have assigned the peculiar title of the Phlogiston, which is indeed no other than a Greek word for the inflammable matter … The inflammability of a body is an infallible sign that it contains a phlogiston …’ The same edition considers ‘element’ to mean earth, air, fire or water, and it has a fascinating analysis of the size of Noah’s Ark, based on its need to contain only a few hundred species.
As chemists investigated gases, and started weighing substances, they made a discovery that spelt doom for the phlogiston theory. Although ash is lighter than wood, the total weight of all combustion products – ash, gas and especially steam – is greater than that of the original wood. Burning wood gains weight. So, if it is emitting phlogiston, then phlogiston must have negative weight. Given enough imagination, this is not impossible, and it would be very useful as an antigravity device if it were true, but it’s unlikely. The discovery of the gas oxygen was the clincher: materials burn only in the presence of oxygen, and when they do, they take up oxygen from their surroundings. Phlogiston was a mistaken concept of ‘negative oxygen’. In fact, for a time oxygen was referred to as ‘dephlogisticated air’.
Significant changes in scientific orthodoxy often occur when new kinds of evidence become available. One of the biggest changes to our understanding of stars came when nuclear reactions were discovered. Before that, it seemed that stars ought to burn up their store of matter very rapidly, and go out. Since they visibly didn’t, this was a puzzle. An awful lot of argument about the Sun’s remarkable ability to stay alight disappeared as soon as scientists realised it shone by nuclear reactions, not chemical ones.
This discovery also changed scientists’ estimate of the age of the solar system. If the Sun is a very large bonfire, and is still alight, it must have been lit fairly recently. If it runs on nuclear reactions, it can be much older, and by studying those reactions, you can work out how much older. The same goes for the Earth. In 1862 the physicist William Thompson (later Lord Kelvin) calculated that on the ‘bonfire’ theory, the planet’s internal heat would have disappeared within 20-400 million years. His approach ignored convection currents in the Earth’s mantle, and when these were taken into account by John Perry in 1895 the age of the planet was revised to 2-3 billion years. Following the discovery of radioactivity, George Darwin and John Joly pointed out in 1903 that the Earth had its own internal source of heat, caused by radioactive decay. Understanding the physics of radioactive decay led to a very effective method for dating ancient rocks … and so it went. In 1956 Clair Cameron Patterson used the physics of uranium decaying into lead, and observations of these elements in several meteorites, to deduce the currently accepted age of the Earth: 4.54 billion years. (The material in meteorites formed at the same time as the planets, but has not been subjected to the same complicated processes as the material of the Earth. Meteorites are a ‘frozen’ record of the early solar system.)
Independent verification has come from Earth’s own rocks; in particular, tiny particles of rock called zircons. Chemically,