Now some people approach that by saying what you should do is teach the theory and the facts will then emerge. But that is wrong for two reasons. The first, as I’ve already said is, that the theories will change as we understand more deeply about the nature of atoms and molecules. The second reason is that it is philosophically wrong; you cannot say sodium chloride has the properties it has because so and so, because you are using the theories to explain the facts, and this is not a very sound way of proceeding. What you should really do is start with the facts and try and generate… it’s going the wrong way round. One person who encapsulated that particular idea very well was, of course, Karl Popper who said there is no point in proving that all swans are white, because you can go anywhere in Europe and find white swans and that is certainly true. But you only have to go to Perth in Western Australia, where the Swan River... it is built on the fact that the swans there are black. Popper’s point was that the one observation of a black swan disproves the fact that all swans are white. And whilst you can carry that idea too far as well, it’s very important to try and think of ways which undermine the existing ideas to get an advance. To take a simple example, you can say, that water boils at 100 degrees Celsius – why? Well, you can think of theories but, in fact, if you then climb a mountain and do the same experiment, you find that water boils at 95, so you’ve got a new fact to explain and you have to think about the pressure that the water is under, and so forth.