Lewis Wolpert: 'As Francis Crick pointed out, a theory that fits all the facts will be wrong, as some of the facts will be wrong'

Francis Crick, who died in July, was one of the great scientists of the 20th century. He was famous for his discovery with Jim Watson of the structure of DNA, as well as other major contributions. His death made me reflect on the characteristics of great scientists. I wondered if there were aspects of Crick's life that could illuminate this. I knew and admired Francis, and once did a radio interview with him.

It is clearly an advantage to like science from an early age. There must be very few great scientists who come into science without an early commitment. Crick was interested in science as a child and even said to his mother that it was a pity everything would be discovered by the time he grew up, as there would be nothing left to find out.

Jim Watson wrote of him: "I have never seen Francis Crick in a modest mood. Perhaps in other company he is that way, but I have never had reason so to judge him." I think he was confusing a major characteristic of Crick, which seems to me to be his self-confidence. This was very important for his success. During the war, he was in the Admiralty, and at the end of the war he was 30 and did not know what to do. While he had studied physics he had not published anything yet. Not a basis for a promising career. He had the courage to decide that he should pursue what he enjoyed gossiping about. He realised that his interest lay in the way the brain worked and the boundary between the living and the non-living. He chose the latter, and became a student again, studying cells, but then again had the confidence to change to the X-ray diffraction of proteins, and worked for a PhD.

He thought that choosing what to work on was a crucial decision. For him it was important to choose a subject not a problem. Then he moved round in the subject until something yielded, something clicked together. He would not say that he was going to solve such and such a problem, since it might be insoluble. It was important to find a problem that was tractable, a great skill for any scientist. Later, he did not work on how proteins folded given their chemical composition, which he considered intractable.

This fits well with Peter Medawar's aphorism that science is the art of the soluble.

He thought it dangerous to get too bogged down in experimental details. It was important to make bold assumptions and try them out. Many students, he thought, failed to do that and so missed making important discoveries. As a theoretician he had to work on the raw material that was the result of other people's experiments. He thought it important to meet those scientists in order to judge the validity of their work. As he famously pointed out, a theory that fits all the facts is bound to be wrong, as some of the facts will be wrong. And as Gerald Holton pointed out, the graveyard of failed scientists is filled with those who gave up their ideas at the first evidence that it was wrong.

Collaboration was very important to him. You had to be friends and also very candid, without being rude. You could then sound aggressive, and say it's all nonsense, without causing offence. Collaboration was thus for him an invaluable way of getting rid of false ideas. This also meant that you must meet frequently. For example, he shared an office with Sydney Brenner for 20 years. They had a rule that they could say anything that came into their heads. Most of the conversations were nonsense, but now and then something of interest emerged and could even lead to experiments.

I always found that he expressed himself with tremendous clarity. He explained that this was not self-conscious, but that he had to explain things to himself in a fairly straightforward way. It was the way he talked to himself. He is sadly missed.

Professor Wolpert is professor of biology as applied to medicine at University College London