Francisco Crick in Paradiso

Richard Dawkins has written another book on evolution. I haven't read it, but I noticed that one reviewer thought that the force of Dawkins' arguments was becoming diluted by a combination of militant atheism and over-flamboyant prose. To conservative scientists like me, the idea of selfish genes, while certainly snappy, leads to ignoring the biology surrounding the genes and, in the end, to a distorted view of evolution. I do sympathize with Dawkins, however: he faces a tough problem in trying to convince people that natural selection explains evolution. The resistance does not come from any profound religious beliefs, but rather from a deep feeling that it can't work. It's very difficult for anybody to believe that making random changes in a television set, or even in the plans for a factory making television sets, will convert it from black-and-white to colour. Our common experience with anything complex is that the most likely result of tampering with it will be to break it. Human-designed systems have certain properties which stem from the nature of engineering and are related to the limitations of our mental processes. We need to impose very severe constraints on complicated designs to get anything to work. Because we are unable to talk or think about more than a very small number of processes taking place simultaneously, we isolate them into subassemblies so each can be treated separately. We also have to be absolutely explicit about how things should act in time; causality must be obeyed, and if X causes Y, then X must appear before Y. We also like hierarchical systems to make explicit the flow of control. I used to think that these principles of modularity, rigorous sequentiality, and hierarchical control might underlie the structure and function of all elaborate systems. They are certainly true for writing a large piece of software or making a watch; in each case, even small departures from the original construction will produce a mess. I now believe that while these principles may be at the heart of artificial engineering, natural engineering is different. Biological systems have processes which are more flexibly organized and capable of displaying more resistance to lethal alterations, and have more versatility in adaptive responses. Thus the evolution paradox resolves itself as follows. If we persist in thinking that natural systems are like artificial ones, we will need a designer to impose the same constraints on …