Reinventing Discovery - Michael Nielsen [41]
The examples of Einstein and of Watson, Crick, and Franklin illustrate the strength of the shared praxis in science. To an extent unusual in many parts of life, in science it’s often the person with the best evidence and best arguments who wins out, and not the person with the biggest reputation and the most power. Pauling may have been widely acknowledged as the world’s leading chemist, but other chemists could see just as surely as Watson and Crick that Pauling’s structure was simply wrong. This strong shared praxis makes science well suited to collective intelligence.
This strong shared praxis doesn’t mean that science is a clean and simple process. The actual day-to-day process of doing science is messy and speculative and filled with error and argument. The scientist Richard Feynman was so full of irrepressible brainwaves and “great” ideas, most of which later proved to be wrong, that according to his biographer James Gleick his cannier colleagues developed a rule of thumb: “If Feynman says it three times, it’s right.” The same could be said for many scientists. Often a scientist begins an investigation with little more than a whiff of an idea, a suspicion that some hypothesis is true. They sketch out a way of testing it, often vaguely at first, gradually filling in more and more details. Experiments often need to be performed many times, with the experimental design gradually changed and improved, as the scientist understands better what evidence is required in order to be convincing. All this is a slow process that involves lots of speculation and argument and false starts, as the scientist gradually moves to more and more robust arguments and evidence. The end goal, though, is a set of st argents and evidence that adheres to the shared praxis of the field. And that is quite unlike a discussion of Bach-versus-the-Beatles, or a political discussion, or a discussion of Shakespeare, where in the end there may remain a fundamental division over basic values. Of course, scientists do still sometimes publish wrong or mistaken or unconvincing papers. But even when a scientist publishes such a result, other scientists can go back and repeat the experiments to find flaws, or point out shortcomings in the arguments. In short, they can retest the results against the shared praxis of the field, and find them wanting. It’s this ability to be wrong in a clear-cut way that enables forward progress. In this sense science is, as I said earlier, already one big collaboration, held together by common standards of evidence and reasoning.
Are there parts of science without a shared praxis, parts more like economics, say, where the problems are so challenging that the field is still a proto-science, with shared knowledge and techniques only starting to emerge? As an example, one of the big open problems of physics is the problem of finding a quantum theory of gravity—a single theory that unifies both quantum mechanics and Einstein’s theory of gravity. It’s one of the toughest problems of physics, a problem that has defeated the best minds for decades. In the 1980s an approach to the problem known as string theory rose to prominence, and gradually came to dominate work on quantum gravity. At the same time, a much smaller number of physicists continued to pursue other approaches to quantum gravity. In recent years a debate called by some the “string wars” has been waged between advocates of the different approaches. Many physicists claim string theory is the only reasonable approach to quantum gravity. Others, including Stephen Hawking, Roger Penrose, and Lee Smolin, believe different approaches are worth pursuing. Remarkably, some prominent string theorists dismiss the non-string theorists not just as wrong, but as misguided, or even as fools. When such a fundamental division occurs, it is nearly impossible for large groups to collaborate across that division. Collective intelligence can only be