Take the example of one of the greatest scientists of the 19th century, Lord Kelvin, who was not immune to overconfidence. In a 1902 interview “on scientific matters now before the public,” he was asked about the future of air travel: “Have you no hope of solving the problem of aeronautics in any way?”
Lord Kelvin replied firmly: “No; I don’t think there is any hope. Neither the balloon, nor the airplane, nor the gliding machine will succeed in practice.” The first successful flight of the Wright brothers was a little more than a year later.
Scientific overconfidence is not limited to technological matters. A few years earlier, Kelvin’s famous colleague, AA Michelson, the first American to win a Nobel Prize in science, expressed a similarly brilliant view about the basic laws of physics: “Most of the basic principles are probably now solid. established.”
Over the next few years – largely due to Michelson’s own work – fundamental physical theory underwent the most significant changes since Newton’s time, with the development of relativity and quantum mechanics “radically and irreversibly” changing our view of the physical universe.
But is this overconfidence a problem? Maybe it really helps the progress of science? I suggest that intellectual humility is a better and more progressive position for science.
Thinking about what science knows
As a researcher in the philosophy of science for over 25 years and a one-time editor of the leading journal in the field, Philosophy of Science, I have studied and reflected on the nature of scientific knowledge across my desk. The biggest questions are not settled.
How confident should people be about the conclusions of science? How confident should scientists be in their own theories?
One thought is the so-called “pessimistic induction,” most advanced today by the philosopher Larry Laudan. Laudan pointed out that the history of science is filled with discarded theories and ideas.
It would almost be tempting to think that now, at last, we have found the science that will not be discarded. It is far more reasonable to conclude that today’s science will be largely rejected by future scientists, or significantly altered.
But the pessimistic induction is not the end of the story. An equally powerful theme, prominently advanced today by the philosopher Hilary Putnam, is the so-called “no-miracles argument.” It would be a miracle, so as they say, if successful scientific predictions and explanations were only by accident, or bad luck – that is, if the success of science did not result from getting something right about nature of reality.
There must be something right about the theories that, after all, made air travel – not to mention space travel, genetic engineering and so on – possible. It would be almost tempting to conclude that today’s theories are simply wrong. It is much more reasonable to conclude that there is something right about them.
A pragmatic argument for overconfidence?
Philosophical theorizing aside, what is best for scientific progress?
Of course, scientists can be wrong about the accuracy of their own positions. However, there is reason to believe that over the long arc of history – or, in the cases of Kelvin and Michelson, in relatively short order – such mistakes will be exposed.
In the meantime, perhaps trust is very important to doing good science. Perhaps science needs people who pursue new ideas hard with the kind of (over)confidence that might lead to rare declarations about the impossibility of air travel or the finality of physics. Yes, it can lead to dead ends, retractions and the like, but maybe that’s just the price of scientific progress.