VIENNA, 1919: a city scarred by lost war and empire. Navel-gazing is in order, and Sigmund Freud’s new ideas of the subconscious and psychoanalysis are all the rage. One young apprentice cabinet-maker is brooding – how could anyone prove them true, when the subconscious is unknowable?
Karl Popper soon abandoned cabinet-making for a loftier pursuit. He wanted to find a way to demarcate ideas like Freud’s from what he saw as a truer kind of knowledge: science.
Advertisement
Get your head around the most mind-bending concepts in science
From black holes to blockchains, from consciousness to the multiverse, we explain it all in this 13-part special
It still isn’t at all easy. By its nature, science can’t rely on logical deduction alone, or build up knowledge purely from incontestable truths (see “How to think about… Logic”). It must make leaps into the unknown, just as Freud did, formulating hypotheses and searching for evidence of their truth.
This is called induction, and it hides a niggle described by philosopher David Hume 150 years before Popper. The classic example involves checking the colour of as many swans as you can find, then extrapolating a rule to say “all swans are white”. That sounds like science. But it can’t lead to reliable knowledge, Hume argued, because you can never know a black swan isn’t in the next pond.
Popper’s resolution seems oddly negative: science is about proving not truth, but falsehood. The crucial thing is that when you find evidence that disproves a scientific hypothesis, you discard or amend that hypothesis. You can never find truth exactly, but by slowly ruling out ideas, you edge closer to it. When at some point the weight of evidence seems overwhelming, your hypothesis becomes a scientific theory, like the general theory of relativity, the theory of evolution by natural selection or the theory of human-induced climate change.
The thing is, scientists don’t stick to Popper’s strict criteria in practice, and often follow their hunches or look to confirm rather than refute their theories. “The thing about scientific methodology is that it defies simple summary,” says , a philosopher of science at the University of Bristol, UK. “You always have to apply the principles in a context-specific way.”
Alternative Facts
With no way for us to say “this theory is definitely true”, the door is left wide open for alternative facts – especially where the conclusions of empirical study are unpalatable, as with climate change. And things can easily sound scientific when they’re not, especially when they play to your prejudices or make the news. Take a study from last year that claimed to have found genes that determine whether people like Marmite, a yeasty breakfast spread popular in the UK. It involved DNA tests, statistics and a sample of 261. So far, so scientific, but the way the study was conducted meant it had no power to connect cause and effect.
In 2015, indeed, found that a third of research papers in psychology appear to have reached spurious conclusions. Ladyman’s answer is that science is not one thing, but a set of related disciplines. When it comes to the ability to purvey truth, we should distinguish clearly between the utterly reliable physics of tide tables, say, and the messier end of psychology.
That said, how did we uncover the crisis in psychology research? Through science, when scientists sought to replicate results and failed. Without blemish it ain’t, but science is at least uniquely willing to stare at itself in the mirror. To corrupt a Winston Churchill quote, it is probably the worst way of seeking truth – apart from all the others that have been tried from time to time.
This article appeared in print under the headline “How to think about… Scientific truth”
