When Karl Popper was a young man in post-Hapsburg Vienna, he was obsessed by the boundary between scientific and nonscientific explanations: what he called “the problem of demarcation.”
Popper was a man of astonishing breadth of intellect. “Among the theories which interested me Einstein’s theory of relativity was no doubt by far the most important,” he wrote without false modesty years later. “Three others were Marx’s theory of history, Freud’s psychoanalysis, and Alfred Adler’s so-called ‘individual psychology.’”
Eddington’s observations of the solar eclipse of 1919, which confirmed Einstein’s predictions based on relativity, “thrilled” the young philosopher. Soon after, he became dissatisfied with the other three theories. They seemed fundamentally weaker than relativity.
He asked: “What is wrong with Marxism, psychoanalysis, and individual psychology? Why are they do different from physical theories, from Newton’s theory, and especially from the theory of relativity?”
The answer, Popper came to believe, had to do with the difference between verifiability and falsifiability.
The scientific method at the time was thought to proceed from observation, to hypothesis, to verification by experiment. The three flawed theories owed their prestige to the fact that they rested on an immense mass of empirical observations: ultimately, every data point within their fields could be made to conform with, and confirm, these theories. In this respect, Popper concluded, they were identical in method to astrology. They could never be wrong.
Einstein, by contrast, had made deductive predictions about the 1919 eclipse which could have proven relativity unsatisfactory or even false. Instead, of course, the opposite happened – Eddington’s findings made the theory more persuasive. It had passed a test. Such tests became for Popper the solution of the problem of demarcation.
The three theories which troubled him weren’t empty of explanatory power, any more than astrology was. But their explanations were not scientific. Conversely, science could never arrive at anything so final as the truth. But by constantly exposing explanations to the test of falsifiability, it weeded out untruths, and allowed only the strongest theories – in terms of prediction and explanation – to survive.
Whether Popper arrived at the scientific method is a question for another time. I have my doubts whether such a thing exists. While science is undoubtedly methodical, there may well be as many methods as there are scientific disciplines; the vast majority are learned in the lab rather than the classroom, and resemble modes of thinking and behavior more than logical formulas.
But in a larger sense, Popper was correct, and falsifiability added a crucial element to our understanding of how scientific explanations shed error. Much like fellow Austrian F. A. Hayek, who discovered a kind of Darwinian selection at work in the marketplace, Popper found science to be driven by an evolutionary series of “conjectures and refutations” – falsifiable theories and tests which either increased the theory’s power or fatally weakened it, making room for a fitter explanation.
Popper, among the last grandees of the rationalist tradition, sought to apply his method toward the improvement of social life. He called this approach “critical rationalism,” and he believed it could assess the rightness of proposed social improvement by evaluating the consequences, avoiding preconceived notions of right and wrong. In the area of “piecemeal social engineering,” the consequences, he thought, were the equivalent of falsifiability.
I have dealt elsewhere with the insurmountable problems raised by Popper’s critical rationalism. In brief, it applies a method for explaining reality to the quest for the good life: an uneasy slide from is to ought. Unlike science, which judges theories by the standard of reality (as best we understand it), Popper judged social change by the moral effect of the consequences – but to do this, he needed a morality which preceded the consequences, making the whole exercise pointless.
Strange for such a brilliantly logical thinker, he fell, like all consequentialists must, down the rabbit hole of an infinite regress.
Still, as commonly happens with the mistakes of geniuses, Popper’s speculations shed light on an important yet long-ignored problem: the maximum speed of cultural evolution. It’s orders of magnitude faster than biological evolution; but much more deliberate than the progress of science, and thus out of sync with social engineering of any kind.
This too is a story for another day.