Carlo Rovelli, in his book There Are Places in the World Where Rules Are Less Important Than Kindness, wants us to take Aristotle seriously as a scientist. There’s something appealing about the idea. Aristotle was intensely curious about the natural world. He observed, classified, compared, and reasoned. Unlike Plato, he wasn’t content to contemplate ideal forms. He wanted to know how nature worked. Some of his observations and conclusions were remarkably good, others strikingly poor.
I think there’s a distinction here that gets lost, particularly when modern physicists undertake the rehabilitation of ancient thinkers. Attentive observation, empirical testing, and theoretical revision are different intellectual accomplishments. They don’t necessarily travel together.
Aristotle was an attentive observer, as Rovelli notes. He collected observations, classified organisms, and made distinctions that others had overlooked. But attentiveness, however admirable, is quite a bit less than science. A sophisticated theoretical reconstruction of Aristotle can make Aristotle look more scientifically sophisticated than his actual investigative methods warrant.
Empirical testing requires something more than attentiveness. It means devising observations or experiments specifically to discriminate between competing explanations. It means looking for evidence that might expose a weakness in a preferred account – disconfirming evidence, we now say. There’s a difference between noticing what happens and arranging things so that competing explanations must make different predictions. The latter puts a theory at risk.
Even that isn’t enough. A person can be an exemplary observer and experimenter yet remain reluctant to accept the implications of the evidence. Tycho Brahe is a useful example (see earlier post). He made observational astronomy a precision science, but his commitment to a physically intelligible, Earth-centered cosmos survived evidence that undermined important parts of the old Aristotelian picture. His observations were revolutionary. His interpretation was less so.
Kepler, working with Tycho’s measurements, eventually crossed a boundary that Tycho did not. The famous eight-minute discrepancy in Mars’s orbit was too large to dismiss. Kepler had spent years trying to preserve circular planetary motion. He finally abandoned it. The observations defeated an idea that had both mathematical elegance and considerable philosophical authority.
Neither man was simply rational or irrational. Both had theoretical commitments. The difference was what happened when those commitments faced troublesome evidence.
This is the part of the Scientific Revolution that interests me most. It wasn’t simply an accumulation of better observations, nor the replacement of one set of explanations with another. Something changed in the relationship between theory and evidence. Theories became answerable to the natural world in a new and more demanding way. They had to survive tests that their authors might not welcome and whose results they could not necessarily anticipate.
Consider Aristotle’s account of falling bodies. He held that heavier objects fall faster than lighter ones. The difference was observable, but so were the complications. A large rock and a feather differ in more ways than weight. Shape and resistance matter. A little systematic investigation could have raised awkward questions about the explanatory power of the theory. The point isn’t that Aristotle should have anticipated Newtonian mechanics. It’s that a theory about falling bodies ought to invite attempts to find where it fails.
Rovelli points out that features of Aristotelian physics can be recognized within Newton’s theory. That’s true, and interesting. But the ability to recover an earlier theory as an approximation within a later one tells us something about the mathematics and the physical domain in which the approximation works. It tells us rather less about how the earlier theory was established, how rigorously it was tested, or how its author responded to contrary evidence.
There is a temptation, particularly among theoretical physicists, to emphasize the continuity of scientific thought. The great theories of the past survive, they say, as approximations, special cases, or useful ways of thinking about particular phenomena. This is mathematically satisfying and often historically illuminating. But it can obscure a more consequential discontinuity: the change in the status of empirical evidence.
The universe hasn’t changed. Our willingness to let it overturn our explanations has. Perhaps this is why the rehabilitation of Aristotle sometimes seems to have a contemporary subtext. Theoretical physicists work at considerable distances from direct observation, conceiving mathematical structures whose empirical consequences may be difficult to test. A history of science emphasizing the continuity of great theoretical ideas offers a reassuring pedigree. Aristotle, suitably reconstructed in modern mathematical language, becomes an ancestor rather than a cautionary example.
I don’t know whether this is what motivates Rovelli. I suspect, from reading several of his essays, that the tendency is more than incidental. The question of motive is less important than the argument itself. Mathematical sophistication and scientific method aren’t interchangeable. A theory can be lovely and intellectually impressive, and even approximately right, without having been established by a process that deserves the name of modern empirical science.
Aristotle deserves our attention. He doesn’t need to be made into a modern scientist to earn it.
