Popper as Historian: The Case of Kant
March 29, 2025
Karl Popper is underrated as a historian. His worldview was centered on the idea of a problem. He thought that knowledge growth begins with problems—not with observations or data or experience. His political philosophy was focused on piecemeal, incremental reforms aimed at specific problems, instead of ideologically driven radical approaches. His critique of historicism was based on the idea that today’s problems lead to solutions, which in turn create new problems that cannot be predicted. “All life is problem solving,” he was fond of saying.
Popper’s obsession with problems led him to read history differently than many people. He would never present the content of someone’s argument without understanding what particular problem they were trying to solve. This gave him tremendous respect for people who were ultimately wrong, but nevertheless solved a problem that were plaguing people at the time. He appreciated both empiricism and logical positivism, philosophical traditions which he ultimately abandoned but which helpfully pushed back against subjectivism and religious dogmatism.
This approach to history sounds obvious—isn’t this what all historians do? But when you read Popper it’s clear that he takes the idea of understanding the context of each thinker more seriously than most. The uniqueness of his claims drips from every page.
Popper wasn’t an uncontroversial historian. His problem-oriented perspective led him to theories that other historians wouldn’t consider kosher. His theory that Plato’s theory of forms rose from troubles with Pythagorean atomism1 is not something I’ve seen corroborated elsewhere, and historians have criticized his readings of Kant, Hegel, and Plato. (He did teach himself ancient Greek, surely he gets some sort of break for that.)
But his perspective is also a refreshing way of understanding history. And it’s often very insightful. Popper changed the way I think about Immanuel Kant and here I want to discuss two of the problems Kant was considering that led him to his critique of pure reason and his notion of a priori knowledge. (If you want to hear some version of this in audio form, episode 80 and episode 59 should keep you entertained.)
Immanuel Kant, presumably either wondering if it's ethical to go to a barber, or wondering how on earth you leave this city with all it's bridges.
Finitude of the universe
Kant was born, lived, and died in Konigsberg (now Kaliningrad). During all the time he spent not leaving Konigsberg, he naturally pondered whether the universe was finite or infinite in time. This was a problem left open by Newtonian mechanics, the going theory of physics at the time. (This was, of course, more than 100 years before Einstein.)
Kant thought he had a logical proof for both. You can read these in detail in his Critique of Pure Reason, but the arguments are roughly as follows:
- Suppose the universe is infinite. Then an infinite number of years elapsed before this moment. But this is impossible, since we’ve reached this moment. Therefore, the universe must be finite.
- Suppose the universe is finite. Consider whatever timeless void was before the birth of the universe. In this void, each “interval” (whatever that means) of “stuff” (whatever that is) is identical in temporal relation to the others, since time doesn’t exist. But the interval that immediately preceded the beginning of the universe was not identical to the others. In particular, it’s demarcated by being closer to the beginning than the others. This is a contradiction, hence the universe must be infinite.
If your eyebrows aren’t somewhere near your hairline, I’d be surprised. Neither of these arguments make sense. The first fails because one-sided infinities exist. (Take the natural numbers—you can count to any number in a finite amount of time, even if the set as a whole is infinite.) And the second argument begs the question. What is an interval in a timeless void?
But put yourself back in the 1700s when this was first-rate argumentation, and try to take both conclusions seriously. You would be flabbergasted—you’ve just logically proved that the universe must be both finite and infinite! How did Kant resolve this paradox? He concluded that there are certain questions that reason alone cannot answer. There were limits of pure reason. As Popper puts it:
What lesson did Kant draw from these bewildering antinomies? He concluded that our ideas of space and time are inapplicable to the universe as a whole. We can, of course, apply the ideas of space and time to ordinary physical things and physical events. But space and time themselves are neither things nor events: they cannot even be observed: they are more elusive.
- Conjectures and Refutations (C&R), page 242
Kant thought the question “is the universe finite or infinite” exceeds the boundaries of human knowledge. The contradiction arises because we’re trying to apply concepts (like space and time) that only make sense within our own experience, and apply them too broadly. This led Kant to what he called “transcendental idealism,” the view that space and time are not objective features of reality, but rather forms of human intuition (a priori knowledge) that help structure our experience. Explained concisely:
Space and time are not part of the real empirical world of things and events, but rather part of our mental outfit, our apparatus for grasping this world. Their proper use is as instruments of observation: in observing any event we locate it, as a rule, immediately and intuitively in an order of space and time.
- C&R, page 242
Newtonian Mechanics
According to Popper, Kant was certain that Newtonian mechanics was true.
Like almost all of his contemporaries who were knowledgeable in this field, Kant believed in the truth of Newton’s celestial mechanics. The almost universal belief that Newton’s theory must be true was not only understandable but seemed to be well-founded. Never had there been a better theory, nor one more severely tested. Newton’s theory not only accurately predicted the orbits of all the planets, including their deviations from Kepler’s ellipses, but also the orbits of all their satellites. Moreover, its few simple principles supplied at the same time a celestial mechanics and a terrestrial mechanics.
- C&R, page 250
Again, at that point, Newtonian mechanics was the only thing on offer. There were no observations that contradicted its predictions. It was “a universally valid system of the world that described the laws of cosmic motion in the simplest and clearest way possible—and with absolute accuracy. Its principles were as simple and precise as geometry itself.” (C&R, page 250).
The idea of finding perfect truth in a scientific theory is difficult to wrap our minds around now—fallibilism has crept into the water. Today, even our most successful theories do not command this status. Despite our inability to refute or improve general relativity or quantum mechanics2, no scientist is tempted by the idea that we’ve reached the end of the story, that we’ve found the final theories of physics.
But Kant was faced with a problem. How did we know Newtonian mechanics was correct? Newton himself thought that his theory was a result of induction. He thought it could be logically derived from observations about rocks falling down hills, apples falling from trees, and the orbits of planets. But Kant had been convinced by Hume that this was impossible (all that dogmatic slumber business). Hume showed that it was impossible to logically prove the truth of general theories from finitely many observations. Moreover, Newton’s theory was precise and mathematical, and observations are imprecise and subjective. The idea that Newtonian mechanics came from induction was, for Kant, both logically and intuitively problematic.
Kant solved this problem by turning once again to a priori knowledge. Our brain, he thought, came pre-equipped with ideas like causation and geometry. We evaluate observations and generate theories which accord with these building blocks. Like puzzle pieces snapping into place, Newtonian mechanics was the unique synthesis of our observations and our intuitions. As Popper explains it:
Kant approached this problem by first considering the status of geometry. Euclid’s geometry is not based upon observation, he said, but upon our intuition of spatial relations. Newtonian science is in a similar position. Although confirmed by observations it is the result not of these observations but of our own ways of thinking, of our attempts to order our sense-data, to understand them, and to digest them intellectually. It is not these sense-data but our own intellect, the organization of the digestive system of our mind, which is responsible for our theories. Nature as we know it, with its order and with its laws, is thus largely a product of the assimilating and ordering activities of our mind. In Kant’s own striking formulation of this view, ‘Our intellect does not draw its laws from nature, but imposes its laws upon nature’.
- C&R, page 244
Not many scientists or philosophers buy the idea of a priori knowledge now (not in the way Kant was thinking about it at least, but Kant was of course correct that we’re not blank slates). But Kant’s philosophy helped break the spell that generating knowledge is easy and passive—that we learn by simply observing the truth which is the world is serving us. Kant helped us realize that we must actively theorize about the world, using observations not as an oracle for truth but instead as tests for our theories. And these insights laid the foundations for Popper’s own philosophy of critical rationalism.
Subscribe to get notified about new essays.
