Philosophy of Science
Why Science Doesn't Actually Advance the Way Scientists Think It Does
Thomas Kuhn published a book in 1962 that made the scientific community deeply uncomfortable — because it suggested that science progresses not through the steady accumulation of facts, but through something that looks a lot like revolution.
The Idea
The standard picture of science is tidy: hypotheses are tested, evidence accumulates, knowledge builds brick by brick toward truth. Thomas Kuhn, a physicist turned historian of science, looked at how science actually developed and found something messier and more interesting. In 'The Structure of Scientific Revolutions', he argued that science operates mostly within what he called a paradigm — a shared framework of assumptions, methods, and expectations that defines what counts as a legitimate question, a valid experiment, and a satisfying answer. Most scientists, most of the time, are doing what Kuhn called 'normal science': solving puzzles within the rules the paradigm sets. This isn't a criticism. It's genuinely productive. The paradigm focuses effort and makes progress possible. The trouble is that it also makes scientists systematically resistant to anomalies — results that don't fit. For a long time, anomalies get explained away, buried in footnotes, or quietly ignored. But they accumulate. And when enough anomalies pile up that the paradigm can no longer contain them, the field enters a crisis. What follows is a paradigm shift: not a refinement of the old framework, but its replacement with a new one. The unsettling implication is that scientists in different paradigms are not simply looking at the same world with better or worse tools — they are, in a meaningful sense, working within different worlds.
In the World
Consider the shift from the Ptolemaic to the Copernican model of the solar system — but not in the way it's usually told, as a simple triumph of observation over dogma. For decades, Ptolemy's Earth-centred model actually worked. Astronomers patched it with epicycles — small circles layered onto the orbits of planets — to account for anomalous movements, and the predictions held up well enough for navigation and calendar-making. The anomalies didn't immediately destroy the old model; they were absorbed into it. What Copernicus offered in 1543 wasn't better data — his system initially made predictions no more accurate than Ptolemy's. What it offered was a different way of seeing, a new framework that made certain puzzles dissolve and raised entirely new questions. It took decades, and the later work of Kepler and Galileo, before the new paradigm became the only defensible home for working astronomers. The people who resisted weren't fools or cowards. They were operating rationally inside a framework that had served well and wasn't obviously broken yet. Kuhn's insight was that this is how revolutions always go: not a single decisive experiment, but a slow-building crisis, followed by a conceptual rupture that can never be fully translated back into the language of what came before.
Why It Matters
Kuhn's framework changes how you watch science unfold in real time. When a new result challenges a well-established theory and gets dismissed or explained away by mainstream researchers, that's not necessarily intellectual cowardice — it might be normal science functioning exactly as designed. The question worth asking is whether the anomalies are accumulating. Fields on the edge of a paradigm shift often feel, from the inside, like bickering over details. From the outside, they look like productive debate. Only in retrospect does it become obvious that something structural was breaking down. There's a broader point too: expertise and paradigm are deeply entangled. The same training that makes someone a world-class physicist also makes it genuinely harder for them to see outside the assumptions that training instilled. This isn't a bug — it's what allows deep expertise to exist at all. But it means that the people most qualified to evaluate a crisis are sometimes the last to recognise one. Knowing this doesn't make you sceptical of science; if anything, it makes you more appreciative of how hard-won real revolutions are.
A Question to Ponder
Is there a belief you hold — not just in science, but in how you understand some part of the world — that functions more like a paradigm than a conclusion: something you use to interpret evidence rather than something the evidence itself could easily overturn?
Get a new one of these every morning.
Start learning with Thinkable