The Structure of Scientific Revolutions: Chapter 8 Summary & Analysis

Though crisis causes scientists to abandon old paradigms, Kuhn believes that—at least according to the various historical examples he has studied—scientists never do so unless they have a new paradigm to replace the old one. To reject a paradigm without having another one is, Kuhn argues, “to reject science itself”; it would mean returning to a scattershot collection of opinions and beliefs.

But while anomalies lead to crises, counterinstances (or moments in which the paradigm does not behave exactly as expected) are an everyday part of normal science. Kuhn thus posits that there is no clean line between what is an anomaly and what is merely another challenging puzzle for scientists to solve. And often, when scientists are unable to solve these puzzles, their colleagues see it not as the failure of the paradigm but as the failure of that individual scientist. Moreover, sometimes anomalies are resolved by discoveries in another field that offer a surprising solution.

 Only very special kinds of anomalies, then, create crises. Sometimes this is because the anomaly has practical importance for society, as in the case of Copernicus (the Catholic Church was struggling to create an accurate calendar, and Copernicus was trying to figure out why). Sometimes, the anomaly grows more glaring as the field advances; sometimes, the anomaly cuts immediately and clearly to the heart of the paradigm. When a great many scientists have to pay attention to one anomaly, their responses to that anomaly start to conflict, and the paradigm begins to collapse.

Sometimes, scientists themselves recognize this breakdown: Einstein once said a moment of crisis in his field was “as if the ground had been pulled out from under one, with no firm foundation to be seen anywhere.” But more often than not, crisis is not explicitly named. And indeed, normal science is often able to resolve crisis. In other cases, scientists give up, deciding they do not have the necessary equipment or knowledge to create a new paradigm. A complete scientific revolution, then, is relatively rare. But each crisis does “loosen” the rules and stereotypes of the paradigm it takes place in.

Moreover, as Kuhn insists, a true paradigm shift is not cumulative; instead, it requires scientists to go back to basics. To illustrate this point, Kuhn uses the idea of the Rorschach test (though he does not call it by this name). If in the old paradigm, scientists looked at a picture and saw a bird, now they have rotated the paper—and so they see an antelope.

Kuhn then begins to describe the process of “extraordinary science.” In contrast to normal science, which tries to reject or resolve the anomaly, extraordinary science works with the anomaly to create a new paradigm. However, Kuhn notes that it is more difficult to describe this process purely with historical fact; here, he admits that he is conjecturing more than he was in other parts of the book.

The first step of extraordinary science, Kuhn argues, is to test out normal science by “push[ing] the rules” of a paradigm as far as they will go. The second step, as practiced by Copernicus and Einstein, is to isolate the anomaly and make it seem clearer and more out of place than it did initially—to “localize and define” what is actually differing from the paradigm’s expectations. 

Most importantly, Kuhn believes that extraordinary science often goes hand-in-hand with new philosophical thought (much of which comes out of the humanities). For example, Einstein’s theory of relativity happened alongside a sea change in moral and social theory; it is not a coincidence, Kuhn writes, that “thought experiments” are a fundamental part of new paradigms.

Kuhn does not claim to understand how a person can eventually arrive at the beginnings of a new paradigm—how such an idea “emerges all at once, sometimes in the middle of the night.” But he does note that such people are usually very young or very new to the field, and thus less indoctrinated into a given paradigm’s rules and expectations.