The Structure of Scientific Revolutions

Almost seven years after the initial publication of The Structure of Scientific Revolutions, Kuhn returns to clarify some of his ideas. Partly, he is responding to readers’ criticisms or misunderstandings. Partly, he is hoping to incorporate his own later knowledge and research.

First, Kuhn reiterates that paradigms are circular—and therefore he wishes that before leaping into this circular narrative, he had begun his original text with a discussion of “the community structure of science.” More than in other professions, Kuhn believes that scientists belong to communities: the members of a community have had very similar educations, and they have very specific sets of shared goals. Within these groups, there may be many sub-groups (and in fact, each sub-group may only have a few hundred members). 

Kuhn also specifies that even in pre-paradigm periods, scientific communities share some basic ideas and beliefs. What really changes in a paradigm shift is that the shared beliefs become more specific—they offer more “challenging puzzles” and supply better “clues to their solution.”

Finally, Kuhn responds to the criticism that he only cares about major scientific revolutions (ones that affect large groups of people). On the contrary, Kuhn believes that the smaller, everyday scientific revolutions—which may affect as few as 25 people—are the most important, as it is these revolutions which most demonstrate the need for Kuhn’s argument. Similarly, he acknowledges that the crises that start paradigm shifts may be introduced from other disciplines or subgroups.

In the next section, Kuhn revises his blanket use of the term paradigm, which he feels he originally used in two contradictory ways. Rather than saying that scientists all share a single paradigm, he coins the new term “disciplinary matrix” to describe the shared set of theories, rules and beliefs that guide a given discipline at a given moment.

One important part of any disciplinary matrix is the “symbolic generalizations” shared by a group of scientists (whether that is a set of rules or a set of definitions). But there is also a deeper aspect to these disciplinary matrices—these frameworks, Kuhn notes, “supply the group with preferred […] analogies and metaphors.” And finally, Kuhn notes that shared disciplinary matrices dictate a set of shared values, whether that is an emphasis on prediction or on accuracy or on plausibility. However, values may also differ (to some extent) between individuals in the group.

Kuhn also redefines the crucial problems of a given paradigm as “exemplars.” These exemplars (usually famous experiments that helped to clarify the overarching disciplinary matrix) help students learn about a field, and they are also the main source of symbolic generalizations. 

In the next section, Kuhn argues that exemplars deserve special attention because “the paradigm as shared example is the central element of what I now take to be the most novel and least understood aspect of this book.” Kuhn argues that when a student tries out several textbook problems using certain rules and assumptions, they begin to assimilate a “time-tested and group-licensed way of seeing.” In other words, they begin to view the world according to their discipline’s framework. 

To exemplify this, Kuhn references the phrase “actual descent equals potential ascent”—this is a law built on Galileo’s experiments rolling a ball down an incline. These words mean nothing to a student who has not had some sort of exposure to “the ingredients of nature” as Galileo understood them. But after the student does several problems involving motion, weight, and inclined planes, they can begin to understand these words as other scientists mean them. There is thus a kind of “tacit knowledge” involved in paradigms, “which is learned by doing science” (trying out textbook-like problems) “rather than acquiring rules for it.”

Next, Kuhn clarifies his claims about intuition. Rather than referencing intuition as a mystical force, he explains that he is talking about the different ways that people can feel and perceive almost identical stimuli.  Because people’s own personal worlds are determined not by stimuli but by the “sensations” they feel in response to those stimuli, everyone does to some extent live in a different world from everyone else. Kuhn then argues that one of the fundamental principles of a paradigm is that it allows various members of a scientific group to feel the same sensations in response to the same stimuli. More than just a shared set of rules, then, this kind of shared seeing—like the shared “tacit knowledge” Kuhn has just discussed—defines a paradigm.

Kuhn thus calls attention to the neural apparatus that governs perception. In particular, he argues that just as certain ways of perceiving allow humans being to survive from one generation to the next, certain responses to stimuli are more effective than others—and are thus easier for one generation of scientists to hand down to the next. 

Speaking mostly to the philosophers of science who criticized his original text, Kuhn clarifies his remarks about the incommensurability of paradigms. Rather than saying that believers of different paradigms can never understand each other, Kuhn specifies that “translation” of different words and concepts gives scientists operating under different paradigms some small ways of understanding one another.

Translation across paradigms is therefore one of the crucial tools of persuasion. However, Kuhn is realistic about the fact that translation is often difficult and complex—especially because it is so foreign to the practice of normal science. He also acknowledges the principle (which he draws from linguistics) that understanding a theory in translation is very different from actually experiencing that theory in its original form. All translation may really do, then, is provide “points of entry” to an otherwise-strange paradigm.

In the penultimate section, Kuhn responds to criticism that he has taken a relativist view of science. To make his case, he argues that one could make a sort of evolutionary tree of all modern scientific specialties. One could then easily formulate a list of criteria—“accuracy of prediction,” “simplicity, scope, and compatibility with other specialties”—that would show how the more recent theories have advanced beyond the first ones.

At the same time, while later scientific theories may be simpler or more accurate predictors than their predecessors, Kuhn reiterates his belief that science is still not getting any closer to an objective truth—to what is “really there.”

In his final section, Kuhn responds to two dominant views of his original book. Critics believe Kuhn is switching back and forth between description (how things are) and prescription (how things should be). Kuhn feels that this is a less clear distinction than many would like to pretend. Indeed, he believes that his argument both describes how scientists do act and suggests how they should act in the future.

Kuhn also is uncomfortable with the many readers who applaud his work because it can be applied to other fields. On the one hand, his work does apply the revolutionary structure of politics or art to science; as Kuhn puts it, “revolutionary breaks in style, taste and institutional structure” are a central part of art history and political history alike.

But on the other hand, Kuhn reiterates that he is most interested in the way that science is different from other fields. There is less room for competing conclusions in science, and scientists speak to a much narrower audience. Most of all, science prioritizes puzzle-solving over creation in a way no other field seems to do.

To close his book, Kuhn calls for more study of intellectual communities as a whole (both scientific and non-scientific). After all, scientific knowledge only exists if it is shared by a group—and so understanding these groups is key to understanding scientific knowledge.