Paradigm

Kuhn compresses his discussion of the centrality of the notion of paradigm into a single chapter entitled "The Priority of Paradigms." Paradigms have priority because there is nothing more basic by which "paradigm" could be defined. In logical terminology, the word paradigm functions as a primitive term. Properties of paradigms can be given and examples of paradigms can be enumerated, but the word cannot be defined, any more than number can be defined in arithmetic. Kuhn justifies his introduction of the term paradigm by arguing that, for the historian, it is a better organizational concept than any other. By looking for paradigms and changes in paradigms, the historian can classify scientists and historical periods in ways that lead to productive research and a better understanding of the history of science. Turning to philosophers to justify the undefinability of paradigms, Kuhn invokes Michael Polyani's (1891–1976) idea of tacit knowledge and Ludwig Wittgenstein's (1889–1951) idea that some human activities cannot be captured by a set of rules, arguing that while paradigms cannot be reduced to a set of methods and beliefs, they are recognizable to the historian as the organizing principle underlying a period of normal science.

The root definition of paradigm as both pattern and example exhibits both sides of a classical philosophical debate over the nature of definition. Plato (c. 428–348 or 347 B.C.E.) argued forcefully that providing examples is not adequate; the real definition of a term must specify what the examples have in common and thus explain why they all properly fall under the concept being defined. In the terminology of later philosophy, Plato argues that a definition must tell us the essence of a thing. For Plato, this is the eidos, the eternal form or idea to which all objects falling under a concept must conform. Kuhn sides with David Hume (1711–1776) and Wittgenstein in rejecting Plato's requirement that the essence be given in the proper definition of a concept. Like Hume, Kuhn argues that it is enough to say that the objects falling under a concept resemble one another in various aspects that can be specified and to take that resemblance as a starting point. Kuhn claims that paradigms do not have an essence, since there is always some disagreement and some difference in emphasis among scientists who are working under the same paradigm. In Wittgenstein's terminology, the historian can find a family resemblance among the views of these scientists rather than single common set of beliefs and methods.

Kuhn also defends his view that paradigms cannot be reduced to a set of beliefs and rules of method by pointing out that scientists learn by working through concrete examples of problems, not by learning rules. Thus learning a paradigm is more like learning a skill than like learning a body of knowledge, a point that Joseph Rouse has rightly emphasized. Kuhn is very close to using paradigm in its original meaning here, since the problems and solutions through which students learn are to be taken as patterns of scientific thought and work.