"In The Structure of Scientific Revolutions (1962) he draws a distinction between ‘normal science’ conducted on a set of fundamentally unquestioned assumptions called a ‘paradigm’, and ‘revolutionary science’, which addicates the paradigm to address problems inconsistent with those assumptions. A new paradigm is thereby formed. Philosophically, Kuhn’s work seems to suggest that science is not timelessly true."
Flew and Priest (2002)
"His The Structure of Scientific Revolutions 1962, 2nd rev. edn 1970, was very widely read and discussed because it presented a view of science very different from those current among philosophers and scientists. Earlier views were that science is cumulative: scientists discover more and more truths about the world. The realization that this is naive, and that sometimes scientists abandon earlier views as mistaken, did not really alter the basic optimism: it was said that an earlier scientific theory is only abandoned in order to put a better theory in its place.
All this seemed threatened by Kuhn, who took a more historical and/or sociological look at science. A science or branch of science is the preserve of a particular community of scientists. These scientists, by virtue of their common education (or, rather, dogmatic initiation) all take for granted a ‘paradigm’, a way of viewing the world and of practicing science in it. In an attempt to force nature to fit their paradigm, they try to solve puzzles defined by the paradigm. If a scientist solves a puzzle, the scientist is congratulated and rewarded. If a scientist fails to solve a puzzle, the scientist is blamed and the puzzle reserved for other, better scientists. Thus, ‘normal science’, which as its name suggests is the norm, is puzzle-solving under the aegis of an unquestioned paradigm. However, every paradigm has its day. Unsolved puzzles accumulate; scientists begin to lose confidence in the paradigm. The community plunges into a ‘crisis state’, into a period of ‘extraordinary or revolutionary science’: nothing is taken for granted any more, alternative paradigms are canvassed. One of these will solve one or two of the unsolved puzzles that the earlier paradigm could not solve. Seeing its promise, more and more scientists will ‘convert’ to the new paradigm. Usually these are the younger scientists: older ones will simply die out, exemplifying the dictum: ‘You cannot teach an old dog new tricks’. A new generation of scientists is taught to recognize the virtues of the new paradigm; a new period of ‘normal science’ begins.
Does science make progress through scientific revolutions? Are later paradigms better than earlier ones? No, Kuhn suggests, they are just different. The scientific revolutions which supplant one paradigm with another do not take us closer to the truth about the way the world is. Successive paradigms are INCOMMENSURABLE. Kuhn says that a later paradigm may be a better instrument for solving puzzles than an earlier one. But if each paradigm defines its own puzzles, what is a puzzle for one paradigm may be no puzzle at all for another. So why is it progress to replace one paradigm with another which solves puzzles that the earlier paradigm does not even recognize?
Many philosophers thought that Kuhn was impugning the rationality, objectivity and progress of science. In his later writings (especially the ‘Postscript’ to the second edition of his book) Kuhn tried to reassure them that this was not so. On the other hand, sociologists of science and others anxious to debunk science respond to Kuhn’ ideas with enthusiasm. So did some social thinkers who took Kuhn to be saying that the way to become truly scientific is to become dogmatically committed to some ‘paradigm’."
Mautner (2000)
Wikipedia: The Structure of Scientific Revolutions: Criticisms of Kuhn and SSR