Kuhn, Thomas S. (1962).
The Structure of Scientific Revolutions.
Chicago: The University of Chicago Press.
Kuhn in a nutshell: "The successive transition from one paradigm to another via revolution is the usual developmental pattern of mature science" (p. 12).
The scientific enterprise presents itself (e.g. through textbooks and professional education) as progressing gradually and cumulatively, leading inexorably to the present state of knowledge. Recent historians of science, however, by studying the historical record of research activity, have begun to realize the inadequacy of this stereotype. More accurately, science proceeds by revolution.
"Normal science" works within established world views and norms -- within a paradigm. Individuals acquire or inherit the paradigm for their chosen fields (e.g. nuclear physics, astronomy) during the course of professional training, through contact with members of he scientific community firmly ensconced in that paradigm.
Normal science seeks to solidify and enhance current undersanding of the paradigm, "which is rarely an object for replication, but usually for further articulation and specification under new or more stringent conditions." The paradigm makes normal science very efficient by framing and directing acceptable avenues of research.
Normal-scientific research does not aim to produce major novelties; rather it aims to achieve an anticipated conclusion in a new or clever way. A good research problem is like a puzzle: certain rules and restrictions govern what is an admissible solution, and the existence of a solution is assured, which helps keep the researcher motivated.
Paradigms function on a level higher than a mere set of shared rules; a set of shared beliefs and attitudes unify a given scientific community, while it is difficult to abstract a consistent set of "rules" that everyone follows. Paradigms also function on various levels, from the scientific enterprise as a whole down to the most esoteric subspecialty.
During the course of normal-scientific research, anomalies arise that seem to challenge the prevailing paradigm. The discovery of an anomaly often can not be localized to a single person at a particular time. The anomalous discovery may turn out to be assimilable into the paradigm (in which case it wasn't an anomaly after all), or it may resist repeated, concentrated attempts to assimilate it.
An accumulation of resistant anomalies will spark a crisis by revealing the inadequacy of the prevailing paradigm. A crisis is characterized by a breakdown of normal-scientific puzzle-solving activity, and by debate between proponents of the old paradigm and proponents of a novel theory that tries to explain the anomalies.
A scientist may be led to reject an old theory if a new one is available to take its place, and if it does a better job of explaining and predicting observable phenomena. A younger scientist is more likely to accept new theories, to "think outside the box", than an older one who has a longer history of commitment to the paradigm.
Scientific paradigm shifts are appropriately called revolutions: "Like the choice between competing political institutions, that between competing paradigms proves to be a choice between incompatible modes of community life" (p. 94). Thos on either side of the paradigm shift cannot communicate effectively because they lack a common world view. The self-perpetuating character of paradigms necessitates revolutions to dislodge them.
While reality remains unchanged, the world looks like a different place for someone looking at it through "new-paradigm-colored glasses". An individual experiencing a paradigm shift undergoes something like a permanent gestalt switch.
"...science textbooks (and too many older histories of science) refer only to that part of the work of past scientists that can easily be viewed as contributions to the past statement and solution of the texts' paradigm problems. Partly by selection and partly by distortion, the scientists of earlier ages are implicitly represented as having worked upon the same set of fixed problems and in accordance with the same set of fixed canons that the most recent revolution in scientific theory and method has made seem scientific" (p. 138)
Competing paradigms will undergo a period of testing which involves both verification and falsification. A new paradigm may gain hold for a variety of interrelated reasons: it dramatically increases the accuracy of explanation and prediction of phenomena, especially the crisis-provoking phenomena; it holds greater problem-solving potential; it is simpler, more elegant; it holds more promise for the future of research. The new paradigm will encounter resistance, and some scientists will simply hold out until they die, but "the transfer of allegiance from paradigm to paradigm is a conversion experience that cannot be forced" (p. 151). "Rather than a single group conversion, what occurs is an increasing shift in the distribution of professional allegiances" (p. 158).
This proposed model for scientific revolutions seems to resonate with a cyclic view of history. Science does make progress, however, within a period of stable, normal science: "both the list of problems solved by science and the precision of individual problem-solutions will grow and grow" (p. 170). Successive paradigm shifts, moreover, will lead us to "an increasingly detailed and refined understanding of nature" (p. 170). But this should not be understood as an inevitable march toward Truth; science makes progress in the sense of non-teleological forward motion.