The Structure of Scientific Revolutions 3rd Edition

I'm not sure if it is still the case, but there was a time when Kuhn's book was _the_ most frequently cited book in scientific literature. With all respect to my fellow reviewers, it might be a tad bit arrogant to dismiss such a book as "puerile."
Before Kuhn, we were taught in school that scientific progress was linear, that it was an unending progression of refinements and developments, with one "truth" leading to the next "truth." Kuhn's insights including pointing out that such a linear progression was mostly a lie. His thesis was that the major developments in science were mostly revolutionary. That some "truths" turned out to be false. Astronomy was revolutionized by Galielo and Copernicus, and man was divested from the center of the universe. Physics was revolutionized by Newton. Biology and Darwin. It didn't hurt that plate tectonics came along shortly after Kuhn published, and Kuhn looked like his model was predictive, too.
Part of Kuhn's impact, I have to admit, was a result of the time which the book was first published. In the middle and late 1960's, questioning authority was the heart of any undergraduate's thinking, and Kuhn's ideas were read by some as a license to question all authority.
Perhaps as a consequence, Kuhn's model has been carried by other writers beyond all reason, with everyone from sociologists to New Age fuzzies usurping his terminology, making "paradigm shift" a nearly instant cliche. But his influence has gone far beyond those who want to mis-apply his ideas to everything from post-modern dance to sociobiology. Uniformitarianism has been bloodied, perhaps permanently. By geologists, evolutionists, archaeologists and more; the influence has been pervasive and real. Stephen Jay Gould may or may not subscribe to "Structure," but he has sure demolished uniformitarianism in evolution.
I disagree with those who regard "Structure" as "the most important" anything. But it unquestionably has been stunningly influential, and any serious student of science or philosphy, I believe, will be reading this book a hundred years from now. And apart from its influence and impact, the book still reads well almost 40 years on. It's fun and, if you enjoy seeing the world stood on its ear, you'll like Kuhn's approach.

Thomas Kuhn performed a signal service for historiography of science by studying how new ideas and new ways of thinking displace the old. He invented the term 'paradigm shift' to describe what happens when 'normal science' runs into 'anomalies' and enters a 'crisis', which in turn leads to a 'scientific revolution'. Nobody had heard of such things before, so Kuhn had a scoop. He sketched some historical examples in iconoclastic style; the result is this short book, first published forty years ago and still wowing Cultural Studies students today.

Much of what Kuhn the historian of science says here is sensible and well taken. It has certainly been influential, perhaps in ways the author never intended, and should be read for that reason. But there are odd omissions. The greatest paradigm shift in physics since Newton - the adoption of fully-fledged quantum mechanics after 1925 - finds no significant place in this study. Eminent physicists, including Einstein, and even Schrodinger, one of its founders, regarded the new paradigm with deep distaste on aesthetic and philosophical grounds. Yet the methodology was adopted universally almost at once. What sociological factors, what structures of power and patronage brought this about? We are not told.

It is when Kuhn puts on his philosopher-of-science hat and tells us about the 'incommensurability of paradigms' that we should question what he means, and more especially what some people have read into it. The idea is that Archimedes or Aristotle, encapsulated in their ancient world-view, would have been unable to see what Newton was getting at in his 'Principia'; and likewise Newton if you gave him a copy of Dirac's 'Quantum Mechanics'. This has been held to have implications for epistemology, viz: it is a mistake to think of the evolution of science (or any rational endeavor) as 'progress' in the sense of bringing us closer to an accurate picture of the world. Kuhn's position can be likened to Darwinian evolution: progress *from*, yes; progress *towards*, no. There is room here for fancy footwork. But the finer points are lost on some who simply cheer it as a poke in the eye for rationality.

If an epochal break can be found anywhere in the history of science, it is in the transition from the Aristotelian to the modern world-view which took place in early modern times. Since then nothing remotely like it has happened. The training of physicists still begins with a detailed study of Newtonian mechanics, which for many purposes, from shooting pool to spaceflight, provides an entirely adequate description. An important part of learning relativity or quantum mechanics lies in understanding how they fit in with Newtonian physics - in fact, precisely how the paradigms are commensurable where their domains overlap. The same people at different times use the paradigm of Newton and the paradigms of Einstein and Bohr/Heisenberg. They don't use the paradigm of Aristotle or the New Age paradigm because - interesting though these are to the historian or the social scientist - they don't work; they are not fruitful for puzzle-solving, Kuhn would say.

A process of generalization of paradigms has been characteristic of physics for the past few centuries, and this seems true of mature sciences generally. At the fundamental level a paradigm that has proven really useful is hardly ever scrapped (Kuhn cites two cases from physics since Newton: the recurring controversy over the nature of light - both sides seem to have won that one - and the caloric theory of heat). Instead, the old paradigm is subsumed into a more developed theory with a broader domain of application, yielding in some sense deeper insights. Kuhn the physicist knew this, of course, though some of his readers don't; so he had to defend the unusual position that e.g. Newtonian mechanics is fundamentally incompatible with Einsteinian mechanics, even though one is a limiting case of the other (Kuhn disputed this) and both are used successfully all the time. This was the only way he could maintain that they are 'incommensurable'.

Where does this leave the incommensurability of paradigms? The concept can be interpreted according to taste along a spectrum: at one end, true but trivial; at the other end, deep but almost certainly false. Indeed - and I'm going to be shockingly naive here - you wouldn't be reading this otherwise; you'd be chipping flints. For what it's worth, my opinion is that Newton, far from 'living in a different world', would be perfectly at home with modern physics and raring to go, given a couple of years to get up to speed; Archimedes might take a little longer, while Aristotle would be a leading light at the Sorbonne.

More problematic even than incommensurability of paradigms in Kuhn's work are occasional gnomic statements such as the following:

"There is, I think, no theory-independent way to reconstruct phrases like 'really there'; the notion of a match between the ontology of a theory and its 'real' counterpart in nature now seems to me illusive in principle"

and

"Scientific knowledge, *like language*, is intrinsically the common property of a group *or else nothing at all*" (my italics).

Taken with the thesis of the book (though Kuhn denied it) remarks like these open the door to all the baggage of so-called radical relativism. Now the baggage is in the hall and halfway up the stairs, as Gross & Levitt, Sokal & Bricmont and others have pointed out. Some of us wish it was out back in the hen-house.

At the heart of modern physics there is indeed an incommensurability, in at least one of Kuhn's senses. It is between the two fundamental theories, general relativity and quantum mechanics. That doesn't stop people from using both paradigms, but it's a great puzzle: no one knows how to fit them together correctly. When we find out (strictly speaking I should say 'if'), it will be as a result of a paradigm that hasn't shifted since the seventeenth century: theoretical structure expressed in the language of mathematics, built on and feeding back into an empirical base. And there will be real, at present unimagined consequences.

You may say that's naive or begs the ontological question. But I say it's the best we've got. No amount of self-regarding talk about hermeneutics and postmodern science - though it comes with a reference list as long as your arm to all the stars of critical 'theory' - will advance our understanding one iota. Whatever the world is, it isn't like that, and Kuhn never really imagined it was.

In spite of the impression I may have given, the book is worth reading and it isn't difficult (some background knowledge of actual science would help). Read it for yourself; don't believe everything people say about it.

Note added: Some readers think that Kuhn was describing a process of successive approximation to truth, incorporating a smart new account of convergence. The point cannot be made too strongly that he was doing nothing of the sort. I recommend reading page 206 from which the remarks about 'really there' were quoted. You don't have to be a relativist and anti-realist to be a Kuhnian, but it helps.