THOMAS S. KUHN'S The Structure of Scientific Revolutions (1962) has been and still is more widely known than anything written by a historian or philosopher of science in our generation. When asked to review it for Science, I was in Paris on leave and accepted eagerly. Tom Kuhn and I had been friends since the academic year 1946-47 when we were resident tutors at Harvard in Kirkland House. We there discovered our mutual, or rather reciprocal, interest in history of science, he from the side of physics and I from history.
What was my dismay, then, when on first reading Structure my reaction was puzzled irritation. What I had expected I do not know, but this was not it. The style is not inviting. There is constant repetition of key words and terms such as paradigm, puzzle, Gestalt shifts, incommensurability, anomaly, normal science, and so on, to each of which a special meaning is attached for the sake of the argument. I was at a loss what to do and almost decided to return Structure to Science on the grounds that I was not the one to review it. Then I thought of Herbert Butterfield on the importance of picking up the other end of the stick when confronted with an intractable puzzle. I put the book aside, and decided to try on a different thinking cap after a couple of weeks. In the meantime I was reading Auguste Comte's Cours de philosophie positive in the Bibliothèque Nationale.
When I came back to Structure, I saw what it was about (understanding a new paradigm?) and wrote the following review. It was among the first to appear. Kuhn was in Copenhagen at the time, interviewing for his oral history of the origin of quantum mechanics. He wrote me how pleased he was. All this came about at an important juncture in our careers. Kuhn had begun teaching history of science in one of the General Education courses inaugurated by James B. Conant at Harvard in the late 19405. His first book, The Copernican Revolution, was the outgrowth from that course. Noel Swerdlow has recently shown how, all unbeknownst to the author at the time, the seeds of Structure were planted in that slim volume.1 Such was Kuhn's technical acumen that, unversed in Latin though he was, he saw deeply into the problem and, unlike many a historian of early astronomy with impeccable latinity, he got the issues right.
On leaving Harvard, Kuhn moved to Berkeley, where he had a joint appointment in the History and Philosophy Departments. There he got on well with his colleagues in history, but was unable to develop a relationship of mutual confidence with the philosophers. For my part I had initiated an undergraduate course in history of science at Princeton in 1956 and a graduate program in History and Philosophy of Science in 1960. Both went well enough that Nassau Hall authorized enlarging the staff in 1962-63. My first thought was to try to persuade Kuhn to move to Princeton. Although Structure had just appeared, the range of its importance was not immediately recognized and certainly not by me. That he accepted strengthened our Program beyond measure. I have had occasion to note elsewhere that ours was a wonderful collaboration during the sixteen years prior to his acceptance in 1979 of what amounted to a research professorship at MIT.2 It might be said of us that, in Isaiah Berlin's contrast, he was the hedgehog concentrating on one deep thing while I was the fox sniffing around on the surface at many tempting things. Different though Tom Kuhn and I were in background and temperament, we saw eye to eye on all academic, human, and practical aspects of the Program and of university affairs in general. Our only difference was of negligible importance personally.
It concerned merely the fundamental nature of science.
1. T. S. Kuhn, The Copemican Revolution: Planetary Astronomy in the History of Western Thought (Cambridge: Harvard University Press, 1966); N. M. Swerdlow, "An Essay on Thomas S. Kuhn's First Scientific Revolution, The Copemican Revolution," Proceedings of the American Philosophical Society (Vol. …