For a book built on a narrative of, among other things, the history of our understanding of electricity, The Structure of Scientific Revolutions, by Thomas Kuhn, has had a remarkable run. It appeared in 1962, and people have been arguing about it ever since. Structure didn’t make the list of 88 “Books that Shaped America” that the Library of Congress put on display last summer, even though Benjamin Franklin, Experiments and Observations on Electricity  is first on the list (that title will return below). For Structure is the book that made the word paradigm, meaning a way of seeing, part of the everyday discourse of nearly everyone who deals with ideas for a living.
In his introduction to a fiftieth anniversary edition that appeared last summer, Ian Hacking, a leading philosopher of science, reports that Structure sold just 919 copies in 1962-63, and 774 the following year. A paperback edition sold 4,825 copies in 1965; by the time of a second edition, in 1971, total sales were more than 90,000. The figure was 650,000 by 1987, and presumably as many again, or more, have been sold since then (he doesn’t say). The University of Chicago Press published a second sleeper in 1962: Capitalism and Freedom, by Milton Friedman, which has sold more than half a million copies since then.
Friedman’s broadside is one thing, but what accounts for Structure’s success? My guess is, already in the 1960s, it was recognized that we were living in an age dominated by science and that Kuhn had supplied a reliable guide to its conduct and cultural significance. That domination has increased greatly since, of course: see Jerome Kagan’s The Three Cultures: Natural Sciences, Social Sciences, and the Humanities in the 21st Century for a roadmap of recent developments.
Before Kuhn, the philosophy of science was boring and the history of science a backwater (though other investigators, notably Robert K. Merton, Michael Polanyi, and Stephen Toulmin were already taking a historical turn.). There was a lot of boilerplate instruction about the steps of the scientific method and the logic of scientific discovery (if you’re not wrong, you might be right) to be found in the first chapters of textbooks, but, as Kuhn wrote at the beginning of Structure, this was no better than an image of national culture drawn from a tourist brochure.
After Kuhn, the focus shifted to the social organization of science: to the textbooks themselves, graduate education, the communities (“invisible colleges”) in which science was done, and the various nexuses in which results were put to work, from scientific journals and legal briefs to corporate laboratories and entrepreneurial start-ups. (For a survey of the present day, see Steven Shapin’s The Scientific Life: A Moral History of a Late Modern Vocation; his own vocation, Shapin says elsewhere, has been “to lower the tone” of science studies over the last forty years.)
How does a science get started? According to Kuhn, the story goes something like this: in the beginning someone contributes a powerful example of how to think about a set of scientific problems: Aristotle’s Physica, Ptolemy’s Almagest, Newton’s Principia, Franklin’s Electricity, Lavoisier’s Chemistry, Lyell’s Geology. The achievements appear, not out of the blue, but they are transformative. A community forms around them because they offer not a finished theory but rather a thinking cap, a pre-analytic way of seeing things and asking questions about them.
This way-of-seeing aspect that each possessed Kuhn designated a paradigm. The word itself is ancient Greek; he borrowed it from language studies, where it described the all-but-unconscious pattern by which one learns to conjugate a verb or decline a noun when learning to speak a language. A successful paradigm is enabling. It both poses plenty of unanswered questions and suggests means by which they might be conclusively answered. (Creationism may possess a paradigm in the Bible’s Book of Genesis. Because it is not a fruitful one, for it does not permit questions to be posed and answered, it’s not a successful science in Kuhnian terms) .
This is the route to what Kuhn called “normal science.” By that he meant successful science, rather like filling in the outlines of a hastily drawn map once a new continent has been discovered. In this metaphor, normal scientists come in all sorts of guises: trailblazers, pioneers, settlers, sodbusters, ranchers, developers. Kuhn, unfortunately, chose two other metaphors to describe the conduct of this phase, and those labels have sometimes caused proud scientists to rebel at his description. Successful normal scientists were “puzzle-solvers,” he said, working away at adducing facts, producing theories and making sure the one dovetailed with the other. Or they were, in essence, engaged in “mopping up” after a big paradigmatic invasion. (As a young radar officer in World War II, Kuhn had investigated captured Nazi radar sites after the Normandy invasion.)
Kuhn was a great student of the Copernican revolution, which meant he thoroughly understood the Ptolemaic system that it overthrew – crystalline spheres arrayed around an earth at the center of the universe. Ptolemy, and the astronomers who worked in his tradition for nearly fifteen hundred years, were excellent normal scientists. They had built a system that cohered; when observation of the heavens produced a troubling fact (anomalies, Kuhn dubbed such facts), they added a sphere or two.
But the troubling facts multiplied. Eventually a scientific crisis was at hand – anomalies with which existing normal science simply could not cope under any circumstances. At that point, a “revolutionary,” usually a young scientist, capable, but with little commitment to the old tradition – in this case, Copernicus – would produce a new paradigm, radically reordering the old facts, ignoring some and adducing new ones. The new paradigm would be resisted for a time, science being an inherently conservative enterprise, but gradually would gain adherents among the young. In time, the new order would be widely accepted.
Summing up, Kuhn wrote in The Copernican Revolution: Planetary Astronomy in the Development of Western Thought (the book that preceded Structure by five years): “A conceptual scheme, believed because it is economical, fruitful, and cosmologically satisfying, finally leads to results that are incompatible with observation, belief must then be surrendered and a new theory adopted; after this the process begins again.”
In Structure, Kuhn went on to make the point that scientific revolutions didn’t have to be huge events with sweeping cultural ramification, such as the Copernican, Newtonian or chemical revolutions. The professional groups affected by them could be far smaller. He showed how the youthful Benjamin Franklin, with an ingenious series of experiments using a Leyden jar – a device conceived in the 1740s for bottling electrical “fluid” that was surmised to exist – supplied a paradigm that attracted a large community of investigators.
Franklin’s Experiments and Observations set forth a body of fundamental propositions about electrical phenomena that permitted investigators to perform experiments that produced questions and more definite answers. These “puzzle-solvers” raced ahead — Cavendish, Coulomb, Volta and Joule the most famous among them – until experiments by Davy and Faraday enabled new technologies that, in due course, supplied new anomalies with which the Franklinian paradigm could not deal. Finally, in 1861 James Clerk Maxwell wrote down a series of equations describing electromagnetism as a field instead of a fluid. Electromagnetic studies were transformed again. “Maxwell’s equations were as revolutionary as Einstein’s, and they were resisted accordingly,” Kuhn wrote in Structure.
Kuhn spent the next fifteen years defending his use of paradigm; the word entered everyday speech not because of anything he did, but because the analysis for which it served as shorthand seemed to explain so much, far beyond the precincts of high science. (I remember becoming convinced of the utility of the distinction between normal and revolutionary change in the course of reporting on the “boxed beef revolution” that shut down the great stockyards of Omaha and Chicago.) In 1978 he produced Black Body Theory and the Quantum Discontinuity, 1894-1912, in hopes of demonstrating the same mechanisms at work in the early stages of a third such revolution in our understanding of electromagnetism. But by that time the debate had, for the most part, left the history of particular sciences behind. The word paradigm didn’t appear in the book. And anyone who wishes to be taken seriously in technical discussions avoids the term today. Yet usage exploded in the 1990s, perhaps owing to semonstration effects — see this useful graphic supplied by Joshua Gans, of the University of Totonto.
Kuhn died in1996. He behind an unfinished book (two friends are still piecing it together); voluminous correspondence; two volumes of collected papers (one of which contained a sophisticated and sympathetic interview); and a veritable avalanche of secondary works. A proper biography presumably will be among the more important such works of the twenty-first century.
As especially fascinating aspect of the story has to do with the reception of Structure. A tendency to mildly disparage it has emerged. Hacking, in his introduction, assures us that science has moved on. The Cold War is over; physics is no longer “where the action is.” Today, he says, “biotechnology rules.” Thus Structure, he writes, “may be – I do not say it is –more relevant to a past epoch in the history of science than it is to the sciences as they are practiced today.”
David Kaiser, a physicist who is a professor in the history of science at the Massachusetts Institute of Technology (where Kuhn spent his last seventeen years), put the case clearly on the eve of a fiftieth-anniversary symposium: “Kuhn had an ambition with the book, which was common at the time: he really thought there was a structure, a hidden key that makes science tick. I think many of my colleagues today in the history and sociology of science would find that ambition wrong-headed. There is not a single magical key that will unlock the way science gets done.”
There is another possibility, of course – that, for one reason or another, it is the historians and philosophers of science, taken as a group, that have got it wrong. They are, after all, “normal” scientists. For as Daryn Leboux, of Queens University, and Jay Foster, of Memorial University of Newfoundland, said in their Science magazine review of the fiftieth anniversary edition, Structure was a revolution of its own, and revolutions are complicated things. They can spark backlash as well as assimilation. It is possible, even likely, that The Structure of Scientific Revolutions is one of those books, like The Origin of Species, that take more than a generation, even two or three, to find its level – a real anomaly in the age of blink. I eagerly look forward to the seventy-fifth anniversary edition.
10 responses to “Paradigms, after Fifty Years”
[…] trouble if I blog too much when family is around for the holidays, so a quick one from David Warsh: Paradigms, after Fifty Years, Economics Principals: For a book built on a narrative of, among other things, the history of our understanding of […]
I think that the success of Kuhn is due more to style – he wrote a short, easy to read book, accessible to non specialits – then to any particular thought in the book.
[…] -David Warsh, “Paradigms, after Fifty Years.” […]
This post is way too charitable toward Kuhn’s work. By the time I was coming of age, 25 after its publication, Kuhn’s work was being heavily criticized. Calling “The logic of scientific discovery” “boilerplate” is superficial at best. At worst, I’d take it as a personal insult. Popper has had a material influence on the practice of science. Is one of the very few epistemologists scientists care about, and for good reasons. Someone told me once that epistemologists are just failed scientists, and this applies to Kuhn (who started as a physicist).
Kuhn’s thesis is a nice narrative, mostly consistent with a non-continental lightweight version of Marxism, and the right book at the right time. It has given ammunition to plenty of forgettable scientists to cast themselves as non-normal scientists (Gleick’s “Chaos”, anyone?). And in Economics, I am sure that post-autistics economists and econophysicists have absorbed its lesson, maybe through hearsay.
But ultimately it is unfalsifiable and predicts nothing, and by construction has nothing normative to say. As a work of historical scholarship it relies too much on secondary sources. Worse still, it has help spawn sociologists of science like Bruno Latour. Need I say more?
I know this comment is venomous, but a reader approaching Kuhn’s for the first time may receive a mistaken impression.
Beware that Google Books Ngrams is case sensitive. The link you supplied is the data for “Paradigm”, but it does not include the data for “paradigm”. Here’s a link that provides both, though (expectedly), the two lines appear to be proportionate: http://books.google.com/ngrams/graph?content=Paradigm%2Cparadigm&year_start=1900&year_end=2008&corpus=15&smoothing=3&share=
You write: “Aristotle’s Physica, Ptolemy’s Almagest, Newton’s Principia, Franklin’s Electricity, Lavoisier’s Chemistry, Lyell’s Geology”
Amusingly, none of these references is strictly correct; all of these books have longer titles which, due to their popularity, have been reduced in the vernacular to a single word, in much the same way that famous entertainers become “mononymous persons” (vide Wikipedia).
A good point – and good example of kuhn’s skill at a rhetorician (the list is his).
As far as I can see, paradigm shifts have a “fractal” property: Large pardigm shifts are rare, but within the stable periods of large paradigm shifts, there are much more frequent smaller paradigm shifts regarding small parts that fit within the larger paradigm. In a sense, all scientific activity is the battle of paradigms, and most of the day-to-day action is battle on the smallest scales.
very nicely put. that seems to me the essence of the matter. it is routinely overlooked.
I will admit that I’ve not read much of Kuhn beyond his landmark book, but from what I have read, he did not intend his book to be the starting point for a new normative system of philosophy of science, unavoidable though that may have been.
However, it seems quite definitionally impossible to derive a proposition from a set of premises, i.e., to say something in the analytic philosophical sense, without having a normative framework which establishes the procedures for doing so, a fact which the reception and impact of Kuhn’s work seems to validate. Insofar as some version of formal logic is true, one can only be so “meta” while saying something with meaning.