David Kaiser marks the 50th anniversary of an exemplary account of the cycles of scientific progress.
The Structure of Scientific Revolutions: 50th Anniversary Edition
- Thomas S. Kuhn
Fifty years ago, a short book appeared under the intriguing title The Structure of Scientific Revolutions. Its author, Thomas Kuhn (1922–1996), had begun his academic life as a physicist but had migrated to the history and philosophy of science. His main argument in the book — his second work, following a study of the Copernican revolution in astronomy — was that scientific activity unfolds according to a repeating pattern, which we can discern by studying its history.
Kuhn was not at all confident about how Structure would be received. He had been denied tenure at Harvard University in Cambridge, Massachusetts, a few years before, and he wrote to several correspondents after the book was published that he felt he had stuck his neck “very far out”. Within months, however, some people were proclaiming a new era in the understanding of science. One biologist joked that all commentary could now be dated with precision: his own efforts had appeared “in the year 2 B.K.”, before Kuhn. A decade later, Kuhn was so inundated with correspondence about the book that he despaired of ever again getting any work done.
By the mid-1980s, Structure had achieved blockbuster status. Nearly a million copies had been sold and more than a dozen foreign-language editions published. The book became the most-cited academic work in all of the humanities and social sciences between 1976 and 83 — cited more often than classic works by Sigmund Freud, Ludwig Wittgenstein, Noam Chomsky, Michel Foucault or Jacques Derrida. The book was required reading for undergraduates in classes across the curriculum, from history and philosophy to sociology, economics, political science and the natural sciences. Before long, Kuhn's phrase “paradigm shift” was showing up everywhere from business manuals to cartoons in The New Yorker.
Kuhn began thinking about his project 15 years before it was published, while he was working on his doctorate in theoretical physics at Harvard. He became interested in developmental psychology, avidly reading works by Swiss psychologist Jean Piaget about the stages of cognitive development in children.
Kuhn saw similar developmental stages in entire sciences. First, he said, a field of study matures by forming a paradigm — a set of guiding concepts, theories and methods on which most members of the relevant community agree. There follows a period of “normal science”, during which researchers further articulate what the paradigm might imply for specific situations.
In the course of that work, anomalies necessarily arise — findings that differ from expectations. Kuhn had in mind episodes such as the accidental discoveries of X-rays in the late nineteenth century and nuclear fission in the early twentieth. Often, Kuhn argued, the anomalies are brushed aside or left as problems for future research. But once enough anomalies have accumulated, and all efforts to assimilate them to the paradigm have met with frustration, the field enters a state of crisis. Resolution comes only with a revolution, and the inauguration of a new paradigm that can address the anomalies. Then the whole process repeats with a new phase of normal science. Kuhn was especially struck by the cyclic nature of the process, which ran counter to then-conventional ideas about scientific progress.
At the heart of Kuhn's account stood the tricky notion of the paradigm. British philosopher Margaret Masterman famously isolated 21 distinct ways in which Kuhn used the slippery term throughout his slim volume. Even Kuhn himself came to realize that he had saddled the word with too much baggage: in later essays, he separated his intended meanings into two clusters. One sense referred to a scientific community's reigning theories and methods. The second meaning, which Kuhn argued was both more original and more important, referred to exemplars or model problems, the worked examples on which students and young scientists cut their teeth. As Kuhn appreciated from his own physics training, scientists learned by immersive apprenticeship; they had to hone what Hungarian chemist and philosopher of science Michael Polanyi had called “tacit knowledge” by working through large collections of exemplars rather than by memorizing explicit rules or theorems. More than most scholars of his era, Kuhn taught historians and philosophers to view science as practice rather than syllogism.
Most controversial was Kuhn's claim that scientists have no way to compare concepts on either side of a scientific revolution. For example, the idea of 'mass' in the Newtonian paradigm is not the same as in the Einsteinian one, Kuhn argued; each concept draws meaning from separate webs of ideas, practices and results. If scientific concepts are bound up in specific ways of viewing the world, like a person who sees only one aspect of a Gestalt psychologist's duck–rabbit figure, then how is it possible to compare one concept to another? To Kuhn, the concepts were incommensurable: no common measure could be found with which to relate them, because scientists, he argued, always interrogate nature through a given paradigm.
Perhaps the most radical thrust of Kuhn's analysis, then, was that science might not be progressing toward a truer representation of the world, but might simply be moving away from previous representations. Knowledge need not be cumulative: when paradigms change, whole sets of questions and answers get dropped as irrelevant, rather than incorporated into the new era of normal science. In the closing pages of his original edition, Kuhn adopted the metaphor of Darwinian natural selection: scientific knowledge surely changes over time, but does not necessarily march towards an ultimate goal.
“Scientists have no way to compare concepts on either side of a scientific revolution.”
And so, 50 years later, we are left with our own anomaly. How did an academic book on the history and philosophy of science become a cultural icon? Structure was composed as an extended essay rather than a formal monograph: it was written as an entry on the history of science for the soon-to-be-defunct International Encyclopedia of Unified Science. Kuhn never intended it to be definitive. He often described the book (even in its original preface) as a first pass at material that he intended to address in more detail later.
To me, the book has the feel of a physicist's toy model: an intentionally stripped-down and simplified schematic — an exemplar — intended to capture important phenomena. The thought-provoking thesis is argued with earnestness and clarity, not weighed down with jargon or lumbering footnotes. The more controversial claims are often advanced in a suggestive rather than declarative mode. Perhaps most important, the book is short: it can be read comfortably in a single sitting.
For the 50th-anniversary edition, the University of Chicago Press has included an introductory essay by renowned Canadian philosopher Ian Hacking. Like Kuhn, Hacking has a gift for clear exposition. His introduction provides a helpful guide to some of the thornier philosophical issues, and gives hints as to how historians and philosophers of science have parted with Kuhn.
The field of science studies has changed markedly since 1962. Few philosophers still subscribe to radical incommensurability; many historians focus on sociological or cultural features that received no play in Kuhn's work; and topics in the life sciences now dominate, whereas Kuhn focused closely on physics. Nevertheless, we may still admire Kuhn's dexterity in broaching challenging ideas with a fascinating mix of examples from psychology, history, philosophy and beyond. We need hardly agree with each of Kuhn's propositions to enjoy — and benefit from — this classic book.
About this article
Cite this article
Kaiser, D. In retrospect: The Structure of Scientific Revolutions. Nature 484, 164–165 (2012). https://doi.org/10.1038/484164a
European Journal of Human Genetics (2021)
Identifying single influential publications in a research field: new analysis opportunities of the CRExplorer