What Was the “Paradigm Shift”?

When did you encounter your first paradigm shift? Not the phenomenon itself, but the term? Perhaps at an airport bookstore, where bestselling authors of books with titles like Change Your Paradigm, Change Your Life and The 15 Commitments of Conscious Leadership: A New Paradigm for Sustainable Success use it more or less as a synonym for “game changer.” Or maybe on the taps of Paradigm Shift Brewery, a craft brewery located in Massilon, Ohio. The goods and services you can purchase with “paradigm” in their name include coffee and crypto, sneakers, and health care management. The corporate website for perhaps the best-known Paradigm, a high-end Canadian speaker company, explains that the founders “decided to, eh-hem, change the prevailing industry paradigm.”

The language of paradigms and paradigm shifts is ubiquitous except among the people most familiar with its source: historians and philosophers of science. Once upon a time—let’s say the late 1960s—a reference to “paradigm shifts” primarily signaled knowledge of Thomas Kuhn’s historicist approach to the philosophy of science. Kuhn’s 1962 classic, The Structure of Scientific Revolutions, transformed our understanding of scientific change and has become a foundational text for historians, philosophers, and social studies of science. It is nonetheless unusual these days for anyone who studies science professionally to invoke the term “paradigm shift.” The concept has become completely unmoored from the term.

The Structure of Scientific Revolutions, in other words, is one of those books that everybody knows but doesn’t read, or reads once and shelves. On rereading my copy, neglected since a first-year graduate seminar in the history of science over 25 years ago, I was struck by Kuhn’s insistence on the power of historical research to puncture idealized claims of scientific progress. Paradigms and normal science? Sure. But the truly radical idea here is that outsiders—in this case, historians—can offer better insight into the inner workings of a profession than the practitioners themselves.

What, exactly, is a paradigm shift? In Structure, Kuhn defines a scientific paradigm through its relation to what he calls “normal science.” A mature scientific community, one that is relatively secure in its methods, intellectual assumptions, and choice of problems, is operating in a period of “normal science.” Collectively, those rules and standards for scientific research constitute “shared paradigms.” These shared paradigms lay a path for scientific communities to work efficiently, allowing individual scientists to focus on the “mop-up work” of collecting data and solving puzzles suggested by the operating paradigm. 

Over time, however, this routine puzzle-solving work will generate “anomalies” that violate the expectations established by the paradigm. At first, scientists who encounter such anomalies tend to assume that they have made some sort of experimental error. Those who insist on the correctness of their divergent findings may be considered cranks. An abundance of anomalies, once acknowledged by the community, throws that community into crisis. When the crisis cannot be resolved by tweaking the existing paradigm, a competing interpretation that casts the data in entirely new ways may gain ascendance. This switching of the paradigms constitutes a scientific revolution. Scientists might reject phlogiston in favor of oxygen, for example, or embrace heliocentrism after centuries of Aristotelianism. 

Kuhn famously described these competing paradigms as “incommensurable.” A belief in phlogiston is simply incompatible with the existence of oxygen. You cannot simultaneously believe that the sun orbits the Earth and that the Earth orbits the sun. Drawing on mid-twentieth-century psychology and theories of mind, Kuhn compared these shifts in attitudes to gestalt shifts—the kind of change in perspective that turns an image of a rabbit into a duck. In perhaps the book’s most controversial chapter, Kuhn argued that changing paradigms changes nature itself, or at least the way scientists perceive it. Change a paradigm, change the world. 

This, in broad outlines, is the gist of the book’s argument. The text itself is fairly technical, as Kuhn, who trained as a theoretical physicist, assumes his readers will have a working knowledge not only of key concepts in eighteenth-, nineteenth-, and twentieth-century physics but also of live debates in mid-twentieth-century philosophy of science. Accordingly, much of the initial critical response came from specialists focused on inconsistencies in Kuhn’s own thinking. One early review, for instance, cataloged at least 22 different ways that Kuhn uses the term “paradigm.” Other readers noted that Kuhn barely bothered to define these “scientific communities” whose assumptions and norms define the paradigms.

Kuhn generously acknowledged the legitimacy of both of these critiques in a postscript published in the 1969 edition of the book, but he held his ground on what he saw as the book’s fundamental point: Science does not progress through the accumulation of incremental facts, theories, and methods. When science does progress, it does not necessarily progress in a particular direction that aligns it more closely with nature. Instead, science advances through ruptures that require that scientists reject the past. As he argued in the original edition: “Cumulative acquisition of unanticipated novelties proves to be an almost non-existent exception to the rule of scientific development. The man who takes historic fact seriously must suspect that science does not tend toward the ideal that our image of its cumulativeness has suggested. Perhaps it is another sort of enterprise.”

Alas, Kuhn did not specify what kind of enterprise it might be instead.

We can gain some clues from the context in which Structure was written. This requires a journey back to the early 1950s, when a young Kuhn was teaching the history of science at Harvard. Kuhn’s tenure at Harvard coincided with the waning years of the presidency of James B. Conant, a chemist who had long left the lab in favor of scientific and political administration. During World War II, Conant had chaired the National Defense Research Committee and advised President Harry Truman on the use of the atomic bomb. About halfway through Kuhn’s tenure at Harvard, Conant retired to serve as the high commissioner for occupied Germany.

The astonishingly productive Conant also wrote several books on the future of higher education in the United States. These books repeatedly turned to a theme that Conant and many scientists of his generation saw as one of the most pressing questions facing the U.S. in the Cold War: What was the role of science, and specifically science education, in a democracy? National defense required that political authorities and scientists work closely together, yet many also feared the rise of a technocracy or that science might bend to political authority. Having examined the recent history of science in Nazi Germany, the Soviet Union, and the U.S., Conant sought a pedagogical strategy that could cultivate a new generation of curious scientists and instill respect for science among nonscientists but that would also carefully calibrate citizens’ questioning of authority.

Conant’s solution was historical method. Shortly after the war ended, Conant began teaching a new course on the history and philosophy of science for Harvard undergraduates. In a turn that was unusual for courses in the history of science at the time, Conant covered scientific wrong turns and dead ends, including alchemy and theories of spontaneous generation. Kuhn’s experience of teaching a version of this course transformed his understanding of the history and philosophy of science.

This crucial context explains the book’s otherwise peculiar obsession with scientific textbooks. The problem with textbooks, as Kuhn repeatedly explains, echoing Conant, is that they obscure scientific debate and prior explanations in the name of inducting students into the reigning understanding of how the world works—that is, a scientific paradigm. These textbooks, he claims, have to be “rewritten” when normal science changes, because they “refer only to that part of the work of past scientists that can easily be viewed as contributions to the statement and solution of the texts’ paradigm problems.”

Kuhn acknowledges that there might be sound pedagogical reasons for this, but he nonetheless remains clearly troubled by the fact that most textbook stories about the development of science—at least the kind presented in 1950s textbooks—were simply not true. Again echoing Conant, Kuhn particularly objected to the fact that students had no access to their own facts that they could use to test the relationship between theory and evidence: “Science students accept theories on the authority of teacher and text, not because of evidence.” Students, in other words, are at the mercy of authority figures, and these authority figures were, via their textbooks, feeding them lies.

Kuhn was troubled by the fact that most textbook stories about the development of science were simply not true.

Kuhn’s critiques of scientists’ own telling of history, particularly as tendered in textbooks, is not subtle. Historically accurate accounts of how science develops, full of messy data, theoretical dead ends, and disproven assumptions, Kuhn suggested, posed a threat to scientific authority by elevating “human idiosyncrasy, error, and confusion.” At one point, Kuhn explicitly compares scientists’ approach to rewriting history to the one depicted in George Orwell’s 1984, albeit with the caveat that scientific paradigms derive their authority from the consensus of the scientific community rather than political fiat.

To modern readers, this sounds like an attack on science, or at least on their textbooks. But this was 1962, just five years after the Soviet Union launched Sputnik, the first artificial satellite. The Soviet Union’s apparent lead in the space race generated a national panic about the state of science education in the U.S. One consequence was the passage of the National Defense Education Act of 1958, which, among other things, provided millions of dollars to redesign high school science curricula, particularly in physics, mathematics, biology, and the earth sciences. In keeping with Conant’s theories of education and democracy, these experimental science curricula typically incorporated at least moderately accurate historical accounts of scientific change. They also encouraged students and teachers alike to embrace the uncertainty produced through experimentation, both as a way to prepare future scientists for the realities of scientific work and to discourage unthinking allegiance to authority.

In 1962, then, few university scientists, many of whom were themselves engaged in creating the new curricula, took offense at Kuhn’s strident commentary on science textbooks. Kuhn’s list of the people who propagate falsehoods about science—textbook authors, popularizers, and philosophers—doesn’t even include scientists (though of course many scientists engage in this kind of work). As more time passed from the date of publication, however, this context was lost. While many scientists found Kuhn’s model of scientific change useful, they bristled at his characterization of the scientific community. No one, it’s fair to say, wants to be compared to the textbook writers in 1984.

Kuhn claimed, both in 1962 and for the rest of his career, that he had not intended any of this as an attack on the scientific enterprise itself. For Kuhn, the entire point of Structure was to elucidate what he and many scientists of his generation saw as science’s distinctive ability to build cumulative knowledge. “Why,” as Kuhn wrote, “is progress a perquisite reserved almost exclusively for the activities we call science?” Part of the answer flows from the practices of normal science, which free scientists up to focus on routine problem-solving. But Kuhn argued that even the disruption of paradigm shifts generally strengthens, rather than weakens, scientists’ ability to solve new problems, for the straightforward reason that scientific communities prefer practices that “ensure the continuing growth of the assembled data that [they] can treat with precision and detail.”

For Kuhn, this definition of scientific progress, with its dedication to routine problem-solving and fidelity to the actual history of science, represented a major step forward from pedagogical fairy tales about science as a process of constant discovery. But Kuhn himself acknowledged one critical way in which his account of scientific progress diverged from common understanding: It had nothing to do with truth. Instead of thinking of science as a process that inevitably draws closer to natural reality, Kuhn suggested that we treat scientific change as an evolutionary process, similar to natural selection, in which various paradigms compete for community advantage.

Critics charged relativism. If science had no inherent orientation toward nature or truth, then facts could be whatever a scientific community agreed them to be. Kuhn deflected this critique by pointing to the supposedly distinctive characteristics of the scientific communities that he hadn’t actually defined. In the 1969 afterword, he called for more studies of scientific communities. Given the works he cited, he likely had in mind projects like sociologist Diana Crane’s notion of an “invisible college” or entrepreneur Eugene Garfield’s approach to citation analysis, each of which highlighted network effects within recognized communities of experts.

Kuhn’s fiercest critics were right about one thing: Kuhn didn’t trust scientists to tell their own stories.

By the mid-1980s, a new generation of historians, sociologists, and anthropologists of science brought a more critical lens to scientific power. They incorporated ideas about race, sex, gender, national and historical context, and, most importantly, power into their analyses of what drove scientific communities to embrace some theories and reject others. Defenders of science took to calling this approach “Kuhnian,” to Kuhn’s everlasting chagrin. The charge was ridiculous then and ridiculous now: Structure’s approach to its topic is in many ways absurdly intellectually conservative, given that it excludes the natural and social sciences and any discussion of technology or, as he puts it, “external social, economic, and intellectual conditions in the development of the sciences.” It is blithely Eurocentric, confidently asserting that “only the civilizations that descend from Hellenic Greece have possessed more than the most rudimentary science.” In Structure, Kuhn sought to explain what made science tick, not to critique science’s cultural power.

But Kuhn’s fiercest critics were right about one thing: Kuhn didn’t trust scientists to tell their own stories. Until Structure, scientists and their historians seemed natural allies, jointly committed to sharing the story of scientific progress. Structure shattered this notion, suggesting that social scientists, with their attention to empirical data, were better situated to explain not only the past but also the future of science. To the extent that scientists resisted Kuhn’s interpretation, they resented his incursion on what they perceived to be their own territory—a dynamic familiar to anyone who’s encountered, say, economists’ response to economic history or teaching bans on the 1619 Project.

Kuhn argued that a warts-and-all approach to history could uphold and even enhance scientific authority by showing science’s remarkable power to progress despite periodic changes in perspective. It’s a complex view of history that is incommensurable with stories of uncomplicated discovery and scientific heroism. For the culture at large, Structure’s greatest contribution has been linguistic—the notion of “paradigm change” as a synonym for “thinking outside the box.” But read in 2024, in light of the current history wars, the real revolution at the heart of Structure was a paradigm shift in scientific narrative. Who gets to tell the story of a field? And how truthful should they be in their telling?