A Brief History of the History of Science

James Bryant Conant created the new world then tried to teach us how it works. A celebrated organic chemist, Conant spent the first World War developing poison gases and the interwar period on the theory of acids and bases. He was named President of Harvard University in 1933, where he overhauled the school’s curriculum, and then chaired the National Defence Research Committee, which oversaw the Manhattan Project. He advised President Truman to drop atomic bombs on Japan. 

Conant witnessed — and was in large part responsible for — the transformation of the United States into a scientific technocracy. While previous generations of scientists were confined to the lab, Conant sat on panels that advised the President on matters of national concern and joined diplomats to evaluate other countries’ nuclear capabilities. This was a truly radical transformation: Before the 20th century, scientific laboratories had been monastic spaces, where ascetic researchers served the master of theory, never practice. When Ernest Rutherford arrived at the Cavendish Laboratory with a hobbyist interest in radios, his advisor J. J. Thompson — discoverer of the electron — gave him a stern admonition: You cannot serve God and Mammon at the same time. 

But in Conant’s day, everyone was serving God, Mamon, and the bureaucratic state at the same time. His was a world where “a citizen, a businessman, a public servant, a lawyer, a teacher, or a writer, may be called upon at some time to evaluate the work of scientists.” ¹ Conant saw that democracies were becoming scientific technocracies, and spent much of his energy trying to prepare us for it.

His big idea: teach everyone the history of science.

Conant believed that it was too complicated to teach everyone quantum mechanics and organic chemistry, but most citizens could still develop a “feel” for science by following scientists of the past through their discoveries. “When a science is in its infancy,” he argued, “and a new field is opened up by a great pioneer, the relevant information of the past can be summed up in relatively brief compass.” We can’t make everyone an Oppenheimer, but we can expect everyone to understand a Newton.

Once the war had ended, Conant assembled a team of research and teaching assistants to scour the archives and write, sometimes for the first time in English, definitive accounts of major episodes in scientific history. Phlogiston, the purported substance inherent to combustible bodies that made them flammable, got a chapter. So did the history of the atom, and the calorie, and electric charge. “I believe,” Conant said, “that the case-history method is the nearest we can come to the more ideal procedure of having every leading citizen spend a year or two looking over the shoulders of research men in various kinds of laboratories.” ²

Conant’s project is an example of what I call naive history of science. Naive history of science comes from the belief that science is a linear march of progress, an ever-growing stack of giants standing on each other’s shoulders, and that by studying how this process played out in the past, we can ensure that it continues into the future. It assumes that by understanding the science of simpler times, we can gain insight into our over-complex present.

As an undergraduate student in mathematics and physics, I was a believer in naive history of science. I knew that every equation in physics represented decades of fighting among scientists — each derivation a museum of passion projects — but this backstory was never revealed to me. I thought the only way to truly understand the textbook physics of the present was to unearth the physics of the past, working our way up from simplicity to complexity.

I assumed that the place to go if I wanted to acquire this knowledge was graduate school in the history of science. Surely if anywhere was  going to teach me the arc of scientific progress in its totality, this was it.

This wasn’t it.

I was surprised to learn that the history of science involves almost no learning about science. Today, the field is interested in situating science in its political, economic, and cultural context, rather than explaining how anyone actually figured anything out. The product I was seeking had been discontinued for 75 years.

This was, in large part, because naive history of science failed as a pedagogical and a philosophical approach.  It failed because teaching people the history of chemistry did not make them more skilled chemists, policy-makers, or citizens. It failed because the science of the past isn’t simpler than the science of the present, and because naive history of science depended on a view of science the American public came to resent. 

The story that follows is an inquiry into naive history of science, what replaced it, and how the field called “history of science” has changed since Conant’s day. Several years into graduate training, I no longer believe in the naive approach I once sought; I now view it as one among many possible uses of history — to inspire, to educate, to critique, to deconstruct, to expand the imagination. We ignore the lessons of history at our own peril: so too the lessons of writing history.

Academic fields are born as insurgencies within universities. They gain legitimacy slowly, through the founding of their own journals, conferences, and eventually academic programs and departments. The central entrepreneur of the history of science field was George Sarton, a Belgian chemist who was  forced to flee his home during World War I. He grabbed as many of his papers as he could, including copies of a new journal he had founded called Isis, ³ and boarded a ship for the United States.

Sarton’s project was to give science the kind of historical stature afforded to politics, war, art, and literature. He believed that “science is the most powerful agency of change not only in the material world but also in the spiritual one” ⁴ — that, in fact,  “all our knowledge is, to some degree, scientific knowledge.” ⁵ The history of science should constitute “the intellectual baggage of every cultivated man” and become “one of the fundamental instruments for the organization of a new and better world.” ⁶ In other words, Sarton was one of the most vocal advocates of the progress studies of his day. 

None of this was exactly unique to Sarton: His views on progress drew from Enlightenment philosophers of the 18th century and the positivists of the 19th. Auguste Comte, the founder of positivism, wrote that human thought naturally evolved in stages, progressing from the theological to the metaphysical until it finally reached the scientific. ⁷ Science was both the culmination of all human knowledge, and the model along which all forms of inquiry should be structured. In shaping the history of science as an academic discipline, Sarton hoped to institutionalize this positivist program and raise science to its rightful place as the center of intellectual life. 

This project was a modest success. The history of science as an academic field grew slowly in the interwar period, and Sarton’s calls to action inspired many scientists’ papers and letters to be translated and archived for the first time. His journal Isis remains the official publication of the History of Science Society. And Sarton paved the way for teaching the history of science at Harvard, where Conant later made it a core part of the university’s general education curriculum.

But Sarton’s ultimate goal — to raise the status of scientists from ascetic intellectuals to history-making figures — was not achieved in the classroom. The Manhattan Project did more to rocket scientists, to superstar status in the United States more than undergraduate classes ever could. Atomic scientists appeared on the cover of Time Magazine; enrollment soared at colleges and universities; Harper’s observed that “No dinner party is a success without at least one physicist.” ⁸ Physicists had won the war, and morale was high.

While Conant was busy reforming the Harvard curriculum, he also sat on the country’s highest-level advisory boards for nuclear policy. Scientists were now bureaucrats, and bureaucrats had to understand science, but both sides could dip into the reserves of history for shared insights about “the vagaries of human nature and…questions involving the individual and society, man’s conflicting desire for freedom and for order, for personal glory and for cooperative teamwork of self-effacing individuals.” ⁹ It was a good pitch: As scientific departments expanded, history of science programs grew with them. 

With more students and more support, the discipline also grew more professional. Sarton himself was a scientist by training, and when he founded Isis in 1912, the journal’s audience was other working scientists. Its contents tended towards the biographical and technical: it was common for early articles to present newly uncovered documents or biographical details with little further analysis. By the time Sarton retired in 1952, the journals’ readers and contributors were trained scientific historians, with new expectations for what history was supposed to do.

Conant and Sarton had believed that science could be understood as a purely intellectual endeavour. Their focus was ideas and the various dramas between idea-havers. While Conant hoped that history could help bring science out of the lab and into society, he maintained a narrow view of how society could affect science. This wouldn’t fly for the trained historians. They wanted to do more than record the history of their own disciplines — they wanted to put it in context. 

At the same time, public sentiment towards science was beginning to change. By the late 1950s, science came to represent less modernism and progress, and more nuclear armageddon, environmental destruction, and an unending Cold War. Synthetic fertilizers, pharmaceutical scandals, chlorofluorocarbons, thalidomide, and myriad other emergencies became a daily news onslaught. In his 1961 farewell address, President Eisenhower warned about the military-industrial complex. American science at scale was no longer the force that won the war against Hitler: it had itself become the threat.

The old histories of science would no longer meet this new political moment. The march of progress narrative felt naive, and something new was emerging to replace it. The charge was led by one of Conant’s former students: a young PhD in physics and professor of History of Science named Thomas Kuhn. 

In 1962, Thomas Kuhn published The Structure of Scientific Revolutions. Even if you are familiar with Structure’s core ideas, it is still worth taking a moment to appreciate just how radical an intervention it was. This is a point which Kuhn makes clear from the book’s very first sentences: 

Kuhn wasn’t interested in raising the standing of scientists or helping laypeople understand technical concepts: He wanted to establish a new model of how science worked. Before writing Structure, he had spent years interviewing the physicists responsible for quantum mechanics, and found them to be unreliable narrators. Historians up to this point had been too willing to take scientists on their own word. Kuhn’s would focus less on what scientists claimed to be doing, and more on how they really behaved when accepting or rejecting scientific theories. 

In this account, scientists are trained to engage in “normal science,” the procedural solving of puzzles within a framework agreed upon by most members of a field. Kuhn calls this framework a “paradigm,”  his shorthand for the accepted methods, tools, equations, techniques, procedures, and standards of evidence at any given time. Every so often, anomalies arise that cannot be explained by the current paradigm. When enough anomalies accumulate, a crisis ensues. A sufficiently serious crisis can lead to a “revolution.”

In a scientific revolution, the old paradigm is replaced by a new one. Critically, this is not because the new paradigm has “proved” the old one wrong. It is impossible to fully prove or disprove one paradigm from within another, since scientists working in different paradigms don’t agree on what kinds of evidence would count as proof. In Kuhn’s terms, the new paradigm and the old are incommensurable. And because scientists, like most people, are reluctant to abandon the intellectual tools which have served them all their lives, younger researchers working in the new paradigm will tend to become unintelligible to their seniors. 

Kuhn repeatedly emphasized this “essential tension” between tradition and innovation within science. Scientists are trained in a particular tradition, and must respect that tradition. Without tradition, science would be totally anarchic, caught up in perpetual foundational disputes. But scientists must also be willing to totally discard paradigms that have outlived their usefulness. For Kuhn, scientific revolutions are “tradition-shattering complements to the tradition-bound activity of normal science.” 

Kuhn’s book was itself tradition-shattering. It broke open the positivist history of science by proposing a way that science could evolve outside the goal-directed forward march of progress. For Kuhn, science is an ever-growing body of knowledge that is constantly shedding its own skin; discarding unnecessary exoskeletons to retain only what is strictly necessary for the training of new students and the discovery of new ideas within the current paradigm. If science is a tower of giants, it’s an unusual tower indeed, because you could remove the base and the whole operation would remain stable.

Structure was an immediate hit, one of few books from its era that academics in history, philosophy, sociology, and beyond could assume that all of their peers and colleagues had read. It was also extremely controversial: Kuhn was accused of being a (gasp!) relativist, of taking science off its sturdy objective footing and reducing it to nothing more than a set of current cultural practices. Kuhn, of course, rejected these accusations: Just because the process of scientific discovery can be explained in sociological terms does not mean that scientific truths are true because everyone believes in them. 

But Kuhn’s work initiated its own paradigm shift. In the decades that followed, many historians and sociologists of science began to follow his example to ends that quickly veered into the relativistic, the social constructivist, and the postmodern. Kuhn opened the door to understanding the scientific community in terms of their social processes and relational dynamics, not just the discoveries and personalities of great men. Anthropologists began studying laboratories as purely cultural spaces, showing up with binoculars and notepads to document this unusual creature “the scientist”; feminist scholars pointed out the ways science reinforced cultural, gendered, and racial biases. In the ‘70s and ‘80s, as the environmental and anti-nuclear movements gained steam, the de-pedestalization of science would only gain momentum. 

The most potent intellectual project to emerge in these decades was a field called the “sociology of scientific knowledge.” It started with the observation that when studying scientific controversies of the past, historians often dismissed errors (phrenology, phlogiston) as the result of politics or biases, but were happy to accept that true theories won out because they are true. The sociologists proposed a “strong programme:” what if we ignored, for the time being, our knowledge of which theories are true and false, and purely examined the social forces that led people to accept certain theories and reject others?

The most famous work to take this sociological approach was Leviathan and the Air Pump by Steven Shapin and Simon Schaffer, still read by all graduate students in history of science. The book is about a debate about the nature of the vacuum between Robert Boyle — chemist, inventor, early member of the Royal Society — and Thomas Hobbes, whom you probably recognize more from writing Leviathan than for his stance on fluid dynamics. Shapin and Schaffer used this controversy to understand the role of experiment in determining scientific truth, at a time when scientific experiment was still emerging as a new kind of social practice.

Boyle was an avowed experimentalist. He hosted demonstrations in his laboratory, bringing influential figures together to show them a hollow sphere which he claimed to be pumped absent of all air. If you couldn’t make it in-person, Boyle was happy to send around elaborate written descriptions of these events, so that “virtual witnesses” could collaborate his experimental results. 

Hobbes viewed this as not just false, but dangerous. To Hobbes, the existence of a vacuum legitimated the possibility of “incorporeal substance,” which priests could use to gain political favor by promising the existence of an immortal soul. This could in turn cause a split in the English public between the Monarch and the Church, which might lead to civil war. Hobbes equally disliked the idea there could be a closed-off, apolitical laboratory space where facts could be worked out bit-by-bit. For Hobbes, knowledge should be developed in plain view, using precise definitions, not a faulty, leaking vacuum pump. Hobbes accused Boyle of trying to persuade his viewers through elaborate experimental theater and a letter-writing campaign.

Shapin and Shafer reject the idea that Boyle’s view of the case won out because Boyle was “right” and Hobbes was “wrong.” From the point of view of the people there, at the time, there was no obvious way to adjudicate between Hobbes’ natural philosophy and Boyle’s experimentation. The validity of experimentation as a source of knowledge was itself under debate, and the experiments themselves were neither consistent nor conclusive (the pump really did leak). 

They invite us, in other words, to take seriously Hobbes’ own perspective: Scientific knowledge is socially constructed. The book concludes with a memorable parting message: “It is ourselves and not reality that is responsible for what we know. Knowledge, as much as the state, is the product of human actions. Hobbes was right.” ¹¹

Of course, Steven Shapin and Simon Shaffer don’t think that Thomas Hobbes was factually correct about whether or not vacuum exists. In the long term, the experimental scientific method remains a highly reliable source of knowledge, and Boyle’s theories have stood the test of time. But Hobbes was right that getting people to believe scientific truths often involves persuasion campaigns similar to those run by politicians; and that the problem of scientific knowledge and the problem of social order are much more intertwined than naive history of science would allow. 

Leviathan and the Air Pump is one of a great many boundary-pushing histories that flowered in the later part of the Cold War era. Unburdened by the need to write biographical stories of great scientists, historians could foreground scientific institutions and funding mechanisms, the role of women in science, the relationship between science and empire, or the importance of patronage relationships. Histories of science could become histories of scientists’ dietary habits, or their religious lives, or their pictures with monkeys. ¹²

One final example to round out the point: Consider two ways of visualizing the history of the bicycle. The naive history approach begins with the bicycle as it is today, and traces back its antecedents — a reverse chronology, backwards through the march of progress. 

The social history approach starts from the past, and asks: At any given moment, what vehicles could have plausibly been part of the category “bicycle”? This exercise yields a much richer picture. All manner of strange devices — low-wheelers, high-wheelers, “manupeds,” “safety ordinaires”— proliferated in the late-19th century. Sporting cyclists wanted big wheels for speed; elderly people wanted to be close to the ground; women didn’t want their dresses to get caught. Out of these many branches eventually emerged the smaller air tire bicycle as we now understand it.

What should be clear from these examples is that the past is not simpler than the present. If anything, studying the history of a “resolved” object like the bicycle or the theory of the vacuum yields far more complexity. Science is not a march of progress, it is branching evolutionary tree. Most branches lead to dead ends, but we do ourselves a disservice as historians if we only consider the rare few that remain intact. When we treat history as an act of contextualizing people within their world, rather than tracing their trajectory to ours, the past comes alive in a new way.

After several decades of humanities scholars critiquing, deconstructing, and re-conceptualizing science, scientists eventually took notice. As it happens, scientists don’t like being accused of creating politically mediated social constructs; they want to be discovering truths about the world. Starting in the early ‘90s, a steady stream of scientist-penned books and articles offered their own critiques of science studies. These disgruntled scientists accused their counterparts in the humanities of “pretentious nonsense, miasmatic New Age drivel, romantic antagonism, unrelenting ecobabble, sophomoric ideology, damaging quackery, and even hermeneutic hootchy-koo.” ¹³ The Science Wars had begun.

For several years during the mid-90s, the temperature was high. Up for grabs: Who gets to speak about science? Does critiquing or deconstructing science delegitimize it? Should we analyze physics or math through the lens of gender or race? Are science studies scholars fanning the flames of anti-scientific thought in America? Are scientists even interested in bridging the two-culture divide?

The wars came to a head in 1996, when the physicist Alan Sokal published a hoax article in a critical theory journal. “Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity” proposed that quantum gravity is a social and cultural construct, and that future work should consider its progressive political implications, not just scientific ones. Science Warriors rejoiced: they had finally proven that social theory journals were a joke, that humanists didn’t really know what they were talking about, and that discussions of science should be left to the scientists.

Take a step back. What exactly was happening here? Why were physicists pulling pranks on social theorists? 

The sociologist Dorothy Nelkin has posited that the Science Wars were really a backlash by a physics community in decline. As the Cold War drew to a close, the massive military R&D budgets of the postwar era no longer made as much sense as they used to. The implicit contract between the state and science — the state providing no-strings-attached funding for basic scientific research, science providing novel insights that could eventually be turned into instruments of industrial or military value — was breaking apart. Corporations were becoming increasingly entangled in university goings-on, and basic research in biology, pharmacology, and later computing was increasingly taking place in the private sector. Massive national labs, once the forefront of research, were now state-sponsored factories making marginal improvements to existing wartime innovations.

Congress began shifting its priorities away from physics too. The watershed moment took place in 1993: Congress pulled the funding from the Superconducting Super Collider, a planned particle accelerator three times larger than the large Hadron Collider that was to be built an hour south of Dallas, Texas. Construction was already underway and 22km of track had already been laid when concerns about cost made the project impossible to justify. This was the final symbolic blow: Physics, the defining science of the 20th century, was now on the decline.

The physicists needed a scapegoat to blame for Congress’ turn against their field, and the social constructivists were an easy target, albeit a nonsensical one. Congress was not reading Donna Haraway and Bruno Latour, they were simply responding to the American public’s concerns about the appropriation of military funds. Physics was an institution under siege, which tend to, in the words of the anthropologist Mary Douglas, “block personal curiosity, organize public memory, and heroically impose certainty on uncertainty.” ¹⁴ Physicists wanted to seal their doors and preserve their purity in the face of their falling national status after a full century as the leading voices in science.

And what of the science studies scholars? They also did not leave the Science Wars unscathed. The critique leveled against them was in part justified: Arguments about the social construction of scientific facts were used to justify climate denial, or downplay the deleterious effects of cigarettes, or argue for ignorance on vaccines. Starting in the early 2000s, the field began a pivot away from pure critique and towards affirmations of scientific expertise. In 2004, Bruno Latour — a founder of social constructionism and notable targets of scientific ire — published an essay called “Why has critique run out of steam?,” encouraging scholars in the humanities to help buttress the legitimacy of scientific expertise in the Bush era.

History of science, for its part, continued as it always had: evolving its focus to meet the political moment. One clear shift is that the field is becoming far more international, and incorporating forms of knowledge outside of traditional Western science. As Lorraine Daston has observed, the history of science is gradually becoming the history of knowledge. All traditions of knowledge are fair game: medieval European alchemy, indigenous Peruvian botany, early modern Caribbean medicine, ancient Chinese technology, global folk craft practices, techniques of bureaucracy and management. In Daston’s words, “the form science might well emerge as  one interestingly distinct species of the genus knowledge, but it is unlikely to be the only one.” ¹⁵

A second clear trend in 21st century histories is a move away from big picture theories and towards many overlapping frameworks. Historians are unlikely to propose grand overarching structures in the mode of Kuhn, and much more likely to explore a subject from several different vantages: class, race, gender, empire, culture, politics, the body.

Consider two recent examples from the field I am most familiar with, the history of AI: Abstractions & Embodiments, an edited volume by Janet Abate and Stephanie Dick, and a 2023 special journal edition titled Histories of Artificial Intelligence: A Genealogy of Power. The edited volume is about juxtaposing the abstract nature of computers — their relationship to boolean logic, formal languages, and mathematical procedures — and their highly tangible, material properties, which are tied up in “global systems of resource extraction, environmental control, and gendered and racialized labor.” ¹⁶ The special edition’s goal is to expand “the scope of what ‘history of AI’ is a history of,” ¹⁷ by situating the history of AI in histories of management, bureaucracy, and surveillance. The authors’ ultimate aim is worth stating in full: “Historical understanding can be a powerful tool for breaking with long-taken-for granted paradigms and assumptions about language, norms and possibility. At its best, history salvages the complexities of past decisions, and decision makers, to populate one’s imagination about potential new social practices.”

Here we have an entirely new purpose for the history of science, one we have not seen in our century-long survey. We have come quite a long way from Conant: Instead of the simpler past serving as a guide to a hopelessly complex contemporary world, it is the complex past which can free us from the imaginative poverty of an overly narrow present.