The History of Science

Not an obvious list of geniuses, but a messy human story that stumbles upon greatness.

The way science is usually taught is as a relay race of geniuses. Aristotle passes the baton to Galileo, who hands it to Newton, who hands it to Darwin and Einstein, each one a lone brilliant mind seeing what no one had seen, and the whole thing marching steadily toward the truth. The real history of science is messier, more human, and far more instructive: a story of error, ego, money, luck, suppression, and slow collective correction, in which the establishment is sometimes the hero and sometimes the villain, the lone dissenter is sometimes a visionary and far more often a crank, and the truth, when it finally arrives, usually does so not because one person was brilliant but because the method was better than the people using it.

That messiness is not a reason to distrust science. It is the reason to understand how it actually works, because the original story leaves you defenceless in two opposite directions. It leaves you unable to see when science has been corrupted, captured by money or ego or institutional pride, and equally unable to see why the lone genius claiming persecution is, almost always, simply wrong. Science is a human enterprise, with all the failure that implies, and with the one feature that nonetheless makes it the most reliable knowledge-finding process we have.

II. The Greek Experiment: Reasons Instead of Gods

Something genuinely new appeared in the Greek-speaking world from around 600 BC, and it is one of the more consequential shifts in human thought. A handful of thinkers in Ionia, on the coast of what is now Turkey, began trying to explain the natural world without reaching for the gods. Thales of Miletus asked what everything was fundamentally made of and answered, wrongly but revolutionarily, “water,” because the point was not the answer but the question: a natural question expecting a natural answer. This is the seed of everything that follows. The Greeks were the first, as one historian puts it, to look for general principles beneath the observations, rather than just cataloguing the observations themselves.

What followed over the next few centuries was an explosion of abstract inquiry. Democritus proposed that all matter was made of indivisible particles he called atoms, a guess that turned out, two thousand years early, to be profoundly right in spirit. Pythagoras and later Euclid built mathematics into a system of proof, deriving certain truths from first principles, an achievement so powerful it still defines what we mean by rigour. Aristotle attempted to classify and explain essentially everything, living creatures, motion, causation, the heavens, and, in doing so, created the first comprehensive system of natural philosophy. Eratosthenes calculated the circumference of the Earth from the angle of shadows in two cities and got remarkably close. Aristarchus suggested, far ahead of his time, that the Earth might go round the Sun.

It was the birth of rational, principled inquiry, of the conviction that the universe is intelligible and that human reason can grasp it. But it was not yet science in the full sense, because it leaned heavily on reasoning from first principles and comparatively little on systematic experiment and testing. Aristotle’s physics, brilliant and internally coherent, was also substantially wrong, and because it carried such immense authority, some of its errors would stand for nearly two thousand years. This is the double edge of the Greek achievement and a lesson the rest of this history keeps teaching: reason unhooked from rigorous testing can build magnificent, self-consistent structures that happen not to be true. The Greeks supplied the indispensable conviction that the world can be understood. The discipline of forcing those understandings to survive contact with evidence was still incomplete.

III. The Long Custodians: How the Knowledge Survived

The popular Western story leaps straight from the Greeks to Galileo, as though the lights went out for a thousand years and then someone flicked them back on. This is one of the more misleading things about the way the history is usually told, and correcting it matters both for accuracy and for fairness.

When the Western Roman Empire fell and Greek learning faded from Latin-speaking Europe, that knowledge did not vanish. It was preserved, translated, debated, and significantly extended elsewhere, above all in the Arabic-speaking world during the Islamic Golden Age, roughly the eighth to the thirteenth centuries. Scholars gathered and translated the Greek texts, and then went further. Al-Khwarizmi developed algebra into a systematic discipline, giving us both the word and the foundation of a field. Ibn al-Haytham, known in Europe as Alhazen, did something especially important for this section’s story: in his study of optics and light, he insisted on systematic experiment and the testing of hypotheses against observation, to the point that many historians credit him as one of the earliest practitioners of something genuinely like the scientific method. Ibn Sina, known as Avicenna, wrote a canon of medicine that would be taught in Europe for centuries. Across astronomy, medicine, optics, chemistry, and mathematics, this was not mere safekeeping; it was active advancement.

This knowledge then flowed back into Western Europe from the tenth to the thirteenth centuries, recovered through translation and contact, reviving the study of natural philosophy in the medieval universities then taking shape. The medieval scholars are easy to mock for their reverence of Aristotle and their entanglement of reason with theology, and certainly some of that reverence later became a cage. But the universities, the preservation of the texts, the careful commentary, and the slow rebuilding of a learned culture were the soil from which the next great shift would grow. The deeper point, easily lost in a triumphalist Western telling, is that the river of knowledge ran through many civilisations and many hands. No one person invented the understanding of nature. It was carried, lost, recovered, and added to across continents and centuries, and the pretence that it sprang fully formed from European genius is itself a piece of the self-congratulatory mythology this page exists to puncture.

IV. The Break: When Looking Beat Authority

Then, in Europe across the sixteenth and seventeenth centuries, something strange happened that genuinely deserves the name revolution. Copernicus published his Sun-centred model of the heavens in 1543, and Newton published his laws of motion and gravitation in the Principia in 1687. Between those dates, the way human beings investigated the natural world changed more fundamentally than it had in all the millennia before, and the change was less about any single discovery than about a new attitude to knowledge itself.

The heart of it was a shift in what counted as authority. For two thousand years, the way to settle a question about nature was to consult the ancients, to ask what Aristotle had written. The revolutionaries made a radical move: they decided that observation and experiment outranked any authority, however venerable. Copernicus put the Sun, not the Earth, at the centre, displacing humanity from the middle of creation. Tycho Brahe made obsessive, precise measurements of the heavens; Kepler used them to work out that the planets move in ellipses, trusting the data even where it broke beautiful theories. Galileo turned the new telescope on the sky, saw moons orbiting Jupiter and craters on a supposedly perfect Moon, and insisted that what he could see outweighed what authority decreed, a stance that brought him, as a later chapter discusses, into direct collision with the Church. Francis Bacon argued forcefully for building knowledge up from controlled observation and experiment rather than down from received assumptions. And Newton fused it all into a single mathematical system that described the fall of an apple and the orbit of the Moon with the same equations, a triumph so complete it set the template for what a scientific theory could be.

Two features made this new way of knowing different in kind from what came before. The first was the marriage of mathematics with deliberate experiment: not just observing nature or just reasoning about it, but forcing precise quantitative predictions to confront controlled tests. The second was that knowledge became collective and open in a new way. The founding of bodies like the Royal Society in London and the Académie des Sciences in Paris created communities that shared methods, published results, and checked one another’s work, the beginnings of the social machinery that later chapters show to be the real engine of scientific reliability. Knowledge stopped being the possession of isolated sages and started becoming a public, cumulative, self-correcting enterprise. That shift, more than any single law or discovery, is what the Scientific Revolution actually was.

V. The Invention of the Scientist

Here is a fact that surprises almost everyone and that quietly dismantles a great deal of mythology: for most of the history just told, there were no scientists. Not the word, and not the role. The people we now claim as the founders of science did not call themselves scientists and would not have recognised the term. The word was coined only in 1833, by the English polymath William Whewell, to describe a new kind of professional, and it caught on slowly, gradually displacing the older “natural philosopher.”

This is not a trivial point of vocabulary. It is a window onto how recent and how constructed our modern picture of “the scientist” really is. The towering figures of earlier centuries were not members of a profession called science; they were courtiers, clergymen, physicians, aristocrats with time and money, and very often practitioners of pursuits we now file under superstition. Kepler earned his living partly casting horoscopes. Newton, the supreme icon of rational science, spent enormous energy on alchemy and biblical prophecy. These were not scientists in lab coats but curious, brilliant, frequently mystical people working without salary, method, or community of the kind we now take for granted. The tidy image of the rational scientist as a distinct and timeless type is, to a striking degree, a nineteenth-century invention, assembled partly to explain and flatter European dominance and projected backward onto people who would not have understood it.

What the nineteenth century actually built was the profession: science as a salaried career, conducted in universities and laboratories, organised into disciplines, funded and institutionalised, with training, journals, and standards. This professionalisation was the engine of staggering progress, Darwin’s theory of evolution, the germ theory of disease, the laws of thermodynamics, the periodic table, the foundations of genetics, all crowded into a single century. But it also planted the seeds of the troubles that later chapters examine. Once science became a career with salaries, prestige, institutional loyalties, and funding to compete for, it acquired all the ordinary human pressures of any profession: the hunger for status, the defence of territory, the temptation of money, the fear of being wrong in public. The very forces that would suppress a Semmelweis or court a sugar industry’s funding entered science not as corruptions of some pure original state, but alongside the professional structures that also made modern science possible. The scientist was invented, and with the scientist came both the extraordinary productivity and the very human fallibility that runs through everything that follows.

VI. When the Establishment Was Wrong

In the 1840s, a Hungarian physician named Ignaz Semmelweis, working in a Vienna maternity ward, noticed that women delivered by doctors were dying of childbed fever at far higher rates than those delivered by midwives. He worked out that the doctors were going straight from performing autopsies to delivering babies without washing their hands, and that something carried on those hands was killing the mothers. He ordered handwashing with a chlorinated solution, and the death rate collapsed. He had, in effect, discovered antisepsis before anyone knew germs existed. And the medical establishment rejected him. His findings were ridiculed, he lost his position, and he died in an asylum in 1865, broken and discredited, decades before Pasteur and Lister vindicated the germ theory that explained why he had been right all along. The episode gave us the phrase “the Semmelweis reflex,” the reflexive rejection of evidence that contradicts established belief, often because accepting it would require admitting fault. Senior doctors did not want to believe their own hands were killing their patients, and so they did not.

The story is genuinely damning, but if you stop there, you have learned the wrong lesson. Semmelweis could not explain why his method worked; he had the result but no mechanism, in an era before germ theory, which made his claim far harder to accept than hindsight suggests. He was also, by many accounts, abrasive and undiplomatic, attacking colleagues in ways that made enemies rather than converts. And the rejection was not total; some physicians did adopt his practices. The reality was not “the establishment were unthinking louts and Semmelweis was a pure martyr,” but something more useful: a correct and important finding was delayed for decades by a combination of institutional ego, a missing mechanism, and the messenger’s own conduct. All of those are recurring features of how science goes wrong.

The pattern recurs. In the early twentieth century, Alfred Wegener proposed that the continents had once been joined and had drifted apart, and was dismissed for decades, partly because he could offer no mechanism, until the discovery of plate tectonics proved him right. In the 1980s, two Australian researchers, Robin Warren and Barry Marshall, proposed that stomach ulcers were caused not by stress and acid, as every authority then insisted, but by a bacterial infection. The idea was treated as faintly ridiculous; to gastroenterologists, the notion of a germ surviving in the acidic stomach was, in one contemporary’s words, like saying the Earth is flat. Marshall’s work was ranked in the bottom fifth of submissions at a conference. Unable to get a fair hearing, he eventually drank a culture of the bacteria himself, gave himself gastritis, and cured it with antibiotics. They were proven right, transformed ulcer treatment from lifelong management to a short course of antibiotics, and won the Nobel Prize in 2005. It is worth noting who profited from the old view: pharmaceutical companies made fortunes from the acid-suppressing drugs that managed ulcers without curing them, a reminder that institutional inertia and financial interest often pull in the same direction.

VII. Why the Establishment’s Caution Is Not Always Villainy

The scientific establishment’s default resistance to new claims is not simply ego, though ego is often present. It is also, most of the time, appropriate. The overwhelming majority of bold claims that contradict established science are wrong, and a field that abandoned its hard-won theories every time someone presented a surprising result would not be a science at all; it would be a weathervane. The same conservatism that wrongly delayed Semmelweis and Marshall is what correctly protects established knowledge from the daily flood of mistaken, misanalysed, and fraudulent claims. The demand that an extraordinary claim come with an explanation and survive replication is not closed-mindedness; it is quality control, and it is right far more often than it is wrong. Even the cases of genuine suppression are, on closer inspection, rarer than the popular telling suggests; historians who have looked find that prematurely rejected correct theories are genuinely uncommon, because more often a surprising-but-correct idea provokes a controversy in which part of the field takes it seriously, rather than a clean unanimous rejection. The establishment being wrong is the dramatic exception we remember precisely because it is an exception. The establishment is sometimes wrong, and the establishment is usually right, and no rule tells you in advance which case you are looking at. Only the evidence, examined carefully, can do that. If only the AI industry had this level of constraint. 

VIII. When Money Writes the Conclusion

If institutional ego is one way science gets corrupted, money is the other, and it is more deliberate, more dangerous, and very well documented. 

The clearest example is the tobacco industry, which, faced with mounting evidence that its product caused cancer, did not try to prove the science wrong. It did something more clever: it funded research and public-relations campaigns designed to manufacture doubt, to make a settled question look open, knowing that a confused public would keep buying cigarettes. The internal strategy, later exposed, was summarised in a now-infamous line from a tobacco executive: doubt is our product. This playbook, attacking the science, funding friendly researchers, emphasising uncertainty, has since been used again and again, and it is documented in detail in the work of historians Naomi Oreskes and Erik Conway.

The sugar industry ran the same play, and we know the details because internal documents surfaced. In the 1960s, as evidence mounted that sugar contributed to heart disease, a sugar industry trade group secretly paid prominent Harvard nutrition scientists to publish a review that downplayed sugar’s role and shifted the blame onto dietary fat. The funding was not disclosed; the industry helped set the review’s objective and supplied material; the scientists delivered the desired conclusion in a leading medical journal. It helped steer decades of dietary advice, public guidelines, and research funding toward fat and away from sugar, an emphasis that some researchers argue worsened the very epidemics of heart disease and obesity the guidance was meant to prevent. As the nutrition scientist Marion Nestle put it, industry-funded research, like that funded by the tobacco and pharmaceutical industries, almost invariably produces results that favour the sponsor’s product, even when independent research reaches the opposite conclusion. The same concern extends to the pharmaceutical industry, where the companies that fund drug trials have a direct financial stake in the outcome, where unfavourable results have at times gone unpublished, and where the statistical games described in Understanding Statistics find some of their most profitable application.

Follow the money. When you encounter a research claim, especially a health claim, ask who funded it and who profits from the conclusion. This is not cynicism, and it does not mean industry-funded science is automatically false, which would be its own lazy error. It means that funding is a powerful and documented source of bias, that conflicts of interest are reliable predictors of which way a study will lean, and that the absence of disclosed funding sources and competing interests is itself information. The corruption of science by money is not a conspiracy theory; it is a matter of public record, exposed repeatedly by other scientists using the methods of science. Which is the detail the conspiracy-minded reading misses: we know about the sugar papers and the tobacco strategy because the self-correcting machinery eventually dragged them into the light.

IX. The Galileo Gambit: The Trap on the Other Side

Now the counterweight, and it is just as important, because everything above can be twisted into a licence for believing anything. The reasoning goes: the establishment suppressed Semmelweis and Marshall, the establishment is corrupted by money, therefore the establishment is not to be trusted, therefore my rejected idea must be the next great vindicated truth. This move is so common among cranks and pseudoscientists that it has a name: the Galileo gambit.

They laughed at Galileo, the argument runs, and he was right; they laugh at me, therefore I am right. The flaw is a simple failure of logic. As Carl Sagan put it, the fact that some geniuses were laughed at does not mean that everyone who is laughed at is a genius; they laughed at Columbus and the Wright brothers, but they also laughed at Bozo the Clown. Being dismissed by the scientific community is not evidence that you are right, because the overwhelming majority of people dismissed by the scientific community are dismissed for the excellent reason that they are wrong. For every Semmelweis who bucked the consensus and was vindicated, there are thousands who bucked it and stayed wrong, and they have vanished from the story precisely because they were wrong. The vindicated rebels are memorable because they are rare. To model yourself on them based on being disbelieved is to mistake a necessary condition for a sufficient one: being doubted is something the great revolutionaries shared with every crackpot who ever lived.

Galileo was not, in fact, suppressed by the scientific establishment of his day; he was censored by the Catholic Church, a religious authority enforcing dogma, which is rather closer to the behaviour of the modern crank’s own movements than to the science they are attacking. And what eventually vindicated Galileo, Semmelweis, Wegener, and Marshall was not their persecution. It was evidence, accumulated and checked by the same scientific community, that eventually became impossible to deny. Persecution did not make them right; the evidence did. The persecution is the part of the story we find dramatic, but it is the evidence that did the work.

X. The Subtler Failure: Fooling Yourself

Beyond ego and money, there is another way science goes wrong, and it catches not frauds and cranks but skilled, honest, well-intentioned researchers. The physicist Irving Langmuir called it pathological science: the science of things that are not so, produced by capable scientists who have unconsciously fooled themselves. Its hallmarks are an effect at the very edge of detectability that never gets stronger no matter how the experiment improves, claims of great accuracy, and explanations that grow more elaborate as the evidence fails to cooperate. Whole research programmes have run for years on effects that turned out not to exist, sustained by genuine scientists who wanted to believe. Even Nobel laureates are not immune; more than one has, late in a distinguished career, embraced an idea their peers regard as plainly mistaken. The point of this is humbling, and it returns to the foundation laid in The Scientific Method: the enemy is not only the dishonest outsider but the self-deceiving insider, and the credential offers no protection against it. This is exactly why the method does not, and must not, rest on trusting the brilliance or integrity of individuals.

XI. Why the Method Survives Its Practitioners

Pull the whole messy history together, and a single conclusion emerges, and it is the one this section has been building toward. Science is done by humans, and humans are biased, proud, tribal, corruptible, and capable of fooling themselves and others. The individual scientist is not especially trustworthy. The establishment is sometimes wrong and sometimes captured. Money distorts. Egos suppress. Honest people deceive themselves. If science depended on the virtue or genius of the people doing it, it would be no more reliable than any other human institution, which is to say not very.

And yet it works, over time, better than anything else we have ever built for finding out what is true. The reason is the thing this history actually demonstrates: science is not the individuals but the process that grinds on across and despite them. Semmelweis was vindicated, eventually. Marshall won his Nobel. The sugar papers and the tobacco strategy were exposed. Wegener’s continents were proven to drift. Pathological science programmes collapsed when their effects could not be replicated. In every case, the correction came not from any one heroic person but from the slow, adversarial, self-checking machinery of replication, accumulating evidence, and generational turnover, the same machinery that demands a mechanism, that eventually digs out the hidden funding, that fails to reproduce the effect that was never real. The method is more honest than the people who practise it, because it is built to be, designed precisely to catch the errors that individuals cannot see in themselves.

This is the resolution of the section’s central tension, and the reason it is neither dogmatic nor naive to trust science. You do not place your trust in scientists, who are as fallible as anyone, nor in institutions, which can be captured, nor in consensus as such, which has sometimes been wrong. You place a calibrated, provisional, eyes-open trust in the long-run self-correcting process, while keeping the tools from this section in hand to check any particular claim yourself, including who paid for it. The history of science is messy because people are messy. That it nonetheless converges on truth, slowly and against the grain of its practitioners’ flaws, is the strongest possible argument that the method is doing something real. The mess is not the failure of science. The mess, corrected over time, is science.

XII. Cross-Links

• Science for the section overview
• The Scientific Method for the logic that does the correcting
• Understanding Statistics for the statistical games money exploits
• Scientific Methodology Cheat Sheet for the practical toolkit
• The Science Rabbit Hole for the deeper questions about scientific knowledge
• Resources for the reading list