Hobbes v. Boyle: Who Decides Scientific Facts
In the 1600s, philosopher Thomas Hobbes and experimental scientist Robert Boyle clashed over a strange new machine, the air pump, and a dangerous question: when should society trust scientific claims, and who gets to decide?
Their disagreement wasn’t just about experiments. It was about evidence, authority, and the role scientists play in shaping public policy.
That same tension still defines our world today, where scientific expertise informs decisions that affect all of us, often in the middle of public distrust and political noise.
There is a sentence that has become almost magical in American public life.
“The evidence shows…”
Sometimes that sentence lands like a relief. Sometimes it lands like an insult. Sometimes it lands like a challenge, with the exclamation, “Prove it.”
If you have ever watched a public debate about a vaccine, a heat wave, air quality, lead in drinking water, or a changing medical guideline, you have seen the same pattern. One side says, “Here is the evidence.” The other side says, “I do not trust the people telling me what the evidence means.”
That is the part that matters.
Because very often the disagreement is not only about data. It is about who gets to turn observations into facts that we share.
And in the United States right now, trust is uneven. Pew Research Center has tracked public confidence in scientists and found that it is powerfully shaped by partisan identity and debates over the role scientists should play in public policy. Their November 2024 report lays out that tension in detail, including how Americans are divided over whether scientists should stay focused on establishing facts or take a more active role in policy debates.
If we want to understand why evidence so often fails to persuade, we need to go back to the early moments of modern science, when its rules were not yet settled, and there was infighting within the science community. Once we see this infighting, we start to recognize it everywhere, even today.
Picture a crowded room in London in the early 1660s. The air smells like wax and wool. People lean forward, shoulder to shoulder, because there is a machine at the center of the room that looks almost magical.
It is a glass chamber attached to a pump. Someone works the handle. The sound is rhythmic and straightforward, like an old bicycle pump. And then, inside the glass, something changes.
A flame shrinks. A living creature struggles. A bell goes quiet. The room reacts because it is watching nature do something it “should not” do.
Robert Boyle, one of the men behind this machine, has a simple message, which is “Watch what happens. We can test it.” The machine is an air pump. And the man who thinks Boyle’s machine is a dangerous way to manufacture “truth” is Thomas Hobbes. Hobbes, who is not in the room, has his own message, and that is, “That is not proof. That is a performance.”
This is the story of their dispute, and why it is not really about air; it is about a question that still divides communities today, which is, if we cannot test something, how do we decide what counts as true?
The Machine That Made Facts
If we want to understand the Boyle–Hobbes dispute, start with something very ordinary. Let’s start with the fact that most people cannot build a scientific instrument. Boyle’s device is usually described as an “air pump,” or a “vacuum chamber.” The words can sound technical, but the idea is relatively simple. Imagine a sturdy glass container connected to a pump that can pull air out. It creates a small world under glass where we can change one thing. We can change the amount of air and watch what changes with it. Robert Hooke helped Boyle with the air pump’s development. The Science History Institute describes Boyle’s air pump as the centerpiece of a new experimental culture and notes that Boyle’s 1660 publication describes his attempts to understand air, respiration, and pressure.
Robert Boyle was one of the founders of modern experimental science. He helped establish the Royal Society in seventeenth-century England and is best known for Boyle’s Law, which describes the relationship between pressure and volume in gases. But more importantly for this story, Boyle believed that carefully designed experiments, repeated and openly described, could produce reliable knowledge about the natural world, even when the deeper causes were not yet understood.
Boyle did not do one dramatic demo with his airpump and call it a day. He conducted 43 experiments and described them in this 1660 book, New Experiments Physico-Mechanicall, Touching the Spring of the Air.
But Boyle had a problem: most people could not be in the room. Most people could not touch the machine, check the seals, or watch the flame shrink. So Boyle was not only doing experiments but also inventing ways to make people believe in experiments they did not personally witness.
What Boyle did was promote a new approach to natural philosophy. Instead of endless arguments about what must be true in principle, he wanted controlled tests that could surprise his viewers. He wanted repeatable events that were carefully observed and meticulously recorded.
And what was groundbreaking at the time was that this was not only science; it was also social engineering. Boyle’s core problem was not to make a flame shrink in low air. But instead, his core problem was how to persuade people who were not in the room.
Now, let’s keep this setting in mind. This is England, not long after civil war and political upheaval. People lived through competing authorities and competing certainties. People not only feared political chaos; they also feared epistemic chaos, meaning that if no one agrees on what is true, then conflict becomes permanent.
Thus, in the seventeenth century, a new scientific culture was forming. A community of investigators began meeting, demonstrating experiments, keeping records, and arguing under agreed-upon rules. And the Royal Society, which hosted Boyle, was beco ming the symbol of this movement.
The Royal Society took shape as a community that settled disputes by experiment, not by deference. The Society’s own history notes that its motto, adopted in the First Charter in 1662, is “Nullius in verba,” often taken to mean, “Take nobody’s word for it.” This framed the Society’s motto as a commitment to verify statements by experiment.
This motto may sound like a hopeful promise. But here is the twist: a motto is not a method, and a process is not trust.
Boyle was asking people to believe results that depended on specialized equipment, specialized skill, and a carefully staged environment. He was asking them to accept a new kind of authority, even while claiming to resist authority.
And that is where Hobbes entered the story. In this scenario, Boyle was saying, not in these exact words, but he was saying, “Watch what happens,” to which Hobbes responded, “Who gets to decide what you saw?”
Hobbes, the Skeptic of the Room
So, who is Thomas Hobbes? He is famous for his political philosophy, especially Leviathan, but he also cared deeply about what counts as real knowledge. He wanted certainty, he trusted deduction, and he valued geometry. Additionally, he was suspicious of claims that rested on fragile instruments and witness testimony, especially when the witnesses were a tight-knit group.
In 1661, Hobbes published Physical Dialogue, or, On the Nature of Air, a work aimed at criticizing Boyle’s air-pump experiments. And in his critique, as quoted from the 1985 book, Laviathan and the Air Pump by Steven Shapin and Simon Shaffer, Hobbes stated, “If indeed philosophy were (as it is) the science of causes, in what way did they [the experimental philosophers] have more philosophy, who discovered machines useful for experimentes, not knowing the causes of the experiments, than this man who, not knowing the causes, designed the machines?”
In other words, Hobbes was saying, “If you don’t understand the cause, then running the experiment doesn’t magically turn you into a better scientist.”
This is not a rejection of evidence. It is an argument about what is evidence.
Hobbes also objected to the idea of a true vacuum. Even though the glass chamber looked empty, Hobbes considered that it might still hold some kind of subtle matter. Hobbes refused to treat the statement “it looks empty” as the phrase “it is empty.”
And this is where the dispute became permanently modern.
Hobbes was pointing to something real: experiments do not float in space. Instead, experiments happen in rooms; people control access to those rooms; people control the instruments; and people decide which results “count” and which are “artifacts.”
So Hobbes is not only arguing about air. He is arguing about power.
He is asking: “Who gets to certify reality?”
Even today, we can hear a version of Hobbes whenever someone says, “I do not trust the experts,” or “Show me the raw data.”
Sometimes that skepticism is responsible and healthy. Sometimes it is an excuse to ignore inconvenient results. The hard part is that the same sentence can do both jobs.
So the Boyle–Hobbes dispute is not a simple morality play of “science versus anti-science.”
It is a fight over the rules of admission. What counts as evidence? Who gets to produce it? Who gets to audit it?
And now Boyle has to answer.
How do we see what we were not there to see?
Here is the trust problem in its simplest form.
If we are not in the room, we did not see it. If we did not see it, why should we believe it?
Think about the difference between these two situations:
- A friend tells you, “Trust me, it happened.”
- A friend tells you, “Here is exactly where I was, exactly what I saw, exactly what time it happened, and here is a photo, and here is a second witness.”
Boyle understood that second approach. He flooded his writing with detail, knowing that detail is a technology of trust.
Boyle’s answer was not to demand faith. Boyle’s answer was to make the reader feel like a witness.
He did this with writing that was almost aggressively detailed. He described the setup. He described the conditions. He described failures, leaks, and what went wrong. He described what other people in the room saw. He tried to make the event so concrete that a reader could picture it clearly, step by step, and say, “If I had been there, this is what I would have seen.”
In his paper, “Pump and Circumstance: Robert Boyle’s Literary Technology,” Historian Steven Shapin refers to Boyle’s explanation as “virtual witnessing.” The idea is straightforward: the report is written so readers can build a mental image of the experimental scene they did not directly witness, thereby expanding the public that can evaluate and accept the result.
Now pause and notice something. This is not a side detail. This is the invention of a significant feature of modern science. Because science does not scale through the machine alone. Science scales through shared descriptions. This is also where Boyle’s method becomes both powerful and vulnerable.
It is powerful because it enables people far away to judge evidence. But it is vulnerable because it creates a new dependency, which is trust in the honesty and competence of the people writing the report.
Shapin and Schaffer make this point in their book Leviathan and the Air-Pump, in a chapter often excerpted as “Seeing and Believing.” They describe how “matters of fact” were produced in a physical space where experiments were collectively witnessed, as well as in an abstract space created through virtual witnessing.
So Boyle’s air pump created “evidence under glass,” but Boyle’s writing created “evidence in your head.”
That is the new trick. And once we see it, we realize we still live inside it.
So, if you have ever trusted a study you did not personally run, you have done virtual witnessing. If you have ever believed a medical guideline because you trusted the process behind it, you have done virtual witnessing. You are relying on methods, peer review, professional norms, and the expectation that fraud and error are eventually caught.
Virtual witnessing is how modern societies function. We cannot personally verify bridge engineering, vaccine trials, or satellite measurements. We read, we trust the chain, and we keep an eye on how that chain corrects itself when it is wrong.
Hobbes understood this and said that the truth depends on writing, witnesses, and a group of people who agree that the experiment happened as they say it did.
But, Boyle countered this argument, telling Hobbes that this is why they have rules for reporting, rules for critique, and a community that checks.
And Hobbes, in not so many words, stated that this is precisely the problem.
Who Actually Won, and What It Cost
If we are expecting a tidy ending where the experimentalist triumphs and the skeptic fades away, the history is messier.
The dispute has now become a foundational case study in how experimental science gained credibility. The Boyle–Hobbes confrontation showed how scientific practices, social order, and political concerns are entangled rather than separate.
Boyle’s approach spreads because it is useful. It produces results that can be extended, traded, and built upon. It gives people a way to coordinate their beliefs across distance. But Boyle’s approach also comes with a cost. It creates a world in which the public must often rely on institutions, instruments, and communities of specialists.
That is not an insult. That is the reality of advanced knowledge.
And it is the pressure point that Hobbes was warning about. Hobbes was saying that if reality is certified inside specialized rooms, what happens when large parts of the public stop trusting the rooms?
That question takes us straight into the United States right now.
The United States and the Return of Hobbes
The Boyle–Hobbes dispute is a dispute about trust infrastructure. Not trust as a feeling, but rather trust as a system.
Pew Research Center has documented that Americans are split over what scientists should do in policy debates. In Pew’s November 2024 report, 51% of Americans say scientists should be more involved in public policy debates about scientific issues, while 48% say scientists should concentrate on regulating reliable scientific facts and stay out of public policy debates. This split is not a minor detail. This division reveals a deep uncertainty about a scientist’s place in civic engagement and community initiatives.
Are scientists referees who call balls and strikes? Or are scientists players who should push policy toward what evidence suggests?
Now connect all of this to Boyle and Hobbes. Boyle’s program works best when the public sees scientific institutions as trustworthy referees. However, Hobbes’s critique grows when the public suspects that referees are also players.
So, what do we do? We do what Boyle did, but with modern tools. We show the method. We show uncertainty honestly. We show what would change your mind. We show how disagreement is handled inside the expert community.
We also separate two different ideas that people constantly mix up:
- Science as a method (how we test and correct claims).
- Scientists as humans (who can have incentives, blind spots, and values).
Boyle’s legacy is not necessarily to trust the experts. But instead, it is closer to the idea of checking the work of the scientists.
And in America’s current era, that difference matters, because if someone hears “Trust the experts,” they hear a demand for obedience. Furthermore, if they hear “Here is how this claim was tested, here is what would falsify it, and here is what competing experts argue,” they are more likely to hear an invitation.
That invitation does not guarantee agreement, but it makes evidence-based reasoning possible again.
So, to recap, Boyle is saying, “Watch what happens. We can test it.”
Boyle represents the experimental mindset. His point is: Stop arguing in the abstract. Put the claim in front of the world and see what happens. If the result repeats, it earns credibility.
But Hobbes is saying, “That is not proof. That is a performance.”
Hobbes represents the skepticism about experiments. He is saying: A demonstration in a controlled room with a complicated machine does not automatically equal truth. It could be a trick of the setup, a faulty instrument, or a biased interpretation. He wants certainty that does not depend on “trust me, it happened.”
And the general public is saying, “If we cannot test it ourselves, what would convince us?”
This is the bridge to today. Most listeners cannot personally rerun climate models, vaccine trials, or lab experiments. So the real question becomes: What kind of evidence, transparency, or independent checking would make me trust a claim I cannot personally verify? So, this is where science brings in ideas like peer review, replication, data transparency, and trusted institutions. And this also explains why trust breaks down: people who do not take the time to understand the process of evidence-based science are being persuaded to distrust the institutions.
If we take one thing from this story, take this:
Science does not run on facts alone. Science runs on facts plus the social machinery that makes facts credible.
Boyle helped build that machinery: experimental reports, repeatable procedures, and communities of checking.
Hobbes pointed out the vulnerability: if the public does not trust the machinery, facts do not settle arguments.
So the modern assignment is to understand that skepticism is a normal human response to complexity. We can’t shame people into trusting science. It is the role of the science communicator to make the trust infrastructure visible and understandable.
And here we are, 365 years later, realizing that the argument is not over. It just has better microphones now.
And that matters, because microphones do not just amplify good information. They amplify everything. Careful studies, sloppy studies, half-understood headlines, and emotionally satisfying rumors all travel at the same speed. A well-designed experiment and a confident meme can land in your feed with the same authority, and this is a brand-new version of Hobbes’s worry.
Hobbes was anxious about a world where truth depends on a room, a machine, and a small group of witnesses. Today, the “room” is the internet. The “witnesses” are whoever gets the most engagement. And the “machine” is not an air pump; it is an attention system that rewards certainty, outrage, and simplicity, even when the honest answer is careful and conditional.
So what do we do about it, especially if we are scientists or science communicators?
First, let’s lead with humility, not hype. Let’s admit something out loud that science does not always make clear, let’s clarify that science is not a vibe; it is a process. It is a method for catching ourselves when we are wrong. That means evidence-based science is not “Trust me.” Evidence-based science is “Here is how you could check me.”
Second, let’s be clear about the method, not just the conclusion, and make the trust infrastructure visible. We cannot assume people know what peer review is, or what replication is, or why one study is not the same thing as a scientific consensus. We narrate the chain as follows: Here is the claim, here is the method, here is the uncertainty, here is what would change the conclusion, and here is how independent groups have tried to break it.
Third, we separate facts from choices and separate facts from recommendations. People often reject science because they think a factual statement is secretly a political instruction. We can counteract that by saying that science is strongest at describing what is happening and estimating risks. Policy is where communities weigh options, costs, and values. That distinction lowers defensiveness.
Fourth, we implement Boyle’s practices by building virtual witnessing for the modern world. We show our work in a way that ordinary people can follow. Not every equation, not every dataset, but enough transparency that a reasonable person can see we are not hiding the ball. We use visuals, explainers, and simple demonstrations that are honest about what they can and cannot prove.
Fifth, we respect skepticism, but do not reward cynicism. In other words, we treat skepticism like a fork in the road, not a moral failure. There is reasonable skepticism, the kind that asks, “What would convince me?” and there is corrosive skepticism, the kind that says, “Nothing will convince me.” Our job is to invite the first kind forward, and to stop accidentally rewarding the second kind with endless debates that confuse audiences.
And finally, we remember the point of this whole Hobbes–Boyle story: the goal is not to “win” arguments. The goal is to build a shared reality strong enough that we can coordinate, protect each other, and make decisions that will still look reasonable five years from now.
Boyle wanted a world where disagreement could produce knowledge. Hobbes feared a world where disagreement would never end.
In the end, the question of “who gets to call it a fact?” has never really changed. The question remains the same.
But thanks to social media and a 24-hour news cycle, the microphones are everywhere now. And the chaos is louder than ever before. Science is trying to speak in complete sentences in a world that rewards slogans. We cannot out-trend misinformation, and we cannot out-produce every viral clip. But we can outlast it by being transparent, consistent, and human.
We stop yelling into the fan. We turn toward the skeptics and invite them closer to the evidence, as Boyle tried to do, one careful witness at a time.
Because this is the moment we have. And what we do with it matters. Carpe diem.
FURTHER READING
Public Trust in Scientists and Views on Their Role in Policymaking — https://www.pewresearch.org/science/2024/11/14/public-trust-in-scientists-and-views-on-their-role-in-policymaking
Full Boyle from Science History Institute — https://www.sciencehistory.org/stories/magazine/full-boyle/
Pumped Up — More than 350 years ago the very first air pump changed how science was done by Carin Berkowitz — https://www.sciencehistory.org/stories/magazine/pumped-up/
The History of the Royal Society — https://royalsociety.org/about-us/who-we-are/history/
Hobbes’ Philosophy of Science — https://plato.stanford.edu/entries/hobbes-science
Leviathan and the Air-Pump – by Steven Shapin and Simon Schaffer – on Amazon https://a.co/d/0sHss4L