A New Philosophy of Science for People Who Hate Philosophy

The Problem with Science, According to a Philosopher

Here is a confession: I have spent years avoiding philosophy of science. It felt like a game designed by people who had never watched a lab coat get dirty. You know the arguments. Realism versus anti-realism. Correspondence theories of truth. The kind of debate where nobody changes their mind and nothing in the world shifts.

Hasok Chang, a philosopher at the University of Cambridge, understands this frustration. His 2022 book Realism for Realistic People (Chang, 2022) is an attempt to rescue philosophy of science from people who like philosophy more than they like science. Chang’s central claim is simple and disorienting: the standard version of scientific realism, the one that says our theories correspond to an ultimate reality, is not just wrong. It is useless.

But Chang does not want to abandon realism. He wants to redefine it. He proposes something he calls “activist realism.” It is a philosophy that cares less about whether your theory is “true” in some cosmic sense and more about whether you can actually do something with it. This is a philosophy for people who build things, fix things, and make predictions that fail or succeed in measurable ways.

The Stale Fight That Nobody Wins

To understand why Chang’s approach matters, you need to see the fight he is trying to end.

For decades, philosophers of science have been locked in a war between two camps. On one side are the scientific realists. They argue that our best scientific theories are approximately true descriptions of a mind-independent reality. Electrons are real. Genes are real. The theory of evolution corresponds to how the world actually is. When you say “this is true,” you are claiming a match between your idea and the furniture of the universe.

On the other side are the anti-realists. They point out that scientific theories have changed radically over time. Newton’s physics looked true for centuries, then Einstein replaced it. What we call “truth” is just whatever currently works. Maybe electrons are useful fictions. Maybe we should stop claiming knowledge of unobservable entities.

Both sides make valid points. Both sides also make the same mistake, according to Chang. They assume that the goal of science is to produce statements that correspond to an ultimate reality. The realists think we are getting closer. The anti-realists think we cannot know. But they share the same picture of what “truth” and “reality” mean.

Chang argues that this picture is the problem. It sets up an impossible standard. Nobody has ever shown that a theory “corresponds” to reality in any precise sense. Philosophers cannot even agree on what correspondence means. Meanwhile, scientists keep making progress. They build better microscopes. They cure diseases. They land rovers on Mars. The philosophical debate is irrelevant to the actual practice of science.

What Operational Coherence Actually Means

Chang replaces the correspondence theory with something he calls “operational coherence.” This is the core concept in his book, and it is worth unpacking carefully.

An epistemic activity has operational coherence when the actions it prescribes are consistent with each other and with the actions prescribed by other well-established activities. That sounds abstract, but it is not. Consider a simple example: measuring temperature.

Early thermometers were crude. They relied on the expansion of air or liquid in a glass tube. But how did scientists know that expansion correlated with temperature? They had to build a device, test it against other devices, and check whether the results were consistent. A mercury thermometer and an alcohol thermometer should agree. If they do not, something is wrong with one of them. The coherence between different measurement methods gives you confidence that you are measuring something real.

Chang’s point is that operational coherence is the foundation of scientific knowledge. You do not need to prove that your thermometer corresponds to “true temperature” in some Platonic sense. You need to show that it produces consistent, useful, and reproducible results. Reality emerges from the coherence of your actions, not from a metaphysical claim about correspondence.

This is a deeply pragmatist idea. Chang draws heavily on the work of Charles Sanders Peirce, William James, and John Dewey. The pragmatists argued that the meaning of a concept is tied to its practical consequences. If two concepts lead to the same practical outcomes, they are effectively the same concept. Chang extends this logic to reality and truth. Something is real if it makes a difference in what you can do.

Activist Realism: A Philosophy You Can Actually Use

Chang calls his position “activist realism.” The name is deliberate. It is a realism that demands action, not just belief.

Here is how it works. Instead of asking “Does this theory correspond to reality?” you ask “What can I actually do with this theory?” If a theory allows you to make successful predictions, build working devices, and coordinate actions with other people, then it is real enough. You do not need to know whether it matches some hidden ultimate reality. You need to know whether it works in the world you actually inhabit.

This shifts the burden of proof. The old realism asked scientists to prove that their theories were true in an absolute sense. That is impossible. The new realism asks scientists to prove that their theories are operationally coherent. That is hard work, but it is achievable.

Chang gives a concrete example from the history of chemistry. In the early 19th century, chemists debated whether atoms were real. The anti-realists, led by Ernst Mach, argued that atoms were useful fictions. The realists, led by John Dalton, insisted that atoms were real entities. Both sides made arguments that seemed plausible at the time.

Chang points out that the debate was resolved not by philosophical argument, but by operational practice. Chemists developed methods for measuring atomic weights. They built models that predicted chemical reactions. They created the periodic table. These activities became operationally coherent with each other. The reality of atoms emerged from the coherence of these practices, not from a philosophical proof.

Why This Changes Everything

If Chang is right, then much of what passes for philosophy of science is a waste of time. The endless debates about scientific realism are based on a false premise. They assume that the goal is to match an ultimate reality. But that goal is incoherent. You cannot check whether your theory matches reality without already having access to that reality. It is like trying to check whether a map matches the territory when you are standing on the map.

The alternative is to embrace operational coherence as the standard. This has practical consequences for how we think about science.

First, it makes science more humble. Scientists do not need to claim that they have discovered the ultimate truth. They can claim that they have built a set of practices that are internally consistent and useful. This is a more honest description of what science actually does.

Second, it makes science more open. If operational coherence is the standard, then there is room for multiple coherent systems. Different cultures might develop different ways of knowing that are internally consistent but incompatible with each other. That is not a problem. It is a feature of a pluralistic approach to knowledge.

Third, it resolves the problem of scientific change. Old theories are not “false” in the sense that they failed to correspond to reality. They are “less coherent” in the sense that they could not accommodate new observations and practices. Newtonian physics is not false. It is operationally coherent within a limited domain. It fails when you push it to extremes. That is a more accurate description of what happened.

What This Does Not Prove

Chang is careful to avoid overclaiming. His book is not a license for relativism or postmodern nonsense. He is not saying that all knowledge is equally valid. Operational coherence is a demanding standard. Many claims fail it. Astrology, for example, cannot produce operationally coherent predictions. Crystal healing cannot be tested in a way that produces consistent results. These are not real knowledge because they do not work.

But Chang also leaves some interesting questions open. How do you decide which operational coherence is better? If two systems are both internally consistent, what criteria do you use to choose between them? Chang suggests that pragmatic criteria like simplicity, scope, and fertility are relevant. But he does not give a definitive answer. That is fine. It is an open problem, not a flaw in the theory.

Another open question is about the limits of operational coherence. Can it handle all of science? What about theoretical physics, where predictions are made about entities that cannot be directly observed? Chang argues that even these cases rely on operational coherence. The predictions of quantum mechanics are tested against measurements. Those measurements must be coherent with other measurements. The theory is real to the extent that it enables action.

But there is a tension here. Some physicists argue that quantum mechanics is fundamentally incoherent. The wave function collapse, the measurement problem, the role of the observer. These are not resolved. If operational coherence is the standard, then quantum mechanics is in trouble. Chang acknowledges this. He suggests that we may need to accept a pluralistic interpretation of quantum mechanics, where different interpretations are used for different purposes. That is a controversial claim, and it will not satisfy everyone.

The Practical Payoff

Chang’s philosophy is not just an intellectual exercise. It has real implications for how we think about science and its role in society.

What This Actually Means

• ▸Stop asking whether a theory is true. Ask whether it works. This shifts the conversation from abstract metaphysics to concrete practice. When someone says “climate change is real,” they mean that the models produce predictions that are operationally coherent with observations. That is a stronger claim than “the models correspond to ultimate reality.” It is also a claim that can be tested.

• ▸Embrace pluralism without giving up standards. Different knowledge systems can coexist as long as they are internally coherent. This is not relativism. It is a recognition that there are multiple ways to build coherent practices. Indigenous knowledge about ecosystems, for example, can be operationally coherent within its own framework. It does not need to match Western scientific models to be valid.

• ▸Judge science by its results, not its claims. The old realism asked scientists to make grand claims about the nature of reality. The new realism asks them to show what their theories can do. This is a more honest and more useful standard. It also makes it harder for pseudoscience to hide. Pseudoscience makes grand claims but cannot produce operationally coherent results.

• ▸Accept that scientific knowledge is provisional without being skeptical. Operational coherence can always be improved. New observations can break the coherence of an existing system. That does not mean the old system was false. It means it was incomplete. This is a more accurate description of how science actually progresses.

• ▸Use this philosophy to defend science against its enemies. When someone says “science is just a belief system,” you can respond: no, science is a set of practices that produce operationally coherent results. That is not a matter of belief. It is a matter of demonstration. You can test whether a thermometer works. You cannot test whether it corresponds to ultimate reality. The first standard is achievable. The second is not.

The Bottom Line

Hasok Chang has written a philosophy of science for people who hate philosophy. It is a book that takes science seriously without taking philosophers too seriously. It replaces impossible standards with achievable ones. It replaces metaphysical debates with practical questions. It is a realism that does not require you to believe in things you cannot see. It only requires you to do things that work.

That is a philosophy worth taking seriously. Even if you hate philosophy.