Should You Believe in Invisible Things Like Electrons?
Can You Believe in Things You Can’t See?
You are at a planetarium. The astronomer points to a glowing dot on the dome and says, “That star is made mostly of hydrogen gas.” You wonder: how can she possibly know that? She has never touched it, never scooped up a sample, never even left the building. She trusts a story told by telescopes and equations — a story about things nobody can see.
Most of us accept that science describes a hidden world. That’s the heart of scientific realism: the view that our best scientific theories are at least roughly true about unobservable entities like electrons, genes, and black holes. Realists say we should believe in those hidden things because the theories are so successful.
But there’s a puzzle. If we can’t see those things, and past scientists were sure they had it right only to be proven wrong later, why should we believe today’s invisible account? Some philosophers think the answer isn’t yes or no — it’s both, in a clever way.
The Miracle That Isn’t One
Imagine your friend claims she can predict exactly which domino will fall next. She writes down a sequence, and day after day her guesses match reality. If she was just lucky, it would be a miracle. A much better explanation is that she knows something real about the chain — its weight, its angles, its wobbles.
The philosopher Hilary Putnam (1926–2016) used a similar idea in 1975. He argued that the only way to make the stunning success of science not a miracle is to accept that its theories are true, or nearly true, even about invisible things. This is called the no-miracles argument. Science predicts eclipses, lands robots on comets, and builds phones that use invisible waves. If its stories about the hidden world were completely made up, that track record would be unbelievable. Realism seems like the honest explanation.
But here’s the snag: history is full of theories that were wildly successful in their day and turned out to be wrong about what things really are. The miracle argument needs a reply to that history.
The History Lesson That Makes Realists Worry
Larry Laudan (1941–2022) aimed a famous challenge at realism in 1981. He listed many theories from the past that were both mature and made novel predictions — predictions confirmed only after they were made — yet those theories were eventually tossed out. Their central invisible ingredients turned out to be fictions.
For example, early chemists believed a substance called caloric was a weightless fluid that flowed between hot and cold objects. Caloric theory successfully predicted how heat would move in many situations. Later, scientists agreed that caloric does not exist; heat is a form of energy. Similarly, the ether of light was thought to be an invisible elastic solid filling all space. Fresnel’s ether theory made stunning predictions about how light bends and spreads, but Maxwell’s later theory replaced it with an immaterial electromagnetic field vibrating in empty space. The ether vanished; the field remained.
Laudan argues that since past theories succeeded while being deeply wrong about what’s really there, we have no reason to think today’s best theories — about dark matter, say — won’t suffer the same fate. This is the pessimistic meta-induction: reflecting on the graveyard of abandoned theories makes us expect our current favorites will also be abandoned. So perhaps we should not fully believe in their invisible stuff.
The Invisible Skeleton: Worrall’s Big Idea
In 1989 John Worrall (born 1941) suggested a middle path, something that lets us keep the miracle without ignoring history. He noticed that when Fresnel’s ether theory gave way to Maxwell’s electromagnetism, something important survived. It wasn’t the imagined stuff — the solid ether vanished. It wasn’t even the full theoretical machinery. But the mathematical equations that described how light behaved stayed almost identical.
Worrall called this structural realism. The idea is: science does discover genuine truths about the unobservable world, but only truths about its structure. Structure means the mathematical patterns, the web of relationships between things, the rules that hold no matter what the hidden stuff is made of. We can believe the skeleton — the equations and correlations — even as we stay agnostic about the fleshy nature of the hidden things. Old theories were right about the skeleton, wrong about the stuffing.
Think of a clock: you might know exactly how its gears must be connected to show the right time, but you could be completely mistaken about whether the gears are made of brass, steel, or invisible cheese. The structure of the gear train would still be correct. For Worrall, that structural knowledge is real, and it’s what gets preserved across revolutions.
This move promises to answer both sides. The no-miracles worry disappears because theories keep getting the skeleton right — so their success isn’t magic. The history-of-science challenge is defanged because we never claimed to know the stuffing, only the scaffolding. We get “the best of both worlds.”
From Skeleton to the Whole Body? Two Kinds of Structuralists
Once Worrall opened the door, philosophers realized that “structure” could mean very different things. Two main camps formed.
Epistemic structural realism (ESR) stays closest to Worrall’s original picture. The French mathematician Henri Poincaré (1854–1912) had already suggested something like it decades earlier. On this view, the world really does have individual objects with inner natures — electrons, fields, whatever — but those natures are forever hidden from us. We can only know how those objects relate to one another. Science gives us the relational blueprint, never the color or texture of the bricks. ESR is a humble realism: the things are real, but their what-it-is-like is off-limits.
Ontic structural realism (OSR) goes further. It says the world just is structure all the way down. There are no objects that have relations; rather, the relations themselves are basic, and what we call “objects” are like knots in a net. One reason to take this seriously comes from quantum mechanics. Two electrons can be in an entangled state where they have all the same intrinsic properties — same mass, same charge, same indistinct position — yet the pair still behaves as two, not one. What makes them distinct seems to be only a relation (their spins are opposite), not any hidden “stuff” inside each electron. If individuality depends purely on relations, maybe individuals aren’t fundamental. The net comes before the knots.
Not everyone buys this. Critics object that relations without things doing the relating sounds like a grin without a cat. Others worry that if everything is just structure, we lose the difference between a real physical web and a purely mathematical one. Debate is lively, and it mirrors the way physicists themselves talk about the universe.
Why It Matters Now
You don’t need a planetarium to step into this tangle. Every time you check a weather forecast, you rely on a theory that treats the atmosphere as a vast mathematical structure — pressures, temperatures, flow equations — without ever touching an individual air molecule. Every GPS navigation uses Einstein’s relativity, which describes gravity as the shape of spacetime, not a force between invisible marbles. We act on structures constantly, and they work.
The structural realist’s question lives in those everyday acts. Are we justified in treating the mathematical patterns as real even if we’re not sure what spacetime is underneath? Or are we running on useful fictions, just like the caloric theorists who built good steam engines? The answer matters every time a government needs to trust a climate model, every time a doctor uses a brain scan to locate a memory, every time a kid like you stares at a star and asks: are we really touching the truth, or just a very good shadow?
Think about it
- If you had a perfect map of the London Underground that told you exactly where every train would be and when, but no one could ever see a train, would you believe trains exist? Why or why not?
- Scientists change their minds a lot. Does that make you trust today’s best ideas more, or less?
- Imagine two things that are identical in every test you can do, but swapping them changes something about the universe. Could they still be genuinely two separate things, and what makes them two?
