Week 10: What happens when you look?

A measurement is not a human blinking at reality. It is a physical interaction that spreads a quantum system's alternatives across an enormous environment until they can no longer interfere. That explains a great deal. It does not explain everything.

12 October 2026

Go back to the slits. Last week, putting a detector at one slit killed the interference pattern stone dead. That fact demands an explanation, and the explanation has nothing to do with anyone’s eyes.

Put a detector at one slit, and what you have added is a physical system — a photon bounced off the electron, a charge shifted, something — that responds differently depending on which slit the electron went through. That detector particle does not stay isolated. It goes on to interact with whatever is around it: the air, the casing of the apparatus, stray light, the enormous crowd of physical degrees of freedom that make up “the surroundings.” Each of those interactions leaves its own trace of which path was taken, spread out over more and more of the environment.

This is what a measurement actually is: a physical interaction that correlates a quantum system’s alternatives with something else, and that something else does not stay contained. The detector becomes entangled with the electron’s path. The environment becomes entangled with the detector. Within an absurdly short time, information about which slit the electron used is smeared across an enormous number of particles that were never part of the original experiment.

That matters, because interference needs the two path-alternatives to still be able to combine — to add their amplitudes together at the screen the way two ripples add on a pond. Once each alternative has dragged a different, enormous tail of environmental particles along behind it, recombining them would mean undoing every one of those interactions, in every one of those particles, in exactly the right order. Nobody can do that. For all practical purposes, it can never be done. The two alternatives stop being able to interfere not because a rule forbids it, but because reversing the spread through an environment that large is off the table. That loss of interference through entanglement with an enormous surrounding system is called decoherence, and it is why the pattern disappears the instant something in the environment carries which-path information, whether or not anyone ever looks at it.

You have seen this shape before. Week 4 asked how a crowd could have properties none of its members carry, and the answer was that enormous numbers of degrees of freedom turn individually erratic behaviour into something stable and reliable. The same move is happening here, aimed at a different target. There, an enormous number of molecules turned unpredictable individual motion into steady pressure. Here, an enormous number of environmental degrees of freedom turns a delicate quantum superposition into something that, in practice, behaves like one definite alternative or the other. The environment is the crowd again, and once more its sheer size is doing the work.

Decoherence explains a great deal. It is why you do not see interference between a cat that is alive and a cat that is dead, or between a pointer that reads one value and a pointer that reads another: any system large enough to be entangled with its surroundings the way a cat or a pointer inevitably is has already lost the ability to show interference between macroscopically different alternatives, long before anyone checks. The world looks classical at that scale because coherence has nowhere left to hide.

It does not explain everything. Decoherence tells you why interference between alternatives becomes practically inaccessible. It does not tell you why a measurement returns one specific outcome rather than leaving every entangled alternative equally real, nor does it tell you what, if anything, is special about the moment a definite result appears in a record somewhere. That question — usually called the measurement problem — is not solved by pointing at a large environment. Physicists genuinely disagree about how to finish this part of the story. Decoherence removes a huge amount of the mystery. It does not remove all of it, and the honest thing to do is say so rather than pretend the problem has quietly gone away.

So a lot has been explained. A quantum system does not need an observer’s gaze to behave differently; it needs to get entangled with enough of the world, and the world always obliges. But the deeper question underneath it — why definite outcomes happen at all, one at a time, rather than every possibility persisting side by side — is still open.

And there is a further question this raises just by looking at ordinary things around you: tables, walls, floors. They are made of atoms held together by quantum rules, in a world that only looks solid and definite because decoherence has already done its work at every scale you can perceive. That still leaves a different problem: why quantum matter forms stable, resistant structures at all.

What we explained#

Measurement is a physical process, not an act of observation: a quantum system’s alternatives become entangled with a measuring device, and that device becomes entangled with an enormous surrounding environment. Because reversing entanglement with something that large is not achievable in practice, interference between macroscopically distinct alternatives becomes inaccessible — a phenomenon called decoherence, and a reappearance of Week 4’s lesson that enormous numbers of degrees of freedom produce behaviour individual pieces do not have. Decoherence does not by itself resolve why a single definite outcome occurs; that part of the measurement problem remains open.

What became stranger#

Why does matter behave like matter?