A thought experiment · Quantum physics · 1935
Schrödinger's Cat
Alive and dead at once, until someone looks.
Put a cat in a sealed box. Inside is a single radioactive atom wired to a nasty contraption: if the atom decays, a hammer smashes a flask of poison and the cat dies. If it doesn't, the cat lives. Within the next hour the atom has a fifty-fifty chance of decaying. So far this is just a grim coin toss.
Here is the twist. In the quantum world, before anyone checks, the atom is not simply decayed or not. It is genuinely in both states at once, a "superposition". And because the cat's fate is chained to the atom, the cat is dragged along: not alive or dead, but somehow both alive and dead at the same time, until you open the box and look. The instant you look, you find one or the other, never both.
Erwin Schrödinger dreamed this up in 1935, and he was not claiming cats really do this. He was poking fun. Quantum rules work beautifully for tiny things like atoms, so he scaled one up to a cat to show how absurd it sounds when you blow it up to everyday size. The question he was really asking is the one nobody has fully answered: why don't big things behave like that?
The key idea is superposition. A quantum system isn't stuck in one state; it can sit in a blend of states at once, which we write as a sum like \(|\text{alive}\rangle + |\text{dead}\rangle\). The radioactive atom really is in a superposition of "decayed" and "not decayed", and that is not a statement about our ignorance. It is the actual physics, and interference experiments prove a superposition is more than a hidden coin.
Now follow the chain. The atom triggers a Geiger counter, which trips a hammer, which breaks the poison, which kills the cat. Each link gets entangled with the atom, so the superposition spreads up the whole ladder until the cat itself is described by
\[ \tfrac{1}{\sqrt{2}}\big(|\text{no decay}\rangle|\text{alive}\rangle + |\text{decay}\rangle|\text{dead}\rangle\big). \]
Open the box and you always find a definite cat, alive or dead, each with probability one-half (the square of the amplitude). The glaring problem: we never see a cat that is visibly both. Where exactly does the fuzzy quantum world hand over to the solid classical one? That gap is called the measurement problem, and the leading modern answer is decoherence: a big object is constantly "measured" by the air, light and warmth around it, which destroys the superposition almost instantly.
The atom, detector and cat form one entangled pure state, a cat state,
\[ |\Psi\rangle = \tfrac{1}{\sqrt{2}}\big(|0\rangle_{\!A}|D\rangle|{\uparrow}\rangle + |1\rangle_{\!A}|D'\rangle|{\downarrow}\rangle\big), \]
with the atom, detector pointer and cat (\({\uparrow}\) alive, \({\downarrow}\) dead) all correlated. The Schrödinger equation evolves this unitarily and never collapses it; yet experiment delivers one definite outcome with Born-rule probability \(|\langle\text{outcome}|\Psi\rangle|^2\). The clash between smooth unitary evolution and the abrupt, probabilistic outcome is the measurement problem.
Decoherence (Zeh, Zurek) is the part we understand. A macroscopic pointer couples to an enormous environment, \(|D\rangle|E_0\rangle \to |D\rangle|E_D\rangle\), and for distinct outcomes the environment states become orthogonal almost instantly, \(\langle E_D|E_{D'}\rangle \to 0\), on timescales as short as \(10^{-20}\,\text{s}\) for a cat-sized object. That suppresses any interference between "alive" and "dead", so the cat's reduced state looks like a classical either/or mixture. Decoherence explains why we never see superposed cats, but not which branch becomes real, so it does not, by itself, solve the measurement problem.
Where it stands. The interpretations diverge on the rest: Copenhagen posits collapse at measurement; Everett / many-worlds keeps the unitary evolution and lets both branches happen; objective-collapse models (GRW, Penrose) add real, testable collapse. Schrödinger's 1935 paper coined Verschränkung ("entanglement") and meant the cat as a reductio against the orthodoxy. The mathematics is settled and superpositions of ever-larger systems keep being realised in the lab; the interpretation remains open.
One atom's fate is chained to the cat; both branches live at once until you look. Watch it collapse to one, then reset to both. The chain entangles, so the box holds ½|alive⟩ + ½|dead⟩. Each outcome comes up with probability one-half. A cat state \(|\Psi\rangle\); the interference between branches decoheres away. Coherence decays: why we never see a superposed cat.