Schrödinger's cat

Schrödinger built the case in 1935 as a reductio: if the equation applies to a decaying atom, and the atom's fate is wired to a cat, the equation seems to apply to the cat. Bohr's answer was that the apparatus — detector, flask, animal — belongs to the classical side of the description, and amplitudes were never meant to be carried across. Critics grant that this works and press the obvious question: what physical fact decides which side of the cut a thing is on? Decoherence supplies most of an answer, as the size control here shows, but decoherence is ordinary quantum mechanics rather than a Copenhagen postulate.

Everett's move is to keep the equation and drop collapse, so the superposition simply extends to the detector, the cat, the box and the observer. Decoherence does the rest: the moment the atom's state spreads into a macroscopic system, the components stop being able to interfere, and each carries a complete, consistent record of its own history. Supporters argue the paradox dissolves — there is no leftover state of affairs to explain. Critics reply that the branching structure and the probabilities still need an account the bare formalism does not supply.

In de Broglie–Bohm mechanics the configuration of every particle is definite at all times, and the wave function guides it. When the atom's two components separate, the cat's particles end up inside one of them; the other becomes an empty packet, still present in the wave but no longer relevant to what the matter does. Once the components stop overlapping — which at this scale is immediate — the empty packet can never influence anything again. The price the theory pays openly is nonlocality in the guidance law, which is a problem for relativity rather than for cats.

GRW-type models give every particle a tiny chance per second of spontaneous localization, and entanglement makes the effect collective — a system of 10²⁶ particles suffers a hit essentially immediately. The quantum–classical boundary becomes a rate rather than a convention, which is exactly what the size control in this scene stands for. Because the dynamics differ from standard quantum mechanics, these models are constrained by experiment: interference with ever-heavier molecules keeps narrowing the allowed parameter space, and no collapse has been seen yet.

For a QBist the puzzle only bites if the wave function is taken as a description of what is in the box. Supporters add that the cat is itself a system that has experiences, so it was never a candidate for that kind of state assignment in the first place. Critics object that this makes any account of what actually happened inside unavailable in principle. QBists answer that quantum mechanics is a manual for acting in a world, and that the world's resistance to being described from nowhere is the discovery rather than the evasion.

Rovelli treats every interaction as establishing values between the systems involved, with no observer-independent ledger over the top. A cat is a system rich enough to interact with itself constantly, so facts accumulate inside the box long before the lid moves; what you lack is a relation to them, not a fact. Wigner's-friend scenarios — one observer measuring another — are where this account is pressure-tested, and how far fact-relativity can be pushed remains unsettled.

Griffiths and Omnès require a family of histories to be free of interference before probabilities may be assigned to it. For a system as large as a cat, coupling to the environment makes the intact-plus-decayed family inconsistent, while the definite-condition family passes easily — so the paradox becomes an error about which questions belong together. The standing criticism is that the rules permit many incompatible consistent families and never say which one describes what actually happened.