What if the equations meant exactly what they said?
In 1957, a graduate student named Hugh Everett asked a startlingly simple question. Quantum mechanics has two rules: smooth evolution, and the sudden collapse tacked on at measurement. The smooth rule is precise and universal; the collapse rule is vague and only fires when a fuzzy thing called 'measurement' occurs. So Everett proposed the boldest possible cleanup: *just delete the collapse rule.* Keep only the smooth evolution, and apply it to absolutely everything — atoms, cats, detectors, observers, the whole universe. No special role for measurement, no mysterious jump. This is the Many Worlds interpretation.
It sounds like a tidying-up — and mathematically it is the *simplest* version of the theory, since it throws a whole rule away. But the price is spectacular. If superpositions never collapse, then when a detector measures an electron that was in a superposition of 'here' and 'there', the detector itself goes into a superposition of 'saw here' and 'saw there'. And when you read the detector, *you* go into a superposition of 'I saw here' and 'I saw there'. The equations, taken literally, say the superposition spreads to engulf everything it touches — including you.
Branching, not splitting the sky
Here is the move that turns this from absurd into coherent. When you become entangled with the electron's two possibilities, the two versions of you stop being able to influence or even notice each other. They are no longer two ghosts haunting the same room; they have become two non-interacting *branches* of one gigantic wavefunction. In one branch, a definite you saw 'here'; in the other, an equally definite you saw 'there'. Each version feels perfectly ordinary, remembers a single past, and has no inkling of the other. Nothing was created or destroyed — the one universe simply has more structure than common sense assumed.
What physically severs the branches so they can no longer interfere? The modern answer is decoherence. A macroscopic detector has astronomically many particles, all getting tangled up with the air, the light, the heat of the room. This runaway entanglement with the environment scrambles the delicate relationships a superposition needs to show interference, so quickly and so thoroughly that the branches can never realistically meet again. Decoherence is the unsung hero of Many Worlds: it explains, using only the smooth rule, why we each experience one crisp branch and never feel the blur — without anyone having to invoke a collapse.
The hardest question: where do the odds come from?
Many Worlds wins enormous elegance — one rule, no special observers, no collapse — but it inherits a sharp puzzle in return. If *every* outcome happens, in what sense is one outcome more *probable* than another? The Born rule says an outcome with a bigger amplitude is more likely. But if both branches simply exist, with a copy of you in each, why should you have *expected* the high-amplitude branch? Both versions of you are equally real. Making honest sense of probability inside a theory where everything happens is the deepest open challenge for this picture, and the subject of ongoing, genuinely hard work.
A second, subtler worry is the preferred basis problem: the bare equations do not, by themselves, single out 'saw here' versus 'saw there' as the natural way to slice reality into branches — you could in principle slice it along weird combinations instead. Decoherence is widely thought to pick out the stable, classical-looking branches automatically, which is a large part of why the modern version of Many Worlds is taken seriously where the 1957 original was mostly ignored.
Living with abundance
It is worth being clear-eyed about the trade. Many Worlds buys the cleanest possible *mathematics* — a single, exact, universal equation with nothing bolted on — at the cost of the most lavish possible *ontology*, an unimaginably vast and ever-growing thicket of branches, almost all of them forever invisible to you. To its admirers this is a bargain: you should believe what your best equation says, even when it says something staggering, just as we eventually believed the Earth moves. To its critics, populating reality with countless unobservable copies of everything is too steep a price for tidiness, and stretches the word 'simple' past breaking. Both reactions are reasonable; which one you feel reveals a lot about what you want a physical theory to *be*.