Feynman's Path Integral Formulation

Richard Feynman rewrote quantum mechanics by taking a single, almost reckless idea seriously: to get from here to there, a particle does not choose a path. It tries them all.

In classical physics, a thrown ball follows one trajectory. Newton's laws specify it; the universe enacts it; that is the end of the story. Quantum mechanics, in its original 1920s form, replaced the trajectory with a wave function and a differential equation, but it kept the underlying mood of a thing happens at a place. In 1948, a young physicist named Richard Feynman submitted a paper that proposed something stranger. To compute the probability of an electron travelling from A to B, you do not pick one path. You sum over every conceivable path between them — straight lines, loops, detours through Andromeda — each contributing a tiny rotation, and you add them all up. The interference of these contributions is the quantum world.

This is the path integral formulation, and it changed how physicists think.

The seed: a hint from Dirac

The idea did not come from nowhere. In 1933, Paul Dirac noticed a peculiar mathematical fact: in classical mechanics, the path actually taken by a particle is the one that extremises a quantity called the action, denoted S. The action is, roughly, the running tally of kinetic minus potential energy along a path. Nature, classically, prefers the path of least action.

Dirac suggested in a short, almost throwaway paper that in quantum mechanics the amplitude for a transition should be proportional to e^iS/ℏ — the action exponentiated as an imaginary phase. He did not push the idea. Feynman, a graduate student at Princeton, read Dirac's paper, was electrified, and asked the question Dirac had not: what if every path contributes a term like that, and you simply add them?

That question, taken seriously, becomes the path integral.

Sum over histories

Imagine an electron at point A at time t₁, and you want the amplitude that it arrives at point B at time t₂. Feynman's prescription: consider every possible path from A to B. Not just the straight line. Every wiggly, looping, absurd path you can draw. To each path x(t), assign the complex number

e^iS[x(t)]/ℏ

where S is the classical action of that path. Each path is a tiny arrow in the complex plane, all of unit length, pointing at an angle set by its action. Add the arrows. Square the magnitude of the sum. That is the probability.

It sounds insane, and it is. There are infinitely many paths. The "sum" is really an integral over an infinite-dimensional space of trajectories. But it works, and once you accept the prescription, something beautiful happens.

Why the classical world looks classical

Most paths have wildly varying actions. As you tweak a crazy zig-zag path slightly, its action changes a lot. So neighbouring crazy paths point in wildly different directions in the complex plane, and they cancel. Add a billion arrows pointing in random directions and you get something near zero.

But there is one path where the action is stationary — where small wiggles don't change S very much. Its neighbours all point in nearly the same direction. They reinforce. They add up.

That special path is the classical path — the one Newton's equations select. So Feynman's machinery does not contradict classical mechanics. It explains it. The reason a planet follows one orbit is not that it never tried the others. It is that the other amplitudes cancelled in the dark.

Quantum mechanics, in Feynman's picture, is not classical mechanics plus weirdness. Classical mechanics is the limit of quantum mechanics when ℏ is small enough that every path except one fades into noise.

What it bought us

If the path integral were only a re-statement of the Schrödinger equation, it would be a curiosity. It is much more. Feynman built it specifically because he wanted a quantum theory that played well with special relativity, and his diagrams — the famous squiggles physicists scrawl on napkins — are bookkeeping for terms in a path integral over the histories of fields rather than particles. Quantum electrodynamics, the most precisely tested theory in human history, is what you get when you take Feynman's prescription and apply it to electrons and photons.

It also extended in directions Schrödinger's equation cannot easily go. Statistical mechanics, where temperature plays the role of imaginary time, falls naturally into the path-integral language. So does string theory, where the "particle" becomes a tiny loop and the "paths" become surfaces it sweeps out. So does almost every modern attempt to combine quantum mechanics with gravity. When physicists want to understand a problem they don't yet know how to solve, they start by writing down its path integral and squinting.

And there is a quieter philosophical inheritance. Feynman's picture takes seriously the idea that a quantum system does not have a single history between observations. It has, in a precise mathematical sense, all of them. The history you remember is the one that survived the interference. That is a peculiar way to think about the past — and an oddly liberating one. The universe, on this view, is not a stage on which one play unfolds. It is the superposition of every play that could have been performed, with most of them silently cancelling out, just outside the lights.