JOVANA
Explore Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
Back to the library
Quantum Physics 1948

Space-Time Approach to Non-Relativistic Quantum Mechanics

Richard P. Feynman

To find where a particle goes, add up every path it could take.

Choose your version
In depth · the introduction

To predict where a quantum particle lands, Feynman said: don't pick its path — add up every path it could possibly take.

The idea, unpacked

In everyday physics a ball follows one path. Feynman's startling claim is that a quantum particle, going from here to there, in effect explores every path at once — the straight one, the looping one, the absurd detour to the Moon and back.

Each path carries a tiny spinning clock hand — a phase. To find the chance of arrival, you add up all those clock hands as little arrows laid tip to tail, and square the length of the result. Most paths point every which way and cancel. The ones that survive are clustered around the single path of 'least action' — which is exactly the path ordinary physics predicts.

Where it came from

As a graduate student at Princeton in the early 1940s, Richard Feynman was hunting for a way to do quantum mechanics from the action — the quantity that, made smallest, gives classical motion. At a beer party, a visiting physicist mentioned that Dirac had written, a decade earlier, that the quantum amplitude was 'analogous to' exp(iL/ħ).

Feynman asked what 'analogous' meant, worked through it at the blackboard that evening, and found it was not merely analogous but proportional — the seed of his whole method. He built it into his 1942 doctoral thesis under John Wheeler, and finally published the polished version in 1948, after the war and the Manhattan Project.

Why it mattered

It gave physics a third way to do quantum mechanics, exactly equivalent to the two textbook versions but often far easier — and, because it is built on the action, the natural language for the quantum theory of fields. It also delivered a picture: the principle of least action, which had long seemed a strange piece of mathematical luck, turned out to be quantum interference in disguise. Out of this approach came Feynman diagrams and much of the modern toolkit of particle physics.

A million clock hands

Imagine every possible route from your door to a friend's house, and on each route a stopwatch whose hand spins as you walk. Lay the final hand positions end to end, as little arrows. Routes that are wildly different have hands pointing all over and largely cancel. But the bundle of routes near the shortest, smoothest one have nearly the same hand position and reinforce. The particle's chance of arriving is the length of the total arrow, squared. Slide ħ in the panel below and watch the arrows curl into a spiral.

A free particle goes from A to B. The top panel shows many candidate paths; the bottom panel adds up one little arrow per path, tip to tail, forming a spiral, with a single arrow across it for the total. Dragging the ħ slider toward zero curls the spiral's ends tight and leaves a straight central stretch — the ordinary, classical path that nature actually follows.

Where it sits

This is the third pillar of quantum mechanics, joining Heisenberg's matrices (1925) and Schrödinger's wave equation (1926) — both in this Library — and it reaches back to the principle of least action that runs from Maupertuis and Euler through Lagrange. Looking forward, it becomes the path integral of quantum field theory and underlies almost everything in modern theoretical physics, including the Higgs work also gathered here.

The original document
Original source text
Richard P. Feynman · Reviews of Modern Physics, Vol. 20, No. 2 · April 1948 · pp. 367–387
Abstract
Non-relativistic quantum mechanics is formulated here in a different way. It is, however, mathematically equivalent to the familiar formulation.
In quantum mechanics the probability of an event which can happen in several different ways is the absolute square of a sum of complex contributions, one from each alternative way.
The central postulate, stated in the body of the paper, is that each space-time path from one point to another contributes an amplitude of equal magnitude, and that the phase of that contribution is the classical action S along the path, in units of ħ. The total amplitude — Feynman's kernel K — is the sum of exp(iS/ħ) over all paths.
[ … ]
What it recovers
Feynman shows that this rule reproduces the Schrödinger equation when the kernel is built up over vanishingly small time-steps, and that ordinary classical mechanics — the principle of least action — emerges in the limit ħ → 0, where only paths near the one of stationary action add up in phase. He credits a 1933 remark of Dirac's, on the role of the Lagrangian in quantum mechanics, as the seed of the whole approach.
Cornell University, Ithaca, New York · 1948