Calculus of Variations

Noether's theorem

/ NUR-ter /

Why is energy conserved? Why is momentum conserved? Why angular momentum? For a long time these were separate empirical facts. Noether's theorem, proved by Emmy Noether in 1918, reveals they are all the same fact wearing different clothes: every continuous symmetry of a physical system's action corresponds to a conserved quantity, and vice versa. Symmetry and conservation are two sides of one coin. It is one of the deepest and most beautiful results in all of mathematical physics, and it lives entirely within the calculus of variations.

The precise statement: if the action S = integral of L dt is left unchanged (to first order) by a continuous family of transformations — sliding all coordinates, rotating them, shifting the clock — then there is a specific combination of the coordinates and momenta whose value stays constant along every solution of the Euler-Lagrange equations. The recipe even tells you what the conserved quantity is, built directly from the Lagrangian and the symmetry's generator. The three headline cases: symmetry under shifting time (the laws don't change tomorrow) gives conservation of energy; symmetry under shifting position in space (the laws are the same here and there) gives conservation of linear momentum; symmetry under rotation (the laws don't care which way you face) gives conservation of angular momentum.

Noether's theorem turned conservation laws from a bag of separate rules into a single organizing principle, and it scales all the way up: in field theory it links the symmetries of the Standard Model to conserved charges like electric charge, and the search for new symmetries is a search for new conservation laws. The Beltrami identity you met earlier is exactly the simplest instance — 'no explicit x' is time-translation symmetry, and the conserved Beltrami quantity is the energy. Two honest qualifications: the theorem requires continuous symmetries (a discrete symmetry like a mirror reflection does not yield a Noether conservation law in this way), and the symmetry must be a symmetry of the action, not merely of the equations of motion or of the trajectory.

If the Lagrangian L does not depend on a particular coordinate q (say an angle, so L has rotational symmetry about that axis), then the corresponding momentum p = dL/dq' is conserved: the Euler-Lagrange equation reduces to p' = 0. Rotational symmetry, conserved angular momentum — Noether in one line.

A coordinate missing from L (a cyclic coordinate) means its momentum is conserved — Noether's theorem in miniature.

The symmetry must be a continuous symmetry of the action itself. Discrete symmetries (like a mirror flip) and symmetries only of the equations of motion do not produce a Noether conserved quantity in this way.

Also called
Noether's first theoremsymmetry-conservation theorem诺特第一定理諾特第一定理