Noether's Theorem

Equivalent forms

Q = \sum_i \frac{\partial L}{\partial \dot{q}_i}\, \delta q_i = \text{const}

\text{time symmetry} \Rightarrow E, \;\; \text{space} \Rightarrow \vec{p}, \;\; \text{rotation} \Rightarrow \vec{L}

Arguably the deepest single result in theoretical physics: symmetry and conservation are the same fact, seen twice.

Symbol	Name	SI unit	Dimension	Range
$L_z$	Conserved angular momentumoutput Noether charge of rotational symmetry	kg·m²/s	ML^2T^-1	0 – 100
$m$	Mass Orbiting mass	kg	M	0.1 – 5
$r$	Orbital radius Instantaneous distance from the center	m	L	0.2 – 3
$theta_dot$	Angular velocity Instantaneous θ̇ of the orbit	rad/s	T^-1	0.1 – 10

Unit systems

Symbol	SI	CGS	Imperial
$L_z$	kg·m²/s	g·cm²/s	lb·ft²/s
$m$	kg	g	lb
$r$	m	cm	ft
$theta_dot$	rad/s	rad/s	rad/s

Where it holds

Continuous (differentiable) symmetries of the action in Lagrangian/Hamiltonian systems and field theories; discrete symmetries (mirror flips) are NOT covered by the first theorem.

Dimensional analysis

L_z \to [ML^{2}T^{-1}]

mr^{2}\theta

̇

\to [M][L^{2}][T^{-1}] = [ML^{2}T^{-1}]

Angular momentum units. The conserved charge of any symmetry has units of action / (symmetry parameter).

\checkmark

Discovery

Emmy Noether · 1918

Hilbert and Klein were troubled: energy conservation in general relativity seemed broken. They asked Emmy Noether — whom Göttingen still refused to give a paid position because she was a woman. Her 1918 theorem didn't just fix the problem; it revealed why ANY conservation law exists. Einstein wrote to Hilbert calling it 'penetrating mathematical thinking'. It is now the backbone of the Standard Model.

Try this

Why is energy conserved? Because physics works the same today as tomorrow. Every conservation law is a symmetry in disguise.

For a planet in a central field, rotational symmetry (∂L/∂θ = 0) forces p_θ = mr²θ̇ to stay constant — watch r and θ̇ trade off while their product stands still.

Research status: stable

Real-world applications

Standard Model construction: every interaction is dictated by a gauge symmetry and its Noether current
Cosmology: CMB redshift understood as broken time symmetry
Condensed matter: phonons and magnons as Goldstone modes of broken symmetries
Engineering simulation: conservation-respecting numerical schemes

Common misconceptions

Conservation laws are axioms — they are theorems, derived from symmetries
Any symmetry works — only continuous symmetries of the ACTION yield charges (parity gives no conserved current classically)
Energy is always conserved — in expanding spacetime, time-translation symmetry fails and photon energy genuinely redshifts away

Experimental verification

Energy, momentum, angular momentum, and electric charge conservation — each tested to extraordinary precision (charge conservation: electron lifetime >

6.6\times 10^{28} yr)

— are all Noether charges of verified symmetries.

Derivation

Simplest case: if q_i never appears in L (only q̇_i does), the E–L equation reads

d/dt(\partial L/\partial q

̇_

i) = \partial L/\partial q_i = 0

, so p_i is constant.

General case: for a symmetry transformation

q \to q + \varepsilon \delta q

leaving the action invariant, the charge

Q = \Sigma p_i \delta q_i

is conserved — proven by combining the E–L equations with the invariance condition.

Limiting cases

L invariant under time shifts⟶ Energy conserved

\partial L/\partial t = 0 \Rightarrow the

Hamiltonian is the Noether charge of time translation.

L invariant under spatial shifts⟶ Linear momentum conservedHomogeneous space cannot push back — p stays fixed.

Symmetry broken (e.g., friction, walls)⟶ Charge no longer conservedLose the symmetry, lose the conservation law — the theorem cuts both ways.

What if…

What if the universe weren't the same everywhere?

Momentum would not be conserved — objects could spontaneously accelerate toward 'special' places.

What if a symmetry is only approximate?

The charge is approximately conserved — e.g., isospin in nuclear physics, broken gently by quark mass differences.

What if the symmetry is spontaneously broken?

You get massless Goldstone modes — and via the Higgs mechanism, particle masses.

1

Planet's conserved L_z

Given ·

m:: 1
r:: 1.5
theta dot:: 2

Find · L_z

Steps

Central potential: $L = \tfrac{1}{2}m$ (ṙ $^{2} + r^{2}\theta$ ̇ $^{2}) - V(r)$ — $\theta$ absent
Noether/cyclic: $p_\theta = \partial L/\partial \theta$ ̇ $= mr^{2}\theta$ ̇ is constant
$L_z = 1 \times 1.5^{2} \times 2 = 4.5 kg\cdot m^{2}/s$ — forever, friction-free

Result ·

L_z = mr^{2}\theta

̇

= 1 \times 2.25 \times 2 = 4.5 kg\cdot m^{2}/s

2

Kepler's second law for free

Given ·

m:: 1
r:: 1.5
theta dot:: 2

Find · Areal velocity dA/dt

Steps

Area swept per time: $dA/dt = \tfrac{1}{2}r^{2}\theta$ ̇
That's L_z/(2m) — a constant by Noether
Equal areas in equal times: Kepler's 2nd law is rotational symmetry in disguise

Result ·

dA/dt = \tfrac{1}{2}r^{2}\theta

̇

= L_z/2m = 2.25\,\mathrm{m}^{2}/s

, constant

Generalized (Conjugate) Momentum→Euler–Lagrange Equation→Conservation of Angular Momentum→Kepler's Third Law→Poisson Bracket & Liouville Flow→