Symmetries and Conservation Laws

A physicist watching two protons smash together in the Large Hadron Collider does not sit and solve the full quantum field theory of the collision — that is hopeless. Instead, before any dynamics, they reach for a much older and far cheaper tool: a bookkeeping sheet. Add up a handful of labelled numbers on the particles going in; add up the same numbers on whatever comes out. If the two columns disagree, the process is forbidden — it simply will not happen, no matter how much energy you pour in. If they agree, it is allowed (and then the dynamics decides how often). Those magic numbers are the conserved quantum numbers: electric charge, baryon number, lepton number, colour, and a few more.

Where do such iron-clad accounting rules come from? Not from tradition — from symmetry. This page teaches one skill, richly: conservation-law bookkeeping — how to look at a proposed reaction like n \to p + e^- + \bar\nu_e and decide, in seconds and with nothing but addition, whether Nature permits it. Along the way we will see the deep reason the rules exist at all (Noether's theorem), meet the symmetries that generate the very forces (U(1), SU(2), SU(3)), and learn which of our conserved numbers are sacred and which are only nearly sacred.

Noether's theorem: why conservation laws exist at all

In 1918 Emmy Noether proved one of the most beautiful results in all of physics. Stated for our purposes:

The dictionary is exact and it runs both ways: a continuous symmetry on one side, a conserved quantity on the other. This is what turns "charge is conserved" from a lucky empirical rule into a theorem. If you ever find a new conserved number in the lab, Noether promises there is a hidden symmetry behind it, waiting to be named.

Gauge symmetry: the symmetries that build the forces

The story gets bolder. Take Noether's global phase symmetry — the freedom to rotate the phase of a charged field by the same angle everywhere — and demand something stronger: that you be allowed to choose the phase independently at every point in spacetime. This is a local gauge symmetry. The equations only survive such freedom if you introduce a new field to compensate — and that compensating field turns out to be a force carrier. Symmetry does not just protect a conservation law; it conjures a force.

The Standard Model is three of these gauge symmetries stacked together:

In one line, the gauge group of the Standard Model is SU(3)_C \times SU(2)_L \times U(1)_Y. Every force you have ever felt is one of these symmetries made local. For our bookkeeping, the headline is simpler: each gauge symmetry hands us a conserved charge to tally.

The scorecard: the numbers we tally

For everyday reaction-checking, three additive quantum numbers do almost all the work. "Additive" means you just sum them over all the particles on a side.

The bookkeeping method, in a picture

Here is the whole skill on one page. Take free neutron decay, n \to p + e^- + \bar\nu_e (this is beta decay, the process that powers a huge fraction of natural radioactivity). Build a little ledger: one row for each conserved number, one column for each side of the arrow. Fill in the values, sum each side, and check the two sums match. Reveal the ledger row by row.

Notice the antineutrino pulling its weight: without it, the right side would carry lepton number L = +1 (from the electron) against the neutron's L = 0, and the decay would be forbidden. Nature must emit that \bar\nu_e (with L=-1) precisely to keep the lepton column balanced. The conservation law is not a description of the decay — it is a constraint that shapes it.

Worked examples: allowed or forbidden?

Example 1 — beta decay n \to p + e^- + \bar\nu_e (the one above, checked in full).

Every column balances, so beta decay is allowed — and indeed it is exactly what free neutrons do (lifetime about 15 minutes).

Example 2 — proton decay p \to e^+ + \gamma (a famous forbidden reaction). Energetically it looks fine: a proton is heavier than a positron, so there is spare energy. But check the numbers:

Two columns fail. Baryon number drops from 1 to 0 and lepton number from 0 to -1, so the decay is forbidden. This is why the proton is (as far as we can tell) essentially stable — its lifetime exceeds 10^{34} years — and why you, made of protons, do not slowly evaporate.

Example 3 — a pion-production reaction p + p \to p + p + \pi^0.

All balance, so this is allowed — and it is a standard way to make pions in an accelerator.

Example 4 — a sneaky lepton-flavour case \mu^- \to e^- + \gamma. Charge (-1 \to -1) and total lepton number (+1 \to +1) both balance — yet this decay has never been seen. The catch is per-flavour lepton number: the left has L_\mu = +1, L_e = 0; the right has L_\mu = 0, L_e = +1. Muon-number and electron-number are each violated, so in the Standard Model it is forbidden. Total L is not enough — you must keep the flavours separate.

Not quite as sacred, and that gap may be the reason you exist. Electric charge conservation is exactly protected: it is tied to the unbroken U(1) gauge symmetry, and a violation would wreck the whole theory. Baryon number B and lepton number L are different — they are accidental symmetries of the Standard Model, conserved by every interaction we have written down, but not guaranteed by any gauge principle.

Many theories beyond the Standard Model (Grand Unified Theories) predict tiny violations — including proton decay, which giant detectors hunt for year after year. And even the Standard Model itself violates B and L individually through subtle "sphaleron" effects at high temperature, while preserving the combination B - L. Cosmologists suspect such B-violation in the early universe is why there is more matter than antimatter — why anything is here at all. So treat B and L as rock-solid for practical reaction-checking, but know that they rest on softer ground than charge.

The trap: "If a reaction releases energy (the products are lighter than the reactants), then it can happen." False. Energy conservation is necessary but nowhere near sufficient. The forbidden decay p \to e^+ + \gamma is energetically downhill — the products are far lighter than the proton — yet it never occurs, because it violates baryon and lepton number. A reaction must satisfy every conservation law at once: energy and momentum and charge and baryon number and lepton number and colour. Fail any one and the process is dead, however much energy is on offer.

A second, sign-flipping trap: antiparticles carry the opposite baryon and lepton number to their particles. An antiproton has B = -1, a positron has L = -1, an antineutrino has L = -1. Forget the minus sign and a balanced ledger will look unbalanced (or vice versa). When you see a bar (as in \bar\nu_e or \bar p), flip the sign of its B and L before you add it in.

Yes — and the broken ones taught us the most. The discrete symmetries are the fragile ones. Parity P (mirror reflection) was assumed sacred until 1956, when the weak interaction was caught violating it maximally — the weak force is left-handed and knows its left from its right. Charge conjugation C (swapping particles for antiparticles) is likewise violated by the weak force. Even the combination CP is broken by a tiny amount, first seen in neutral kaons in 1964 — a subtle preference for matter over antimatter baked into the laws themselves.

The last refuge is CPT — charge, parity and time-reversal together — which is believed to be exact for any sensible quantum field theory. So the hierarchy runs: charge, energy, momentum exactly conserved; B and L conserved in practice; C, P, CP only approximate. Knowing which drawer a symmetry sits in is a real part of the physicist's craft.