← Wonderland

Thermodynamics and the Arrow of Time

Drop an ice cube into warm water. It melts. It never spontaneously refreezes. From this banal fact comes one of the deepest puzzles in physics — why time runs the way it does.

The fundamental laws of physics are, almost without exception, time-symmetric. Newton's equations, Maxwell's, Schrödinger's, even general relativity — run them backward and they remain perfectly valid descriptions of reality. Film a swinging pendulum and play the reel in reverse: nothing looks wrong. A planet's orbit can be traced forward or backward through Kepler's laws with equal ease.

And yet at the scale we actually live in, time has a direction so obvious we never think about it. Eggs break but do not unbreak. Coffee cools but never spontaneously reheats. Memory runs from past to future and not the other way. Films played in reverse look absurd — water rising into a faucet, smoke gathering into a chimney. This is the arrow of time, and the puzzle is sharp: how can directionless microscopic laws give rise to such an emphatic directional macroscopic world?

The answer, worked out across the nineteenth century, is one of the most beautiful in science.

The Second Law and entropy

In 1850, Rudolf Clausius distilled the asymmetry into a principle. The First Law of thermodynamics says energy is conserved. The Second Law says that in any spontaneous process, a quantity he named entropy — symbol S — never decreases. It can stay the same in a perfectly reversible process; in any real one, it grows.

“The energy of the universe is constant; the entropy of the universe tends to a maximum.” — Clausius, 1865

Entropy began life as a bookkeeping device for steam engines. Clausius defined the change in entropy as heat exchanged divided by temperature, and noticed that in any real cycle this quantity always grew. Lord Kelvin, working in parallel, showed why it mattered: heat flows from hot to cold of its own accord, never the reverse. From this asymmetry he concluded that the universe is running down — that all of its useful energy is being relentlessly degraded into uniform warmth. He called the endpoint the heat death.

For half a century the second law felt almost magical. The microscopic laws made no distinction between past and future. The macroscopic world ran one way only. Nobody could say why.

Boltzmann's insight

The bridge was built by Ludwig Boltzmann in the 1870s. He proposed a reading of entropy so simple it sounds like a slogan, and so deep it is engraved on his tombstone:

S = k log W

Here W is the number of microscopic arrangements — the microstates — consistent with a given macroscopic description. Entropy is just (the logarithm of) how many ways the world could be, in detail, while looking the same in bulk.

low entropy: few microstates time high entropy: many microstates

Remove the divider and the gas spreads. Not because a force pushes it, but because there are vastly more ways to be spread out than to be bunched.

A gas confined to one half of a box has very few possible microstates. The same gas spread evenly throughout the box has astronomically more. Nothing in the laws of motion forbids the gas from spontaneously bunching into one corner again — it is just that the number of microstates corresponding to “bunched” is overwhelmingly smaller than the number corresponding to “spread out.” Toss N coins and ask how many configurations correspond to “all heads” versus “roughly half heads”: one versus an immense binomial. With Avogadro's number of particles, “overwhelmingly” understates it; “essentially never” is closer.

So entropy increases not because some metaphysical force pushes it upward, but because there are simply many more ways to be high-entropy than low-entropy. The second law is a statistical near-certainty, not a strict law. The arrow of time, on this reading, is the arrow from the improbable to the typical.

The arrow and the cosmos

This explanation is incomplete, and the gap is one of the most important open problems in physics. If high-entropy states are typical, why is the universe not already in one? Why is there structure, gradient, life? Why was yesterday lower in entropy than today?

The only answer that works is: because the day before yesterday was lower still, and the day before that lower still, all the way back to the Big Bang. The universe began in a state of extraordinarily low entropy. Everything we call the arrow of time — the cooling of coffee, the metabolism of cells, the formation of memories — is downstream of that one initial condition.

This is sometimes called the past hypothesis. It is not derivable from physics as we have it. It is a boundary condition imposed on the universe at its birth. Why the universe started in such an improbable state is not known. Roger Penrose has estimated the smallness of the initial entropy at one part in 1010123, a number so absurd it is hard to write. Whatever the explanation, our entire experience of time-flowing-forward is the consequence of that primordial improbability gradually being spent.

time → entropy S Big Bang extraordinarily low entropy now stars, life, structure heat death maximum entropy, no more change

The universe's history as a slow climb from an improbably low-entropy origin toward equilibrium.

So the arrow of time is not a feature of the laws of motion. It is a feature of the initial conditions. The microscopic world is reversible. The macroscopic world is directional only because the cosmos began in an extreme corner of its phase space, and has been slowly, statistically, drifting back to the middle ever since.

The end of change

When that drift completes — when every gradient is gone and entropy is maximal — there will be no more change. No process. No memory. No tomorrow. Two regions at the same temperature can exchange no useful energy. A universe in thermal equilibrium is a universe in which nothing can happen, because nothing can be distinguished. The arrow will have run its full length.

That is what Kelvin called the heat death. It is, if the universe persists long enough, the future we are all moving toward. Every cup of coffee that cools, every ice cube that melts, every breath you take is a tiny step in that direction — a small payment on the original cosmic improbability.

The arrow of time is not a mystical force. It is the universe spending the strangeness of its beginning. We notice the spending because we are made of it: every memory we form is itself a local decrease in entropy paid for by a larger increase elsewhere. To remember the past is to participate in the arrow. The same statistical asymmetry that makes coffee cool makes the future feel different from the past. There is no other arrow. There only ever was this one.


Further reading

  1. Clausius, R. (1865). Ueber verschiedene für die Anwendung bequeme Formen der Hauptgleichungen der mechanischen Wärmetheorie.
  2. Boltzmann, L. (1877). Über die Beziehung zwischen dem zweiten Hauptsatze der mechanischen Wärmetheorie und der Wahrscheinlichkeitsrechnung.
  3. Penrose, R. (1989). The Emperor's New Mind, chapter 7 — Cosmology and the Arrow of Time.
  4. Carroll, S. (2010). From Eternity to Here: The Quest for the Ultimate Theory of Time.
  5. Eddington, A. (1928). The Nature of the Physical World, lecture IV.