Thermodynamics
The second law applies to closed systems, and the earth continuously receives energy from the sun.
5 min readUpdated
The second law of thermodynamics is regularly cited against evolution: order does not arise from disorder, so life could not have become more complex without an outside intelligence.
The law does not say that. It concerns entropy in closed systems, and the Earth is not one.
What the laws state
| Law | Statement |
|---|---|
| Zeroth | If two systems are each in thermal equilibrium with a third, they are in equilibrium with each other |
| First | Energy is conserved; it changes form but is neither created nor destroyed |
| Second | The total entropy of an isolated system never decreases |
| Third | Entropy approaches a constant as temperature approaches absolute zero |
The second law is stated for isolated systems — those exchanging neither energy nor matter with their surroundings. The formal statement is that for such a system, dS ≥ 0.
Entropy is not a synonym for untidiness. In statistical mechanics it counts the number of microscopic arrangements consistent with a system's macroscopic state, and it is measured in joules per kelvin.
Why the Earth is not a closed system
The Earth receives roughly 174 petawatts of solar radiation continuously and radiates heat to space.
The exchange is not merely of energy but of entropy. Sunlight arrives as high-energy photons from a 5,800 K source; Earth radiates away many more low-energy infrared photons at about 255 K. Roughly twenty infrared photons leave for each visible photon absorbed, and each carries entropy.
The net effect is that Earth exports entropy to space at a much higher rate than it imports it. That export is what pays for local decreases in entropy at the surface.
Ordinary local decreases in entropy
Entropy decreasing locally is not unusual or rare.
| Process | Local entropy change | Paid for by |
|---|---|---|
| Water freezing | Decreases — molecules form an ordered lattice | Heat released to surroundings |
| A seed growing into a tree | Decreases | Sunlight, water, soil nutrients; heat exported |
| A snowflake forming | Decreases | Latent heat released to the air |
| Crystals precipitating from solution | Decreases | Heat and increased solvent entropy |
| A refrigerator cooling its interior | Decreases inside | Greater increase outside; electrical work |
| An embryo developing | Decreases | Maternal metabolism; heat exported |
If the objection held, none of these could happen. Every one is observed routinely, and the second law is not violated in any of them because the surroundings gain more entropy than the system loses.
The development of an embryo is the case that makes the point sharpest. A single cell becomes a highly organised organism in months, without any intelligence directing the process, in an open system supplied with energy.
The relevant quantity
For a system at constant temperature and pressure, spontaneity is governed by the Gibbs free energy:
ΔG = ΔH − TΔS
A process proceeds spontaneously when ΔG is negative. This can occur with a positive ΔS, or with a negative ΔS when the enthalpy change ΔH is sufficiently negative.
This is why chemistry produces ordered structures without violating anything. Lipids self-assemble into membranes because the entropy gain of released water molecules exceeds the entropy loss of the ordered lipids — the total still rises.
Common objections
"Adding raw energy increases disorder — an explosion doesn't build a house"
Made by Henry Morris and Duane Gish, this objection identifies a real requirement. Energy alone is not sufficient; a mechanism is needed to couple the energy to the process.
Living systems have that mechanism. Photosynthesis captures photons in specific molecular structures, and the energy is stored in chemical bonds and released through enzyme-mediated reactions. The coupling is the machinery of metabolism.
The objection therefore applies to the origin of the first such system rather than to evolution, which operates on populations that already possess metabolism. That question is abiogenesis and is genuinely unsolved.
Non-biological examples also exist. Convection cells, hurricanes, and crystal growth all produce organised structures from undirected energy input, with no machinery involved.
"There is a law of increasing genetic entropy"
John Sanford's argument in Genetic Entropy (2005), and it is a different claim from the thermodynamic one — it concerns the accumulation of deleterious mutations rather than physics.
The two are frequently conflated, and conflating them lends the genetic argument the authority of a physical law it does not have.
The genetic claim is addressed in Genetic Information. In brief, the predicted fitness decline is not observed, and Lenski's populations show fitness rising after 75,000 generations.
"The universe as a whole is a closed system, so entropy must increase overall"
Correct, and this is standard cosmology. The universe's total entropy does increase, and the eventual state — heat death — follows from it.
Nothing in evolution requires otherwise. Local decreases within a globally increasing total are exactly what the law permits.
The early universe was in a low-entropy state, which is itself a live question in physics known as the past hypothesis. That the universe started with low entropy is what allows any structure to form, and it is a separate question from whether biology violates the law.
"Evolution requires information to increase, which entropy forbids"
This merges thermodynamic entropy with Shannon information entropy. The two share a mathematical form but measure different things and are not interchangeable.
Shannon entropy measures uncertainty in a message and is dimensionless. Thermodynamic entropy has units of joules per kelvin and refers to microstates of a physical system.
Higher Shannon entropy means more information capacity, not less, so the argument reverses its own terms — random sequences carry maximal Shannon entropy.
What the physics shows
The second law applies to isolated systems. Earth continuously receives concentrated solar energy and radiates diffuse heat, exporting entropy at a higher rate than it imports it.
Local decreases in entropy occur routinely in freezing, crystallisation, and embryonic development, in each case paid for by a larger increase elsewhere.
The version of the objection with real content — that energy requires a coupling mechanism — concerns the origin of the first metabolising system rather than evolution, and points to abiogenesis, which remains unsolved.