Where Does Complexity Come From? (Big Picture Ep. 3/5)

The universe’s march toward higher entropy doesn’t prevent complex structures from appearing—it often helps explain why they show up in the first place. The key distinction is that entropy and complexity measure different things: entropy tracks how many microscopic arrangements produce the same macroscopic state, while complexity tracks how difficult it is to describe the macroscopic state in detail. That difference resolves the apparent contradiction between the Second Law of Thermodynamics (increasing disorder overall) and the emergence of intricate systems like stars, life, and even cats.

Entropy can decrease locally without violating the Second Law, as long as the rest of the universe pays the cost. Cooling water to form ice is the classic example: order increases in the water, but the surrounding environment gains at least as much disorder. Yet even with that accounting, the deeper question remains—why do complex, information-rich patterns ever arise if the overall trend is toward greater disorder?

The transcript’s answer hinges on how complexity evolves during mixing. Consider a cup initially split into half coffee and half milk. At first, the setup has relatively low entropy because swapping coffee molecules with each other (or milk with each other) doesn’t change the macroscopic appearance much, but swapping coffee with milk would. The system is also simple to describe: milk sits on top, coffee on the bottom.

As the liquids mix, entropy rises because the macroscopic state becomes less sensitive to which specific molecules occupy which micro-positions. Swapping coffee and milk molecules becomes increasingly unnoticeable once the mixture is thoroughly blended. Complexity, however, behaves differently at first: describing the evolving, interwoven “tendrils” of coffee and milk requires increasingly detailed information. In other words, entropy increases while complexity can initially grow because the system’s macroscopic features become harder to summarize.

Eventually the mixture reaches equilibrium, where coffee and milk are thoroughly mixed. At that point, entropy is high and complexity decays again: the system looks like a uniform blend, so there’s little intricate structure left to specify. The general pattern is therefore cyclical: as entropy increases, complexity tends to rise early, peak, and then fall as equilibrium simplifies the macroscopic description.

This framework is then applied to the universe. The early universe is described as smooth and dense—low entropy and extremely simple. The far future is expected to be smooth again but very dilute—high entropy and simple once more. The “complex” era is the middle ground: medium entropy, where structures like stars, galaxies, mineral veins, swirling clouds, amino acids, proteins, and living beings can form. Just as the coffee-and-milk mixture becomes simpler at equilibrium, the transcript suggests that in the far future complicated structures will be simplified out of existence as the universe approaches a more uniform, high-entropy state.