We show that the concept of entropy is inescapable even in “pure” branches of mathematics such as algebra, topology and category theory. Specifically, we describe an entirely general categorical construction which, when given the real line and the standard simplices as inputs, produces Shannon entropy as the output. The construction involves operads and their algebras. We then show how this categorical line of thinking leads naturally to the entirely concrete and explicit characterization of information loss in Chapter 10.

Ordinary probabilities are real numbers, and ordinary entropy is a real number too. Building on ideas of Kontsevich, we develop an analogue of entropy in which both the probabilities and entropy itself are integers modulo a prime number p. While the formula for entropy mod p is quite unlike the formula for real entropy, we prove characterization theorems for entropy mod p and information loss mod p that are very closely analogous to the theorems over the real numbers, thus justifying the definition. We also establish a sense in which entropy mod p is the residue mod p of real entropy.

We give an overview of the whole book. We explain the problem of measuring diversity, summarizing the mathematical concepts with which it connects (including entropy and measures of size such as cardinality, volume and Euler characteristic). We indicate some of the branches of mathematics that will be involved (information theory, geometry, probability theory, abstract algebra) and the techniques that will be used (functional equations and a little category theory).

Any deterministic process loses information, and one can quantify the amount of information lost. Information loss is a generalization of entropy, and in some ways is a better-behaved quantity, being more functorial. We give a simple axiomatic characterization of information loss.

I distinguish between two versions of the black hole information-loss paradox. The first arises from apparent failure of unitarity on the spacetime of a completely evaporating black hole, which appears to be non-globally hyperbolic; this is the most commonly discussed version of the paradox in the foundational and semipopular literature, and the case for calling it `paradoxical' is less than compelling. But the second arises from a clash between a fully statistical-mechanical interpretation of black hole evaporation and the quantum-field-theoretic description used in derivations of the Hawking effect. This version of the paradox arises long before a black hole completely evaporates, seems to be the version that has played a central role in quantum gravity, and is genuinely paradoxical. After explicating the paradox, I discuss the implications of more recent work on AdS/CFT duality and on the `Firewall paradox', and conclude that the paradox is if anything now sharper. The article is written at a (relatively) introductory level and does not assume advanced knowledge of quantum gravity.