ADR-019: game-theory joins the topics

Context

Filing the morality-as-cooperation literature (ADR-018) brought in its evolutionary sources: Maynard Smith & Price on animal conflict, Axelrod & Hamilton on the evolution of cooperation, Nowak’s five rules, and Curry’s founding chapter, which builds each of its seven domains on a game. Two agents filing different parts of the batch each found the same gap and proposed the same word. Neither existing candidate fits:

• probabilistic-modeling is Bayesian inference and statistical models;
• mathematics is “mathematics read for itself”, with frameworks applied to other fields as a clause, which is how the evolutionary papers were tagged as a stand-in.

The record already held game-theoretic works with nowhere to put the fact: Lewis’s Convention, Hindriks & Guala’s rules in equilibrium, Bicchieri’s norms, and Debreu’s social equilibrium existence theorem.

Decision

Add game-theory to topics, beside mathematics, in no group. Its blurb: “strategic interaction and its equilibria — evolutionary stability, repeated games, bargaining, coordination, signalling and conventions — in biology, economics and philosophy”.

It is a method and a body of results, so like phenomenology (ADR-016) it is not expected to lead. A work carries it when its argument runs through a game, not merely when it mentions cooperation.

Tagged. No tag was removed.

• After the subject: LIT-505 (Maynard Smith & Price 1973), LIT-498 (Axelrod & Hamilton 1981), LIT-501 (Nowak 2006), LIT-473 (Lewis, Convention), LIT-481 (Hindriks & Guala), LIT-493 (Bicchieri) and LIT-067 (Debreu 1952).
• Last: LIT-503 (Curry 2016).

Hamilton 1964 is population genetics, not a game, and stays without it. Peters & Adamou’s ergodicity solution of cooperation (LIT-035) pools resources under multiplicative growth with no strategic choice, and stays without it too.

Alternatives considered

• Keep mathematics as the stand-in. It says the papers are mathematical, not what the mathematics is, and Lewis and Bicchieri are not mathematics papers at all.
• Widen probabilistic-modeling. Its blurb is samplers, priors and topic models; a reader browsing it for Bayesian methods would meet hawks and retaliators.
• A narrower word, evolutionary-game-theory. It would leave out Lewis’s conventions and Debreu, which are classical, and the record holds both kinds.