Double negation
ProvedAJGames.dual_dualFor every game , the dual of the dual of is itself:
This is a sanity check of the definition of duality.
import Mathlib import Definitions.Def_AJGames_Game
namespace AJGames /-- **Double negation** `A⊥⊥ = A` (Abramsky and Jagadeesan, *Games and Full Completeness for Multiplicative Linear Logic*, J. Symbolic Logic 59(2) (1994), §3.2: "clearly `A⊥⊥ = A`"). AJ axioms used: G1, G2, G3 only (the fields of `Game`); no tensor axiom. -/ theorem dual_dual (A : Game) : A.dual.dual = A := by sorry end AJGames
Read-back
What the Lean code literally says, in plain math · claude-opus-5-5
Setting. There are two players, and , and swaps them (, ). A game is a tuple where is a type of moves (in a fixed universe), labels each move with a player, and is a set of finite sequences of moves (the legal positions) such that: the empty sequence is in ; is closed under taking prefixes; and every position alternates, i.e. any two consecutive moves in it carry different labels. (The three conditions are proof fields; two games are equal exactly when their move type, labelling and position set agree.)
The dual has the same move type and the same set of positions, and the swapped labelling:
Statement. For every game (in any universe),
as an equality of games. Unfolding, the left side has move type , positions , and labelling ; so the statement says this game is the same object as itself. There are no hypotheses; it holds for every game, including the game with only the empty position and games with an empty move type.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.