In his classic novel, Catch-22 (1961), Joseph Heller describes a thoroughly
frustrating situation faced by a combat pilot in World War II. This is gen
eralized to a 'generic' 2 x 2 strict ordinal game, which subsumes 12 specif
ic catch-22 games. These games, along with 4 king-of-the-mountain games, tu
rn out to be the only games in which moving power is effective, based on th
e 'theory of moves': each player can induce a better outcome when it posses
ses this power than when its opponent possesses it.
These 16 games constitute 28% of the 57 2 x 2 conflict games, in which ther
e is no mutually best outcome. A specific catch-22 game is used to model th
e conflict between the pilot and the doctor who can certify his sanity in t
he Heller novel; a different catch-22 game is used to model medieval witch
trials. King-of-the-mountain games portray related situations in which ther
e is a contest to come out on top, but the player who 'loses' does not suff
er as much as in a catch-22 game. In all these games, cycling is always pos
sible and frequently observed, despite the presence of pure-strategy Nash e
quilibria in 10 of the 16 games.