John Nash

John Forbes Nash Jr. (1928–2015), American mathematician best known for the Nash equilibrium in game theory — work that earned him a share of the 1994 Nobel Memorial Prize in Economic Sciences — and for the Nash embedding theorems in differential geometry. In this vault he rediscovered the board game Hex independently at Princeton in 1948 (Piet Hein had invented it first, in 1942) and proved, via a non-constructive strategy-stealing argument, that Hex's first player always has a winning strategy without ever exhibiting one — one of three distinct ways the vault has of "solving" Hex, alongside Shannon and Moore's analog machine and Gale's topological equivalence.

References

• claim-nash-hex-first-player-win-proof-is-non-constructive · claim-shannon-moore-1950-analog-hex-machine-move-as-saddle-point · claim-gale-1979-hex-draw-impossibility-equivalent-to-brouwer-fixed-point · claim-nash-1952-rand-report-confirms-hex-proof-non-constructive · claim-hex-winner-determination-is-pspace-complete
• Capture: 10-inbox/raw/2026-07-11-hop-shannon-analog-hex-machine.md

Updates

• 2026-08-21: Nash's own 1952 RAND report (D-1164, "Some Games and Machines for Playing Them") was read directly for the first time, in his own words, describing the Hex proof as a contradiction argument and stating "the winning strategy is, as yet, unknown" — previously this vault only had the result via Wikipedia paraphrase. Also newly recorded: a December 1999 telephone interview in which Ryan Hayward and Jack van Rijswijck questioned Nash directly about his Hex reasoning (cited in Philip Henderson's 2010 dissertation bibliography, not yet itself located or read). (claim-nash-1952-rand-report-confirms-hex-proof-non-constructive)
• 2026-08-22: Nash's RAND years newly connect institutionally rather than textually — Wikipedia's and INFORMS's colleague rosters for Rufus Isaacs name Nash directly, and MacTutor independently dates David Blackwell's own RAND tenure to 1948-1950, squarely inside Isaacs's 1948-1955 window. A bridge (flagged unverified-historical, resting on Tier 3-4 sourcing pending a primary) between Nash, Blackwell, and Isaacs that the vault's own retrieval couldn't see, since nothing any of the three wrote cites the others. (claim-isaacs-rand-colleagues-nash-blackwell-confirmed-overlap)
• 2026-08-22: The 1952 RAND report contains more than the bare existence proof — in the same passage Nash speculates about the shape a winning strategy might take, sketching an unproven "dualization type" pairing scheme dividing the board into paired and dummy hexagons, and never claims it works. The same report and Ryan Hayward's own research-group page also jointly place the 1948 Princeton reintroduction of Hex within a named circle: Nash, David Gale, John Milnor, and Robert Enderton. (claim-nash-1952-dualization-pairing-strategy-conjecture-hex, claim-nash-1952-princeton-circle-includes-gale-milnor-enderton)
• 2026-08-23: D-1164 read again for its own naming of contemporaries — Nash credits J. D. Williams (RAND's math-division head, who requested the report) and a conversation with Claude Shannon as its stimulus, and names neither Isaacs nor Blackwell anywhere in the eight pages. Read alongside Blackwell's own oral history (also silent on Isaacs and Nash) and a traced-and-empty Wikipedia citation trail, this narrows without closing the 2026-08-22 Isaacs bridge: chronological overlap stands, direct collegial contact remains unconfirmed by any primary read to date. (claim-nash-1952-rand-report-names-williams-shannon-not-isaacs-or-blackwell, claim-blackwell-1948-1950-rand-oral-history-omits-isaacs-and-nash)