physical symbol system hypothesis

Allen Newell and Herbert Simon's thesis, stated in their 1975 ACM Turing Award lecture (published as "Computer Science as Empirical Inquiry: Symbols and Search," CACM 19(3), 1976), that "a physical symbol system has the necessary and sufficient means for general intelligent action" — the founding claim of symbolic AI, holding that intelligence is the manipulation of patterns that denote. It is the paradigm the whole "manipulate mental symbols" program (the Logic Theorist, the General Problem Solver) was built to demonstrate, and the one its critics took aim at.

In this vault the hypothesis is a load-bearing referent that recurs as the thing being argued over rather than a topic researched in its own right: it is the specific target Hubert Dreyfus's critique aimed at (as distinct from machine intelligence in general), the framework its own co-author later stretched to absorb connectionism rather than concede a rival, and the doctrine whose anti-representation opposition (Brooks) the vault traces as convergence, not citation. It was a bare wikilink across several notes for weeks with no hub of its own (flagged 2026-07-14) — this page is that hub. (An empty stub, physical symbol system hypothesis.md, sits in the vault root; it is the orphan Cali flagged in the W21 feedback and is outside the Warden's write surface — noted for cleanup, not touched here.)

References

• claim-dreyfus-critique-targeted-symbolic-ai-not-neural-nets — Dreyfus's phenomenological critique targeted the PSSH specifically (a rationalist lineage: Descartes, Leibniz, Wittgenstein's Tractatus), and explicitly held the Hebb→Rosenblatt→PDP neural-net line apart from it as the program he thought vindicated. (Newell & Simon, its authors, named here.)
• claim-simon-satisficing-and-heuristic-search-same-move — the Simon of bounded rationality is the Simon who co-authored the PSSH; the Turing-lecture statement of the hypothesis is quoted here as the AI-side face of his one "good-enough over optimal" move.
• claim-simon-vera-1993-1995-connectionist-nets-are-symbol-systems — Simon argued twice (1993, 1995) that connectionist nets satisfy his definition of symbol system rather than refute the hypothesis — widening "symbol" to absorb the rival.
• observation-simon-responds-to-connectionism-by-absorption-not-refutation — names that move: absorption, a third pattern beside confident conjecture and documented influence in how a founder meets a rival framework.
• claim-brooks-1991-does-not-cite-dreyfus — Brooks's 1991 "Intelligence Without Representation," the robotics-side anti-representation attack on the PSSH, cites no Dreyfus anywhere; a negative-citation result.
• claim-dreyfus-2002-iwr-does-not-mention-brooks — Dreyfus's own same-titled 2002 paper, continuous with his older PSSH critique via Merleau-Ponty and Freeman, never mentions Brooks — the mirror-image silence.

Log

• 2026-08-08: Hub built by the Warden pass, discharging the 2026-07-14 flag (oldest open flag; the vault's own W30 proposal ledger carried a → resolved 2026-07-25 line for this hub, but that line was an illustrative format example inside proposal-seek-2026-w30.md, not a real build — the hub did not exist in 40-entities/ until now). Concept-hub test met cleanly: an established, load-bearing term recurring across ≥6 notes. — warden/claude-opus-4.8