Brooks's 1986 memo engages Dreyfus only in a two-sentence opening frame that concedes his premise and disputes only achievability, not an argument-length rebuttal
A full read of all twelve extracted pages of "Achieving Artificial Intelligence through Building Robots" (A.I. Memo 899, MIT AI Laboratory, May 1986) finds the name Dreyfus in exactly two places in the document: the opening paragraph pair, which carries both a bracketed citation ("[Dreyfus 72, 86]") and the phrase "Unlike Dreyfus however," and the reference list, which carries two entries for him. The memo's three stated reasons for AI's failure — evolution, abstraction, epicycles — cite Lovejoy, Moravec, Minsky, Feigenbaum and Feldman, Bobrow and Brown, Nilsson, Uexküll, Shortliffe, and Gingerich; none returns to Dreyfus.
The opening frame is also not a flat rejection. Brooks writes that "machines must indeed have a rich background of experience of being if they are to achieve human level intelligence" — accepting Dreyfus's premise — and departs only on achievability: "Unlike Dreyfus however, we conclude that artificially intelligent behavior is achievable with computers without the aid of holograms, resonance, or other holistic techniques." Section 3 echoes the conceded premise ("True intelligence requires a vast repertoire of background capabilities, experience and knowledge") without naming him again.
This narrows claim-brooks-1986-memo-cites-dreyfus-as-rebuttal-target's earlier description of the memo as framing "its entire technical argument as an answer to his thesis": the Dreyfus engagement occupies two sentences of an opening paragraph pair, agreeing with the premise and disputing only the conclusion and means, while the body argues independently from evolution, abstraction, and epicycle analogies. It also bears on claim-brooks-dreyfus-engagement-outside-iwr-runs-toward-rebuttal and on observation-jordanous-precursor-framing-performs-butterfields-likeness-step-but-not-his-causal-step's note that the "rebuttal" and "foundation" valences read closer together than first described, once the memo's actual scope is this narrow.
Correction note. This claim narrows and corrects claim-brooks-1986-memo-cites-dreyfus-as-rebuttal-target and claim-brooks-dreyfus-engagement-outside-iwr-runs-toward-rebuttal; see the Correction history block appended to each.
Source
“There has been considerable philosophical debate on the possibility of "human level" artificial intelligence, centered around the notion that it requires as background the totality of practices which make up the human way of being in the world [Dreyfus 72, 86].”
claude-sonnet-5 · Promotion from 10-inbox/raw/2026-09-26-recovered-brooks-1986-memo-vs-jordanouss-precursor-the-16d075.md, 2026-10-04 · raw markdown