Brooks 1986 memo vs Jordanous's 'precursor': the memo's Dreyfus engagement is one introduction paragraph that concedes Dreyfus's premise — a disagreement with the 2026-09-24 sibling's 'rebuttal' valence
Sibling exists. 10-inbox/raw/2026-09-24-bipartite-two-things-this-vault-knows-in-different.md already tested exactly this pair (claim-brooks-1986-memo-cites-dreyfus-as-rebuttal-target and observation-jordanous-precursor-framing-performs-butterfields-likeness-step-but-not-his-causal-step). It was promoted; the pair's verdict there is "partial mechanism, not a false friend": Jordanous performs Butterfield's likeness-labelling step and stops before the causal-line step. This capture does not re-investigate the Butterfield/Jordanous half; I did not re-fetch either of the sibling's two sources (Butterfield ELTE reproduction, Jordanous MDPI PDF), whose quotes the vault's 2026-09-25 audit already re-verified.
What I did: read the other endpoint, Brooks's 1986 memo, end to end (the vault's memo note carries no source_sha; its audit says the promotion "did not re-fetch"). Fetched with extract_pdf (method: ocr, 12 pages, TLS verified). No addressed-to-AI language, override language, authority claims or credential requests were present. No safety flags.
Verdict on the pair: the sibling's mechanism verdict stands as far as it goes, but the memo's actual text disagrees with one premise the sibling and the memo-note both lean on: that the memo's relation to Dreyfus is a straightforward "rebuttal" that contrasts with Jordanous's "foundation" wording. Recorded below as an open contradiction, not silently resolved.
Claim: In AIM-899, Dreyfus is named only in the Introduction (two mentions); the body and the argument sections never mention him
Claim type: what a document says (historical/definitional at the document level), read directly from a Tier 1 primary; the quoted phrases are exact. I read all twelve pages of the extracted text: the Introduction, sections 2-4 (evolution, abstraction, epicycles, bottom-up decomposition, insects, progress, pragmatics) and the reference list. The string "Dreyfus" occurs in the Introduction twice (the bracketed citation and "Unlike Dreyfus however"), and in the two reference entries. Nowhere else.
"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]."
The three arguments the memo gives for why AI has failed (evolution, abstraction, epicycles) cite Lovejoy, Moravec, Uexküll, Gingerich, Bobrow and Brown, Minsky, Nilsson, and others, but not Dreyfus.
- source_url: https://people.csail.mit.edu/brooks/papers/AIM-899.pdf
- source_sha: 7f700b7aca0b7d927f8a84d8eb907f05cb3ee5771762aa38712294836947e7f5
- source_title: "ACHIEVING ARTIFICIAL INTELLIGENCE THROUGH BUILDING ROBOTS"
- source_author: Rodney A. Brooks
- source_date: 1986-05
- source_venue: MIT Artificial Intelligence Laboratory, A. I. Memo 899
- source_quote: "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]"
- source_tier: 1
Tension with the vault: claim-brooks-1986-memo-cites-dreyfus-as-rebuttal-target says the memo "frames its entire technical argument as an answer to his thesis" and titles it "frames its entire argument as a rebuttal to him." The text supports "cites Dreyfus twice by name" (with the reference list, four occurrences in all) but not "entire argument." The Dreyfus framing occupies two sentences of an opening paragraph pair; the body is argued from evolution, abstraction, and epicycle analogies without returning to him.
Claim: The memo concedes Dreyfus's premise and disputes only achievability, so "rebuttal" is a partial description
Claim type: what a document says. The sentence that follows the Dreyfus citation:
"In this note we use a technical rather than philosophical argument that machines must indeed have a rich background of experience of being if they are to achieve human level intelligence. Unlike Dreyfus however, we conclude that artificially intelligent behavior is achievable with computers without the aid of holograms, resonance, or other holistic techniques."
The memo thus agrees with the Dreyfus-attributed premise ("must indeed have a rich background of experience of being") and disagrees on the conclusion and the means. The body echoes the premise at section 3 ("True intelligence requires a vast repertoire of background capabilities, experience and knowledge").
- source_url: https://people.csail.mit.edu/brooks/papers/AIM-899.pdf
- source_sha: 7f700b7aca0b7d927f8a84d8eb907f05cb3ee5771762aa38712294836947e7f5
- source_title: "ACHIEVING ARTIFICIAL INTELLIGENCE THROUGH BUILDING ROBOTS"
- source_author: Rodney A. Brooks
- source_date: 1986-05
- source_venue: MIT Artificial Intelligence Laboratory, A. I. Memo 899
- source_quote: "machines must indeed have a rich background of experience of being if they are to achieve human level intelligence. Unlike Dreyfus however, we conclude that artificially intelligent behavior is achievable with computers"
- source_tier: 1
Consequence for the bridge. The sibling's cross-check says the precursor relation "exists in the documents, but its valence differs from the 'foundation' wording." On the memo's text, the valence difference is smaller than that: Brooks takes Dreyfus's premise as the starting position and departs on the conclusion. That is compatible with Jordanous's claim that Brooks's views "had been given some foundation in philosophical AI literature" as a reception claim, so the two notes are closer than "rebuttal vs foundation" implies. Whether the memo shows influence remains a different question from either label: the memo shows Brooks reading and positioning himself against Dreyfus in 1986, which is a link of awareness, not evidence of derivation.
Claim: The Dreyfus works cited in the 1986 memo are not the work Jordanous cites as Dreyfus's precursor contribution
Claim type: bibliographic fact from two Tier 1/2 reference lists. Memo references (verbatim from the extracted text):
"[Dreyfus 72] "What Computers Can't Do: A critique of Artificial Reason", Hubert Dreyfus, Harper Row, 1972." / "[Dreyfus 86] "Why Computers May Never Think Like People", Hubert and Stuart Dreyfus, Technology Review, January 1986."
The vault's audit of the Jordanous note (in watch_flag, resolved 2026-09-25) records Jordanous's ref [10] as Dreyfus and Dreyfus, "Making a Mind versus Modelling the Brain" (1988), and [11] as What Computers Can't Do (Harper and Row, 1979). I have not re-read Jordanous's reference list myself; that half rests on the vault's audit [unverified by this session]. On those two lists the Dreyfus works do not coincide: the memo's [Dreyfus 86] (Technology Review) is not Jordanous's [10] (1988), and 1972 vs 1979 are different editions of the same book. So Jordanous's "precursor" is built on Dreyfus texts different from the ones Brooks cited, which weakens any reading of her framing as a report of Brooks's actual line of sources.
- source_url: https://people.csail.mit.edu/brooks/papers/AIM-899.pdf
- source_sha: 7f700b7aca0b7d927f8a84d8eb907f05cb3ee5771762aa38712294836947e7f5
- source_title: "ACHIEVING ARTIFICIAL INTELLIGENCE THROUGH BUILDING ROBOTS"
- source_author: Rodney A. Brooks
- source_date: 1986-05
- source_venue: MIT Artificial Intelligence Laboratory, A. I. Memo 899
- source_quote: "Why Computers May Never Think Like People", Hubert and Stuart Dreyfus, Technology Review, January 1986."
- source_tier: 1
Further leads
- Read Jordanous's reference list [10]/[11] first-hand to confirm the bibliographic mismatch above (currently rests on the vault's 2026-09-25 audit).
- Dreyfus and Dreyfus, "Why Computers May Never Think Like People," Technology Review, January 1986: the actual text Brooks cited; not located this session.
- Whether the vault's claim-brooks-dreyfus-engagement-outside-iwr-runs-toward-rebuttal synthesis leans on the "entire argument" wording; check before promotion of any correction.
- Related vault notes not re-examined: claim-butterfield-1931-text-warns-historian-forgets-descent-line-is-his-own-construction-and-takes-it-for-causation, observation-brooks-dreyfus-forgetting-bridge-is-method-level-not-mechanism-level, observation-brooks-dreyfus-title-echo-is-an-untagged-jingle-fallacy-instance.
Entity candidates
- Herbert Butterfield — person — foundational figure the Whig-history mechanism (the "precursor" test) rests on; already paged in the vault per the sibling, flag for linking.
- Hubert Dreyfus — person — the older figure against whom Brooks's memo positions itself and whom Jordanous calls a precursor; has a page (entity-hubert-dreyfus).
- Stuart Dreyfus — person — co-author of both the 1986 Technology Review piece and the 1988 paper.
- Rodney Brooks — person — author of AIM-899.
- Anna Jordanous — person — author of the "precursor" framing.
- C. Owen Lovejoy — person — cited in the memo's evolution argument (Lovejoy 81); an unacknowledged major source in the section where Dreyfus is absent.
- Hans Moravec — person — cited by the memo for mobility-first intelligence; an alternative ancestry for Brooks's argument.
- Jakob von Uexküll — person — source of Merkwelt in the memo's abstraction section.
- A.I. Memo 899 — term — the memo itself, "Achieving Artificial Intelligence through Building Robots."
- Awareness link vs derivation link — concept — the distinction the memo evidence makes between citing an author and being derived from him.