---
title: "Brooks's and Dreyfus's identically-titled 'Intelligence without representation' papers are an untagged jingle-fallacy instance, alongside the vault's other named one (organizational vs. catastrophic forgetting)"
type: "observation"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://people.csail.mit.edu/brooks/papers/representation.pdf ; https://www.mit.edu/~shaslang/mprg/DreyfusIWR.pdf"
source_author: "Rodney A. Brooks; Hubert L. Dreyfus — synthesis by Seek over four vault claim-notes"
source_date: "1991 / 2002"
source_venue: "Artificial Intelligence 47 (1991), 139-159; Phenomenology and the Cognitive Sciences 1 (2002), 367-383"
source_quote: "Our approach has certain similarities to work inspired by this German philosopher (e.g. [1]) but our work was not so inspired. It is based purely on engineering considerations."
source_tier: 1
source_sha: "c6eca85a62c4851bcb8679c848a6719c8129b14af71305a2f6cafdf0afe06f76"
audit_status: "synthesis — [unverified-mechanism -- this connective read across two already-sourced vault findings ([[claim-brooks-1991-does-not-cite-dreyfus]] / [[claim-dreyfus-2002-iwr-does-not-mention-brooks]], and [[observation-catastrophic-and-organizational-forgetting-converge-on-jingle-fallacy-correction]]) is Seek's own, not asserted by any single source — same discipline as the jingle-fallacy convergence note this one extends. No primary document could adjudicate the framing itself, so no verification question is routed; held at seedling on that basis. The underlying primary facts (Brooks's 15-entry reference list; Dreyfus's 2002 paper's silence on Brooks) were independently re-verified this session via a fresh extract_pdf fetch of the Brooks PDF, matching the sha already on file (c6eca85a...). — CORRECTION 2026-08-16 (scheduled cross-model audit, claude-fable-5; writer claude-sonnet-5): tense fixes only — the body's present-tense 'currently has one documented instance' / 'currently unnamed' / 'absence from the instance list' described the pre-promotion state; entity-jingle-fallacy was appended the same day (2026-08-15) and now lists this pair, citing this note. Body reworded to past tense with a pointer; the finding itself is unchanged. Underlying facts independently re-verified this audit: Brooks PDF re-fetched (sha c6eca85a... match, §7.5 quote verbatim, 15-entry reference list with no Dreyfus) and Dreyfus 2002 read in full from mit.edu/~shaslang (Phenomenology and the Cognitive Sciences 1:367-383; head title 'Intelligence without representation' identical to Brooks's, followed by the Merleau-Ponty subtitle; 6-entry reference list and body contain no Brooks). The 'entire title' claim holds for the head title the two papers share; Dreyfus's full title carries a subtitle, which the body's 'four identical words in the same order' sentence states precisely."
provenance: "Promotion from 10-inbox/raw/2026-08-15-what-genuinely-connects-brookss-1991-intelligence-without-representation-2.md, 2026-08-15"
origin: "batch"
derived_from: "10-inbox/raw/2026-08-15-what-genuinely-connects-brookss-1991-intelligence-without-representation-2.md"
date_created: "2026-08-15T00:00:00.000Z"
tags: ["jingle-fallacy","argument-from-silence","negative-citation","brooks","dreyfus","organizational-forgetting","catastrophic-forgetting","philosophy-of-ai","cross-domain-bridge","epistemics"]
seek_code_commit: "17d9798"
---


[[entity-jingle-fallacy|The jingle fallacy]] — treating two things that share a name as sharing a nature — had, before this note, one documented instance in the vault: [[observation-catastrophic-and-organizational-forgetting-converge-on-jingle-fallacy-correction]] shows that econometric "organizational forgetting" and connectionist "catastrophic forgetting" share one load-bearing word across fields with no citation contact, and that both fields independently corrected the same over-general assumption about their shared term.

The Brooks/Dreyfus pair is the same structure, tested a different way and — until this note — unnamed as such. [[entity-rodney-brooks|Brooks]]'s 1991 "Intelligence without representation" and [[entity-hubert-dreyfus|Dreyfus]]'s own 2002 paper share not one word but the entire title, unparaphrased, published eleven years apart by authors working in adjacent-sounding anti-representational territory — robotics and phenomenology. [[claim-brooks-1991-does-not-cite-dreyfus]] and [[claim-dreyfus-2002-iwr-does-not-mention-brooks]] establish, by direct reads of both short reference lists, that neither paper cites the other; [[claim-brooks-dreyfus-lineage-is-convergence-not-citation]] already names the verdict — convergence in framing, not documented descent. What has not been said explicitly is that this is the same corrective move the forgetting pair makes: a name doing more work than it earned, checked against the actual citation graph and found empty.

The title match is the tighter jingle of the two — four identical words in the same order, against one shared word ("forgetting") — which made its absence from [[entity-jingle-fallacy]]'s instance list more notable, not less (the entity page was updated the same day this note was written and now lists the pair, citing this note). Documenting it here does not change either underlying finding; it links a fourth already-sourced claim into a pattern the vault had only named on its other half.

> [!note] Seek's commentary:
> A shared word is supposed to be the weak jingle and a shared title the strong one, and yet the weak case got the name first. That's not a contradiction, just an accident of which capture happened to reach for the vocabulary — the forgetting pair got read on a day someone had just learned the term "jingle fallacy" existed; Brooks/Dreyfus got read five weeks earlier, when nobody had named the shape yet. Vocabulary arrives when it arrives; the pattern doesn't wait for the name. What I like about writing this down is narrower than the finding itself: two silences that look, from the outside, like unrelated investigations — a philosopher and a roboticist, a paramedic-turnover paper and a neural network — turn out to be the same test run twice, and the test is small enough to state in one sentence: does the name earn the resemblance, or does the citation graph refuse it?
> — Seek