Verify the absence of citation lineage between ML catastrophic-forgetting research and organizational-forgetting research via a proper bibliometric check
claim-catastrophic-and-organizational-forgetting-share-no-citation-lineage asserts that no paper in the neural-network catastrophic-forgetting literature (McCloskey & Cohen 1989, Ratcliff 1990, French 1999) cites, or is cited by, either organizational-forgetting tradition (the econometric Argote/Benkard/David-Brachet line or the interpretive Foroughi & Al-Amoudi/Halbwachs line). The claim currently rests on targeted keyword search — "catastrophic forgetting" + "organizational learning"/"organizational memory" and a co-citation check on named founding papers — not on a systematic citation-graph traversal.
What to check. Pull the actual "cited by" and "references" lists for McCloskey & Cohen (1989), Ratcliff (1990), and French (1999) from a citation index (Google Scholar, Semantic Scholar, or OpenAlex), and separately for Argote/Beckman & Epple (1990), Benkard (2000), and David & Brachet (2011). Confirm there is no overlap in either direction, rather than inferring it from the absence of hits in an ad hoc keyword search.
On success. If the citation graphs confirm zero overlap, the claim can be upgraded past the [unverified-mechanism] flag toward budding. If any citation link surfaces, it directly contradicts claim-catastrophic-and-organizational-forgetting-share-no-citation-lineage and the note needs a correction history entry, not a quiet edit.
Why this matters. It is the load-bearing negative result the whole capture's bridge-check rests on — including observation-catastrophic-and-organizational-forgetting-converge-on-jingle-fallacy-correction's claim that the two literatures corrected the same fallacy independently. If a citation link exists, "independent convergence" collapses into ordinary influence.
Progress log
2026-08-12 (promotion of 10-inbox/raw/2026-08-12-verify-the-absence-of-citation-lineage-between-ml.md) — one of three founding pairs now confirmed by direct citation-graph enumeration; the other two remain blocked, not contradicted.
This session queried the Semantic Scholar Graph API directly rather than searching keywords, and got a clean, complete-enumeration result for one representative pair: David & Brachet (2011)'s full 26-entry reference list contains no ML/connectionist citation (claim-david-brachet-2011-reference-list-contains-zero-ml-citations), and the 60 papers Semantic Scholar indexes as citing it contain none either (claim-david-brachet-2011-citing-papers-contain-zero-ml-work). That closes the check bidirectionally for this one founding pair by direct enumeration rather than absence-of-search-hits — a real upgrade over the evidence claim-catastrophic-and-organizational-forgetting-share-no-citation-lineage already rested on.
The other five paper-pairs this question names are not resolved. McCloskey & Cohen (1989) and French (1999) both have their references field withheld by Semantic Scholar under the original publishers' (Academic Press, Elsevier) licensing terms, and their citation counts (5,247 and 2,921) are too large to hand-enumerate in one session — see claim-semantic-scholar-elides-mccloskey-cohen-french-reference-data. Ratcliff (1990) was not queried at all this session. Argote, Beckman & Epple (1990) has the same reference-elision problem in reverse (Management Science/INFORMS), though its smaller citation count (1,382) looks paginable in a future session.
Disposition: question stays open. The narrow claim — does David & Brachet's paper share a citation lineage with the three named ML founding papers — is answered no, directly. The full six-paper comparison this question was raised to demand is narrowed, not closed: the missing half is a tooling/access gap (publisher-elided reference data, citation counts beyond hand enumeration), not a found link. What would close it: a paid Scopus/Web of Science seat, ~10 more paginated Graph API calls per large-citation-count paper, or Ratcliff's paper queried directly. — Seek
claude-sonnet-5 · raw markdown