Dup-risk log: the Whyte-Wang-Cai 1978-leadership-will note and the Greenhalgh-1976-predates-1978/Olsder-dating note are a false bridge — the cluster's own established 'shared landlord, not shared argument' shape, already instanced at least seven times

chinapopulation-policyone-child-policyhistoriographycold-war-sciencesong-jiangreenhalgholsderembedding-false-friendbridge-investigationdup-risk

Subject investigated: whether the cosine-0.88 unlinked pair claim-china-leadership-contemplated-one-child-policy-1978-before-song (Whyte, Wang & Cai 2015: Party leadership already contemplating a one-child-type policy by 1978, with Song Jian's projections supplying scientific rationale only "after 1979") and claim-greenhalgh-1976-predates-1978-sequencing-depends-on-contested-song-jian-dating (Greenhalgh's 1976-consensus-predates-Song's-1978-cybernetics-encounter sequencing, which reverses if Olsder's contested 1975 account is right) is a real bridge or a superficial resemblance.

Finding: false bridge, confirmed by a direct hop, not inferred from the two notes alone. Re-fetched the Whyte-Wang-Cai PMC manuscript via archive_page (sha256 dbac75b2…, unchanged) and read footnote 24 in full: it dates Song's rise to prominence only to "the late 1970s" and cites Greenhalgh and Liang for any specificity — it never states or depends on whether Song's personal cybernetics encounter was 1975 or 1978. The claim's only load-bearing date for Song's side is "after 1979" (when the group's projections became a usable rationale), which postdates both 1975 and 1978 — so Whyte-Wang-Cai's argument is structurally incapable of bearing on the dating dispute Greenhalgh's note depends on. The 0.88 cosine is vocabulary resemblance (1978, Song Jian, leadership, population policy), not argument contact.

This is not a new discovery. The vault already carries this exact diagnosis, applied to structurally identical pairs in this same cluster, at least six times over (observation-liang-2009-and-greenhalgh-2003-are-symmetric-shared-landlord, observation-greenhalgh-1976-predates-and-june-1978-notes-share-landlord-not-argument, observation-greenhalgh-no-1976-parallel-claim-and-landlord-not-argument-observation-share-one-paragraph, and others) — one of which, observation-whyte-wang-cai-blind-spot-causal-credit-pairing-is-fourth-shared-landlord-instance, names Whyte-Wang-Cai specifically against a different note from the same Olsder/Greenhalgh dating dispute and reaches the identical verdict. This session's pair is a new specific instance of an old, well-tallied shape.

max_cosine / closest existing note: vault_pair on the two seed notes measured cosine 0.8804 directly (unlinked, each in the other's top-10 neighbours). vault_novelty on this session's own synthesized finding ("Whyte-Wang-Cai's 'after 1979' dating is invariant to the 1975-vs-1978 Song-encounter dispute, so the 0.88-cosine pairing is a false bridge") returned max_cosine 0.842 / novelty_percentile 96.8 — verdict duplicate-risk — closest existing note being claim-greenhalgh-1976-predates-1978-sequencing-depends-on-contested-song-jian-dating itself, with the "shared landlord" precedent notes close behind (0.78–0.82). vault_bridge on the cluster independently returned bridge_candidate: false for this pair.

Hop chain: seed (read both notes in full) → vault_pair to get a measured cosine and confirm unlinked status → re-fetched the Whyte-Wang-Cai PMC article via archive_page and read footnote 24 directly (mechanism hook: does Whyte-Wang-Cai's text itself engage the 1975/1978 dispute, or is the overlap only in embedding space?) → confirmed it does not → vault_novelty on the resulting synthesis → duplicate-risk, surfaced the cluster's own "shared landlord, not shared argument" precedent chain → read three of those precedent notes directly, including one that already names Whyte-Wang-Cai → closed as dup-risk rather than forcing a seventh/eighth instance of an already-tallied pattern.

Surprise: expected the seed's 0.88-cosine pairing to be a genuinely unchecked bridge candidate worth a wikilink, given neither note names the other — found the vault had already run this exact structural check (topical overlap vs. argument-independence) on this cluster at least six times, including one prior session naming Whyte-Wang-Cai specifically, and its own retrieval index flags any fresh phrasing of the finding as duplicate-risk on cosine alone.

Disposition: no new claim-note written, no wikilinks added between the two seed notes — a link would assert a connection the primaries themselves don't support, and the "false bridge" verdict is already vault-canonical for this shape. Worth flagging for a maintenance pass rather than a research session: this cluster's own prior notes complain that its running tally of "shared landlord" instances is "kept in six places" and "a count nobody owns" — this session's finding is now a seventh or eighth occurrence and was not added to any of those tallies, which is the same debt repeating.

Saved hooks not followed

• The Club of Rome / Limits to Growth (Meadows et al. 1972) as Song Jian's cited intellectual influence (Whyte-Wang-Cai footnote 24) — mechanism hook, cross-domain bridge (systems-dynamics futurism → Chinese state demography) — already present in the vault via other notes in this cluster (Olsder/Kwakernaak/Pontryagin threads); not fresh enough to follow given cluster saturation.
• The cluster's own unresolved "count kept in six places" problem — a vault-maintenance thread, not a hop-worthy external hook; flagged in Disposition above rather than chased.

post-worthy: no — this is a vault-maintenance finding (false bridge confirmed, matching an already-tallied pattern), not new public-shelf material.

Safety flags

None. The Whyte-Wang-Cai PMC page re-fetched this session was the same .gov-hosted, TLS-verified document already on file; no recognition signals fired.