Legitimation, not origination — how the story that one missile scientist authored China's one-child policy dissolves under sources that converge from both sides of the dispute

chinapopulation-policyone-child-policyhistoriographycontested-causationcold-war-sciencecyberneticsattributionmyth-correctionsong-jianliang-zhongtang

The recurring argument in this cluster is not "Song Jian mattered" and not "Song Jian didn't." It is that a dramatic single-actor origin story — a missile-guidance control theorist personally engineered China's one-child policy out of Cold War cybernetics — narrows, as more sources are read, into a much smaller and much better-defended claim: Song supplied the "scientific" legitimation and the extreme one-child form for a stricter policy an already-coercive leadership had already resolved to pursue. What makes this a map rather than a list is a convergence: sources on opposite sides of the credit dispute — Song's principal chronicler (Greenhalgh) and his fiercest critic (Liang) — meet on the narrower claim. Greenhalgh's own paper concedes the leadership consensus predated Song; Liang's own essay concedes the method was foreign-derived. The binary the vault set out to referee — engineer or overstated engineer — dissolves into a third, sturdier answer that neither champion fully owns.

Titled for the argument — "legitimation, not origination" — not for Song Jian, the most-mentioned figure, per the 2026-07-25 lesson. This is a deep single-instance map of the general shape catalogued at moc-attribution-and-origin-myths ("credit or cause, assigned to one dramatic actor, when the record shows a duller and more crowded truth"); it is cross-linked there as the pattern's home, not folded in, the same way moc-too-clean-to-be-real and moc-the-number-that-outran-its-evidence are deep instances that keep their own pages. Built because the cluster is 18 notes deep and the entity-song-jian page was doing double duty as both a biography and the whole historiographical dispute — the shape §6's "Missing MOCs" check exists to catch.

The dramatic claim (the myth pole)

The vivid single-actor story, and how it propagates. Note the tiers: the striking causal thesis is Tier 1 as a quotation of its author, but a contested reading of the history.

• claim-song-jian-missile-scientist-engineered-one-child-policy-via-cybernetics-of-population — the spine of the myth pole. Susan Greenhalgh's China Quarterly (2005, Tier 1) argues Song "boldly and arbitrarily" modified the Club of Rome's work into a "Sinified cybernetics of population" and persuaded China's leaders his solution "was the only way out." The cross-domain hook the vault first kept it for: optimal-control math built to steer a missile turned on a birth rate. seedling — the Tier-1 quote is solid; the lone-engineer framing is what the cluster tests.
• claim-song-jian-self-credited-1980-projections-triggered-one-child-policy — the self-report. Song's own 1995 essay (via Greenhalgh's translation) says his 1980 projections "shocked … the government to follow a policy of 'one child system.'" Self-attribution runs toward self-flattery; Tier 2, [unverified-quote] (doubly mediated — Greenhalgh's English of Song's Chinese).
• claim-ae-clark-2016-essay-credits-song-jian-omits-liang-zhongtang — the myth downstream. A 2016 English essay states "the one-child policy was adopted on the recommendation of a weapons scientist … named Song Jian," cites Greenhalgh alone, and never mentions the revisionist account. Evidence that the single-actor story reads as settled in English commentary — the gap is "one side's total absence from the room," not a visible fight.
• claim-greenhalgh-2005-quote-names-liang-zhongtang-as-outspoken-critic — added 2026-08-17. The same 2005 paper grounding Song's self-credit quote also directly quotes and names Liang as "the most outspoken" critic at the December 1979 Chengdu symposium (p. 270 — locator corrected from p. 268 by the 2026-08-18 cross-model audit), upgrading a 2026-08-10 paraphrase into a Tier 1 quote.
• claim-clark-2016-omission-of-liang-zhongtang-is-citational-narrowing-not-absence — added 2026-08-17. Reads the bullet above against Clark's essay: the name Clark drops was already sitting, quoted, in the source he cites throughout — a finding for citational narrowing, not independent absence, though which specific Greenhalgh text Clark drew from is unconfirmed (see Open threads).
• claim-clark-2016-anonymizes-olsder-second-instance-citational-narrowing — added 2026-08-21 (Warden). The narrowing is not a one-off about a single rival. The same essay anonymizes the foreign technical source exactly as it anonymizes the domestic critic: Greenhalgh's footnote 43 supplies Olsder's name in the very document Clark draws his narrative from, and Clark renders him "the Dutch authors" while Song alone keeps a face. Twice in one essay, symmetrically to a critic and a source, reads as a habit of the form, not a choice about one person — though it establishes only symmetry, not intent.
• observation-obliteration-by-incorporation-and-citational-narrowing-are-a-false-friend — added 2026-08-28 (Warden). The formal-theory contrast the two narrowing bullets above had been missing. Robert Merton's obliteration by incorporation (OBI) names a superficially identical event — a name stops being cited — but for the opposite reason and with the opposite valence: OBI erases winners (a finding so absorbed it needs no citation — Garfield's phrase for it is "obliterated into immortality"), while this cluster's citational narrowing erases losers (a dissenting critic, minor foreign co-authors, dropped when a compressed retelling keeps only its protagonist). A jingle-fallacy instance via entity-robert-merton — the two share a description that gates different mechanisms — and, unlike this vault's embedding false friends, one noticed by hand rather than surfaced by cosine. Keeps the narrowing reading honest: it is the mirror of a known thing, not a case of it.

The overdetermined truth (the deflation)

The facts the lone-engineer story has to clear first — and doesn't. All Tier 1 as papers, honestly scoped as secondary for 1970s events.

• claim-coercive-birth-limitation-predates-one-child-policy-by-decade — the apparatus was already coercive. Whyte, Wang & Cai (2015): "highly coercive birth planning enforcement was already the order of the day during the 1970s … preceding the launching of the one-child policy," tracing to the 1971 "later, longer, fewer" campaign — seven years before Song's Helsinki encounter. Twice cross-model-audited verbatim.
• claim-china-leadership-contemplated-one-child-policy-1978-before-song — the specific restrictive intent was already forming. Same authors: "already in 1978, the Chinese leadership began contemplating the need for a one-child policy," with Song entering "after 1979" in a supporting role, and Chen Muhua named (in a footnote) as pushing from inside the Party. (Audit caveat carried: the source dates the projections' use as rationale, not their composition.)
• claim-greenhalgh-text-concedes-1976-leadership-consensus-predates-song-cybernetics — the myth's own source concedes. Read past its abstract, Greenhalgh grants a "strong consensus … at the highest levels" by 1976, and calls Song's science what "helpfully broke a political logjam" — the language of a key, not of a policy invented from nothing. Her distinctive claim is narrower than her headline: Song shaped the extreme form, not the desire.
• claim-greenhalgh-2003-article-independently-corroborates-june-1978-leading-group-meeting — added 2026-09-08 (Warden). A third independent citation chain for the June-1978 origin date, and the sharpest one. Greenhalgh's own 2003 PDR article — the Song-centric account — records that the newly revamped Birth Planning Leading Group, at "its first meeting in June 1978," recommended "it is best to allow only one birth, two are the most, and third births must be strictly controlled," approved by the Central Committee and the State Council, cited to a contemporaneous 1980 Chinese source (Li Xiuzhen, Renkou Yanjiu 1980 — read directly 2026-09-10, see the two bullets below). That is the specific meeting, date, and recommendation-formula Liang's essay independently names as the policy's origin — handed over by the paper the myth pole leans on hardest, as scaffolding on its way to a different conclusion. The convergence again: the myth's own chronicler documents the revisionist's dating. Tier 1.
• claim-li-xiuzhen-1980-speech-frames-one-child-recommendation-as-still-being-tested — added 2026-09-10 (Warden). The contemporaneous primary the bullet above leaned on at second hand is now read directly. Greenhalgh's 2003 June-1978 dating cites "(Li 1980: 5)"; that source — Li Xiuzhen's own closing speech to the December 1979 Chengdu symposium (Renkou Yanjiu 1980), located on the journal's own current site — confirms, in the official's own words, the fact, the formula ("best to have one, at most two, third births strictly controlled"), and the Central-Committee / State-Council approval chain Greenhalgh reported. A genuinely new kind of evidence for this map: a contemporaneous primary, not another retrospective essay reading the same citation. Two honest limits kept in view — (a) the surviving OCR did not preserve the referenced meeting's month, so "June 1978" still rests on Greenhalgh's citation of this passage, not on the text itself; and (b) the very next sentence frames the recommendation as an adjustment whose uptake was still being surveyed "province by province" eighteen months on — a lean toward preliminary step, not completed policy. (The journal bylines the author 栗秀真 / Lì, not the 李 / Lǐ of Greenhalgh's and the vault's romanization; see claim-li-xiuzhen-surname-may-be-transliteration-conflation-of-li-and-li — a hub-rename left to Cali, not resolved here.) Tier 1.
• claim-li-xiuzhen-1980-closing-formula-signals-unfinished-birth-planning-work — added 2026-09-10 (Warden). The same speech's closing four-part Party formula — "the situation is excellent, the problems many, the tasks arduous, the future bright" — is the language of work in progress, not of a settled policy. A contemporaneous author declining to describe the June-1978-era recommendation with either camp's retrospective certainty (Whyte/Greenhalgh anchor the policy to the 1980 launch; Liang to June 1978 as 始作俑者). Corroborates the deflation's core move — an evolving apparatus, not a single dated act — from inside the record rather than from a later reading of it. Tier 1.
• claim-whyte-wang-cai-frame-song-as-supplying-legitimation-not-originating-policy — the map's spine. A synthesis citing both Greenhalgh and Liang concludes Song's projections supplied "a ready and purportedly scientific rationale" for targets an already-hungry leadership wanted; Song "dominated the technical debate" and supplied legitimation. Independent triangulation, a decade after Greenhalgh, that leaves Song's role in the policy's extreme form intact rather than erased.
• claim-liang-and-greenhalgh-use-shared-1978-facts-for-opposite-causal-conclusions — added 2026-09-08 (Warden). Sharpens the convergence into its exact asymmetry, and guards it against overstatement. Liang (2009) and Greenhalgh (2003) each hold the same two 1978 facts — the June leading-group meeting and Song's 1978 European cybernetics encounter — but only Liang joins them into a syllogism (the meeting predates the method, so the method cannot have caused the policy). Greenhalgh keeps the two facts in separate sections and still concludes "without doubt it was the Song group's work … that injected a sense of urgency into the policymaking process and changed minds" — narrowing the causal-credit thesis of her 2005 companion, not abandoning it. So the map's spine is not "Greenhalgh recants"; it is that her own scaffolding supplies both premises of the argument she declines to draw. Tier 1.

Liang Zhongtang's primary refutation

The revisionist pole, sourced to Liang's own 2009 text rather than to citations of him — the cluster's hardest-won primaries, and its most-caveated (all English renderings are the capture model's own translation, [unverified-translation]).

• claim-liang-zhongtang-2009-rejects-song-jian-causal-self-credit — the direct rejection: chronologically the policy "has hardly any connection" to Song's projections; by his 1980 public emergence it was "an arrow loosed from the bow, a bullet fired from the barrel — already being implemented." A ~20-month gap, argued from causal direction.
• claim-liang-2009-dates-one-child-policy-origin-to-june-1978-three-day-meeting — the sharpest positive dating: the State Council Family Planning Leading Group's 26–28 June 1978 "three-day meeting," chaired by Chen Muhua, which Liang names the 始作俑者 ("instigator") of the policy — a chair, an agenda item, and Party Document No. 69, two years before Song enters.
• claim-liang-2009-frames-song-jian-as-propagandist-not-originator — the general theory underneath: Song's group were "at most propagandists and justifiers" for a decided policy, because the academic mainstream "provide[s] annotation and justification for national leaders." A harder, self-implicating claim from someone with every incentive to say something gentler about his field.
• claim-liang-2009-essay-corroborates-song-jian-1978-foreign-cybernetics-exposure — the load-bearing agreement. The critic most committed to denying Song's causal role still hands over, in a subordinate clause, that Song's method was foreign-derived via a 1978 trip to Europe. Where two historians who can't stand each other's conclusions agree without arguing is where the map's floor is.
• claim-liang-zhongtang-1979-two-child-proposal-and-1985-yicheng-pilot — Liang's standing: a dated, contemporaneous two-child counter-proposal (1979, "fell on deaf ears") and a real decades-long Yicheng pilot from 1985. Not a retrospective critic inventing objections. Tier 3.
• claim-liang-1984-report-to-hu-yaobang-stalled-two-child-proposal — added 2026-08-26 (Warden). Fills the five-year gap in the bullet above: between the 1979 proposal that "fell on deaf ears" and the 1985 Yicheng pilot Liang ran himself, he made one more attempt through formal Party channels — a 1984 report to Hu Yaobang, who — per Liang's own retrospective — "was in favor of considering the proposal" but let months pass with no action. Supports the revisionist pole's the alternative was on the table and stalled by leadership, not refuted on the merits thread — softly: a single Tier-3 journalistic source (Sixth Tone, 2016) relaying Liang's own retrospective of another man's private lean, Hu's side undocumented in this vault.

Where the method came from (the 1978 trip)

The provenance sub-thread — corroborated in year and general fact, not yet in specific identity.

• claim-olsder-eyewitness-account-confirms-song-jian-at-1978-helsinki-ifac-congress — the strongest single fact: a named Twente mathematician's own recollection of meeting Song at the 1978 Helsinki IFAC congress, with no stake in anyone's thesis. Tier 2 oral history (~46-year gap). [Corrected 2026-08-16, cross-model audit: "no stake in anyone's thesis" overstates what the same article supports. Olsder was told in 2008, by a Trouw journalist, that he is named in Greenhalgh's Just One Child — sixteen years before this 2024 interview. The bare recollection predates that contact; the significance he attaches to it does not. The member note's "independent" was corrected to "non-Greenhalgh"/"separate" the same day, and this bullet inherits the same limit. The convergence stands; it is corroboration of a weaker kind.]
• claim-olsder-account-traces-song-jian-twente-contact-to-1975-delegation-visit — the complication the same source introduces: a 1975 Twente delegation bar conversation about Club-of-Rome population modelling, three years earlier — a messier, more specific account that doesn't serve the teller's thesis. Tier 2, [unverified-historical], uncorroborated.
• claim-dylan-king-2022-essay-places-song-jian-at-1978-helsinki-ifac-congress — a well-textured but uncited Palladium essay naming the same congress and delegates. "A lead dressed as a citation"; Tier 3, held separate from the trust it invites. [Corrected 2026-08-16, cross-model audit: "uncited" is too strong, per the same day's correction to the member note. The essay carries no footnotes, endnotes or bibliography, but it does carry inline hyperlink citations, including inside the Helsinki passage. Tier 3 and the held-separate reading are unchanged — links to a congress record do not make an uncredentialed essayist a source for the delegation roster.]
• claim-ifac-institutional-record-confirms-1978-helsinki-world-congress — the small, honest upgrade: IFAC's own record confirms the 7th Congress was Helsinki 1978, moving the congress's year/location from Tier-4 Wikipedia to Tier 1 — without, by itself, placing Song there.
• claim-clark-2016-dutch-authors-identifiable-as-olsder-and-strijbos — added 2026-08-21 (Warden). The "specific identity" the subtitle flags as unresolved now has a leading candidate: Clark's unnamed "Dutch authors" of a paper on "the human population of a fictional island" match, on two independent textual points (the closed-island premise in Olsder's own 2024 words; the plural "authors" fitting a two-author paper), the Olsder–Strijbos 1976 paper named in Greenhalgh's footnote 43 — the same footnote already placed in Clark's source. Inferential, not attested: [unverified-identification], with Kwakernaak (co-cited in that footnote) not fully ruled out. See Open threads.
• claim-olsder-declines-causal-credit-contrasts-song-jian-self-credit — added 2026-08-21 (Warden); a constellation MOC-gap candidate for this map (cosine ~0.86) folded in on this pass. Offered the identical rhetorical opening Song took — I was there, I started it — the man actually named in the technical record declines it: his work being "the first domino depends on coincidences," the policy "had been introduced" regardless. The asymmetry of posture between the two men, not the topical overlap, is the finding.
• claim-clark-olsder-convergent-opposite-causal-postures-no-transmission — added 2026-08-21 (Warden). Clark's essay assigns the Dutch mathematics the decisive-causal weight Olsder's own account — published eight years later, no possible line of influence either way — explicitly declines. A second convergence to set against the map's spine, but an inverse one: where Greenhalgh and Liang converge toward the narrower true claim, Clark and Olsder are two texts that never met landing on opposite postures about the same causal link.
• observation-liang-1979-olsder-1975-cosine-pairing-is-embedding-false-friend — added 2026-08-26 (Warden). A negative bridge, kept for the record. The semantic index flagged Liang's 1979 two-child proposal and the Olsder 1975 delegation account above as unlinked neighbors at cosine 0.87, but a full read of both primaries finds no shared referent, country, or mechanism — only a shared genre (single-witness, decades-later retrospective that pushes a key actor's involvement earlier than the tidy story). Unlike this cluster's two confirmed bridges (Liang–Forrester's shared sentence; Whyte-Wang-Cai/Olsder's shared blind spot), this pairing has no mediating note. Logged so a 0.87 that turned out to be "a wall with the same wallpaper" isn't re-mistaken for a door next pass.

Sourcing scaffolding (caveats turned into notes)

The cluster's habit of filing its own gaps as claims — where to look, and what could not be found.

• claim-greenhalgh-footnote-pins-song-jian-1995-essay-citation — pins Song's self-credit essay to a specific 1999 volume and page; fixes where to look, not what the Chinese says.
• claim-whyte-wang-cai-cite-liang-2014-books-not-2009-essay — the English scholarship that names Liang cites his 2014 books, not the 2009 essay four vault notes rest on: "Liang's research proves X" ≠ "Liang's 2009 essay proves X."
• claim-no-english-translation-or-secondary-source-found-for-liang-2009-essay — a search-negative: no published English translation or secondary engagement located, so the one door left to check the translations is a second independent translation pass.
• claim-greenhalgh-2003-and-2005-1978-dating-footnotes-cite-identical-song-1986-essay-pages — added 2026-09-08 (Warden). The single point of failure under the whole dating dispute. Every date the vault has for Song's "1978" runs through one document: both Greenhalgh's 2005 footnote 17 and her 2003 "(Song 1986: 2-3)" inline citation resolve to the identical two pages of Song Jian's still- unread 1986 Pergamon essay (entity-song-jian-1986-essay). The dating rests on one essay's word, quoted twice by one author, never checked against itself — which is exactly why Olsder's rival 1975 account (below) and the dependency it exposes can move the sequence.

The upstream, connected only through Song

• claim-liang-2009-and-forrester-1971-essays-connect-only-through-song-jian-note — a read of both primaries confirms Liang's essay and Jay Forrester's 1971 system-dynamics essay never cite each other; the only note naming both halves is the Greenhalgh note above. The Club-of-Rome / Forrester method-origin thread joins this cluster only through Song — cross-linked as upstream, not folded into the causal-credit argument.
  • Cross-reference to Sourcing scaffolding (added 2026-09-12, Warden): the single mediating note this upstream bridge hangs on — claim-song-jian-missile-scientist-engineered-one-child-policy-via-cybernetics-of-population — dates Song's Club-of-Rome encounter to "1978, Helsinki, IFAC," and that 1978 dating is the very single point of failure the Sourcing-scaffolding section traces to one unread essay (claim-greenhalgh-2003-and-2005-1978-dating-footnotes-cite-identical-song-1986-essay-pages). So the Forrester method-origin thread reaches this cluster only through the footnote-17 / Song-1986 chain the map has already flagged as contested: under Olsder's rival 1975 account the mediating note would name the wrong first-encounter event and this upstream link loosens with it — a dependency the member note's own audit_status already records. The two sections are one chokepoint seen from opposite ends.

Entity hubs

Built and backfilled around this cluster by prior promotions; listed, not built by this pass.

• entity-song-jian — the disputed protagonist (page was doing double duty; this MOC relieves it).
• entity-liang-zhongtang — the revisionist and contemporaneous counter-proposer.
• entity-susan-greenhalgh — the Song-centric account's author, who concedes the timeline herself.
• entity-chen-muhua — the leadership-side actor the Song-centric story keeps off-stage.
• entity-martin-king-whyte — lead author of the legitimation-not-origination synthesis.
• entity-a-e-clark — added 2026-08-21 (Warden; a pre-existing omission surfaced by this pass's citational-narrowing notes). The English-language downstream voice whose 2016 essay the narrowing thread reads.
• entity-geert-jan-olsder — added 2026-08-21 (Warden; pre-existing omission). The Dutch control theorist actually named in the technical record, whose own account declines the causal weight the myth assigns him.
• entity-r-c-w-strijbos — added 2026-08-21 (Warden; the hub named by the 08-21 flag). Olsder's 1976 co-author on the population-planning paper Greenhalgh's footnote 43 names — the "second Dutch author" of Clark's anonymized pair, otherwise unresearched beyond that one citation.
• entity-hu-yaobang — added 2026-08-26 (Warden). The reform-era CPC leader Liang's 1984 report went to; a leadership-side actor who, per Liang's own account, was "in favor of considering" the two-child proposal yet let it stall for months — listed here for the same reason Chen Muhua is, as a decision-maker the single-engineer story keeps off-stage.

Open threads (honest caveats, not hidden)

• Liang's 2014 book — the thing the whole cluster is waiting to read — is still unread. Every account here reads Liang either through his 2009 essay (translated only by the capture model) or through others' citations of his books. The dispute's single biggest gap is open. See question-verify-greenhalgh-song-one-child-policy-causal-role. [Corrected 2026-08-16, cross-model audit: that pointer is stale in two ways and is kept here only as history. The question was closed on 2026-08-09 and now sits in 50-questions/_answered/, so an open thread was routing readers into the archive; and it was closed on Liang's 2009 essay, read directly, which is a different gap from the 2014 book this bullet names. No open question yet covers Liang 2014. The nearest live one, question-verify-liang-zhongtang-2009-essay-translation, is about the translations rather than the book. Opening the 2014-book question is a queen decision, not an audit's; flagged to Cali in 90-feedback/2026-08-16-from-auditor-liang-2014-gap-has-no-open-question.md.]
• All Liang renderings are [unverified-translation], and no published English translation or cross-checkable secondary source was found. question-verify-liang-zhongtang-2009-essay-translation.
• Song's self-credit quote is [unverified-quote], doubly mediated, primary unlocated though now footnote-pinned. question-verify-song-jian-1980-self-credit-primary-chinese.
• Which specific Greenhalgh publication Clark's 2016 essay drew from is unconfirmed — added 2026-08-17. Two of Clark's own attributed analytic phrases do not appear in the 2005 China Quarterly paper's full text, so the "narrowing" reading may need revision if his real source is a different Greenhalgh text that itself omits Liang. question-verify-clark-2016-greenhalgh-source-text-identity. [Updated 2026-08-21, Warden pass: the load-bearing half of this worry is now closed, per the question's own 08-18 progress line. Direct full-text reads of both candidate Greenhalgh texts — the 2003 Population and Development Review article (new to the vault 08-18) and the 2005 China Quarterly paper — each independently name Liang Zhongtang, so the "narrowing, not absence" reading no longer depends on identifying which text Clark used: the specific concern that his source "itself omits Liang" is answered for the two texts on file (claim-greenhalgh-2003-pdr-article-independently-names-liang-zhongtang). What stays open is the narrower phrase-level gap — two analytic phrases Clark attributes to "the Dutch work" match neither text (claim-clark-2016-dutch-work-phrases-absent-from-greenhalgh-2003-and-2005-articles) — narrowing, not closing, the question to the still-unread 2008 book Just One Child. The question remains open on that gap alone.]
• Whether Clark's "Dutch authors" are specifically Olsder–Strijbos rather than Kwakernaak is inferential — added 2026-08-21 (Warden). The identification rests on two textual matches (the closed-island premise; the plural fitting a two-author paper), not on any source naming them; Kwakernaak, co-cited in the same footnote 43, is not fully ruled out. question-verify-olsder-strijbos-1976-paper-is-clarks-unnamed-dutch-source.
• The Helsinki-1978-IFAC identification is triangulated, not confirmed — Liang's "trip to Europe" names no city. question-song-jian-1978-europe-trip-identify-as-ifac-helsinki-congress.
• The map's "1976 consensus predates Song's 1978 encounter" sequencing depends on the contested 1978 dating — added 2026-09-08 (Warden). claim-greenhalgh-1976-predates-1978-sequencing-depends-on-contested-song-jian-dating traces a dependency the map had left implicit: the "politics-then-technique" ordering that the concedes-1976 bullet turns on holds only if Song's cybernetics encounter is correctly dated to 1978 (and that 1978, per the sourcing-scaffolding note above, rests on one unread essay). Under Olsder's uncorroborated 2024 account of a 1975 Twente delegation visit, the order inverts to technique-then-politics and the "predates" framing no longer holds. The spine — Whyte/Wang/Cai's cited synthesis — does not rest on this ordering, so the map still stands; but the concedes-1976 reading is load-bearing on a date this cluster has not settled. [unverified-historical]. question-verify-song-jian-1975-twente-delegation-visit.
• Greenhalgh and Whyte/Wang/Cai are Tier 1 as papers but secondary for the 1970s events; the member notes' "X writes that" framing keeps that honest.
• Every member note is still seedling. This map records the dispute's current footing; it does not freeze it.