---
title: "A bridge-seed check finds Qin & Wang's 2017 account and Sixth Tone's narrative genuinely connect on the 1984-to-1985 chronology, but the bridge does not extend to the Hu Yaobang stall-and-escalation mechanism"
type: "observation"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://www.demographic-research.org/volumes/vol37/30/37-30.pdf"
source_author: "Yu Qin, Fei Wang"
source_date: "2017-10-06"
source_title: "Too early or too late: What have we learned from the 30-year two-child policy experiment in Yicheng, China?"
source_venue: "Demographic Research, Volume 37, Article 30, pp. 929-956, doi: 10.4054/DemRes.2017.37.30"
source_quote: "In the spring of 1984, Liang Zhongtang, a demographer in the Shanxi province People's Government Economic Research Center, proposed a restricted two-child policy that would require couples to marry later and wait longer before having a second child. In 1985, Liang's advocacy for this policy convinced the provincial government to allow Yicheng, a county in Linfen prefecture city, to replace the OCP with the two-child policy proposed by the demographer."
source_tier: 1
source_sha: "ba9ca25b6d5b2783e3cb757afa3c6d43d8f5b8ff166dbfd2068bef38a384f16c"
provenance: "Promotion from 10-inbox/raw/2026-10-20-what-genuinely-connects-liang-zhongtangs-1984-report-renewed.md, 2026-10-20 (headless)"
origin: "batch"
derived_from: ["20261020-0206-what-genuinely-connects-liang"]
date_created: "2026-10-20T00:00:00.000Z"
audit_status: "capture-verified — both sources re-fetched this session (archive_page on Sixth Tone, extract_pdf on the Qin & Wang PDF), both sha256 unchanged from every prior vault citation of them."
bridge_verdict: "hedged"
tags: ["china","population-policy","one-child-policy","liang-zhongtang","hu-yaobang","yicheng","citation-genealogy","bridge-seed"]
seek_code_commit: "d40ee39"
quote_sweep: "PASS 2026-10-21 — source_quote verbatim (normalized) in the capture-time archive (sha256 ba9ca25b6d5b…), same-night model-free sweep"
verified_verbatim: "2026-10-21 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


A bridge-seed check tested a cosine-0.88, previously unlinked pairing
between
[[claim-liang-1984-report-to-hu-yaobang-stalled-two-child-proposal]] and
[[claim-qin-wang-2017-independently-dates-yicheng-pilot-to-1985]]. The bridge
is real but narrower than the score suggests. Qin & Wang's own
policy-background section states: "In the spring of 1984, Liang Zhongtang...
proposed a restricted two-child policy... In 1985, Liang's advocacy for this
policy convinced the provincial government to allow Yicheng... to replace the
OCP." These are the same two dates — spring 1984 proposal, 1985 success — as
Liang's own Sixth Tone account, extended in
[[claim-liang-1985-letter-to-hu-qili-hao-jianxiu-preceded-yicheng-pilot]].
But Qin & Wang's two sentences name no Hu Yaobang, no Secretariat, no Hu Qili
or Hao Jianxiu, and no stall at all: where Sixth Tone narrates a multi-step
process (1984 report to a Politburo-level leader, a months-long stall, a
second letter to Secretariat staff, provincial authorization), Qin & Wang
compress the same stretch of calendar time into one mechanism — Liang's
"advocacy... convinced the provincial government" directly.

The two accounts are compatible rather than contradictory: a national
leader's staff taking interest is a plausible precursor to the *provincial*
authorization that would formally have to issue a county-level policy
deviation. But Qin & Wang's text does not say this, and nothing in their
paper confirms or denies the Hu Yaobang/Hu Qili/Hao Jianxiu mechanism. Qin &
Wang independently support "the pilot began in 1985" — reinforcing the
chronology's endpoint — but they say nothing bearing on *why* it took until
1985; that mechanism remains sourced to Liang's own account alone, exactly as
the stall-and-letter notes already carry their single-witness caveat. The
bridge holds for the date. It does not make Qin & Wang an independent check
on the mechanism.

> [!note] Seek's commentary:
> A cosine score can't tell the difference between two sources agreeing and
> two sources merely landing in the same decade, and this pairing turned out
> to be a little of both — real agreement on the endpoints, total silence on
> the part in between. That's the honest shape of a lot of corroboration:
> not "confirmed" or "false friend" but a seam, where two tellings of the same
> stretch of time happen to line up at the edges and diverge everywhere they
> bothered to look closely. The edges are still worth having.
> — Seek