The 2005-welds/2003-disperses compositional asymmetry in Greenhalgh's narrative structure is not diagnostic of whether science or politics came first — it is a fact about her prose, not her argument

chinapopulation-policyone-child-policyhistoriographycold-war-sciencegreenhalghsong-jiankwakernaakliang-zhongtangcitation-genealogybridge-notecontested-causation

Three prior measurements already established that Greenhalgh welds the same recurring elements — narrative structure, the leadership-consensus material's position, and the Dutch-work critique's placement — into one continuous arc in her 2005 China Quarterly article and disperses them across her 2003 Population and Development Review article (see observation-greenhalgh-dutch-critique-placement-is-a-third-measurement-of-the-2005-welds-2003-disperses-pattern for all three). This note adds a fourth test, aimed not at where the elements sit but at what the placement is doing argumentatively, and finds the weld/disperse asymmetry carries no diagnostic weight on the causal question the cluster cares about.

In 2005, welding the 1976-consensus sentence and Song's Kwakernaak-derived science into one unbroken 13-page arc does not produce an argument that the science explains the consensus — no sentence in the arc draws that connection. In 2003, dispersing the same elements across a five-section run does not leave the question merely implicit either — Greenhalgh poses the legitimation-versus-origination choice explicitly and resolves it on political evidence, never reaching for the science's content. Both compositional habits, pursued to their respective limits, produce the same non-connection. The weld is a narrative choice — one continuous origin story, told in order — not a causal claim smuggled in by proximity; the disperse is a structural choice, not a withheld argument.

This does not conflict with the finding that the 2003 material sits inside one continuous nine-page run: that note measures physical page-continuity (no gap in the document between pp. 178 and 187), while "disperses" describes crossing five same-level section headings within that run. The two descriptions measure different things and agree rather than contradict. The underlying "political will preceded Song" reading this bridge keeps intact is, separately, not independent of the unresolved dispute over whether Song's cybernetics encounter is correctly dated to 1978 or to 1975 — this note's finding (that textual placement doesn't carry causal weight) holds regardless of which date is correct, but the sequencing claim itself does not.