Highlights From The Comments On Lynn And IQ
Read the original on Astral Codex Ten →
Summary
Highlights-from-the-comments companion to Scott's post on Richard Lynn's national-IQ estimates. Engages a fraught topic with care: the IQ/g (Flynn-effect) gap and whether very-low sub-Saharan estimates reflect g or test-proxies; his explicit 5/5/40/50 apportionment of the 'Malawians seem smarter than their measured IQ' puzzle across data-error / trivial-test-bias / IQ-concept-breaks-down-at-extreme-low-education / personal-miscalibration; the reassurance that uncontroversial proxies (World Bank learning outcomes, GDP) reproduce Lynn's pattern; the genetic-diversity-of-'black' caveat; salt iodization; conceding Lynn's racism; and the substantive thesis that abstract/symbolic and practical IQ components come apart at the tails (Lyman Stone's counter that subsistence farming requires real planning; the Reich PGS-vs-IQ-deviation correction).
Why this score
Quality 69 · Strong. Strong/Solid edge -- top of the Highlights-companion cohort. Careful, honest engagement with a genuinely hard topic, advancing a real thesis (abstract vs practical IQ coming apart at the low tail). Capped as a curated comment-digest companion rather than a standalone essay.
Claude’s paradigm shift 45 · Moderate. Moderate, low. The components-come-apart-at-the-tails framing is a fresh-ish refinement, but the post is mostly responding to and reconciling commenters.
Real-world impact 2 · Minor. Minor. Within the intelligence-research discourse; conceptual. 2.