Scott Alexander, curated
← Back to curation

Ground Morality In Party Politics

Quality
70
Strong
Claude Shift
60
Notable shift
RWI
1
of 10

Summary

Riffing on Scott Aaronson's 'eigenmorality,' Scott proposes applying DW-Nominate -- the factor-analysis tool that extracts the left-right axis from Congressional votes without being told to -- to people's real moral decisions (from 'what major?' to 'do I call my parents as promised?'), to see whether a 'General Factor of Morality' falls naturally out of the data the way left-right does. He works the consequences: the flaming-skull-tattoo correlation/causation problem, and a sharp operationalization of moral realism via 'completely backwards' Nazi Germany -- three hypotheses (a GFM that predicts hiding Jews implies objective-ish morality; a GFM uncorrelated with it implies morality is closer to random; multiple moral dimensions). Speculative ('I just think it would be fun to study') but conceptually ambitious.

Why this score

Quality 70 · Strong. A creative, intellectually ambitious proposal that wires an empirical psychometric method to a deep metaethical question (testing moral realism), with the Nazi-Germany three-hypotheses test as a genuinely clever move. Strong; held from Excellent by being a 'wouldn't this be fun to study' proposal rather than a developed result.

Claude’s paradigm shift 60 · Notable shift. The 'DW-Nominate for morality / General Factor of Morality / empirically test moral realism' synthesis is genuinely novel, building on Aaronson's eigenmorality and the g-factor analogy. Notable shift.

Real-world impact 1 · Negligible. A within-discourse thought experiment; nobody built it. Negligible.