Scott Alexander, curated
← Back to curation

Scientific Freud

Quality
74
Strong
Claude Shift
56
Moderate
RWI
2
of 10

Summary

A sharp meta-science essay. Opens with the metaskepticism insight (the ability to debunk 'X is true' doesn't generalize to debunking 'X has been debunked' — the Heroic Foundation Myth of psychiatry he never checked). Then: a large RCT finds psychodynamic (Freudian) therapy non-inferior to CBT, alongside many others — two conclusions: either Freud was right, or (modus tollens) CBT doesn't really work either. Scott argues the latter via the Dodo Bird Verdict (all bona-fide psychotherapies work equally, via a caring charismatic listener + ritual, not the specific theory), and the killer point: CBT got its 'evidence-based' reputation not because it works and psychoanalysis doesn't, but because 'the CBT people did studies and the psychoanalysts didn't' — the two-types-of-'no-evidence' problem. Qualifies that simple symptom-targeting therapies (exposure) do work.

Why this score

Quality 74 · Strong. High-Strong: a sharp, cited essay with several portable insights (metaskepticism as a distinct skill; the Dodo-Bird popularization; 'CBT won because it did studies first'; two-types-of-no-evidence), but single-topic and building on the Dodo-Bird literature. 74.

Claude’s paradigm shift 56 · Moderate. The metaskepticism framing + the CBT-reputation reframe are original. B56.

Real-world impact 2 · Minor. Influential within rationalist/psychiatry discourse ('metaskepticism', Dodo-Bird-for-therapy get cited); no material reach. RWI2.