Seems Like Targeting
Read the original on Astral Codex Ten →
Summary
A concise argument that investigative journalism functions as a targeted weapon. The pattern: Claudine Gay's plagiarism surfaced only once she was already under fire for her anti-Semitism testimony; Neri Oxman (Bill Ackman's wife) got a retaliatory plagiarism expose; EA had two 10+-year-old scandals dredged up within weeks of FTX's collapse; and Scott's own NYT profile turned negative after he protested the doxxing — with journalist Elizabeth Spiers insisting that journalists never target people and that believing so reveals narcissism. The deeper, generalizable point (the other half of 'Against Learning From Dramatic Effects'): everyone has a similar distribution of skeletons, so WHOSE gets packaged into a dramatic accusation depends on the whims of the packagers (journalists), which is why accusations cluster on the already-unpopular — a reason to trust priors/surveys over single dramatic data points.
Why this score
Quality 71 · Strong. Strong (lower end). The 'whose distribution gets packaged depends on the packagers' insight is genuinely useful and connects cleanly to his dramatic-data-point epistemics, and the examples are apt. Held down by brevity and a partly personal-grievance framing (the NYT/Spiers score-settling). 71.
Claude’s paradigm shift 42 · Moderate. Moderate. A sharp framing built on his own 'Against Learning From Dramatic Effects' and general media-criticism rather than a new concept. 42.
Real-world impact 2 · Minor. A concise argument that investigative journalism functions as a targeted weapon (the scandal surfaces once you're already under fire) — 'whose distribution gets packaged depends on the packagers,' connecting to his dramatic-data-point epistemics. Conceptual influence within discourse, short and partly personal-grievance, no material change — low RWI.