Judea Pearl: ordinary language is pre-scientific — it helped alchemists make glass, not turn lead into gold

yudapearl · x · 2026-08-28

Statistician Frank Harrell pushed back on Pearl's recurring claim that statisticians lacked causal inference for RCTs, arguing ordinary-language experiment design has ruled out alternatives for a long time. Pearl countered that ordinary language is pre-scientific: it let alchemists craft colorful glass vessels, yet couldn't convince them lead can't become gold — formal causal frameworks matter.

Original post →

More from Research

Research channel →