Vague Quantitative Language
Words like “significant,” “substantial,” and “conservative” create an illusion of rigor while obscuring actual quantities.
On the Columbia analysis slide, “significant” appeared 5 times with 5 different meanings: detectable in a calibration study, enough to cause damage, statistically meaningful, large enough to matter, and a 640-fold difference. None referred to actual statistical significance. The word felt quantitative while communicating nothing precise.
This wordplay substitutes verbal fudge factors for actual numbers. “Conservative estimate” sounds rigorous but means nothing without knowing the assumptions. “Substantial improvement” begs the question: compared to what baseline?
The pattern thrives in constrained formats. When there’s no room for numbers, tables, or methodology, vague quantifiers fill the space. They sound technical to non-technical audiences and provide plausible deniability for technical ones.
Watch for: significant, substantial, conservative, robust, considerable, meaningful, notable, marked, appreciable. When these words appear without numbers attached, ask what quantity they’re replacing.
Related: 02-atom—hierarchical-bullets-bury-uncertainty, 02-atom—format-shapes-cognition