Study reveals language meaning is built on power, danger, and safety

A new study challenges 70 years of assumptions about language, finding meaning is organized around power, danger, and structure rather than emotion alone, with a hidden preference for safety. This discovery could reshape linguistics, psychology, and AI development. Published in Science Advances, University of Vermont researchers analyzed billions of word uses across 20,000 words, finding their framework explains over 90% of meaning variation versus 72% for the traditional VAD model. The safety bias may explain the Pollyanna principle's positivity tendency. The findings suggest language reflects survival instincts, as distinguishing safety from danger is essential. This could improve AI sentiment analysis and offer new interpretations of human communication, though researchers note evolutionary causation is not definitively established.