The Dangerous Method of Smooth Prose

Smooth prose is prose from which all roughness, all qualification, all evidence of struggle has been removed. It reads well. It flows. The arguments are clean. The transitions are seamless. The sentences are balanced and rhythmically satisfying. And precisely because it reads so well, it is dangerous — it triggers the fluency heuristic, the cognitive bias by which ease of processing becomes a proxy for truth. Peter Elbow argued that good-sounding prose is the enemy of genuine thinking, because the satisfaction of a well-turned sentence can substitute for the harder satisfaction of having actually thought something through. The writer who produces smooth prose on the first pass is manufacturing, not discovering. She is assembling pre-approved ideas into pre-approved structures, designed to withstand criticism rather than to produce insight. AI has perfected this method. Large language models generate prose that is smooth by computational design, optimized for coherence and plausibility. The output arrives pre-polished, and the polish conceals gaps in substance that rough prose