Author: Vladisav Jovanović
Status: Journal article
Version: Latest archived (Apr 2026)
This paper argues that the Chinese Room remains philosophically important, but that the standard syntax-versus-semantics framing is now too narrow. Large language models show that the deeper issue is the answerability condition of understanding. The paper reads the Chinese Room as a model of closed cognition: a system that lacks not only semantic contact in the abstract, but consequence-bearing correction, revision under pressure, and any witness to the cost of mismatch with reality. Understanding is therefore not merely symbol manipulation plus reference. It is symbol use inside a system that can be revised by reality in ways that matter. The paper connects this argument to language models, distinguishing statistical correction from lived consequence-bearing answerability and arguing that formal success without answerable contact remains structurally incomplete.
Chinese Room; language models; understanding; Structural Intelligence; contact; consequence; answerability; closed cognition; syntax; semantics; AI philosophy; LLMs; corrigibility; symbol manipulation; reality-contact