Peirce's revenge on the Chinese Room

Area of Science:

• Philosophy of Mind
• Artificial Intelligence
• Linguistics

Background:

• John Searle's Chinese Room argument (1980) is a key objection to computational theories of mind.
• The argument questions whether machines can achieve genuine understanding or consciousness.
• Previous critiques often focused on the computational aspects, overlooking pragmatic dimensions.

Purpose of the Study:

• To re-evaluate the Chinese Room argument by analyzing its assumptions about linguistic interaction.
• To demonstrate that the argument relies on an impoverished view of communication.
• To extend this critique to modern large language models (LLMs) and "stochastic parrots".

Main Methods:

• Analysis of the Chinese Room argument through the lens of Peircean semiotics (icons, indices, symbols).
• Examination of pragmatic, indexical, and rule-governed aspects of natural language use.
• Introduction of the "Chinese Token-Training Room" as an analog for LLM critiques.

Main Results:

• The Chinese Room argument's validity hinges on isolating systems from pragmatic language use.
• Allowing pragmatic dimensions into the thought experiment either grants semantic understanding or restricts the test.
• Critiques of LLMs face a similar dilemma: restrict competence or allow pragmatic embedding.

Conclusions:

• The Chinese Room argument does not prove functional or computational organization is insufficient for conscious semantic understanding.
• The argument, and its analogs, only demonstrate that isolated symbol or token manipulation is insufficient.
• A more nuanced understanding of linguistic interaction is required to assess machine understanding.