The Chinese Room Argument, by John Searle, is one of the most important thought experiments in 20th century philosophy of mind. The point of the argument is to refute the idea that computers (now or in the future) can literally think. In short, executing an algorithm cannot be sufficient for thinking. The method is to focus on the semantics of our thoughts. The thought experiment proceeds by getting you to imagine yourself in the role of the central processor of a computer, running an arbitrary computer program for processing Chinese language. Assume you speak no Chinese language at all. Imagine yourself locked in a room with a program (a set of instructions written in, say, English) for manipulating strings of Chinese characters which are slid under the door on pieces of paper. If a note with string S1 (in Mandarin, say) is put under the door, you use the program to produce the string S2 (also in Mandarin), which you then slide back out under the door. Outside the room, there is a robust conversation going on Chinese history. Everyone outside the room thinks that whoever is inside the room understands Chinese. But that is false. By assumption, you have no idea what S1 and S2 mean (S2 is unbeknownst to you, an insightful reply to a complicated question, S1, about the Ming dynasty). But you are running a computer program. Hence, there is no computer program such that running that program suffices for understanding Chinese. This suggests that computer processing does not suffice for thought.
|Key works||The paper that got all of this started is John Searle's famous Searle 1980. See also the initial replies to his paper in the same journal issue. Since its appearance, a large literature has been produced trying to answer Searle's challenge. Leibniz, in his Monadology (1714), Leibniz 1902, suggested something similar by asking his readers to consider stepping into a mill. One of the best replies to the argument is Churchland & Churchland 1990. One theory of computational processes that attemtps to avoid the argument by construing semantics as an explanatory construct is given in Dietrich 1990.|
- The Turing Test (444)
- Godelian Arguments Against AI (264)
- Machine Consciousness (311)
- Machine Mentality, Misc (323)
Using PhilPapers from home?
Create an account to enable off-campus access through your institution's proxy server.
Monitor this page
Be alerted of all new items appearing on this page. Choose how you want to monitor it:
David Bourget (Western Ontario)
David Chalmers (ANU, NYU)
Rafael De Clercq
Ezio Di Nucci
Jonathan Jenkins Ichikawa
Jack Alan Reynolds
Darrell P. Rowbottom
Learn more about PhilPapers