Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cwjcc!tut.cis.ohio-state.edu!rutgers!att!ihlpb!arm From: arm@ihlpb.ATT.COM (Macalalad) Newsgroups: comp.ai Subject: Re: Question on Chinese Room Argument Message-ID: <9679@ihlpb.ATT.COM> Date: 21 Feb 89 17:57:51 GMT References: <4298@pt.cs.cmu.edu> <51157@yale-celray.yale.UUCP> Reply-To: arm@ihlpb.UUCP (55528-Macalalad,A.R.) Organization: AT&T Bell Laboratories - Naperville, Illinois Lines: 79 In article harnad@elbereth.rutgers.edu (Stevan Harnad) writes: [a long, condescending diatribe against Sean Engelson] Stevan, I've tangled with you before, and I'll probably regret doing it again, but there are a few issues that I want to address, and I'd love to hear your response. 1. The distinction between Searle and (Searle + rules) As I understand from your previous postings, you argue that (Searle + rules) = Searle. The systems' argument, as I understand it, argues that (Searle + rules) > Searle. I, of course, tend to agree with the systems' argument, which seems more reasonable and intuitive to me. After all, (Searle + rules) is fluent enough to converse with a native Chinese speaker, whereas Searle can't even begin to speak Chinese. Even further, I would guess that if Searle internalized those rules, he _would_ be able to speak Chinese! The point is that this distinction is far from being trivial or just a matter of preconceptions, and I think you need a stronger argument than appealing to the common sense of the uneducated (or at least, the pre-Yalie), making vague analogies to Darwin and evolution, and hurling cheap insults at Sean Engelson. As far as I can see, the way this problem is posed acknowledges the difference between (Searle + rules) and Searle. After all, this problem isn't interesting at all if we assume that (Searle + rules) = Searle, since we know from the outset that Searle doesn't know Chinese. Central to Searle's argument is the collapsing of this distinction, which I think is fair game for criticism. No circular reasoning, nothing up my sleeves. 2. Determining the understanding of (Searle + rules) Now for the sake of argument, let's assume that there is a distinction between Searle and (Searle + rules). If we all acknowledge the other minds' problem, we can safely agree that the only entity able to decide if (Searle + rules) really understands Chinese is (Searle + rules). Not you or me or any outside observers or even Searle himself. Only (Searle + rules). The issue I now want to take up is your justification of the Total Turing Test. As I understand it, you argue that it is useful to assume that certain entities, specifically humans, are intelligent and can interact with you in intelligent ways that non-intelligent entities can't. If a robot (which I guess would be (metal + rules), but that's a whole other kettle of fish) can interact with you in such a way that you could not guess that it was a robot, then it must be intelligent, too. Of course, I can argue that this is a little too anthropocentric, but the same argument can be made against the Turing Test, as well. What I want to explore is the usefulness of the Total Turing Test. I could argue that it would be just as useful to characterize a system which could converse in a natural language as "intelligent" and capable of "understanding" what I was saying, regardless of whether it was right in front of me tap-dancing, or talking with me via a computer terminal. Remember, Searle's arguments don't really apply here, since this is a question of pragmatics, and not a question of whether there really is any understanding taking place. Of course, if you'd rather offer an objective definition of intelligence and understanding, please feel free.... 3. Conduct on the net Now I understand that you are an important person with important things to say, but that does not give you license to insult anyone else, especially on such a public form as the net. We don't all have the right answers, and often we don't even ask the right questions. I see the net ideally as stimulating discussion, not provoking mud slinging. If I said some things that were uncalled for above, I apologize. I think that a few other apologies are due. 'Nuff said. -Alex