Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Other bodies, other minds
Harnad S. Minds and Machines1 (1):43-54,1991.Type:Article
Date Reviewed: Nov 1 1991

Not too many computer scientists worry about the “other minds” philosophical problem: “How can one be sure that anyone else has a mind, rather than merely behaves as if he had a mind?” As applied to machines, however, the question is of considerable interest to both computer scientists and philosophers, who make up the audience of the excellent new journal Minds and Machines. The other minds problem is, after all, precisely the question that Alan Turing posed for machines. As Harnad puts it, “Is anyone ‘at home’ in an ‘intelligent’ machine?”

Turing answered this question by posing the Turing Test: if one cannot distinguish the machine from a human being in an arbitrarily long conversation, the machine can think (has a mind). Searle, however, has shown with his Chinese Room argument [1] that mere excellence on such a linguistic test need not show understanding: the machine may actually be no more than an extremely clever simulator of a mind. It is Harnad’s thesis that given full robotic capacities--say, all human sensorimotor functions--the Turing Test (now the Total Turing Test or TTT) will succeed in convincing almost anyone that the machine has a mind. In other words, “If it looks like a duck, walks like a duck, quacks like a duck…it’s a duck.”

The TTT is immune to Searle’s argument because more than (very clever) symbol manipulation is required. It is critical for the candidate to be able to make connections between linguistic expressions and their referents in the world of objects, something that Searle’s Chinese Room simulator does not have to--and cannot--do. Harnad grants that even the TTT does not prove that someone is “at home” in the machine, but he makes a convincing argument that passage of the TTT is all that we would normally require to label the machine candidate a mind (or intelligent, or conscious--to Harnad these are all variants of the same idea). After all, we do infer that other human beings have minds from their human-like behavior.

Harnad’s paper is interesting, important, and well written. It should interest the entire AI community and a good part of the larger computing community.

Reviewer:  Joseph S. Fulda Review #: CR115508
1) Searle, J. R. Mind, brains, and programs. In Mind Design, J. Haugeland (Ed.), MIT Press, Cambridge, MA, 1985, 282–307.
Bookmark and Share
  Featured Reviewer  
 
Philosophical Foundations (I.2.0 ... )
 
Would you recommend this review?
yes
no
Other reviews under "Philosophical Foundations": Date
Rethinking smart objects
Rasmus D. (ed), Cambridge University Press, New York, NY, 1999. Type: Book (9780521645492)
Mar 1 1999
Do the right thing
Russell S., Wefald E., MIT Press, Cambridge, MA, 1991. Type: Book (9780262181440)
Aug 1 1992
The practical intellect
Göranzon B. (ed), Springer-Verlag New York, Inc., New York, NY, 1993. Type: Book (9780387197593)
Jul 1 1994
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy