Social Knowledge and Intelligent Machines
280 pp., 6 x 9 in,
- Published: October 30, 1990
- Publisher: The MIT Press
- Published: November 13, 1992
- Publisher: The MIT Press
In Artificial Experts, Collins explains what computers can't do, but he also studies the ordinary and extraordinary things that they can do. He argues that the machines we create are limited because we cannot reproduce in symbols what every community knows, yet we give our machines abilities by the way we embed them in our society. He unfolds a compelling account of the difference between human action and machine intelligence, the core of which is a witty and learned explanation of knowledge itself, of what communities know and the ways in which they know it.
Over the past fifteen years, Harry Collins has been in the forefront of a handful of mostly British sociologists who are revolutionizing our understanding of science by taking seriously a commonplace suggestion, namely, that science should be studied just like any other social phenomenon.
Times Literary Supplement
Collins is a master of the new sociology of science, which watches science as it actually works. Artificial Experts is his masterpiece. Witty, readable, and wise, it introduces the fields of artificial intelligence and the sociology of science to each other and to the outsider. It shows what we do with tools and what they do to us.
Donald McCloskey, Author of If You're So Smart: The Narrative of Economic Expertise, Professor of Economics and of History, University of Iowa
This work significantly advances the debate over machine versus human intelligence. It puts a new twist on the old question of what computers can't do, by asking instead how it is that they are able to do those things that they can do. The answer according to Collins is that intelligent machines work in virtue of their use by socialized human beings.
Dr. Lucy Suchman, System Sciences Labratory, Xerox Palo Alto Reaearch Center
One of Collin's main contributions to the sociology of scientific knowledge is to stress (following Wittengenstein, Winch, Polyani et al.) the tacit element in scientific knowledge and technical skill, and to draw out the implications of this insight. It is an invaluable and powerful point, with wide-ranging ramifications.
David Edge, Director, Science Studies Unit, University of Edinburgh
This book is intended for AI researchers and practitioners as well as sociologists... H. M. Collins makes his case without the polemical excesses of Hubert Dreyfus or John Searle. Unlike them, he seems to have genuinely attempted an experiment in knowledge engineering.
Kamesh Ramakrishna, IEEE Expert
H. M. Collins makes his case without the polemical excesses of Hubert Dreyfus or John Searle. Unlike them, he seems to have genuinely attempted an experiment in knowledge engineering.