theory of human cognition with a suitable knowledge base.2 That is not possible, of course, because a simple, durable theory of human cognition is not available. But from the standpoint of those who would like to use an expert system, a system based on a mishmash of theory is better than no system at all. Expert systems users are, like the ancient mariners, relatively indifferent to the sources of their practical devices.
I want to end this chapter with an optimistic note by returning to the mind and motivation of Kepler. Aficionados of Kepler and his work might raise an objection to Kepler's behavior as I portrayed in my fantasy. They might challenge the supposition that Kepler would capitulate to the computer's seductive powers. Such an objection might well be justified. As I said earlier, Kepler was a fanatical believer in the existence of underlying truths. He believed that the simple laws were there to be found if you were just willing to do the work required to find them. Much of what we know about Kepler, his motivations, and his personality leads us to believe that he would have resisted the temptation to build an enormously complex theory designed solely to account for data, computer or no computer. I would like to hope that social scientists will similarly come to resist the temptation to build complex computer simulation models designed with the purpose of accounting for complex data derived from complex situations. We can call that enterprise fun, we can call it a game, we can even call it a rough shortcut to the solution of practical problems. But we should not think of it as the ultimate goal.
Banaji, M. R., & Crowder, R. G. ( 1989). "The bankruptcy of everyday memory". American Psychologist, 44, 1185-1193.
Commoner, B. ( 1979). The politics of energy. New York: Knopf.
Church, R. M. ( 1983). "The influence fo computers on psychological research: A case study". Behavior Research Methods & Instrumentation, 15, 117-126.____________________
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Book title: A Handbook for Data Analysis in the Behavioral Sciences:Statistical Issues. Contributors: Gideon Keren - Editor, Charles Lewis - Editor. Publisher: Lawrence Erlbaum Associates. Place of publication: Hillsdale, NJ. Publication year: 1993. Page number: 490.