Herbert A. Simon, professor of psychology at Carnegie Mellon University in Pittsburgh, responds:

"I have followed the development of chaos theory with some interest over the past decade or so, but without direct active involvement, so I may or may not be a reliable witness. The only serious efforts of which I am aware to apply chaos theory to social phenomena have been in economics--possibly because only in economics does one find the kinds of time-series data that would be required to test whether or not phenomena are chaotic. There have been a number of such tests for economic data--although even for economic phenomena, time series of the length and accuracy needed for testing are hard to come by, except in the case of some kinds of monetary series and series of stock prices. The verdict would seem to be that there is some indication of chaos in some of the series that have been examined. If that conclusion sounds vague, it is because the evidence looks vague to me.

"As a consequence, I would hesitate to say that chaos theory has had any major impact on the social sciences to date, although the idea that economic trends are difficult to forecast because they are chaotic remains an attractive one. There could be lots of other reasons for our inability to predict, however, all of which would have the same practical consequences.

"In psychology, where most of my own research resides, I have observed nothing of importance that relates to chaos, and I would be rather astonished if it plays any significant role in behavior extending over tens of milliseconds or longer--that is, the kind of behavior we observe and explain in most psychological experiments. But then, astonishment is what we are always seeking in good science, isn't it?"

James P. Crutchfield, research professor at the Sante Fe Institute and a research physicist at the University of California at Berkeley, adds a small plea:

"The phrase 'chaos theory' was invented, as far as I can tell, by the science press around the time of the publication of James Gleick's book Chaos. The proper descriptor, for better or for worse, is 'dynamical systems theory.' I hope I can disabuse people of using the popularized term, which I and many of my colleagues consider inappropriate.