Wolfram's free plug

Stephen Wolfram must be delighted at the criticism of his New Kind of Science . You exaggerate its importance yet further with free front and back-page publicity, as well as by giving J. Doyne Farmer of the Santa Fe Institute, Wolfram's disgruntled former collaborator, space in Soapbox ( THES , September 6).

Farmer insists that "a key demand of scientific theories is that they make predictions". If the theory embraces key unpredictables that go under the headings of "vagaries of human nature" and "acts of God", both necessary features of a system that keeps human beings guessing, it cannot make sensible predictions on points of crucial importance.

Farmer tries to devalue "conceptual models" because they "have been a staple of complex systems theory for decades". Santa Fe's "conceptual models" have failed to produce anything of significance because they are the wrong concepts and fail the simple test of mathematical "necessity and sufficiency". The importance of the right concept that passes that test cannot be over-exaggerated.

J. Pick
Croydon