Herbert A. Simon: The Bounds of Reason in Modern America (review)
Buy Issue for $25 at JHUPHunter Crowther-Heyck's biography of Herbert Simon arrives with perfect timing. A welcome historiographical current is now bringing to light long-neglected links among the histories of technology, science, and economics. In particular, Philip Mirowski, beginning with his 1989 study linking nineteenth-century physics and the origins of neoclassical economics, More Heat Than Light, has consistently made a strong case for studying such connections. However, work on post–World War II developments has tended to stumble in the tangled barbed wire lining the borderlands of computer technology, economics, mathematics, and brave new programs in the biological, psychological, and social sciences. Historians venturing into this territory often gesture toward the existence of shadowy hybrid [End Page 877] entities, sometimes called cyborgs, that arose from a postwar primordial soup of servomechanisms, digital logic, game theory, military patronage, and the political mentality of the cold war.
Simon's career, passing freely between public administration theory, management science, computer science, and psychology, is well situated to lend some much-needed detail to our understanding of this area. Crowther-Heyck seizes the opportunity, arguing that Simon built his reputation as a mediator between "sciences of control" and "sciences of choice." The former refers to fields, such as psychology, concerned with the forces that condition people to behave a certain way; the latter refers to fields, such as economics, concerned with how people make free and rational choices. Simon's response to the tension between the two is embedded in his concepts of bounded rationality and satisficing, which describe how decisions are made freely, but within a set of limited options, based upon limited knowledge and under the pressure of limited time and investigative resources, giving rise to satisfactory—not optimal—decisions.
Taking this outlook as a driving worldview, Crowther-Heyck follows its various manifestations within different institutional contexts (an analytical approach, he acknowledges, with similarities to Simon's own theories). One quickly gets a sense, correct or not, that Simon exhibited little personal growth: it was the scenery that changed. The book begins with Simon's training in the prewar "liberal managerialism" (p. 43) of the Chicago School of Political Science and early employment with the University of California at Berkeley's Bureau of Public Administration. It then discusses Simon's postwar work with the economists at the Cowles Commission and the nascent operations research community, the ways in which his research in the social sciences was supported and shaped by the funding structures of cold war America, and how he moved to the Carnegie Institute of Technology, where he swiftly became a preeminent intellectual force and where he remained for the rest of his career. A clear turning point in Simon's life did occur in 1952, however, when he first encountered digital computers and met his future collaborator, Allen Newell, at the RAND Corporation's Systems Research Laboratory.
Once the computer arrives on the scene, the narrative moves from a sometimes-convoluted structure into an expertly crafted exposition of Simon's program of nested hierarchical model-making, which he applied to both computer science and human reason. Bypassing the standard computer-as-scientific-metaphor claim, Crowther-Heyck presents a deeper look at Simon's philosophy of science wherein mimesis tests the value of a model as heuristic tool rather than as a representation of a concrete phenomenon. Thus, a particular computer model would not be considered an ontologically faithful representation of the human brain, in general, but one of many possible (incomplete) descriptions of a particular rational [End Page 878] function, such as how humans play chess. Understanding thereby derived could then be used to further modify the model or to make inroads into other rational functions.
While it is obvious that Simon's work was deemed important—he was awarded the 1978 Nobel Prize in Economics—...
