Scholarship on Presidential Decision Making
The scholarship on presidential decision making is broad in scope, deep in quality, and varied in texture. A brief overview of some of the scholarly literature is here divided into decision-making theory, advisory structures, and the analysis of individual presidents. Most of the literature mentioned in this introduction is treated in much more depth in the articles that follow.
Much of the decision-making literature after World War II focused on the limitations of the rational model. At first glance, the rational approach to decision making--stating objectives, ranking values, analyzing alternatives, examining consequences, and making choices--seems to accord with common sense (who would want to be irrational?). But the reality of making decisions under conditions of complexity (virtually all public policy decisions) is much more problematic. Much of the scholarly decision-making literature emphasizes just how complex making decisions really is.
The rational or deductive model has been analyzed by, among others, Lindblom (1959), Braybrooke and Lindblom (1963), and March (1994) in general and Schelling (1966, 1980) and Snyder (1961) with respect to nuclear deterrence. Much of the literature on rational decision making and its limits was marshaled and summarized by Graham Allison in Essence of Decision (1971; Allison and Zelikow 1999), which analyzed the Cuban Missile Crisis through the lenses of three different models. His Model I lays out the rational approach in general and applies it to analyze the Cuban Missile Crisis.
Herbert Simon argued that economists attributed a "preposterously omniscient rationality" (1976, xxvii) to individuals, and he distinguished between "facts" and "values" in organizational decision making (1945). He concluded that organizational decision makers operated within "bounded rationality" and tended to "satisfice because they have not the wits to maximinze" (1976, xxviii, italics in original). He also characterized the context of most governmental decision making: "A decision is not a simple, unitary event, but the product of a complex social process generally extending over a considerable period of time" (1965, 35).
Charles Lindblom (1959) attacked the rational decision paradigm by arguing that, in principle, values could not be separated from concrete policies. The informational, resource, and time constraints on decision makers were so substantial that only alternatives marginally different from the status quo could be realistically examined (1959). March and Simon (1958) applied the ideas of satisficing, sequential search processes, and repertoires of actions to decision making in formal organizations and emphasized the cognitive limits on rationality. Cyert and March (1963) challenged the traditional theory of the firm by arguing that firms (and by implication, other complex organizations) did not in fact maximize net revenue or act from perfect knowledge, but rather used organizational slack to make side payments in order to keep together the coalition of subunits in large organizations. Each subunit imposes its own constraints on overall organizational goals.
Allison (1971) summarized much of the scholarly literature on decision making, and his critiques of the rational model in his Model II (organizational process) and Model III (bureaucratic politics) have become staples in the political science literature.
Another stream in the scholarly literature that attempts to understand presidential decision making focuses on presidential advisory systems. The assumption is that no one individual can hope to understand all of the ramifications of decisions facing the president and that staff structures are thus necessary but can help or hinder good decision making. Of course, well-organized advisory systems cannot guarantee good decisions. As President Eisenhower (1965, 630) put it, "Organization cannot of course make a successful leader out of a dunce, any more than it should make a decision for its chief. But it is effective in minimizing the chances of failure and in insuring that the right hand does, indeed, know what the left hand is doing." The study of advisory systems has included the influence of formal organizations, small groups, and role expectations on decision making.
Perhaps the most basic model of structuring decisions is the idea of bureaucracy developed by Max Weber. In observing the development of bureaucracies at the turn of the twentieth century, Weber (1946) abstracted into his "ideal type" the characteristics of typical bureaucracies. Principles such as the necessity of rules, files, monocratic hierarchy, division of labor, continuous organization, salaries, professional staff, and so on, as described by Weber, seem obvious and commonplace to us now, but only because of his pioneering observations when the generic characteristics of emerging bureaucracies were not obvious.
Alexander George (1972, 1980) argued that presidents needed to assure that their advisory systems provide them with a range of alternatives for any important decision and that the best way to assure this was a system of "multiple advocacy." He argued that the mere presence of differing views among White House staffers did not guarantee the effective presentation of alternatives to the president. Thus, the system had to be consciously structured so that the representatives of different alternatives possessed similar intellectual and bureaucratic resources.
Richard Tanner Johnson (1974) divided presidential approaches to White House organization into formalistic (Ike, Nixon), competitive (FDR), and collegial (JFK). White House staffer and scholar Roger Porter (1980) analyzed the operation of the Ford administration White House and characterized different approaches to advisory systems as adhocracy, centralized management, and multiple advocacy. Haney (2002) uses the Johnson and George models to analyze crises in the Truman, Eisenhower, Johnson, and Bush (41) administrations.
Irving Janis (1982) analyzed the effects of small-group solidarity in situations where the stakes are high, pressure is great, and secrecy is important. The danger in these situations is that the group will develop the illusion of invulnerability and inherent morality, underestimate the enemy and chances of failure, and fail to reexamine their initial assumptions. Janis used the term "groupthink" to characterize such situations and analyzed cases of presidential decision making to illustrate the syndrome as well as cases when it was avoided. One possible way to mitigate the groupthink syndrome, Janis suggested, is to have one advisor play the role of "devil's advocate," by systematically criticizing the group consensus and articulating alternative approaches to the problem.
Small-group dynamics are also central to analyses by Garrison (1999) and Ponder (2000), who provide insights into the causes and consequences of the internal staff dynamics of advice to the president, Garrison in foreign policy and Ponder in domestic policy. Newmann (2004) argues that most presidents begin their terms with a relatively formal policy development process, but eventually narrow their sources of input to a few trusted advisors.
The idea of roles played by different staffers, as in Janis's devil's advocate, has also been useful in understanding the relationship among presidential staffers and between the president and staff. The neutral (or honest) broker concept implies that in any important decision-making situation, the staffer presents to the president in a neutral way the most important policy alternatives and represents faithfully the views of the advocates of different policy alternatives. The president can thus have confidence that the dice are not loaded in favor of one or another alternative (or staffer). Playing the role of neutral broker does not preclude the staffer from giving his or her best advice to the president, but it assures that this judgment will not unfairly subvert the judgments of other staffers.
The neutral broker role and its variants are taken up by John Burke in this issue, and have been analyzed with respect to the...