There is a long tradition in the analysis of U.S. foreign policy to study how presidents assemble teams of advisers to help make decisions. Starting especially in the 1970s, a series of new works helped push our understanding of how presidents put these teams together, how they do their work, and what effects the structuring of decision making has on the process of reaching decisions and policy. Exploring the links among decision structure, process, and outcomes has been a consistent theme in the literature. While we have seen a diverse set of approaches to studying these questions in the last twenty or so years, they all trace themselves back in one way or another to the seminal works by Graham Allison (1971), Essence of Decision; Richard T. Johnson (1974), Managing the White House; Alexander George (1980), Presidential Decision-making in Foreign Policy: The Effective Use of Information and Advice; and Irving Janis (1972), Victims of Groupthink.
In this article, I try to highlight some of the conceptual developments we see in the literature in understanding presidents and their advisory structures since these works. I do so not in the form of an exhaustive review of the literature but rather by pointing to several examples of the kind of scholarship that has developed around these issues in the domain of foreign policy. Then I use two heuristic cases from the present Bush administration to show how the literature helps us understand some things about the policy process and the role of the president in it but frankly leaves us unable to fully grasp some others. U.S. policy toward Cuba and policy in Iraq, I argue, are interesting because many of the models from the literature are helpful in understanding how President Bush and his advisers work in a broader policy environment to craft foreign policy. But they also show the limits of what we can understand. We have models that help us see the president caught in a vast bureaucracy or porous policy environment rife with congressional and interest group activism where presidents must lead through persuasion; and we have models that help us see the president as the dominant foreign-policy player who leads by flat. In short, though, what we have yet to develop are theories that help us understand how the president can be both of these things at the same time in the same policy domain. Analysts of foreign-policy decision making have long understood small-group processes, and have perhaps more recently come to appreciate the impact of institutions and domestic politics on decision making, but now must take into account the increasing power of the president to act alone, even when in the midst of a dizzying array of political forces that constrain the White House.
Models of the Foreign-Policy Process
Research that attempts to understand how U.S. presidents have organized the White House for policy making has long explored the possible effects of those structures on policy making. Probably the most well-known study of the process of decision making is Janis's Victims of Groupthink (1972). Janis was motivated to explain performance failures, such as the American fiasco at the Bay of Pigs, by examining the internal dynamics or group processes that lead ultimately to group decisions. "Groupthink" is when individuals within cohesive groups seek unanimity or concurrence to such an extent that they cease to vigilantly perform the tasks of decision making. Janis hypothesized that the presence of groupthink during the process of-decision making might lead to policy failures. While a psychological phenomenon that cannot be directly observed, Janis argued that groupthink produces behavioral consequences or symptoms that can be observed--and that can be avoided with proper planning.
Richard Johnson takes a more structural approach to these issues; he explores how the president "manage[s] a team of men to provide him with information, staff out his alternatives, and otherwise extend his reach" (1974, xxii) so that the president can be successful at leadership and policy making. Johnson focuses on how the White House is organized for general policy making, he identifies three generic models of organization that presidents have used--a formalistic, a competitive, and a collegial model of decision making. Alexander George (1980) picks up here and applies Johnson's models to the foreign-policy domain to discuss how modern U.S. presidents structure advisory networks and the impact of that on information processing.
Graham Allison's (1971) approach is to explore how organizational structures and bureaucratic games shape the policy-making process and direct policy outputs or outcomes. This bureaucratic politics approach, or family of approaches, opened the "black box" of decision making to analysis and to explore the ways that internal political processes affect foreign-policy making. An emphasis on the ways that organizational processes and procedures (Allison's Model II) and bureaucratic bargaining and infighting (Allison's Model III) affect the decisions reached by groups is central to this approach. Allison examines these components in American decision making during the October 1962 Cuban Missile Crisis.
A variety of recent approaches to understanding how presidents construct a foreign-policy process are worth mentioning, each representing a broader strain of theoretical and empirical research. In Organizing the Presidency (2002), Hess and Pfiffner track the ways that modern U.S. presidents have structured White House operations. Besides describing the organizational styles of each administration, they discuss how presidents learn from the perceived organizational mistakes of each former president in an effort to fine-tune the structure of policy making. Burke and Greenstein (1989) examine the importance of advisory groups as well as presidential personality and the political environment during two cases of American decision making about Vietnam: Eisenhower in 1954 and Johnson in 1964-65. They seek to explain why two presidents who were faced with very similar problems responded in such very different ways. Their analysis indicates that the way presidents organize advisory groups may have an important impact on the process of decision making, but that the individual president's style and the political climate also affect the process of decision making. In my own work (Haney 2002), I tried to take the models developed by Johnson (1974), applied to foreign policy by George (1980), and apply them specifically to foreign-policy crises, showing how presidents constructed hybrids of the ideal types to suit their needs.
In The Institutional Presidency (1992), Burke argues that studies of the U.S. presidency need to begin to examine in more depth the nexus between the enduring institutional (structural) features of the presidency and the management strategies and styles of particular presidents, and the implications of each for the other. He attempts to move in this direction by showing how modern U.S. presidents have dealt with these issues. In a similar vein, Ponder explores how presidents have tried to control policy making by centralizing the process inside the White House. He focuses particular importance on the use of staff to help the president. Bringing some of the "new institutionalism" into this domain, he also views the White House as an institution whose rules are determined endogenously, thus open to presidential influence (2000, 176-77). Garrison (1999) also follows in this general approach, examining what she calls the structural, procedural, and interpersonal influence maneuvers used in policy making by different advisers.
Picking up on the evolutionary model of Hult and Walcott (2003), Newmann tracks the development of policy-making structures over time in an administration. He shows how the initial decision-making structures represent what he calls a president's administrative theory--the president's preferences about the methods and goals for foreign-policy making, and the president's beliefs about what is the proper relationship to have with advisers. He shows how these structures change over the course of an administration, and argues that while each president is unique, there are patterns of evolution across administrations (Newmann 2003, 24-25). Newmann shows a general narrowing of participants as the president seeks advice from an ever smaller number of individuals. He also shows that for policies on which the president places a high priority, he is likely to try to exert more control--by managing the pace of the process; defining policy content; and by excluding participants.
Preston (2001) pursues these issues by asking what difference the personality, leadership style, and policy experience of the president makes for foreign policy. Or, under what circumstances do these factors matter? He uses the "personality at a distance" technique developed by Margaret G. Hermann to "type" leaders along two dimensions of personality and political experience: (1) the need for power or control and the level of presidential involvement in policy making, and (2) the president's need for information (cognitive complexity) and the president's attentiveness to the external environment. This approach also feeds off the earlier effort to show how different foreign-policy decisions are likely made by different kinds of "ultimate decision units." By "decision unit," we mean whether a decision can be reached by a single dominant leader, by a small group, or the result of bargaining among multiple autonomous actors (Hermann, Hermann, and Hagan 1987), and what difference that may make not just for understanding the processes involved but potentially what impact it has on outcomes as well.
In a study that picks up from Janis's...