Read an Excerpt
Foreign Policy and the Bureaucratic Process
The State Department's Country Director System
By William I. Bacchus PRINCETON UNIVERSITY PRESS
Copyright © 1974 Princeton University Press
All rights reserved.
ISBN: 978-0-691-07565-5
CHAPTER 1
The Context of Foreign Policy Making
There will always be the dark and tangled stretches in the decision-making process — mysterious even to those who may be most intimately involved, JOHN F. KENNEDY
The new trends and complexities confronting American foreign policy makers and executors since World War II discussed earlier imply what is perhaps the fundamental feature of current policy making in foreign affairs: It is a shared prerogative of a great number of individuals and organizations standing in uncertain relationship to each other. Policy making can thus be viewed as a complicated process by which these actors work with and against each other to evolve and carry out proposed courses of action. Since the primary focus here is the role played by one group of individual actors, the overall characteristics of this policy process need to be examined in some detail. This chapter is devoted to that task, so that those that follow can more intelligibly illuminate the place of the country director (CD) within the larger process.
Policy Making, Decision Making, and Problem Solving
A first concern must be concept clarification, for the terms policy making, decision making, and problem solving are frequently used indiscriminately. The most common definition of policy links goals and desired outcomes. In this usage, "policy operates in the future," defining a nation's interests, objectives, or responsibilities, or, more bluntly, asking the question, "What difference does it make to us?" Built on broad goals (e.g., a world of independent nations, a world free of aggression, a world where personal freedoms are secured, a world of equal rights and opportunities) policy is seen as "any governing principle, plans, or course of action" designed to achieve such goals. Similarly, policy may be more specific (e.g., to improve the nation's balance-of-payments position), without distorting the future orientation or linkage of goals and guiding principles. More formally, policy might be seen as "an authoritative government decision to seek X or do Y."
A number of practitioners and scholars, however, find such a definition too limited, arguing it implies too direct a link between intention and action. For them, a particular policy can only be determined by looking back on a whole series of related decisions and actions which together define a nation's approach to a particular geographic or substantive area. (CD #1) This interpretation allows policy to be conceptualized as the result of action taken by a number of individuals and organizations, rather than as a single pronouncement or decision. Thus, Braybrooke and Lindblom have suggested that decision making and policy making can be differentiated in that
the latter term encompasses both decision making and the course that policies take as a result of interrelations among decisions and/or in which the latter term incorporates certain political processes, in addition to analytical processes, into the determination of courses of action.
Unfortunately, this interpretation deprives us of the primary usage of practitioners, who constantly refer to the existence or absence of a "policy" for a particular situation or nation. The incompleteness of either of these approaches suggests a combination usage is needed. Snyder, Bruck, and Sapin argue that two components are involved: action (past, current, or projected), and rules to guide action (which may outline the substance of response to some future situation, the occasion or conditions under which a particular response will be made, or an interpretation which will be placed on some future event). Thus,
policy can be anticipatory, cumulative, specific, and general. "To have a policy" means action and/or rules with respect to a problem, contingency or event which has occurred, is occurring, or is expected to occur.
In other words, policy making is the most general of these terms. The term "policy-making process" will be used to label the entire pattern of complex interactions between officials and organizations which result in decisions and in specific courses of action.
At the other extreme, problems and problem solving refer to the occasion for action or decisional situation, the stimulus that leads one or more officials to reexamine a situation with a view toward action or preparation of guidance for future action. The problem can be viewed as a threat or as an opportunity, in that innovations as well as reactions can result from the same sorts of circumstances. The advantage of this usage is that it parallels that of most officials, including the country directors interviewed for this study.
The third term, perhaps most frequently used by students of foreign policy, is decision making, but it differs conceptually from the other two. While decisions (choices among alternatives) play a part in both policy making and problem solving, they can be differentiated from the former in that they are limited in time (when a decision is "made" or becomes official), while policy is ongoing and composed of a series of decisions and actions. Similarly, decision making can be distinguished from problem solving:
Decision making is that thinking which results in the choice among alternative courses of action; problem solving is that thinking which results in the solution of problems.
While a choice among alternatives may be an important element in confronting a problem, the two concepts are separable in theory and practice. Much activity is reactive, and in such situations little if any choice among alternatives may be involved. Once a problem is defined, a decision often exists only in the sense of "making up one's mind" to take or refrain from taking a particular action. This is analogous to Simon's "programmed decision," where action follows from an established routine once a problem is assigned to a particular category. Even if a problem is not routine, constraints may be such that to speak of meaningful alternatives is misleading. As a former ambassador once explained,
a foreign policy problem is like a boulder rolling down a hill. If you try to get under it and hold it back, you'll get crushed; but if you can put a small stick or stone in its path, you may be able to influence where the boulder finally comes to rest. (CD #2 — paraphrase)
Concepts of Choice: Rationality vs. "Muddling Through"
At times, of course, choice among alternatives in solving problems or setting policies is possible, and it is to these instances that rational models of decision making, generally derived from economic theory, have been applied. Lindblom has suggested (without subscribing to) a picture of "rational" decision making which is representative of many such models:
1. Faced with a given problem,
2. a rational man first clarifies his goals, values, or objectives, and then ranks or otherwise organizes them in his mind;
3. he then lists all important possible ways of — policies for — achieving his goals
4. and investigates all the important consequences that would follow from each of the alternative policies,
5. at which point he is in a position to compare consequences of each policy with goals
6. and so chooses the policy with consequences most closely matching his goals.
This describes the process by which an individual rational man reaches a decision (makes a choice). Only minor variations are suggested by those who subscribe to the rational approach in cases where a decision is made by a collective process. One recent presidential aide, well aware of the differences between ideal and practice, puts the rational approach into the context of national security policy:
Theoretically, it would be desirable to undertake, for each important decision, a series of carefully measured, carefully spaced steps, including ideally the following:
First: agreement on the facts;
Second: agreement on the overall policy objective;
Third: a precise definition of the problem;
Fourth: a canvassing of all possible solutions, with all their shades and variations;
Fifth: a list of all possible consequences that would flow from each solution;
Sixth: a recommendation and final choice of one alternative;
Seventh: the communication of that selection;
Eighth: provision for its execution.
The numerous critics of rational models can be divided, broadly speaking, into two schools: those who would modify the model to account for lack of certainty, incomplete information, and other unrealistic assumptions; and those who suggest that even the modified rational model is neither descriptively accurate nor prescriptively desired. Most prominent of the first school are Herbert Simon and his associates. Their two principal alterations of the classical rational models are the ideas of "bounded rationality" and "satisficing." The major elements of their modifications are:
1. Optimizing is replaced by satisficing — the requirement that satisfactory levels of the criterion variables be attained.
2. Alternatives of action and consequences of action are discovered sequentially through search processes.
3. Repertoires of action programs are developed by organizations and individuals, and these serve as the alternatives of choice in recurrent situations.
4. Each specific action program deals with a restricted range of consequences.
5. Each action program is capable of being executed in semi-independence of the others — they are only loosely coupled together.
The other school, represented by Lindblom, agrees with Simon that the rational-deductive model is incapable of being fully realized, but goes even further to argue that it is "in most circumstances and most connections, fruitless and unhelpful as [an] ideal." A further complication is that subscribers to the rational model assert policy analysis or problem solving is comprehensive or synoptic, not surprising in view of the rational model's use of deductive propositions which logically should cover all situations. Unfortunately (in Lindblom's view), the synoptic ideal has a number of weaknesses: It is not adapted to man's limited problem-solving capabilities, to inadequacy of information, to the costliness of analysis, to failures in constructing a satisfactory evaluative method, to the closeness of observed relationships between fact and value in decision making, to the openness of the systems of variables with which it contends, to the analyst's needs for strategic sequences of analytical moves, or to the diverse forms in which policy problems actually rise. While this critique parallels that of Simon in many ways, it does not lead to a modified rational model, but instead to an approach which is in some senses nonrational, and certainly noncomprehensive. This strategy — disjointed incrementalism — has the following characteristics:
1. The analyst compares and evaluates only policies differing marginally from existing policies (margin-dependent choice).
2. A restricted number or variety of alternatives are considered.
3. Only a restricted number of important consequences are considered for any given alternative.
4. The problem is continually redefined, with ends adjusted to means as well as the other way around — this is a means of simplification, since the incremental approach allows for countless adjustments of both kinds, making the problem more manageable.
5. Serial analyses and evaluations are possible, meaning that there is no one correct solution for all time, but that any problem is subject to a never-ending series of attacks.
6. As a consequence, disjointed incrementalism has a remedial orientation, which is geared more to alleviation of present, concrete imperfections than to promotion of future goals.
Thus far, we have been dealing with individuals, and with "intellectual factors" which come into play in arriving at a decision. Once the restriction that a decision is made by a single official is dropped, however, it is necessary to come to terms with the complications introduced by multiple participants.
One approach, associated with the rational-deductive school, is to posit essentially the same process for groups and individuals. Downs, for example, suggests that an individual decision maker, faced with a perceived need for new action, goes through a series of steps related to formulation, analysis, and evaluation of alternatives, development of a strategy of action, action impact, feedback of results, and performance reassessment. While the terminology differs somewhat, the model is still one of rationality. He argues that organizational decision making varies from this model because of the involvement of many individuals, with the following consequences:
— the various steps in the decision and action cycle are carried out by different persons.
— an organization must generate numerous conflict-controlling and consensus-creating mechanisms because its members have widely varying perception apparatuses, memories, images of the world, and goals.
— organizational decision making involves the following significant costs of internal communications that have no analogs within an individual:
a. losses of utility due to errors of transmission
b. losses of utility (for the ultimate uses of the data) due to distortion
c. resources (especially time) absorbed in internal communications
d. losses of utility due to overloading communications channels in the short run.
The steps of the rational model still exist, although performed by different individuals within the organization.
But if actions taken and problems solved are the topics of interest, it may be very misleading to concentrate on the cognitive/emotional elements of the individual decision process as generalized to a group context. These outputs may emerge as much by social processes of accommodation and adjustment as they do from rational (ends-means) analysis.
In addition to substituting disjointed incrementalism for rationalism in characterizing cognitive aspects of individual decision making, Lindblom has suggested that social processes of policy making can be positioned on a continuum from complete centralization (the group analog to individual rationality) to total absence of central coordination. Outcomes result from "partisan mutual adjustment" (PMA), in which a multiplicity of interdependent decision makers each pursues his own rather than generally shared goals, except as controlled by other decision makers ("partisans") or by central supervision. The end result, descriptively, is incremental policy change, brought about by application of one or more techniques of adaptive and manipulative adjustment. Thus PMA is the collective counterpart of the individual strategy suggested by Lindblom and Braybrooke discussed earlier. Real life situations, according to Lindblom, exhibit a mixture of central coordination and partisan mutual adjustment, and "which of the two alternatives is more suitable varies from circumstance to circumstance."
Without duplicating Lindblom's several discussions, two points are of particular importance here. First, partisan mutual adjustment may be more desirable than central coordination, because it avoids centralization of power and may be more efficient:
Central coordination may impose analytic and regulatory tasks on a central authority beyond its capacities. ... Each limit [on man's analytic capacities] is also a limit on a central coordinator's capacity to analyze the desired relations among subordinate policy makers and to regulate them.
This preference for PMA directly opposes the argument that a totally rational policy-making process, while unattainable, should still remain the goal.
Second, rational policy analysis, however extended, is inadequate in collective, group, or systemic applications because it is impossible "through analysis to find policies that are everywhere accepted because proven to be correct." Yet problems must be solved, decisions made, and policies developed for society. This is done by the "play of power." Since power is always held by many individuals, "policy is made through the complex processes by which these persons exert power or influence over each other," such activity being part of the overall process of partisan mutual adjustment. While the "play of power" features cooperation among specialists and proceeds according to rules, a third characteristic is most important here:
The play of power is not a substitute for policy analysis, simply resolving those issues left unsettled by analysis. Instead, policy analysis is incorporated as an instrument or weapon into the play of power, changing the character of analysis as well.
Thus any model that merely carries over individual decision processes, rational or otherwise, into a group context is likely to be incomplete and incorrect, because analyses of policy by individuals will differ from that by collectivities.
In short, any description of the policy process must account both for the limits of rationality faced by the individual and for the social and political aspects of complex collective interactions. As Bauer argues, policy making must be conceived of as "a social process in which an intellectual process is embedded." Because of this social and intellectual intermingling, the rational and mutual adjustment models can be visualized as ideal types near the extremes of a continuum, with real world situations tending toward one or the other.
(Continues...)
Excerpted from Foreign Policy and the Bureaucratic Process by William I. Bacchus. Copyright © 1974 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
