Changing Course: When Governments Choose to Redirect Foreign Policy

Author(s): Charles F. Hermann

Source: International Studies Quarterly , Mar., 1990, Vol. 34, No. 1 (Mar., 1990), pp. 3-21
Published by: Wiley on behalf of The International Studies Association

Stable URL: https://www.jstor.org/stable/2600403

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

The International Studies Association and Wiley are collaborating with JSTOR to digitize,
preserve and extend access to International Studies Quarterly

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
International Studies Quarterly (1990) 34, 3-21

Changing Course: When Governments Choose

to Redirect Foreign Policy

CHARLES F. HERMANN

The Ohio State University

We are in a period of profound change in international relations and for-

eign policy. These developments call attention to the state of our knowledge
about change processes in governmental decisionmaking. This essay re-
views the contributions of several areas of conceptual literature and pro-
poses a scheme for interpreting decisions in which a government decides to
change policy direction. Foreign policy changes can be placed on a contin-
uum indicating the magnitude of the shift from minor adjustment changes,
through both program and goal changes, to fundamental changes in a
country's international orientation. These degrees of change are examined
with respect to four change agents: (1) leader driven; (2) bureaucratic advo-
cacy; (3) domestic restructuring; and (4) external shock. The phases of
decisionmaking mediate between sources of change and the magnitude of
change in policy. The essay concludes with an examination of propositions
that suggest conditions under which the phases of decisionmaking can in-
crease the likelihood of major change.

Introduction

Change is a pervasive quality of governmental foreign policy. Although no era is

immune to dramatic changes in world affairs, we are living in a period with numer-
ous remarkable examples. The countries of the European community have commit-
ted themselves to an integrated, single market by 1992. Spain and several previously
authoritarian governments in Latin America have begun new experiments with de-
mocracy. Regional conflicts in southern Africa, Afghanistan, and between Iran and
Iraq have undergone promising shifts toward conflict reduction as a result of major
changes in the foreign policies of the involved parties. Of course, to many concerned

Editor's note: Charles F. Hermann is President of the International Studies Association 1989-90. This article is an
edited version of his presidential address, delivered at the 30th Annual Convention of the International Studies
Association, London, UK, March 30, 1989.

Author's Note. This essay is the presidential address delivered at the International Studies Association Annual
Meeting held jointly with the British International Studies Association, London, March 28-April 1, 1989. A
number of people commented on the ideas presented in this paper and in an earlier draft. These people included
an informal study group that met at the residence of James N. Rosenau, January 6-7, 1989; a conference held at
the Mershon Center, The Ohio State University, February 17-18, 1989; and a Mershon Center seminar group on
the psychological dimension of decisionmaking. My appreciation is extended to all these people and especially to
Margaret Hermann, Richard Herrmann, Dwain Mefford, James Voss, Rober-t Billings, Brian Ripley, Mar-garet
Karns and the Editors of ISQ. I am also indebted to Sandy Jones for her work as my graduLate resear-ch assistaint.

? 1990 International Studies Association

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 4 Changing Course

with East-West relations, an air of excitement and anticipation has resulted from the
prospect of profound changes in the relationship between the United States and the
Soviet Union and in the domestic upheavals in Eastern Europe. As Hoffmann
(1989:84) has observed: "There are periods of history when profound changes occur
all of a sudden, and the acceleration of events is such that much of what experts write
is obsolete before it gets into print. We are now in one of those periods."
The prospect of fundamental change in the relationship between the western allies
and the communist nations illustrates the importance of making concerted efforts to
understand profound social, economic, and political change and the conditions that
foster them. The severe antagonism, threats, and confrontations of the Cold War
have structured much of international relations for the last four decades. Foreign
policy agendas not only between East and West but between countries in each alli-
ance and between developed and developing nations have often been driven by Cold
War concerns. At least for some governments, other problems have been indefinitely
deferred or reinterpreted into East-West security issues.
For those of us concerned about Soviet-American relations, the recent changes
raise basic questions: How profound are the changes underway? How will policy-
makers and others recognize that they are indeed fundamental and not mere win-
dow dressing? If the problem that has preoccupied security and foreign policy issues
since World War II is being transformed, what is the nature of the new problem or
problems? As Allison (1989) recently noted, if the Cold War ends, it poses a major
challenge: "Perhaps most difficult of all, we will have to think again, to stretch our
minds beyond the familiar concepts and policies of containment." The conceptual
reorientation to which Allison refers is an integral part of any fundamental change
of course in foreign policy.
Changes that mark a reversal or, at least, a profound redirection of a country's
foreign policy are of special interest because of the demands their adoption poses on
the initiating government and its domestic constituents and because of their poten-
tially powerful consequences for other countries. Wars may begin or end. Economic
well-being may significantly improve or decline. Alliances may be reconfigured.
Sometimes the entire international system is affected, as when the Cold War began
after 1945. Many of these dramatic changes in course occur when new governments
with different perceptions of the environment and new agendas come to power. The
current changes in the Soviet Union are an example. At least initially the changes
have been associated with the emergence of Gorbachev and his associates. A basic
question for the future is whether the changes would continue if Gorbachev no
longer held power.
In fact, there may be a tendency to conclude that regime change is virtually the
ony way to achieve profound shifts in a nation's foreign policy. Reflection, however,
will reveal cases in which the same government that initiated a course in foreign
policy recognizes that significant changes must be undertaken. After engaging in war
with Israel, President Sadat of Egypt dramatically changed direction with his trip to
Jerusalem. The administration of Lyndon Johnson, after having Americanized the
Vietnam War, changed course and began to negotiate U.S. withdrawal.' Fidel Castro
of Cuba, once an advocate of the export of revolution by military means, changed his
mind. President Nixon, who for most of his career saw a monolithic aggressive

I Some debate exists as to when, if ever, President Johnson changed course with regard to the Vietnam War.
Karnow (1984:493-94, 565-66) notes that Johnson authorized negotiations with the North Vietnamese in 1967,
although it is unlikely that he was prepared to reverse his administration's policy on the war at that time. It is more
defensible to argue thatJohnson had shifted his position by the end of March, 1968, when he annouLnced he wouLld
not seek another term as president. Negotiations with the North Vietnamese did open in May 1968 in Paris, with
the United States essentially calling for mutuLal withdrawal.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 5

communist movement, initiated a dramatic change in United States policy toward

China. Under what circumstances do these kinds of changes occur in which an
existing government recognizes that its current course is seriously inadequate, mis-
taken, or no longer applicable? What are the conditions under which self-correcting
change may arise? The answers to such questions have important practical and
theoretical implications for the study of foreign policy. This is so not because major
changes in foreign policy are always changes for the better-clearly they sometimes
are for the worse. These issues are important because of the magnitude of practical
consequences they generate for people and institutions and because understanding
the conditions for change poses one of the most difficult theoretical problems for
scholars and policymakers alike.

Defining Major Foreign Policy Change

Both the foreign policy and the international system perspectives have much to
contribute to our understanding of change. This essay examines change from the
perspective of the actor in the system; that is, it deals with foreign policy change. Let
us be clear about other ways in which this essay is focused and bounded. Reference
has already been made to the distinction between foreign policy redirection that
results from regime change or state transformation, and change that occurs when
the existing government elects to move in a different policy direction. We are con-
cerned here with the latter-what might be characterized as self-correcting
change-when the current actors change their course in foreign policy.
Just as Hereclitus observed that we cannot step in the same stream twice, so also is
a government's foreign policy constantly changing-usually with minor adjustments
or modifications in nuance. Our concern is not with such changes, but rather with
fundamental redirections in a country's foreign policy. Establishing what constitutes
fundamental foreign policy change poses many challenges.
Beginning with the concept of foreign policy, let us stipulate that it is a goal-
oriented or problem-oriented program by authoritative policymakers (or their rep-
resentatives) directed toward entities outside the policymakers' political jurisdiction.
In other words, it is a program (plan) designed to address some problem or pursue
some goal that entails action toward foreign entities. The program presumably speci-
fies the conditions and instruments of statecraft.
With this definition, foreign policy can be viewed as subject to at least four gradu-
ated levels of change:

(1) Adjustment Changes. Changes occur in the level of effort (greater or lesser) and/or
in the scope of recipients (such as refinement in the class of targets). What is done,
how it is done, and the purposes for which it is done remain unchanged.
(2) Program Changes. Changes are made in the methods or means by which the goal
or problem is addressed. In contrast to adjustment changes, which tend to be
quantitative, program changes are qualitative and involve new instruments of
statecraft (such as the pursuit of a goal through diplomatic negotiation rather than
military force). What is done and how it is done changes, but the purposes for
which it is done remain unchanged.
(3) Problem/Goal Changes. The initial problem or goal that the policy addresses is
replaced or simply forfeited. In this foreign policy change, the purposes them-
selves are replaced.
(4) International Orientation Changes. The most extreme form of foreign policy change
involves the redirection of the actor's entire orientation toward world affairs. In
contrast to lesser forms of change that concern the actor's approach to a single
issue or specific set of other actors, orientation change involves a basic shift in the

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 6 Changing Course

actor's international role and activities. Not one policy but many are more or less
simultaneously changed.

At different times, U.S. policy toward Vietnam illustrated all four levels of change.
Following the defeat of the French in Indochina and the separation of North and
South Vietnam, the United States Government pursued a goal of keeping South
Vietnam independent of North Vietnam and aligned with the United States. In the
late 1950s and early 1960s the American policy to pursue that goal involved military
and economic assistance to South Vietnam, including American military advisers.
The increase in military assistance during this period would be defined as adjust-
ment changes. With the introduction of American combat forces in 1965, a program
change occurred, followed by further adjustment changes as the level of American
military effort increased. The determination to return the fighting gradually to
Vietnamese troops and withdraw American forces marked a second program
change. A goal change occurred when American policymakers concluded that the
continued ability of the South Vietnamese to resist the North Vietnamese forces was
questionable, and when the U.S. elected to accept that outcome rather than reintro-
duce American combat forces.
Whether the highest form of foreign policy change-international orientation
change-occurred is more debatable. It can be argued, however, that after the
collapse of Vietnam, the United States experienced at least a decided shift in its
willingness to use large-scale force in its conduct of foreign policy. Together with its
changing relative economic status, this produced a decline in its hegomonic leader-
ship role. Certainly increased caution with respect to the use of force altered Ameri-
can policies toward Angola, Central America, and the Middle East. It is unclear
whether the invasion of tiny Granada represents the end of this American reluctance
to use force or simply further evidence of the lopsided military advantage that the
United States must have before it will introduce force.
In this essay major foreign policy redirection will be defined as the last three forms
of change-that is, change in means (program), ends (goal/problem), or overall
orientation. Reliable empirical differentiation is not always easy. In program change,
however, one would expect to find changes in the configuration of instruments, in
the level of commitment, and probably in the degree of expressed affect. All these
developments, plus policy statements and policy actions incompatible with prior goal
or problem stipulations-if not open rejection of prior goals-accompany goal/
problem changes. International reorientation involves dramatic changes in both
words and deeds in multiple issue areas with respect to the actor's relationship with
external entities. Typically, reorientation involves shifts in alignment with other
nations or major changes of role within an alignment.
We can now inquire about the conditions that promote major redirection in for-
eign policy; that is, changes in program, changes in goal/problem, or-even more
drastically-changes in international orientation. At least four areas of scholarship,
to some degree located in different academic fields of inquiry or disciplines, can
potentially contribute to this exploration. They are (1) domestic political systems, (2)
bureaucratic decisionmaking, (3) cybernetics, and (4) learning.

Some Applicable Areas of Scholarship

Domestic Political Systems

At the core of this perspective on foreign policy change is the assumption that the
rulers and their regime, that is, those who create governmental foreign policy, de-
pend for their continuance on the support of certain constituencies. Constituencies,
as defined here, are those entities whose endoresement and compliance are neces-

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 7

sary to legitimate and sustain the regime. They may be members of a ruling political
party (or a faction within it), the political clients in a client-patron system, a dominant
religious or ethnic group, military officers, major land owners, interest groups and
associations, or the leaders of key sectors of society.
Changes in the policy preferences or in the dominant alignment of these constitu-
encies, or-more drastically-changes in the nature of the political system itself
(such as in who constitutes constituencies or in the relationship between them and a
regime), presumably trigger changes in foreign policy. Certainly the nature of the
political system and its relationship to constituencies strongly influence the way in
which changes in that system affect foreign policy. Boyd (1987) differentiates the
way political change must operate in Third World states, in large communist sys-
tems, and in industrialized democracies (separating the last category into post-indus-
trial societies, pluralistic political economies, and neocorporatist political economies).
In most studies of political change, however, the focus has been on a single country,
most often on the United States. In the American case the emphasis has been on such
topics as public opinion, interest groups, political parties and opposition, elites or, at
a slightly different level, on coalitions and social/economic cleavages (see Almond,
1950; Cohen, 1973; Hughes, 1978; Holsti and Rosenau, 1984). After reviewing the
political change literature pertaining to three countries (United States, West Ger-
many, Soviet Union), Goldmann (1988:44) concludes that three dimensions influ-
ence the extent to which foreign policy is likely to change:

1. The degree of institutionalization, or roughly the extent to which the government

has become committed to the policy.
2. The degree of support, or roughly the extent to which the various actors in
domestic politics support, are indifferent to, or oppose the policy.
3. The degree of salience, or roughly the significance of the issue in the domestic
power struggle.

Domestic politics may affect foreign policy through several different dynamics:

1. Issues become a centerpiece in the struggle for political power. Competing politi-
cal leaders and their supporters use a foreign policy position to differentiate
themselves from opponents. If those out of power succeed, then the foreign policy
changes. Alternatively, an existing regime may change its foreign policy to distin-
guish itself from opponents or to prevent defeat.
2. The attitudes or beliefs of the dominant domestic constituents undergo a pro-
found change. Attitude change becomes the underlying source of explanation and
some profound stimulus presumably creates a realignment in the views of many.
3. A realignment occurs of the essential constituents of a regime, or a revolution or
other transformation of the political system takes place. Thus, for example, when
a military junta seizes power from civilian political parties, the relevant constituen-
cies change. After the fall of the Shah, the entire political system changed in Iran.
The results are similar to those described in item one, except that foreign policy is
a consequence of political realignment rather than the centerpiece. Also, a change
in the system is always necessary in this case, whereas in item one the threat to a
regime's continuance may be enough to cause it to change policy. It is important to
recognize that restructuring or transforming the economic system also can be a
source of foreign policy change.

Bureaucratic Decision Making

The premise of the bureaucratic source of foreign policy change is that contempo-
rary foreign policy-even in small, weak, or authoritarian governments-tends to be
conducted by individuals in organizations. Information and intelligence is collected

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 8 Changing Course

and analyzed by individuals, usually professionals, working in governmental organi-

zations. Options and choices are made by ministers, agency heads, cabinet members,
or their representatives, and policies are implemented by one or more government
departments. Thus, to understand foreign policy change, it can be argued, one must
examine the bureaucratic conditions governing the policy process.
Intriguing evidence for the importance of this source of foreign policy change
comes from eight case studies assembled by Holsti (1982a). Holsti and his associates
examined nations that changed their foregin policy alignment (through shifts in
isolation, self-reliance, dependence, or diversification). These regimes engaged in
the most far-reaching type of change as categorized in this essay. Among the various
external and domestic sources of explanation the authors considered, only one fac-
tor was regarded as a powerful explanation of change in seven out of the eight cases:
the decisionmaking variables of personality and perception (Holsti, 1982b:2 11). In
the eighth case this factor was judged to have been somewhat less powerful, but was
still rated a "moderately significant explanation." (The next most frequently cited
source-non-military threats-was noted as a powerful explanation in only four of
the eight cases.).
Juxtaposed to the Holsti studies that found decisionmaking variables to be a major
source of foreign policy change are other works which suggest that the structure of
governmental organizations and the nature of political leadership normally act to
resist change. Etheredge (1985) reviewed a series of American foreign policies to-
ward Central America. From his analysis of American political processes and institu-
tions as well as political leadership, Etheridge responds negatively to the question
posed by the title of his book: Can Governments Learn? Goldmann (1988) develops a
broader framework using "stabilizers" that operate to maintain the direction of a
government's policy once established. Goldmann (1988:54-62) accounts for the con-
tinuation of policy by positing "administrative stabilizers."
The theoretical insight of these studies is that any foreign policy change must
overcome normal resistance in political, administrative, and personality structures
and processes. Among these resisters to change, bureaucratic inertia and standard
operating procedures were frequently cited. Of course, the studies of foreign policy
change and those concerned with resistance to change are not necessarily contradic-
tory-we simply have not yet postulated with specificity those conditions under
which the resisters are overcome and the foreign policy redirected.
To extrapolate further from the bureaucratic decisionmaking literature may be
risky. I am struck, however, that Holsti (1982b) found in cases of successful foreign
policy realignment that personal qualities and shifts in perception were powerful
explanations. Furthermore, the only factors he cited as resistant to policy changes
were bureaucratic initiatives that operated in two of the eight cases. The importance
of decisionmaking structures and processes seems confirmed by Holsti's work as well
as by the resistance studies of Etheredge (1985) and Goldmann (1988). We might
infer that change in established foreign policy will normally be resisted by various
structural elements of government, and the greater the shift the stronger the resis-
tance. Thus, program changes will be easier to adopt than goal/problem changes,
which in turn will be easier to achieve than reorientation changes in policy-where
resistance will be greatest. To overcome organizational resistance requires the pres-
ence of key individuals with the knowledge and the ability to circumvent normal
organizational constraints.

Cybernetics

Cybernetics and control theory approaches to dealing with change, of course, have
developed elsewhere (see Wiener, 1948; Ashby, 1954; Ashby, 1956), but they have
attracted periodic interest from those concerned with foreign policy (see Deutsch,

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 9

1966; Steinbruner, 1974; Miller and Thorson, 1977; Marra, 1985). An essential
feature of these approaches is that an agent, attempting to pursue some standard or
goal, continuously monitors a select stream of information from the environment
that indicates where he is in relation to that goal and how the relation has altered
across intervals of time. The agent engages in incremental self-corrective action in an
effort to close on the goal or remain in close proximity to the standard. This process
accounts for the association of cybernetics with the concepts of information (feed-
back) and control (steering).
The elaboration of such a.process would appear to be attractive for interpreting
adjustment changes in policy. The system of control might also be extended to cover
program changes as well. It is precisely to pursue a goal or to keep performance
within acceptable boundaries with respect to that goal that basic cybernetics seems
most applicable. The specification of appropriate indicators for monitoring feedback
and then initiating new actions parallels what has been characterized as maintaining
a foreign policy plan or program. Of course, this is not the kind of major change we
seek to understand.
A key problem in the cybernetics of complex systems is the means of dealing with
discontinuous changes in the environment. What happens when the conditions that
have prevailed up to a given point suddenly change so that the previous indicators
become inadequate or no longer apply? Imagine Europeanists who in the early
1950s thought the way to achieve greater integration that included West Germany in
an acceptable fashion was through the creation of a European Defense Community.
They monitored activities that would be important for achieving that goal and took
corrective steps to ensure that the process remained on track. Then the French
National Assembly rejected the idea. The environment had changed. Other correc-
tive steps to establish and maintain support for a European Defense Community in
other countries were no longer applicable. Can cybernetics handle such major envi-
ronmental changes?
The international environment in which most foreign policies operate is a very
large system with enormous uncertainty. Ashby (1956:244) has suggested that the
challenge for cybernetics is not normally size itself but "the variety in disturbances
that must be regulated against." Simon (1968) and others have suggested that an
approach to the problem is to decompose the environment into various subsystems
that are hierarchically arrayed-presumably according to the priority of one's goals.
Marra (1985:361) constructs a cybernetic model for U.S. defense expenditures using
the umbrella goal of survival, which he divides into a hierarchial set of subgoals:

In the context of the US defense expenditure policymaking process the concept

of "survival" is neither limited to nor primarily concerned with the physical
continuation of various decisionmaking groups. Survival can and does encom-
pass many different dimensions: e.g., national survival, political survival, fiscal
survival, survival in a bureaucratic sense, etc.

Policymakers attend to the highest priorities first: if everything is currently satis-

factory with the highest ordered goal (national survival, for example), then they can
move to the next level (political survival, for example). If feedback indicates trouble
with performance in that area, then policymakers scan key subareas within the do-
main of political survival (such as satisfaction of support groups or strength of
opposition) until the area of the problem is identified. Then corrective measures are
attempted.
This would appear to be the area of cybernetics with the greatest promise for
exploring the kind of major changes that are of current concern. It would seem to
postulate a decision system that can order goal preferences, recognize challenges to
each, and shift effort accordingly. That may impose a very demanding set of require-
ments on most foreign policy systems.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 10 Changing Course

Learning Approaches

Currently, considerable activity is underway in psychology and cognitive science in

an effort to understand learning processes. The general approaches of a generation
or so ago-associated with names like Hull-Spence, Skinner, and Mowrer-in many
respects have been replaced by a variety of more specific learning problems and
alternative explanations. Instead of general principles about reinforcement of re-
sponses through reward or punishment, researchers now have advanced alternative
conceptualizations addressed to different kinds and methods of learning. A common
point of reference among these varying approaches is the idea that learning involves
the acquisition of some new skill, ability, or knowledge that can be recalled and used
on some future occasion. Thus, change does not always imply learning. This assump-
tion of acquisition can be contrasted with cybernetic approaches where corrections
are performed according to previously established rules or operations.
Humans are expected to learn how to do a great many different kinds of things,
from riding a bicycle to solving algebra problems, from forming ways to get along
with family members to negotiating successfully with representatives of different
cultures over vital issues in dispute. At the present stage of scholarship different
models of the learning process have been advanced for understanding different
types of activities (see Glaser and Bassok, 1989). Perhaps most relevant to foreign
policy change are models for solving complex, ill-defined problems lacking any
correct answer. Problem-solving in this context is presumed to involve the organiza-
tion and management of structures of knowledge about different domains, which a
person assembles as a sort of road map or schemata. It also involves the ability to
make inferences from that knowledge. Learning entails not only the acquisition of
knowledge, but also the restructuring and modification of existing schemata. Sche-
mata can be envisioned as a series of mental models that a person uses to interpret
experiences and that enables the individual to cope with them. Learning can involve
various operations on these mental models, such as adding, deleting, and merging-
that is, processes that restructure them (see Stevens and Collins, 1980).
What triggers new learning? Again, different theories exist. Perhaps one of the
most basic explanations is failure. One's mental model does not adequately explain
experiences; coping is impaired. But beyond the recognition of failure is the need to
account for how new knowledge is acquired to revise or replace existing schemata.
Are repeated trials or exposures necessary? Must there be some kind of pattern-
recognition? Or can the observation of a single case provide the basis for what is
called explanation-based learning (DeJong, 1988)?
Learning in the form of problem solving may be quite instructive in considering
major foreign policy change. Leaders may replace goals in foreign policy or the
means for achieving them as a result of a basic restructuring of their own schemata
about the problem or the policy designed to address it. Thus, better understanding
of the conditions under which such reformulations occur could have significant
payoffs. Furthermore, as will be suggested below, the research on problem solving
may be instructive in considering the stages of decisionmaking that foster significant
foreign policy change.

Insights From Literature Review

Abstracting broad conclusions from such diverse literature risks serious misrepre-
sentation. Nevertheless, it may be instructive to offer a personal perspective sug-
gested by these areas of research.
In the domestic political system, two things are necessary to effect change in
foreign policy. First, there must be a change in that system and, second, that systemic

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 11

change must trigger a change in the government's foreign policy. One of the changes
that would fulfill these requirements is a fundamental alteration in the attitudes of
most members of the politically relevant segments of society. In the United States,
such changes have included shifts in beliefs as to what countries posed the greatest
threat to American security in the mid to late 1940s, and in evaluations of the merits
and costs of the Vietnam War effort. Attitudes may now be changing with regard to
whether there is likely to be a continuing Cold War with the Soviet Union. There are
other ways by which political system changes affect foreign policy. Among them is
the transformation of the entire political system, including the political-economic
system, as when a society moves from an agricultural-based economic system to an
industrial one.
In what have been characterized as studies of bureaucratic decisionmaking, major
foreign policy change appears to depend on mobilizing sufficient specialized human
talents to overcome or circumvent the organizational structures and processes com-
mited to the maintenance of existing policy. It is little wonder, therefore, that politi-
cal scientists and others adopting this framework tend to conclude that major foreign
policy change most often occurs with a change in government accompanied by the
appearance of new leaders. New leaders are sometimes best able to create organiza-
tional changes and make key new appointments.
Cybernetics, which can readily capture adjustment changes, can also interpret
more profound reconfigurations. What must occur for major change, however, is
the activation of some kind of "meta regulators" that move policymakers from one
problem to another or to higher goals necessary to maintain the system. In non-
mechanistic terms, it is not obvious what these regulators are. Perhaps they entail
powerful political and social forces capable of setting and resetting agendas and of
redistributing the government's effort and priorities.
Finally, learning approaches suggest that major foreign policy change can occur
when key policymakers who are confronting a problem restructure their mental
models or schemata of the problem. This can lead to a redefinition of the problem or
to a new understanding of the relationship between it and their policies.
These different areas of inquiry can be seen as suggesting different sources of
major foreign policy change. I have labeled these sources of change leader driven,
bureaucratic advocacy, domestic restructuring, and external shock.
Leader driven change results from the determined efforts of an authoritative policy-
maker, frequently the head of government, who imposes his own vision of the basic
redirection necessary in foreign policy. The leader must have the conviction, power,
and energy to compel his government to change course. Anwar Sadat's decision to
pursue a peace settlement with Israel following the 1973 war illustrates this source of
change. It would be instructive to explore whether Sadat underwent some significant
reconceptualization of his problem with Israel that contributed to this dramatic shift.
Certainly learning approaches might well provide insight into changes triggered in
this manner.2
Bureaucratic advocacy as an agent of change may seem a contradiction in view of
what has been described as the resistance of bureaucratic organizations to major
redirection of policy. The suggestion is not that an entire government becomes
seized with the need for change, but rather that a group within the government
becomes an advocate of redirection. This group may be located in one agency or
scattered among different organizations but with some means for regular interac-

2 AuLerbach (1986) pr-oposes a cognitive fr-amewor-k to accounllt fol tulr-inig-point decisionis that enitails fil-st a
perceptioni of failulr-e that genierates dissonance, which theni cain result in attituLde change alnd a(l ecisioll to chalnlge
policy. It is possible, but not cer-tain, that Sadat viewed the 1973 Walr as at failure.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 12 Changing Course

tion. To be effective, the advocates have to be sufficiently well placed to have some
access to top officials.
Unless the evidence from the external environment is very dramatic, officials in
the middle levels of government may be in a better position than their superiors to
receive signals that current policy is not working. They also may have the advantage
of a keen understanding of how the government works and what, therefore, must be
done to overcome resistance. For these reasons, the decisionmaking approach can be
an important means of interpreting the necessary conditions for bureaucratically
induced change. The Canadian government's move for greater independence from
the United States in the early 1 970s appears to be an example of change generated at
least partially in this fashion. In his case study, Holsti (1982c) attributes major impor-
tance to groups within the Canadian government. At one point he notes that "it is
clear that initiative lay with two subdivisions of the Department of External Affairs
and that the Cabinet basically responded" (Holsti, 1982c:97, emphasis in original).
Domestic restructuring refers to the politically relevant segment of society whose
support a regime needs to govern and the possibility that this segment of society can
become an agent of change. In writings on American foreign policy and domestic
politics much has been made of the breakdown of the Cold War consensus on the
basic goals of policy following the Vietnam War, Watergate, and the political activism
of the sixties and early seventies. More narrowly, one might link sharp reductions in
the American foreign agricultural aid program-with its supports for American
farmers-to the change in rural and urban demographics and to Supreme Court
rulings insisting that all Congressional districts represent roughly equal numbers of
people. These trends eliminated the powerful Congressmen with high seniority
from rural districts who engineered such aid programs. In different political systems
the dynamics of domestic restructuring will vary, but at their core is a common
theme: Foreign policy redirection occurs when elites with power to legitimate the
government either change their views or themselves alter in composition-perhaps
with the regime itself.
External shocks are sources of foreign policy change that result from dramatic
international events. Presumably most foreign policy change results from a percep-
tion by government leaders of some change or initiative (or lack of it) in the external
environment. Normally these events are hardly traumatic. Typically, their scope is
modest; they may be ambiguous; their immediate impact is limited. In short, unless
they are repeatedly reinforced by other events, most foreign stimuli are easy to miss,
misinterpret, ignore, or treat routinely. By contrast, external shocks are large events
in terms of visibility and immediate impact on the recipient. They cannot be ignored,
and they can trigger major foreign policy change. The Vietnam Tet Offensive,
Sadat's visit to Jerusalem, the 1971 action of the Nixon Administration to terminate
the convertibility of dollars for gold and devalue the dollar are all examples of
external shocks. In each case major events contributed to significant foreign policy
change in one or more governments. With a cybernetic approach such cases repre-
sent the kind of feedback that would trigger foreign policy system regulators for
monitoring and coping with discontinuous change. Accordingly, they would appear
to be candidates for explanation from this conceptual approach.
Obviously, there is likely to be interplay between these sources of foreign policy
change. They may work in tandem, or one (such as an external shock) may activate
another (a leader driven initiative) which in turn creates the redirection of foreign
policy. Just as the actual agents of change may interact with one another, so also may
it be possible to integrate some of the theoretical explanations that provide the
underpinning for various interpretations of change.
Consider a final observation from the review of various approaches to the explana-
tion of change. All the material examined seems explicitly or implicitly to assume

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 13

that change is driven by failure. The inadequacy of an existing schemata to account

for some critical experience forces the individual into a learning mode that may lead
to a restructuring of his or her mental model. Information that existing policy is not
performing properly motivates advocates of change in a government and serves as
ammunition in their struggle with the bureaucracy. Realignment of the domestic
system often results from large-scale discontent with the existing government or the
failure of the system to meet the demands of present or newly powerful constituents.
Particular types of negative feedback or discrepant information must trigger the
overarching cybernetic control system.
If one assumes that the redirection of foreign policy must result from explicit
decisions to change course, then a shared task for each theoretical perspective is to
determine how information about failure or potential failure enters the decision
system and under what conditions it actually triggers major change.
An assumption of this essay is that governments change their foreign policy
through a decision process. Up to this point we have considered various degrees of
change in policy that need to be explained and we have proposed various agents of
change. But intervening between agents and change is decisonmaking. At the risk of
implying a more linear relationship than is likely to exist, the linkage can be diagram-
med as follows:

Primary Government Extent of

Change Decison to Foreign Policy
Agents Change Course Change

leader driven program change

bureaucratic advocacy - decisionmaking - - goal/problem change

process
domestic restructuring international
orientation change
external shock

FIG. 1. The mediating role of decision processes between change agents and degree of
policy change.

In other words, to effect a change in governmental foreign policy, agents must act on
the governmental decision process. The decision process itself can obstruct or facili-
tate change. To pursue the possible theoretical underpinning of foreign policy
change one step further, let us consider stages of the decision process and the
conditions under which discrepant information (information about possible failure)
can produce decisions to alter the course of foreign policy.

Decision Making Stages in Changing Policy

Foreign policy begins with a problem-a threat or opportunity-that motivates con-

cern. Typically, foreign policy problems can be characterized as what Simon (1968)
and others (such as Reitman, 1965, and Voss and Post, 1988) have called "ill-struc-
tured" problems. Ill-structured problems, as contrasted with those that are well
structured, are not clearly defined at the beginning of the decision process. The
boundaries or dimensions of the problem are ambiguous and the preferred goal or
outcome may be vague. Much of the effort of policymakers must be devoted to
defining the nature and implications of the problem. For example, it will be neces-
sary to determine from a great deal of potentially relevant information, what will
actually be required to define the problem. In the process of structuring the prob-

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 14 Changing Course

lem, the basic nature of a possible solution will be increasingly constrained. Problem
solving of this sort-and of course we are speaking of a continuum-has substan-
tially different properties from that associated with well structured problems. Re-
search on the distinctions between such problems continues, and could be important
for our understanding of the conditions for fundamental foreign policy change. For
significant change in policy to occur, the decision process must operate to promote
actively a reformulation and to overcome the usual resistance. We must conceptual-
ize the decision process for the management of foreign policy problems so as to gain
insight into how changes in course occur.
The stages of decisionmaking have been variously conceptualized (see Mintzberg,
Raisinghani, and Theoret, 1976), but perhaps more important than any particular
configuration of the sequence is the recognition that the process is not linear. For
major problems, decisionmaking frequently involves cycles and pauses rather than
an orderly process in which each stage occurs only once and always leads directly into
one subsequent step until an outcome is reached. With that caveat in mind, let us
consider a possible configuration of the phases of the decision process and suggest
what developments at each stage are necessary for major change to occur. The seven
proposed stages are:

1. Initial policy expectations

2. External actor/environmental stimuli
3. Recognition of discrepant information
4. Postulation of a connection between problem and policy
5. Development of alternatives
6. Building authoritative consensus for choice
7. Implementation of new policy

Phase 1: Initial Policy Expectations

Changing course in foreign policy assumes an a priori foreign policy. Any enumera-
tion of the decision stages in foreign policy change must include the policymakers'
existing expectations concerning the effects to be produced by the existing policy.
These expectations form standards against which the policy's performance can be
evaluated. What had the policymakers anticipated their existing policy would achieve
with respect to the problem confronting them? When a major power provides eco-
nomic assistance to a developing country, what results do the donor country's leaders
expect? Will it reduce the immediate suffering of the recipient's population? Will it
provide necessary infrastructure to improve the economic growth of the recipient?
Will it induce the recipient government's loyalty to the donor or the greater loyalty of
the recipient population to its government? Will it be widely recognized at home and
abroad as a generous humanitarian act? Obviously, how the policymakers character-
ized the original problem when the policy was adopted shapes their expectations for
the policy's performance-the policy should solve the problem or reduce its effects.
Alternatively, it should realize an opportunity.
If the original problem is poorly specified, then it is unlikely that the expectations
for policy can be very clear. If disagreements continue to exist among policymakers
over the course they should be following, even after the policy is initiated, then
various policymakers may hold different expectations. Furthermore, if subsequent
events negate certain expectations, policymakers may then consciously or uncon-
sciously shift their expectations to emphasize other outcomes that cannot be charac-
terized as failures. However characterized, policy expectations, generated either by
the policymakers themselves or imposed upon them by others, create standards for
subsequent judgments of success or failure.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 15

Several propositions about the effect of expectations on the conditions that can
lead to change suggest directions for possible inquiry.

(a) The more clearly specified and articulated the problem, the more likely are the
expectations for policy performance to be specific and clear.
(b) The more clearly specified the policy expectations, the less ambiguity arises in
judging whether the policy fulfills or fails to fulfill the expectations.
(c) When policy expectations are not fulfilled, policymakers tend to change expecta-
tions for the policy or to attribute its lack of success to external events.
(d) The more numerous the agencies of government who must agree upon a policy
for its adoption and implementation, the more policy performance expectations
will be general and nonspecific and the more likely are the expectations to be
multiple and not necessarily consistent. Also, the more numerous the agencies,
the greater the resistance to acknowledging policy failure.
(e) If policy expectations are specific and the effects of policy publicly visible, then it
is more difficult for policymakers to avoid the evaluation of the policy against
those expectations.
(f) The shorter the time between the implementation of policy and observable
actions by the objects of the policy, the more likely are the policymakers to
attribute the actions to their policy-to see a causal linkage: conversely, the
greater the time between implementation and the object's action, the less likely is
the behavior to be attributed to policy.

Policy can be changed for various reasons (for instance, its success may eliminate
any further need for it), but if one invokes cybernetics or a form of failure-induced
learning to account for change, then the policymakers must accept some kind of
causal connection between what their policy will do and the state of the problem of
concern to them.

Phase 2: External Actor Responses and Other Environmental Stimuli

In any information-based interpretation of foreign policy change, developments

arise in the environment-most likely they are unexpected-that produce a stimulus
for examining an existing policy. The international occurrence may be a direct
response to the original policy or a completely independent event that can be under-
stood to have implications for that policy. In addition to activities by the intended
objects of an actor's policy, there may be actions by third parties or natural events
that transform the problem. The effectiveness of the United States grain embargo
against the Soviet Union following its invasion of Afghanistan in late 1979 was
influenced not only by dissent on the part of some of America's allies, but also by
previous good weather in several parts of the world that resulted in bumper grain
crops. Some foreign policy changes require joint action, and therefore cannot occur
without the appropriate external response. For example, Sadat's trip to Jerusalem
required the consent of the Israeli government.
Most important is the characterization of the external stimulus as it affects the
actor's policy and its purposes. Is it consistent or inconsistent with the actor's policy
expectations? Does it pose a threat or an opportunity? Is it time-urgent or open-
ended?
To explain a decision to change policy, the characteristics of the environmental
stimuli must be understood in terms of how they are perceived by policymakers. The
following propositions are offered to that end:

(a) Stimuli from the environment that are discrepant with the actor's policy are
more likely to occur when that policy threatens the interests of other actors.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 16 Changing Course

(b) Stimuli from the environment that are discrepant with the actor's policy are
more likely to occur when that policy expects major commitments from other
actors.
(c) Stimuli from the environment that are discrepant with the actor's policy are
more likely to occur when that policy is perceived by others to be ambiguous in
its expectations or require resources that others do not believe they cain allocate.
(d) Stimuli from the environment that are discrepant with the actor's policy are
more likely to occur when that policy requires action in concert by several other
actors.
(e) Stimuli from the environment that are discrepant with the actor's policy are
more likely to occur when that policy creates demands that are inconsistent with
existing norms, principles, and structural arrangements governing those to
whom the policy is addressed.

Phase 3: Recognition of Discrepant Information

External stimuli serve as cues or signals that potentially form feedback to the policy-
makers. Presumably when an external development generates information that is
inconsistent with policy expectations, or offers new evidence about the nature of the
problem, it becomes a signal rather than background noise for the policymakers.
The classic difficulty is whether the cue will be recognized and how it will be inter-
preted.
Given that individuals and organizations must deal with potentially discrepant
information all the time, the issue is when it becomes impossible to ignore it or
accommodate it within an existing schemata. Of course, it may be possible to recog-
nize that some difficulty has arisen that requires only some adjustment of the policy,
but no major redirection. For major foreign policy change to occur, it is necessary for
authoritative policymakers to conclude that their prior formulation of the problem,
their mode of dealing with it, or both, no longer accomodate information received
from the environment.
Some of the conditions affecting detection of environmental signals are suggested
by the following propositions:

(a) Policymakers are more likely to recognize discrepant information concerning a

policy if they anticipate that there is an available practical solution or if they are
generally confident of their success in managing similar problems in the past.
(b) Recognition of a policy discrepancy is more likely if those in a position to monitor
the environment do not have a personal or professional investment in the contin-
uation of the present policy or if they have continuing reservations about its
merits.
(c) Recognition of a policy discrepancy is more likely when the polity has multiple
entities with differing missions that monitor the environment.
(d) Recognition of a policy discrepancy is more likely when the policymakers' sche-
mata is not strongly connected to other major policies or to an ideological belief
system.
(e) Environmental stimuli are more likely to be recognized if they involve extensive
use of resources, persist over a period of time, are reinforced promptly by
similar or parallel events, or come from a friendly or trusted source.
(f) Crisis stimuli (involving high threat, time pressure, and surprise) are more likely
to be recognized than noncrisis stimuli.
(g) A government engaged in a crisis or coping with several major problems is less
likely to recognize a stimulus offering an opportunity.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 17

Phase 4: Postulation of a Connection Between Problem and Policy

If policy is to change, a critical step is the identification of a causal connection

between the deepening problem and existing policy. The policymakers must con-
clude that their government's policy is either ineffective in dealing with the problem,
making the problem worse, generating new problems of substantial concern, or
costing much more than anticipated. Having committed themselves to the existing
policy, government leaders may find it extremely difficult to perceive its flaws or
failure. This widely acknowledged attribution bias suggests that policymakers will
attribute success in foreign affairs to their policy, but will attribute failure to external
factors (Tetlock and Levi, 1982). In cases of either failure or success, the connection
between current policy and a foreign policy problem may be incorrectly drawn.
Many developments in the international environment that occur after a policy is
implemented are totally independent events, but they may be perceived as associated
with the policy. Nevertheless, it is forging the connection-regardless of whether it is
actually correct-that becomes essential if reevaluation of policy is to occur.
If a link between a problem and current policy has been drawn by policymakers,
they may be inclined to look for explanations that are least disruptive to their exist-
ing schemata of the problem and their understanding of how their policy should
affect it. Thus, they may attribute difficulties to the policy's implementation while
preserving the policy goals and the problem schemata.
With these considerations in mind, we offer the following propositions:

(a) The shorter the time between the implementation of a policy and the recognition
of subsequent difficulties with the foreign policy problem it was designed to
address, the more likely are policymakers to perceive a causal association be-
tween them.
(b) Policymakers will more readily recognize a basic flaw in policy if they have made
repeated unsuccessful efforts to improve the implementation of that policy.
(c) Policymakers are more likely to recognize a causal connection between existing
policy and a foreign policy problem if that connection is advanced by a trusted
and respected source or was forcefully disputed when the policy was originally
being considered.
(d) Policymakers are more likely to recognize a causal connection between existing
policy and a foreign policy problem if the goals of that policy are lower in a
hierarchy of policy goals and if higher priority goals are being threatened by the
problem-that is, if the problem threatens to disrupt more important goals.
(e) Policymakers are more likely to recognize a causal connection between existing
policy and a foreign policy problem if discrepant information about the problem
fits within the categories of some other problem schemata known to the policy-
makers.

Phase 5: Development of Alternatives

No matter what difficulties they may recognize with existing policy, if policymakers
cannot find a means to reduce the problem then change is unlikely. For coping with a
problem two broad approaches exist. First, there can be changes in policy intended
to address the problem. Second, there can be changes in the definition of the prob-
lem.
The more frequently considered course is changing policy. To minimize the re-
structuring that people and governments must undergo, policymakers can be ex-
pected initially to explore ways of changing the conduct of the policy while retaining
the original goals. More drastic change occurs when the goals of the policy them-
selves are rejected. In Vietnam, the United States changed its goal from retaining an

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 18 Changing Course

independent South Vietnam allied with the U.S. to withdrawal of American combat
personnel under circumstances that would enable the South Vietnamese to continue
waging the war. In such cases, a new set of goals will replace prior ones. In other
difficult cases, the existing set of goals will be replaced by only vague ideas about the
preferred new direction. A period of incrementalism follows.
If major reconceptualization of the policy occurs, then the representation of the
foreign policy problem itself may come under examination. A problem may be
redefined, or simply declared no longer to be a problem. Thus, the basic postwar
security problem for the United States was at some point transformed from contain-
ing the Soviet Union to promoting peaceful coexistence.
One of two means to generate alternatives is search. Mintzberg and associates
(1976) suggest that alternative solutions can also be developed by designing new
options. Essentially, processes are undertaken to seek out already existing options
that can be borrowed and applied directly, or in a readily modified way, to the
present problem. The more laborious task-in most cases-of designing or invent-
ing a completely new option is more likely when search routines fail to generate
plausible options.
Illustrative propositions concerning change in this stage of the decision process
include the following:

(a) Reexamination of the goals of a foreign policy and the definition of the problem
they are intended to address occurs only when compelling evidence suggests that
modifications of the policy itself will not change the situation.
(b) Internal actors without vested interests in the current policy and with knowledge
of policy instruments other than those presently involved will more likely gener-
ate new policy alternatives that they can implement.
(c) More policy options are likely to be considered if they can be discovered through
search routines than if they are designed especially for the present problem.
(d) The more sensitive to contextual information and the more conceptually com-
plex are the policymakers who perceive a need for a change, the more likely are
multiple options to be perceived.

Phase 6: Building an Authoritative Consensus for New Options

Philippine President Ferdinand Marcos enjoyed strong support from the United
States Government until August 21, 1983, when Benigno Aquino was assassinated at
the Manila airport as he returned from exile in America. As evidence mounted not
only of the involvement of Marcos's government in the murder but also of widescale
corruption, an increasing number of American policymakers turned against him.
Evidence of the need to change American policy with regard to Marcos accumulated
over the ensuing months. But President Reagan was personally close to Marcos and
his wife, and no authoritative action could be taken to change U.S. policy until it had
the support of the American president. Not until February 23, 1986, when President
Reagan met in the Situation Room of the White House with his senior advisors-
most of whom had been persuaded for some time of the need to withdraw support
from Marcos-was the President finally convinced (see Karnow, 1989). A year and a
half had been devoted to building the necessary consensus for a policy change.
In many cases in foreign policy, an authoritative consensus involves more than one
policymaker but, regardless of the number, policy change cannot proceed until it has
been established. The choice stage of the decision process has been one of the most
thoroughly studied. Certainly this phase frequently triggers recycling back to early
stages in an attempt to resolve differences. It is well understood that some of the
means used to resolve differences in the choice stage can lead to bargaining and log-
rolling strategies that alter the direction, if not the very feasibility, of policy change.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 19

Again, a few propositions illustrate the range of considerations that affect the
likelihood of policy change at this stage:

(a) The more cognitively complex are authoritative decisionmakers, the more likely
are they to be willing to consider alternative policies.
(b) When power is shared among multiple actors, consensus is more likely when all
actors accept a common set of norms governing the political process and the
political legitimacy of the other participants.
(c) The more ideologically similar are the policymakers who must approve a change
in policy, the more likely will a consensus be reached on a change in policy.

Phase 7: Implementation of New Measures

The study of organizational behavior and bureaucratic politics emphasizes that the
decisionmaking process does not end with an authoritative selection of some new
policy. Unless the entire plan can be realized in a relatively simple act, such as a
speech by the head of government, the process of implementation can alter the
nature of the intended policy dramatically. If the previous policy had strong advo-
cates even after a change in course has been ordered, or if there are powerful
proponents of another option not adopted, then resistance to the new policy may
manifest itself during implementation. Thus it becomes important to establish
whether those individuals and agencies charged with carrying out a policy are fully
committed to it.
Lack of strong commitment to new policy may not be the only difficulty in imple-
mentation. Equally important is whether the objectives and procedures required for
implementation are clear and whether those charged with policy execution have
resources sufficient to the task.
We return to where we began. What are the policymakers' expectations for the
new policy? Are they clear as to how they expect the policy to alter the problem and
have they communicated those expectations unambiguously to those who must im-
plement the policy? Even more basic, has the problem that the policy is designed to
address been well defined in the course of the policy process? Sometimes the old
problem definition is rejected without adequate specification (particularly to others)
of how it has been revised.
These observations invite a final round of propositions:

(a) The greater the participation in all stages of the decision process by those
charged with policy implementation, the more likely are they to activate the
policy in a manner consistent with the policymakers' intentions.
(b) The more specific and explicit are the policymakers' expectations for the policy,
the more likely it is to be implemented in a manner consistent with those expecta-
tions.
(c) The more those charged with policy implementation are committed to the new
policy direction and prefer it over previous policy (or other policy options of
which they are aware), the more likely is the policy to be implemented fully in a
manner consistent with the policymakers' preferences.
(d) The more consistent the new policy is with the a priori mission and with available
resources of the implementing people and agencies, the more committed they
will be to the execution of the policy.
(e) The more attention the top leadership gives to follow-up and review of new
policy in the post-choice stages of the process, the more likely it is to be imple-
mented fully.

Of course, a characterization of the analytical stages of the decision process that

may be necessary for the emergence of a new direction in foreign policy does not

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 20 Changing Course

provide a theory explaining such changes. Nor do isolated propositions about the
conditions in those stages that may foster major change offer such a theory.
What we hope to have provided here is a platform or orientation that may pro-
mote theoretical inquiry. For example, do various sources of change (and potentially
associated theoretical approaches) primarily concern different stages of the decision
process? Theories of leader driven change may address initial expectations and how
the recognition of discrepant information results in a reformulation of the causal
linkage between policy and problem. The concept of bureaucratic advocacy may
focus attention-and suggest explanations for-the development of alternatives, the
construction of an authoritative consensus, and the implementation of new policy.
Theory focusing on external shocks invites further examination of external stimuli
as well as of the policymakers' expectations that proceed them and the recognition
that follows. In other words, various ways of conceptualizing the sources of change
address various analytical points in the process of policy redirection.

Conclusions

My intent in this essay is to join the ranks of those urging attention to the conditions
giving rise to major changes in foreign policy. We need a perspective that views
major change not as a deterministic response to large forces operative in the inter
tional system, but rather as a decision process. Of course, major shifts in interna-
tional political and economic systems can pose significant requirements for the modi-
fication of foreign policy. But policymakers can either anticipate these international
changes, respond just in time, or only after suffering dramatic consequences. Fur-
thermore, policymakers can act as agents of change in the absence of any over-
whelming systematic force. "We would rather be poor and our own masters than
slaves to a foreign power. We have sipped that bitter tea before" (quoted in Holsti,
1982d: 105). Guided by these words, General Ne Win and his associates on the
governing Revolutionary Council redirected Burmese foreign policy and plunged
the country into deep international isolation. Their action was taken without any new
international force compelling such change, but rather as the leadership's policy
preference.
We may have entered a period of human history where not only is the rate of
change accelerating in political, social, and economic arenas of domestic and interna-
tional life, but where we are on the dividing line between epochs. The performance
of existing policies to meet present and emerging needs must be carefully and
creatively examined. Both the opportunities and the dangers that can result from
failing to deal with the changes that beset us are too great to be ignored.
Under these circumstances it is not enough for those engaged in international
studies and foreign policy studies to examine regularities and patterns of association
under assumed conditions of ceteris paribus. We need a much more vigorous effort
to characterize the conditions that can produce decisions for dramatic redirection in
foreign policy. If there is a very real possibility that major dimensions of foreign
policy may need to undergo significant change, then we scholars urgently need to
improve our understanding of the conditions that can enable such changes as well as
to promote the exercise of wisdom in the redirection of policy that may result.

References

ALLISON, G. (1989) Success is Within Reach. New York Timles February 19.
ALMOND, G. A. (1950) The Amtier-ican Peop)le anid Foreign Policy. New York: Harcourt, Brace.
ASHBY, W. R. (1954) Design for a Brain. London: Chapman and Hall.
ASHBY, W. R. (1956) An Introduction to Cybernetics. London: Methuen.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms
 CHARLES F. HERMANN 21

AUERBACH, Y. (1986) Turning-Point Decisions: A Cognitive-Dissonance Analysis of Conflict Reduc-

tion in Israel-West German Relations. Political Psychology 73: 533-50.
BOYD, G. (1987) The Foreign Policy Consequences of Political Change. In Political Change and Foreign
Policies, edited by G. Boyd and G. W. Hopple, pp. 1-31.
BOYD, G. AND G. W. HOPPLE, EDS. (1987) Political Change and Foreign Policies. London: Pinter.
COHEN, B. C. (1973) The Public's Impact on Foreign Policy. Boston: Little, Brown.
DEJONG, G. (1988) An Introduction to Explanation-Based Learning. In Exploring Artificial Intelligence,
edited by H. E. Shrobe, pp. 45-81. San Mateo, CA: Morgan Kaufmann.
DEUTSCH, K. W. (1966) The Nerves of Government. New York: Free Press.
ETHEREDGE, L. S. (1985) Can Governments Learn? London: Pergamon.
GLASER, R. AND M. BASSOK (1989) Learning Theory and the Study of Instruction. Technical Report No.
11. Pittsburgh, PA: Learning Research and Development Center, University of Pittsburgh.
GOLDMANN, K. (1988) Change and Stability in Foreign Policy. Princeton, NJ: Princeton University Press.
HOFFMANN, S. (1989) What Should We Do in The World? Atlantic Monthly October, 84ff.
HOLSTI, K. J. (1982a) Why Nations Realign. London: Allen and Unwin.
HOLSTI, K. J. (1982b) Restructuring Foreign Policy: A Comparative Analysis. In Whay Nations Realign,
edited by K. J. Holsti. London: Allen and Unwin.
HOLSTI, K. J. (1982c) From Dependence to Diversification: Canada, 1972-8. In Why Nations Realign,
edited by K. J. Holsti. London: Allen and Unwin.
HOLSTI, K. J. (1982d) From Diversification to Isolation: Burma, 1963-7. In Why Nations Realign,
edited by K. J. Holsti. London: Allen and Unwin.
HOLSTI, 0. R. AND J. N. ROSENAU (1984) American Leadership in World Affairs. London: Allen and
Unwin.
HUGHES, B. B. (1978) The Domestic Context of American Foreign Policy. San Francisco: Freeman.
KARNOW, S. (1984) Vietnam. New York: Penguin.
KARNOW, S. (1989) Setting Marcos Adrift. New York Times Magazine. March 19, 50ff.
MARRA, R. F. (1985) A Cybernetic Model of US Defense Expenditure Policymaking Process. Interna-
tional Studies Quarterly 29: 357-84.
MILLER, R. A. AND S. J. THORSON (1977) Control Concepts in Theories of Governmental Decison
Making. In Mathematical Systems in International Relations Research, edited by J. V. Gillespie and
D. A. Zinnes, pp. 60-90. New York: Praeger.
MINTZBERG, H., D. RAISINGHANI AND A. THEORET (1976) The Structure of "Unstructured" Decision
Processes. Administrative Science Quarterly 21: 246-75.
REITMAN, W. (1965) Cognition and Thought. New York: Wiley.
SIMON, H. A. (1968) The Sciences of the Artificial. Cambridge, MA: MIT Press.
STEINBRUNER, J. D. (1974) The Cybernetic Theory of Decision. Princeton, NJ: Princeton University Press.
STEVENS, A. L. AND A. COLLINS (1980) Multiple Conceptual Models of a Complex System. In Aptitude,
Learning, and Instruction, edited by R. E. Snow, P. Federico and W. E. Montague, pp. 177-98.
Hillsdale, NJ: Erlbaum.
TETLOCK, P. E. AND A. LEVI (1982) Attribution Bias: On the Inconclusiveness of the Cognition-
Motivation Debate. Journal of Experimental Social Psychology 18(1):66-88.
Voss, J. F. AND T. A. POST (1988) On the Solving of Ill-Structured Problems. In The Nature of
Expertise, edited by M. T. H. Chi, R. Glaser, and M. J. Farr, pp. 261-85. Hillsdale, NJ: Erlbaum.
WIENER, N. (1948) Cybernetics. New York: Wiley.

This content downloaded from

146.75.160.29 on Mon, 20 Nov 2023 02:51:52 +00:00
All use subject to https://about.jstor.org/terms