How do nations act in a crisis? This book seeks to answer that question both theoretically and historically. It tests and synthesizes theories of political behavior by comparing them with the historical record. The authors apply theories of bargaining, game theory, information processing, decision-making, and international systems to case histories of sixteen crises that occurred during a seventy-five year period. The result is a revision and integration of diverse concepts and the development of a new empirical theory of international conflict.
Originally published in 1978.
The Princeton Legacy Library uses the latest print-on-demand technology to again make available previously out-of-print books from the distinguished backlist of Princeton University Press. These editions preserve the original texts of these important books while presenting them in durable paperback and hardcover editions. The goal of the Princeton Legacy Library is to vastly increase access to the rich scholarly heritage found in the thousands of books published by Princeton University Press since its founding in 1905.
Read an Excerpt
Conflict Among Nations
Bargaining, Decision Making, and System Structure in International Crises
By Glenn H. Snyder, Paul Diesing
PRINCETON UNIVERSITY PRESSCopyright © 1977 Princeton University Press
All rights reserved.
INTERNATIONAL CRISES AND INTERNATIONAL THEORY
This book may be viewed from two different perspectives. On the one hand, it is a study of how states and statesmen behave in international crises, and it uses several kinds of theory to describe and explain that behavior. From this perspective, the purpose of the analysis is to increase our understanding of crises, and the theories are used as analytical tools. From another point of view, the purpose is to improve and integrate the theories — chiefly systems, bargaining, and decision-making theories — using crises as an empirical source for the testing, revision, augmentation, and synthesis of theory. Here the empirical phenomenon — crisis — functions more as a means than an end of analysis.
We did not feel a need to choose between these two perspectives, since they are not opposed but complementary. When one theorizes about any empirical phenomenon, the theory will have broader prima facie applicability to the extent the phenomenon has some elements in common with other phenomena in its general empirical field. And when one tries to improve general theory in the field by focusing on a particular aspect of the field, one is bound to learn a good deal about that aspect. Thus we conceive of our work both as a theory of crisis behavior and as a contribution to the theory of international conflict or international politics generally. Our conscious attempt to integrate several types of theory in an empirical context strengthens, we hope, the latter contribution.
Progress toward a general theory of international politics will proceed fastest, one supposes, by focusing empirical research initially on phenomena that lie at the center of the field. Theory thus developed can then be extended outward, with appropriate qualifications, to the more peripheral aspects. Crises admirably meet this criterion of centrality. Conflict is central to all politics, especially international politics, and crises are conflict episodes par excellence. Lying as they do at the nexus between peace and war, crises reveal most clearly and intensely the distinguishing characteristic of international politics and the logical starting point for theorizing about it: the pervasive expectation of potential war, which follows from the "anarchic" structure of the system. Since war is always possible, the implicit or explicit threat of war is the ultimate form of political pressure and the ultimate means to security and other values. What most urgently needs theoretical description and explanation is how the perpetual shadow of war affects the behavior of states, and how they manipulate that shadow to advance and protect their interests. In placid peacetime this expectation of potential war remains in the background of the statesman's consciousness and its effects are muted, diffuse, and not easily observable. In a crisis, the expectation is dramatically elevated and its behavioral effects stand starkly revealed. Related core elements, such as power configurations, interests, images, and alignments tend to be more sharply clarified, to be activated and focused on a single well-defined issue. The policy choice between coercing or conciliating adversaries, which is a perennial and central dilemma in diplomacy even in ordinary times, comes to a head in a crisis and urgently demands resolution. Thus a crisis distills many of the elements that make up the essence of politics in the international system. It is a "moment of truth" when the latent product of all these central elements becomes manifest in decision and action.
The case for using crises as a data source for the empirical development of general theory rests also on the generality of the primary form of interaction between states in a crisis — that of bargaining, broadly defined to include coercion and deterrence as well as mutual accommodation. Bargaining is pervasive in political life, and especially so in international politics, where centralized authority is absent. The external and internal parameters of crisis bargaining, namely the power structure of the international system and the domestic politics and decision-making processes of the actors, are also parameters of non-crisis bargaining. Thus, whatever theory we can develop about crisis bargaining, as bounded and influenced by these parameters, should have some validity, with appropriate modification, for diplomatic bargaining in other contexts.
Of course, we cannot claim to be presenting here a "general theory" of international politics. Theories about international systems, bargaining, and decision making are not the only theories that would have to be linked on the road to that (possibly illusory) goal, though we contend that they must be central to any general theory. And despite what we have just said about the generalizability of theory developed from the study of crises, such wider relevance pertains mostly to the "strategic" dimension of international politics — i.e., the search for security and promotion of other interests by sovereign states via the accumulation, use, or threatened use of military force. The theory is less relevant, for example, to such things as the politics of international economics, revolution, "transnational" relations, international organizations, and other elements of international community. And of course crises have certain special characteristics — notably a high degree of conflict, the perception of a dangerous probability of war, and a high emotional content — which limit the general relevance of theory based on them. We would argue, nevertheless, that it is possible to "factor out" empirical features peculiar to strategic crises and hence to increase the general applicability of the present theory. Some of the non-strategic aspects mentioned above do often appear tangentially in our crisis cases — e.g., revolutionary disturbance of some kind occurs in almost all cases, international organizations occasionally get involved, and community restraints such as international law often moderate the coercive behavior of the parties. While this does not mean that a theory centered on crises contributes much to a general theoretical understanding of such matters, it does help to show their relationship to the strategic dimension. And it further supports the point made above that a strategic crisis tends to galvanize, clarify, and concentrate many important elements in international politics, and to reveal the interaction between them more explicitly than in other empirical contexts.
Our methodology was quite straightforward. We took certain abstract, deductive theories and models about bargaining, decision making, and international systems and tested their empirical validity, or at least their plausibility, against 16 case studies of crises, ranging from Fashoda, 1898, to the Yom Kippur "alert crisis" of 1973. Much of the theory was available in the existing literature; some of it we invented. We also developed a list of specific hypotheses — some of them deduced from the general theories, some abstracted from "conventional wisdom," others merely hunches — and also a checklist of theory-relevant items for our researchers to look for in the cases. We sought to maximize comparability of the cases by defining the concept "crisis" as precisely as possible and by choosing cases that essentially matched the definition. The range of cases spans two different international systems — the multipolar, non-nuclear system that preceded World War I and continued during the interwar years, and the bipolar, nuclear system that followed World War II. Thus we were able to observe the effects of international systemic variables on bargaining and decision making, and to determine what aspects of such behavior seemed to be similar across both systems.
The results of this confrontation of theory and history amount to a set of judgments about the empirical validity and usefulness of the various theories, and explicit or implicit suggestions for their revision, augmentation, and combination, at least for the study of crises and potentially for the study of international politics in general.
THE ANATOMY OF INTERNATIONAL CRISES
The term crisis has been used in so many different ways, in personal and domestic social contexts as well as in international affairs, that it has no generally accepted meaning. Consequently, we must stipulate a definition that suits our research purposes.
An international crisis is a sequence of interactions between the governments of two or more sovereign states in severe conflict, short of actual war, but involving the perception of a dangerously high probability of war.
We use the term sequence of interactions rather than situation because of the ambiguity and emptiness of the latter term. Sequence of interactions is more meaningful, first, because it is the kind of interaction going on between the states that gives their relations the character of "crisis" and because the term interaction ties in nicely with our dominant theoretical theme of bargaining. Second, the word sequence clearly denotes a span of time and also a certain relatedness between the specific instances of interaction — each instance is affected by the instances just past and by the contemplation of possible following instances.
Note that our definition says nothing about the amount of time covered by a crisis. Most previous analysts have emphasized shortness of decision time as one of the defining conditions of crisis, along with a related sense of urgency. While the notion of urgency is supportable in terms of a sense of danger and risk that the parties feel must be alleviated as soon as possible, short decision time is not a necessary characteristic of crisis. Many historical crises lasted for months, even a year or longer — e.g., the Morocco crisis of 1905-1906 or the Berlin crisis of 1958-1960.
The term governments of sovereign states needs no elaboration except to note what it excludes. Principally, it excludes revolutions and internal war except when intervention by outside governments leads to a danger of war between the intervenors or between one intervenor and the incumbent government. Thus the Dominican "crisis" of 1965 was not a crisis by our criteria because it was a revolutionary situation in which only one outside power intervened and essentially on the side of the incumbent government. However, a crisis under our definition — one between sovereign governments — may involve some revolutionary elements or even be precipitated by them, as the 1914 crisis was precipitated by Austria's fear of revolution fomented by Serbian intrigues. Quite obviously, a crisis always involves "severe conflict." There is, first, a deep conflict of interest between the parties. However, conflict of interest in itself is not sufficient to bring about a crisis. One of the parties must initiate some form of conflict behavior in an attempt to resolve the underlying conflict of interest in its favor. Usually, a crisis erupts when one party attempts to coerce the other with threats of violence and the other party resists.
The centerpiece of our definition is "the perception of a dangerously high probability of war" by the governments involved. Just how high the perceived probability must be to qualify as a crisis is impossible to specify. But ordinary usage of the term crisis implies that whatever is occurring might result in the outbreak of war. The perceived probability must at least be high enough to evoke feelings of fear and tension to an uncomfortable degree.
Several corollaries are implied here. First there are various forms of disputation (conflict behavior) between great powers that cannot be called crises because, if there is any chance of war at all, it is below the "crisis threshold." Such low-intensity forms of conflict behavior may be referred to as "disputes," "disagreements," "press wars," or simply "bad relations."
Second, the probability-of-war element excludes so-called "crises" between allies, such as the Skybolt affair of 1962. Our crises are between governments who identify each other as enemies or at least potential enemies, and the kinds of political processes we shall be analyzing are primarily adversary processes. However, the adversaries usually have allies, so we include some analysis of alliance relations.
Third, the term probability of war excludes war itself from the concept "crisis," although minor forms of violence "short of war" are included as potential instruments of coercive bargaining. In the modern age, when the line between peace and war has become increasingly blurred, it may in some empirical cases be difficult to determine precisely when "crisis violence" becomes transformed into "war." And, as Thomas Schelling has emphasized, war itself, especially limited war, has become increasingly an affair of bargaining, similar in many respects to crisis bargaining. Despite such occasional ambiguity, we simply stipulate that cases of large-scale violence (war) lie outside the class of events we call "crises," although war may be a consequence of crises. The distinction corresponds closely enough to ordinary language usage and to empirical reality.
The exclusion of war, per se, from the concept, does not exclude crises that accompany or arise out of war between states that are aligned or allied with the warring protagonists, or between one of the states at war and a third state. For example, the 1973 Yom Kippur war in the Middle East triggered a brief crisis between the United States and the Soviet Union. Similarly, almost a hundred years earlier, the Russo-Turkish war of 1877-1878 set off a crisis between England and Russia. The Korean war stimulated a crisis between the United States and China that eventually led to China's entry into the war.
Finally, the term probability, in the loose, subjective sense in which we use it here, suggests the element of uncertainty, an element stressed by Thomas Schelling: "The essence of the crisis is its unpredictability." One kind of unpredictability is that arising from the participant's lack of full control over events, the possibility of "things getting out of hand." But even if the parties do have firm control (over their own behavior at least), uncertainty also arises from their very imperfect information about the other party's values and intentions. To a considerable extent it is this element of uncertainty, in both the forms mentioned, that lends to an event its "crisis atmosphere," i.e., to feelings of fear, tension and urgency. If each party knew what the other intended to do — in simple terms: yield, stand firm, or fight — and also knew its own intentions in the light of that knowledge, there could be no crisis. Either no coercive challenge would be issued, or, if issued, it would be followed inexorably either by the opponent's capitulation or by war. Even if some length of time occurred between challenge and outcome, it would be characterized not by feelings of "crisis" but by the parties' preparation to do what their values and certain knowledge dictated. Thus, it is largely because of the lack of complete information that crises occur at all. A corollary is that if one of the parties thinks it does have accurate information about the other's intentions, the situation does not become a crisis for that party until it realizes it has misestimated those intentions or loses confidence in its initial estimate. For example, the Cuban crisis of 1962 did not become a crisis for the Soviet Union until President Kennedy issued the U.S. challenge in a television speech and confounded Soviet expectations. It had already been a crisis for the U.S. government for a week while secret decisions were being made.
Our definition is quite different from those typically employed by students of the effects of crisis on decision-making behavior. A definition of this latter type is the one advanced by Charles F. Hermann: "a crisis is a situation that (1) threatens high-priority goals of the decision-making unit, (2) restricts the amount of time available for response before the decision is transformed, and (3) surprises the members of the decision-making unit by its occurrence." While this definition usefully points to certain characteristics of crisis that presumably affect decision making, it obscures the state-to-state interaction aspect of crises, which we regard as fundamental. Only when this aspect is central to the definition do all the dimensions of crisis as an international political phenomenon come into view.
Excerpted from Conflict Among Nations by Glenn H. Snyder, Paul Diesing. Copyright © 1977 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.
Table of Contents
- Frontmatter, pg. i
- CONTENTS, pg. v
- List of Figures, pg. vii
- List of Tables, pg. ix
- PREFACE, pg. xi
- GLOSSARY, pg. xv
- I. International Crises and International Theory, pg. 1
- II. Formal Models of Bargaining, pg. 33
- III. Crisis Bargaining: Strategies and Tactics, pg. 183
- IV. Information Processing, pg. 282
- V. Decision Making, pg. 340
- VI. Crises and International Systems, pg. 419
- VII. Summary and Synthesis, pg. 471
- APPENDIX: CASE SUMMARIES, pg. 531
- INDEX, pg. 571