Read an Excerpt
War Termination as a Bargaining Process
By Paul R. Pillar
PRINCETON UNIVERSITY PRESSCopyright © 1983 Princeton University Press
All rights reserved.
Patterns of War Termination
An inquiry into the process of negotiating peace must begin by identifying the part that negotiation plays in war termination generally. The present chapter presents a brief but systematic tabulation of past war endings, revealing patterns that suggest how future wars will be likely to end. The tabulation also reveals which past wars can yield the most insights about how peace agreements are likely to be negotiated in the future.
Following some preliminary remarks about the basis of the tabulation, the first section presents a typology of war endings. The next section uses this typology to summarize how past wars have ended. The results reveal changes in patterns of war termination which suggest that most future international wars will end with negotiated settlements. This argument is then extended, showing why most of these settlements will be negotiated prior to an armistice. Based on these findings, war termination is then described as a bargaining problem. The final section notes the implications of the results for the study of peace negotiations and identifies the sources of material to be used in subsequent chapters.
The tabulation below includes wars that ended between 1800 and 1980. It thus roughly covers the period of the modern mass army, which was born in the wars of the French Revolution. The period is also sufficiently long to reveal trends.
Several lists of wars have been published, though none were compiled to study how wars have ended and none include any data on that subject. The list in this chapter is based principally on the most recent compilation of wars of the nineteenth and twentieth centuries, found in Melvin Small and J. David Singer's Resort to Arms. Their criteria for inclusion as an international war involve the number of casualties and the status of the entities that fought. They include a conflict in their list if it involved at least one member of the interstate system and caused at least 1,000 battle deaths among all the members of the interstate system that were belligerents. Further, if a war involving only one member of the interstate system lasts longer than one year, that member's battle deaths must reach an annual average of 1,000. The list in this chapter includes all 111 international wars ending between 1816 and 1980 identified by Small and Singer. The list is extended back to 1800 by drawing on Quincy Wright's earlier compilation, which adds 10 interstate wars.
The peace negotiations discussed in this book have mainly occurred in international wars. We should also take note of how civil wars have ended, however, not only because of their intrinsic importance but also because this will permit some later comparisons that will clarify the conditions in which negotiated settlements are most likely. Small and Singer present a list of civil wars in Resort to Arms, but their definition is a broad one which includes instances of mass violence that some others would describe as riots rather than civil wars. For our present purpose we will thus use Wright's more restrictive list, taking from it all conflicts that Small and Singer, in an earlier version of their work on international wars, state that they exclude solely because they are civil wars. This adds 11 wars through 1965 (when Wright's list ends), and applying what appears to be Wright's standard for civil wars to the portion of Small and Singer's list for 1966 to 1980 adds 10 more. The final list includes 142 wars that concluded between 1800 and 1980, from the French Revolutionary Wars ending in 1801 to the civil war in Zimbabwe ending in 1979.
Two minor problems with the list should be mentioned. The first is that the list includes several multilateral wars which were really clusters of struggles grouped together under a single name, not all of which ended together. These are the French Revolutionary and Napoleonic wars, the Latin American wars for independence, the two world wars of the twentieth century, the Bolshevik regime's struggle against several opponents during its first years, and the various Arab-Israeli wars. Rather than attempting to break these conglomerate wars into individual components, which would in effect mean constructing a new list, the end of each will be taken to be the end of fighting on the last front to fall quiet. Given the small number of these conglomerate wars in comparison with the overall total of 142, this procedure is unlikely to distort any patterns. Besides, most of the conflicts in each conglomerate war tended to have similar endings anyway.
The other minor difficulty results from the way Small and Singer use their casualty criteria to determine the duration of a war. Some colonial and imperial wars were stop-and-go affairs or dragged out over several years at a low level of violence. The purpose of the 1,000-deaths-per-year standard is to include only the violent phases of such conflicts. This is acceptable if the concern is to measure the duration of violence (one of Small and Singer's concerns), but less desirable in examining how violence has ended, because in a couple of cases the Small-Singer cutoff point was only a temporary interruption of a war that concluded definitively sometime later. To adjust their dates, however, would mean losing the precision of the casualty criterion and substituting something else that might entail begging the very questions which the exercise is designed to answer. Therefore, the casualty criterion will be retained to identify when a war ended as well as whether it belongs on the list in the first place, and the cases in which this raises difficulties will be noted individually.
How Wars Can End
This chapter is concerned with the structure or form of war endings, and not with the substance or legal status of peace settlements. The distinctions we are exploring involve whether and how the conclusion of a war is related to political decisions, who takes such decisions, and when they are taken. Figure 1 depicts a typology of war endings that makes these distinctions. The initial basis for classification is whether fighting ends when the war does. The question is genuine because some of the wars that appear on a list such as Small and Singer's never reached their own conclusion. Instead, they became part of a larger war (which they may have helped to trigger), with additional participants, its own place on the list and its own settlement. In such cases, the concluding date of the smaller war only marks the start of the larger one. The smaller war may be said to have ended by absorption (A). (One might prefer to think of this not as the end of a war but only as a quirk of this kind of list. At least it serves as a reminder that the level of violence can go up as well as down.) If the fighting does cease, we may next ask whether it did because both belligerents decided that it should cease. If one side stops fighting without having taken a decision to stop, it means that its opponent has rendered it incapable of continuing the fight, either through extermination (of an organized force, not necessarily of individuals) or expulsion (E) from the country or theater in which the war is fought. When a war ends instead through decisions taken by both parties, it may be without any explicit agreement — i.e., through a simple withdrawal (W) from combat — or through acceptance of a written accord. The parties could have drafted such an accord themselves or accepted one written by a third party. Virtually the only kind of third party that has served this function is an international organization (10). If the agreement was not the work of a third party, it might have been imposed, without discussion or modification, by one side upon the other. This would be capitulation (C). Capitulations may include "unconditional" surrenders as well as agreements that are much less one-sided; all that matters is how the agreement was reached. The alternative to imposition by one side is negotiation by both sides. Finally, negotiated peace agreements differ according to the sequence of negotiation and cease-fire. In some cases, the principal peace agreement is negotiated after (NA) the establishment of an armistice. In others, the only peace agreement reached is the one which establishes the cease-fire; the negotiation which led to it took place before (NB) the armistice. This distinction depends solely on whether or not there was a post-armistice agreement that was part of the process of war termination, not on the substance or legal form of an agreement.
Not every war fits neatly into one of the boxes. For example, when an insurrection gradually loses strength and finally ends without any explicit agreement, it may be unclear whether the end came because the remaining leadership of the insurrection decided to give up the struggle (withdrawal) or because they lost all of their troops (extermination/expulsion). When there is an explicit agreement, the line between capitulation and negotiation can be fuzzy. In common parlance the two terms are usually distinguishable; negotiation suggests at least the possibility of give-and-take, whereas capitulation is a matter of one side signing on the dotted line. The ambiguity that does occur reflects the lack of a generally accepted definition of negotiation. This in turn is at least partly the result of a confusion — often deliberate — of the concept of negotiation itself, with the concept of negotiating behavior that is "reasonable" or "in good faith." An easy way to criticize one's opponent is to accuse him of not negotiating at all. However, to define negotiation so that it requires not just that reciprocity be possible but that it actually occur would mean that in many encounters commonly described as negotiations, true negotiation is not taking place most of the time. Furthermore, such a definition would confuse or prevent some comparisons between individual negotiations that are useful in studying questions of successful versus unsuccessful negotiating strategies, easy versus difficult negotiations, etc. Accordingly, a peace settlement is considered to have been negotiated if it emerged from a face-to-face session which at least one of the parties entered with the apparent intention not of capitulating, but of jointly writing an agreement.
How Wars Did End
Table I lists the 142 wars which ended between 1800 and 1980, in order of their ending dates. The table provides two pieces of information about each war, in addition to the dates. One is a classification of the wars themselves into three types, which is taken from Small and Singer: (1) interstate wars, in which a member of the state system fought on each side; (2) extra-systemic wars, which were colonial or imperial conflicts involving a member of the state system on only one side; and (3) civil wars. These distinctions will be referred to again in a moment. The other is the classification of war endings, following the typology in Figure 1. The sources consulted are listed in the bibliography and keyed to the numbers in this table. The most questionable cases are asterisked and explained further in notes. In addition to the seven basic categories, an additional coding scheme in parentheses indicates wars which, though not falling within the NB category, share some of its attributes. The notation (I) indicates that some political issues were settled through indirect negotiations (conducted through intermediaries, broadcasts, or written messages) prior to the armistice. (P) means that the armistice agreement itself incorporated substantial political provisions, usually as a "preliminary peace treaty." (U) indicates that face-to-face negotiations were conducted on the terms of settlement prior to an armistice but did not succeed in producing an agreement.
The Prevalence of Explicit Agreements
The totals at the bottom of the table show that a substantial majority of the wars ended with an explicit agreement between the two sides. In most of the ones that did not, one of the parties was exterminated or expelled. As long as both belligerents remain, in other words, it is rare for combat to end without a written agreement. It is perhaps even more rare than the total of seven withdrawals indicates, because five of these seven are questionable or special cases. An agreement to end the Spanish-Santo Dominican War was negotiated and signed in 1865, but it was effectively discarded when the Dominican leaders repudiated their own negotiators and word arrived that Madrid had decided to evacuate its forces from the island. Spain and Chile also signed an armistice agreement and peace treaty after their war in 1865–1866, but these came so long after the actual end of hostilities that the case is classified as a withdrawal. The French adventure in Mexico during the same decade ended when Napoleon decided to withdraw his troops, but the imperial forces within Mexico put up some further resistance before capitulating. The fighting between Britain and the followers of the Mahdi in the Sudan in the 1880's ended with a British withdrawal back to Egypt. However, this omits (because British deaths did not meet the casualty criterion) the further fighting in the 1890's, when the British under Kitchener returned to crush the dervishes at the battle of Omdurman and exterminate their rebellion. Finally, the civil strife in Colombia which ran from the 1940's through the early 1960's is one of those instances of an insurrection gradually petering out, and which could be considered either a withdrawal or an extermination. There are only two wars on the list which unambiguously ended through withdrawal: China's border clashes with India in 1962 and with Vietnam in 1979.
Withdrawals probably would not be quite so scarce if the casualty criteria for inclusion in the list were relaxed. At low levels, military force is sometimes used demonstratively, to assert a claim or to issue a warning. The Chinese Communists' periodic shelling of the offshore islands in the Taiwan Strait is an example, as perhaps were their larger actions against India and Vietnam. When force is used only to demonstrate, and not to eliminate an adversary or to change his will, the advantages of explicitness in an agreement are irrelevant because no agreement is sought.
An instance of armed conflict growing out of more ambitious aims but which would also be classified as a withdrawal if it qualified for the list is the Communist insurgency in Greece between 1946 and 1949. That struggle ended when the rebels announced that they were ending military operations and withdrew what was left of their army and government to Albania. The conflict in Greece is worth mentioning because, despite its small scale, it influenced the thinking of decision-makers, at least in the United States, who directed later and larger wars. In Congressional testimony in May 1951 (less than two months before the start of the Korean truce talks), General Omar Bradley, the Chairman of the Joint Chiefs of Staff, acknowledged that the fighting in Korea might end in the Greek pattern rather than with a formal agreement. And during the early part of the Vietnam War, American officials again were thinking in terms of a unilateral withdrawal from combat by the Communists, rather than an explicit agreement with them. But the important point here is that, despite such thinking, these later wars nevertheless ended with formal agreements.
Excerpted from Negotiating Peace by Paul R. Pillar. Copyright © 1983 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.