The second is to identify the policy implications of these results: when a
revolution
occurs, what precautions should other states take?
Revolution and War_nodrm
192 U.
S.
fears of Chinese-backed subversion also underlay the Americans' growing involvement in Vietnam, and the vi- olence and xenophobia unleashed during the Cultural Revolution in China merely confirmed the image of the PRC as an irrational and unpredictable adversary.
193
This impression was at best a caricature and at worst extremely mislead- ing. Although its leaders were willing to implement far-reaching and ill- conceived domestic programs such as the Great Leap Forward, China's foreign policy was acutely sensitive to external constraints and its use of
188 See Shu Guang, Deterrence and Strategic Culture, chaps. ? ; Gurtov and Hwang, China under Threat, chap. 3; Jan Kalicki, The Pattern ofSino-American Crises: Political-Military Interac- tions in the 1950s (Cambridge: Cambridge University Press, 1975), chaps. 3, 6, 8; and Allen S. Whiting, "New Light on Mao: Quemoy 1958, Mao's Miscalculations," China Quarterly 62 (1975)?
189 Foot, Wrong War, 27-28.
190 Quoted in Shu Guang, Deterrence and Strategic Culture, 239?
191 ForMao'sstatementsaboutnuclearweapons,seeAliceLangleyHsieh,ComunistChina's
Strategy in the Nuclear Era (Englewood Cliffs, N. J. : Prentice-Hall, 1962), 1, 52, 132; and Stuart Schram, Mao Tse-tung (Harmondsworth: Penguin, 1966), 291. On China's nuclear program, see John Wilson JLewis and Xue Litai, China Builds the Bomb (Stanford: Stanford University Press, 1988); and Avery Goldstein, "Understanding Nuclear Proliferation: Theoretical Expla- nation and China's National Experience," Security Studies 2, nos. 3-4 (1993).
192 See Chang, Friends and Enemies, chap. 8.
193 In 1966, Assistant Secretary of State William P. Bundy referred to the PRC as "a gov- ernment whose leadership is devoted to the promotion of communism by violent revolu- tion," and Secretary of State Dean Rusk told a congressional hearing that "a country whose behavior is as violent, irascible, unyielding, and hostile as that of Communist China is led by leaders whose view of the world and of life itself is unreal. " See Franz Schurmann and
Orville Schell, eds. , The China Reader, vol. 3: Communist China (New York: Vintage, 1967), 378-79- 508.
? ? The American, Mexican, Turkish, and Chinese Revolutions
force was cautious and restrained. 194 Mao was willing to raise tensions in the Taiwan Straits in order to convince the United States that the PRC could not be intimidated, but he went to some lengths to limit the risks and stayed well clear of the brink. 195 With the exception of Korea and Vietnam (where . Chinese strategic interests were directly involved), Beijing's support for "wars with national liberation" was more rhetorical than real, and Mao stressed that foreign revolutionaries must rely on their own efforts. 196 Simi- larly, Mao's seemingly provocative statements about nuclear weapons were intended primarily to bolster Chinese morale and to convince the United States that China could not be cowed by nuclear threats; significantly, Mao's private remarks were quite prudent, and his public statements became more
moderate as China's nuclear capability increased. 197
China's overriding concern for security was also evident in its sensitivity
to shifts in the global balance of power and its willingness to ignore ideo- logical principles in responding to them. 198 During the 1950s, the PRC saw the United States as its principal threat and chose to "lean to one side" with the Soviet Union. By the end of the decade, dissatisfaction with Soviet sup- port led Mao to adopt a neutral position between the two superpowers, while China expanded its ties with Western Europe and Japan and tried to unite the Third World in an "international united front. " The latter effort did not attract many adherents, however, and the onset of the Cultural Revolu- tion in 1966 initiated a brief period of self-imposed isolation. 199
194 See Allen S. Whiting. "The Use of Force in Foreign Policy by the Peoples' Republic of China," A n n a l s of t h e A m er i ca n A c a d e m y of P o l i t i c a l a n d Soci al S c i e n c e 4 0 2 ( 1 9 7 2 ) , 5 5 ? , an d Th e C h i n e s e C a l - culus ofDeterrence: India and Indochina (Ann Arbor: University of Michigan Press, 1975).
195 During the 1954 crisis, Mao prohibited "offensive actions against foreign warships and airplanes," and the Central Military Commission told Chinese troops that they could retali- ate to a U. S. attack only "if this involves no grave risks. " When the GMD evacuated the off- shore island of Dachen in February 1955, the Central Military Commission denied a local commander's request to occupy the island for fear that it would entangle them with U. S. forces. Mao acted cautiously in 1958 as well, rejecting a proposed air strike on Quemoy in Au- gust and ordering his artillery not to fire on U. S. ships. See Shu Guang. Deterrence and Strate- gic Culture, 198, 219, 236-37; and Chang. Friends and Enemies, 72.
197 In 1965, Mao claimed that his calling the atomic bomb a "paper tiger" was "just a figure of speech. " He suggested that a nuclear war "would be a catastrophe for the whole world" and stated, "If one must fight one should confine oneself to conventional weapons. " See Schurmann and Schell, China Reader, 364-65. See also Avery Goldstein, "Robust and Afford- able Security: Some Lesson from the Second-Ranking Powers During the Cold War," Journal ofStrategic Studies 15, no. 4 (1992), 492-{)6, 5oo-503; and Lewis and Xue, China Builds the Bomb, 34, 215-18, and app. A.
198 According to A. Doak Barnett, "when security interests have been at stake, Chinese Communist leaders have generally . . . given them priority over other interests, including ide- ological and economic ones. " China and the Major Powers in East Asia (Washington, D. C. : Brookings Institution, 1977), 254-55.
199 See Harry Harding. "China's Changing Roles in the Contemporary World," in Harding. China's Foreign Relations, 186-87; and Armstrong, Revolutionary Diplomacy, 77-90.
? 196 See Van Ness, Revolution and Chinese Foreign Policy, 72.
? ? ? Revolution and War
By the late 1960s, the growth of Soviet military power in the Far East and a series of border clashes between Soviet and Chinese troops convinced Mao that the main threat was "Soviet hegemonism. "200 Just as the PRC had once "leaned to one side" to balance the perceived threat from the United States, it now turned to the United States to balance the threat from the So- viet Union. The rapprochement became overt after President Richard Nixon's 1972 visit to Beijing, and given the aura of suspicion that had sur- rounded Sino-American relations since the late 1940s, it would be hard to find a more vivid example of unsentimental realpolitik. Moreover, this shift ? occurred in period when ideology was especially important in shaping pol- icy within China. Thus, although the Sino-Soviet alliance can be seen as the result of either balancing or ideological solidarity, the Sino-American rap- prochement represented strictly the former. Z01
China's willingness to subordinate ideological principles to its security in- terests was evident in other relationships as well. Despite their public com- mitment to "Asian solidarity" and "peaceful coexistence," recurring border disputes and mutual suspicions led to a major deterioration of Sino-Indian re- lations in the late 1950s. The two states eventually fought a brief border war in 1962, and China's forces acted both cautiously and defensively in repelling the Indian challenge. India's subsequent tilt toward the Soviet Union was matched by a de facto alliance between China and Pakistan, despite the lat- ter's nonsocialist character and its close ties to the United States. 202 Similarly, the U. S. withdrawal from Indochina in the early 1970s allowed the long-sup- pressed rivalry between China and Vietnam to reemerge; Vietnam strength- ened its alliance with the Soviet Union while China backed the Khmer Rouge in Cambodia, and China and Vietnam eventually fought a brief but intense border war following the Vietnamese invasion of Cambodia in 1979.
In addition to providing evidence of China's willingness to sacrifice prin- ciples to the requirements of balance-of-power politics, these developments can be viewed as part of the gradual process of learning and adaptation. Al- though the United States and China continued to exaggerate each other's hostility after the Korean War, they avoided a direct clash in subsequent crises in part because each had learned that such a war would be costly. 203
200 See Thomas W. Robinson, "The Sino-Soviet Border Conflict," in Stephen S. Kaplan, DiplomacyofPower:SovietArmedForcesasaPoliticalInstrument(Washington,D. C. :Broko ings Institution, 1g81).
202 See Anwar H. Syed, China and Pakistan: Diplomacy ofan Entente Cordiale (Amherst: Univer- sity of Massachusetts Press, 1974), chap. 4; and Ya'acov Vertzberger, The Enduring Entente: Sino- Pakistani Relations, 196o-t980, Washington Papers no. 95 (New York: Praeger, 1983), 15-24.
203 China reportedly suffered over 26o,ooo dead in Korea, while the United States lost around47,00. 0 AwarenessofwhatanotherclashmightcosthelpedkeeptheUnitedStatesout of Indochina in the 1950s, and the fear of Chinese intervention constrained U. S. behavior dur- ing the height of its subsequent involvement there.
201 I am indebted to Avery Goldstein for this insight.
? ? The American, Mexican, Turkish, and Chinese Revolutions
Over time, the PRC gradually evolved from a radical state to a reformist state to a conservative state, and although the CCP remains in power as of this writing, its policies bear little resemblance to the ideals of Marx or Mao, and its foreign policy aims are largely indistinguishable from the goals of other great powers.
The Chinese Revolution fostered intense perceptions of threat with other major powers, based on the shared belief that the other was inherently ag- gressive and that the threat could be reduced or eliminated through the use of force. Over time, however, the impact of these early views faded and the new regime and its neighbors gained a more accurate understanding of their respective interests and capabilities. Despite its revolutionary origins, revolutionary China generally behaved in a restrained and prudent manner after the Korean War. My theory thus receives further support.
Although the comparison is hardly novel, the similarities between the Chinese and Russian experiences are worth noting. Both regimes came to power after a world war, a situation that helped shape their early percep- tions and gave them a "breathing space" in which to consolidate their power. Both fought and won a civil war, faced pressure from hostile imperi- alist powers, and exaggerated their opponents' willingness to attack them directly. Having fought wars shortly after gaining power (though China's involvement in Korea was far more costly than Russia's war with Poland), both states gradually overcame their initial isolation and reemerged as ac- cepted (albeit suspect) players on the world stage. Finally, although both China and Rutssia were willing to compromise their ideological principles in order to preserve their security, neither abandoned these ideals completely, and they continued to interpret world events through the distorting prisms of a revolutionary worldview. This tendency helps explain why other states kept them at arm's length even when there were obvious incentives to forge a closer relationship.
CONCLUSION
Each of the revolutions examined in this chapter increased the intensity of security competition, and each of these states came close to war on one or more occasions. The conflicts arose for reasons that are consistent with the mechanisms identified in chapter 2 and bear more than a passing resem- blance to the dynamics observed in the French, Russian, and Iranian cases.
First, foreign reactions to each of these revolutions were heavily influ- enced by the potential impact on the balance of power. To France, the Amer- ican Revolution was an opportunity to weaken Britain; several foreign states saw the turmoil in Mexico as both a threat to their existing interests
? ? ? Revolution and War
and a chance to improve their positions at others' expense; Great Britain and Greece considered the revolution in Turkey a "window of opportunity" in the Near East; and the Chinese revolution was viewed as a major gain for the Soviet Union. In each case, foreign powers sought to take advantage of the power vacuum produced by the revolution, and their efforts led to con- flict with the new regime, other foreign powers, or both.
Second, after the revolution each of these states experienced spirals of hostility with other states that brought them to the brink of war at least once. The revolutionary state and other powers tended to suspect eaclh other's intentions, and each new regime was jealous of its sovereignty and obsessed with issues of security. Although each of these states did face real security threats after the revolution, the threats were usually exaggerated and, at times, wholly illusory. This paranoia was most pronounced in the case of China, but misunderstandings and exaggerations marred Anglo- Turkish relations during the Nationalists' bid for power, exacerbated U. S. relations with Britain and France in the 1790s, and disturbed revolutionary
Mexico's relations with the United States as well. As expected, lack of infor- mation and poor channels of communication generally made these prob- lems worse.
These states also exhibited the expected combination of vulnerability and! optimism, although the intensity varied greatly. U. S. leaders were deeply fearful of foreign interference yet confident that their new nation would! eventually control the entire continent, and their long-term optimism dis- couraged the use of force on several occasions. The Nationalists in Turkey and the Constitutionalists in Mexico were equally opposed to foreign inter- ference and willing to use force to prevent it, yet each ultimately preferred to negotiate rather than fight; moreover, foreign intervention was clearly discouraged by an awareness of what such a campaign might entait Thus, the predicted mixture of fear and overconfidence was muted in these three
. cases. It was clearly present in China, however, and helps explain Mao's de- cisi? n to intervene in Korea in 1950. Intervention risked U. S. retaliation and! jeopardized the reconstruction of his war-torn country, but Mao was con- vinced that the threat was too great to ignore and that his army could ac- complish nts mission at an acceptable cost.
Given the presence of these familiar sources of conflict, it is not surprising that each of these states came close to war. Yet as we have seen, the Ameri- can, Mexican, and Turkish revolutions did not go over the brink. The expla- nation lies in the strong condition of defense dominance that accompanied each; unlike in the four other cases examined in this book, both the internal character and external circumstances of the American, Mexican, and Turk- ish revolutions made the use of force less attractive. Although conflicts and crises did occur, both sides ultimately concluded that going to war would not be worth the cost and risk.
? ? The American, Mexican, Turkish, and Chinese Revolutions
This condition of defense dominance was attributable to three main fac- tors. First, the American, Mexican, and Turkish revolutions were all com- paratively moderate, relative to our other cases. In particular, none was based on a explicitly universalist ideology (the Founding Fathers being at best ambivalent about the ability of other societies to replicate the U. S. ex-
perience), and none of these states did much to export their principles to other countries. This greatly reduced the potential threat they posed and re- duced the need for other states to take countermeasures to contain them. Another consequence was to allow each of these states to reenter the exist- ing system of states more rapidly than postrevolutionary France, Russia, Iran, or China could do.
Second, each of these three states was both large and relatively weak-es- pecially in comparison with the other great powers-making a war by mis- calculation less likely. U. S. weakness helped limit pressures for war in the 1790s, as both Washington and Adams ultimately chose to negotiate with their foreign adversaries rather than risk a war for which they were clearly unprepared. Similarly, early in this century Mexican and U. S. leaders both understood that Mexico was by far the weaker of the two (and thus not much of a threat), though its sheer size would have made it difficult to oc- cupy and subdue. As a result, both sides had ample reason to act with fore- bearance. The Turks were willing to fight the Greeks, Armenians, or Georgians (the latter two being even weaker than they), but Kemal recog- nized that a war with the Entente would be counterproductive and possibly
fatal. By contrast, the combination of China's vast population, Soviet sup- port, and the proximity of Korea made it possible for Mao to contemplate the overt use of force, especially as he believed that the United States was preparing to attack as soon as it established its position in the Far East.
Third, these three revolutionary states were favored by geographic isola- tion and fortuitous timing. The United States lay an ocean away from the other great powers, and its potential opponents were preoccupied by events closer to home. Mexico enjoyed similar advantages; although the outbreak of World War I encouraged a certain amount of foreign meddling, the war in Europe distracted its potential opponents and played a key role in dis- suading the United States from more active interference. The Turks profited. from the revolution in Russia and the war-weariness afflicting the other Eu- ropean powers, which discouraged Western intervention and allowed the Turks to set the two sides against each other. Had the timing of these revo- lutions been different, it is easy to imagine a less favorable outcome.
Finally, the cases examined in this chapter suggest a reciprocal connection between revolution and war. Revolutionary change will make war more . likely, but the onset of war will shape the revolutionary process itself. In France, the outbreak of war in 1792 radicalized the revolution and led to the founding of the republic. Foreign intervention encouraged the Bolsheviks to
? ? Revolution and War
take more extreme measures, and the outbreak of war with Iraq strength- ened the radical elements in Iran as well. By contrast, the American, Mexi- can, and Turkish revolutions had more moderate outcomes in part because these states managed to avoid war until the revolution was essentially com- plete.
The American, Mexican, Turkish, and Chinese revolutions enhance our confidence in the propositions developed elsewhere in this book. That con- fidence is increased by the diversity of the cases and by the fact that even the apparent exceptions demonstrate that revolutions create strong pressures for war. In the next and final chapter, I consider whether more recent events fit this pattern as well and summarize the lessons that national leaders might draw from these results.
? ? ? ? ? ? Conclusion
"Revolutions . . . occurred and will always occur so long as human na- ture remains the same. "
-Thucydides
"In a revolution, as in a novel, the most difficult part to invent is the end. " -Alexis de Tocqueville
This boolk has explored some of the ways that revolutions affect interna- tional politics, focusing primarily on the relationship between revolution and war. I argued that revolutions alter the balance of threats between states, leading to more intense security competition and a heightened prob- ability of war. I tested and refined this argument by examining three major revolutions in detail-those of France, Russia, and Iran-as well as four ad- ditional cases where the fit between theory and reality was less obvious.
Four tasks remain. The first is to summarize and compare the results of the seven case studies, in order to highlight the principal theoretical conclu- sions we may infer from these events.
The second is to identify the policy implications of these results: when a revolution occurs, what precautions should other states take? What actions should they avoid? The third is to sketch what the theory tells us about the recent collapse of the Soviet empire and its effects on the likelihood of war. The final task is to consider the long- term relevance of this study: is mass revolution a fading phenomenon, or are the problems caused by past revolutions likely to occur in the future?
REVOLUTION, SECURITY COMPETITION, AND WAR
The cases examined in this book confirm that revolutions increase the in- tensity of security competition between states and raise the probability of war. Although war did not occur in every case, each regime came close to war soon after gaining power and each revolution fostered greater security competition among the other major powers. The occurrence of revolution
[331]
? ? Revolution and War
was hardly the only source of competition and conflict, but in each case the level of tension was higher than it would have been without the revolution. In the French case, the collapse of the old regime caused competition be- tween Austrfta, Prussia, and Russia to rise, triggered a spiral that placed France and Austria on a collision course, and brought the rivalry with En- gland to the point of open warfare in 1793. Although war might have oc-
curred even if the revolution had been averted, the internal turmoil in France was directly responsible for the war that did break out and for its rapid expansion.
Similarly, although the collapse of the Romanov dynasty temporarily lowered the level of conflict by taking Russia out of World War I, the estab- lishment of Soviet Russia ultimately increased it. The revolution encour- aged other states to compete over the lands of former tsarist empire and brought to power a regime that supported the violent overthrow of other states. The tsarist regime would never have acted as Soviet Russia did, and the other great powers would never have seen it as a subversive force. Again, it is safe to infer that the revolution was directly responsible for much of the enmity and insecurity that characterized relations between So- viet Russia and most other powers.
This pattern is even more apparent in the Iranian case. The revolution alarmed Iran's Arab neighbors, led directly to a protracted war with Iraq, and transformed U. S. -Iranian relations from close alliance to bitter rivalry. Relations with most other countries deteriorated as well, leaving the Islamic Republic largely isolated. Although the shah's ambitions had alarmed his neighbors on occasion, they never provoked the level of international en- . mity that the Islamic Republic has. Thus, Iran's present position is simply incomprehensible outside the context of the revolution.
The same effects were present in the other four cases, albeit to varying de- . grees. The United States experienced repeated conflicts with both its former British rulers and its putative French ally, coming close to war with each on one or more occasions. The revolution in Mexico exacerbated the competi- tion for influence between the United States, Britain, and Germany, led the United States to intervene in 1914 and again in 1916, and continued to mar relations between Mexico and the United States until the late 1930s. The Na- tionalist revolution in Turkey challenged European ambitions in the Near East, brought the new regime to the brink of war with Great Britain in 1922, and eventually forced the Entente to withdraw its troops and abandon the Treaty of Sevres. Needless to say, it is hard to imagine the Ottoman sultan achieving like results. Finally, the Communist triumph in China ended
decades of foreign interference, moved Beijing firmly into the socialist camp, and placed the new regime on the path to war in Korea. Once again, it is unlikely that Chiang Kai-shek and the GMD would have acted the same way or provoked similar responses.
? ? Conclusion
Hypothetical scenarios can never be tested, of course, but the possibility that these states would have experienced equally high levels of conflict in the absence of revolution seems remote. Although other causes of conflict were undoubtedly present, these cases show that revolutions have indepen- dent causal effects on the level of security competition and the probability of war.
Balance-of-Threat Theory
Why do revolutions make competition and war more likely? As described in detail in chapter 2, balance-of-threat theory suggests that revolutions cause security competition by altering the perceived level of threat between the revolutionary state and its main adversaries, on one hand, and by en- couraging both sides to believe that the use of force can overcome the threat at an acceptable cost, on the other.
The evidence strongly supports this general argument. The crises or wars that followed each of the revolutions examined here resulted from a combi- nation of opportunism and insecurity, based on misjudgments about the balance of power, overly malign perceptions of intent, and (in the worst cases) exaggerated beliefs about the likelihood of contagion or counterrevo- lution. Compounding the problem were uncertainty and misinformation, which reinforced each side's prior beliefs and made peaceful settlements more elusive.
TheBalanceofPower. IntheFrench,Russian,andIraniancases,therevo- lution's effect on the balance of power was a central cause of war. Yet the im- pact of a revolution on the balance of power does not cause war by itself. In particular, these effects cannot explain why some states try to exploit the op- portunities while others remain aloof, nor can they account for aggressive behavior on the part of the revolutionary state. For example, although Prus- sia saw French weakness after 1789 as a chance for expansion, most Euro- pean states welcomed the erosion of French power and did not use force to exploit it. Instead, the war began when, reacting to Austria's efforts to in- timidate the French, the Girondins convinced the Assembly to declare war
in April 1792. Similarly, Iran's weakness after the revolution does not fully explain why Iraq saw a military attack as desirable or necessary; the appar- ent collapse of Iranian power might just as easily been considered an op- portunity for Iraq to turn its attention to other problems. ' The four cases examined in chapter 6 reinforce this conclusion: although the American, Mexican, Turkish, and Chinese revolutions created significant power vacu-
1 Furthermore, focusing on power alone cannot explain why Iraq decided to go to war but ? Iran's other neighbors (the Soviet Union, Turkey, Afghanistan) did not.
? ? ? ? ? Revolution and War
urns and fostered greater security competition between other powers, only China was subsequently involved in war. If the belief that a revolution has weakened its victim is part of the link between revolution and war, it is hardly the whole story.
PerceptionsofIntent. Eachoftheserevolutionsproducedsharpdepartures from the foreign policies of the old regime, in tum creating sharp conflicts of interest with other states. Furthermore, both the revolutionary state and tlhe other great powers tended to exaggerate one another's hostility and aggres- siveness. One of the most characteristic features of postrevolutionary for- eign policy, this tendency is also one of the most pernicious.
These cases also confirm that spirals of suspicion can arise from several distinct sources. The most obvious source is ideology: if the worldview of a revolutionary movement stipulates that certain regimes are inherently hos- tile, the new regime is likely to interpret the behavior of foreign powers in the worst possible light. Asecond source is domestic politics. As the French and Iranian cases suggest, factions within a revolutionary movement may dramatize foreign dangers in order to consolidate their own positions. Spi- raling may also be fueled by testimony from emigres or foreign revolution- aries, whose desire for support gives them an obvious incentive to foster conflicts between the revolutionary state and other powers. This tactic played a modest role in several of these cases; French emigres echoed Euro- pean fears of the revolution in France, Russian exiles stiffened Allied resis- tance against normalizing relations with the Bolshevik regime, and Iranian exiles (including the shah) contributed to the deterioration of Iranian for- eign rellations. The activities of the "China lobby" and the misleading testi- mony provided by some U. S. diplomats after the American Revolution further illustrates this danger.
Offense, Defense, and the Export ofRevolution. The cases examined in this study also confirm that revolutions cause war by affecting perceptions of the offense-defense balance, conceived in both military and political terms. In the French, Russian, Iranian, and Chinese cases, decisions to go to war were encouraged by a combination of fear and overconfidence, usually based on each side's calculations of the likelihood that the revolution will spread or be reversed. Expecting that their example would be contagious, the revolutionary state was more confident, less willing to compromise, and prone to support revolutionary efforts abroad. At the same time, the general belief that the revolution might also be easily reversed made exporting the revolufrion seem necessary to the revolutionary state and helped convince its adversaries that they could eliminate the threat with little effort.
Ironically, history suggests that both these beliefs are usually misguided. Although each of these revolutions was accompanied by evidence of discon-
? ? Conclusion
tent in other societies, none of them spawned successful imitators during the decade after the seizure of power, and efforts to spread the revolution via propaganda or contagion only angered and alarmed other states. The Girondins' predictions of a universal crusade for liberty were disappointed; "Bolshevik" uprisings in Finland, Hungary, and Bavaria collapsed quickly, the Polish proletariat welcomed the Red Army with bayonets instead of flowers; and Soviet attempts to spark Communist revolutions in the Near East and China all failed. Efforts to export the Iranian Revolution have been equally abortive thus far, despite the universalist pretensions of Khomeini's Shiism and Iran's support for fundamentalist groups throughout the Muslim world. We should not be surprised at these results, however, because would- be propagators of revolution face several significant obstacles.
First, although a revolution often comes as a? surprise to virtually every- one (including the revolutionaries themselves), it also provides a timely warning to others. As a result, potential targets will be less prone to the mis- takes that let the old regime be toppled. Thus, the French example alerted the other European powers to keep a close watch on potential "Jacobins" and to make a number of modest reforms. These measures were universally successful except in areas conquered by French troops. Similarly, states fac- ing a Bolshevik challenge acted vigorously to suppress potential uprisings after 1917, just as the Persian Gulf states suppressed, coopted, or expelled anyone suspected of spreading Iran's revolutionary message. Louis XVI in Paris, Nicholas II in St. Petersburg, and Mohammad Reza Pahlavi in Tehran may have dithered their way to their own destruction, but their contempo- raries in other countries seem to have learned from their mistakes.
Second, potential victims of a spreading revolution also learn to balance against this frightening possibility. The growing danger from revolutionary France led to the formation of a large (if unruly) coalition by the summer of 1793, and the Entente maintained a common front against Bolshevism until the mid-192os. In the same way, the threat from Iran prompted greater co- operation between Iraq, Saudi Arabia, Kuwait, and the United States. 2
Third, the failure of revolutions to spread underscores the advantages that incumbent regimes ordinarily enjoy. Even states with severe internal problems usually retain some shreds of legitimacy, as well as a substantial' asymmetry of power over their internal rivals. Moreover, the combination of favorable circumstances and coincidences that make the first revolution possible are unlikely to occur elsewhere in precisely the same fashion. Thus, pro-French radkals in England were quickly overcome by prompt govern- ment action, the German Communists proved to be no match for the Reichs-
2 By contrast, because foreign powers were not especially worried that the U. S. , Mexican, or Turkish revolutions would spread, they did not band together in strong opposing coali- tions.
[335]
? ? ? Revolution and War
wehr in 1919, 1921, and 1923, and the Shiites in Iraq were crushed when their leaders tried to duplicate the Iranian experience. Given the barriers to a revolution's spreading beyond a single state's borders, the real mystery is why anybody believes that it wilV
The evidence also suggests that reversing a revolution is nearly as diffi- cult as spreading one. Revolutionary regimes survive because they are usu- ally adept at mobilizing military power; whatever their other failings, revolutionary movements are especially good at persuading people to run grave risks and make large sacrifices for the sake of an ideal. 4 Foreign inter- ference can facilitate this task by providing the legitimacy that a revolution- ary regime needs, and a foreign threat can make it easier for leaders to eliminate rivals in the name of "national unity. "
When foreign powers do not possess reliable sources of information and do not discount the testimony of emigres sufficiently, they are likely to end up backing far-fetched counterrevolutionary schemes that hold little chance of success-as illustrated by English support for the Quiberon expedition in 1795, British and French aid to the White armies in the Russian Civil War, Iraq's support for various Iranian exile groups in Baghdad, and covert U. S. efforts to support anti-Communist groups in China. Supporting counter- revolutionary forces may be a cost-effective means of pressuring a revolu- tionary regime, but it is unlikely to remove it from power and will almost
certainly fuel its perceptions of threat.
These obstacles do not mean that counterrevolutionary efforts never suc- ceed, but success will be more difficult than other states expect. Foreign in- tervention did reverse the Dutch revolt of 1787 and the Polish "revolt" of 1791, to cite but two examples, and both superpowers intervened to reverse unwanted upheavals in their own spheres of interest throughout the Cold War. 5 These examples suggest that outside intervention can work when
3 Exceptions arise when the new regime is able to defeat and occupy its adversaries and impose its system upon their populations by force. Thus, France's "sister republics" and the Soviet empire in Eastern Europe resulted from military expansion rather than the spread of revolutionary ideals. Examples of "spontaneous" Communist revolutions include those in the People's Republic of China, North Korea, Vietnam, Cuba, Ethiopia, Angola, and Nic- aragua, but these "triumphs" occurred decades after the 1917 revolution.
4 See Theda Skocpol, "Revolutions and Mass Military Mobilization," World Politics 40, no. ? 2 ( 11}88); and Ted Robert Gurr, "War, Revolution, and the Growth of the Coercive State," Com-
parative Political Studies 21, no. 1 (1988).
5 The Soviet Union intervened in East Germany in 1953, in Hungary in 1956, and in
Czechoslovakia in 1968. The Polish Army declared martial Jaw (with Soviet support) in 198o, and the Red Army invaded Afghanistan in 19Bo in an attempt to prop up a faltering Marxist regime there. The United States helped topple the Mossadegh regime in Iran in 1953 and the Arbenz regime in Guatemala in 1954 and played a subordinate role in removing the Allende regime in Chile in 1968. In 1983, a U. S. invasion ousted the New Jewel Movement in Grenada, and the U. S-backed contras forced the Sandinista government in Nicaragua to agree to new elections (in which they were voted out of office) after a protracted and bloody civil war.
? ? ? Conclusion
there is a large disparity in power between the two states involved. Even in such cases, however, the effort often turns out to be greater than the inter- vening power anticipated.
In short, the perceptions of a mutual offensive advantage that accompany most revolutions are especially dangerous, because the combination of in- security and overconfidence that leads to the use of force is usually ill- founded. In fact, revolutions are hard to export and difficult to reverse, and in most cases, both sides are more secure than they think and would be bet- ter off remaining at peace.
The tendency for revolutions to foster perceptions of offense dominance is not universal, and the absence of this factor is the main reason why the American, Turkish, and Mexican revolutions did not lead to war. These rev- olutions altered the balance of power and generated both real conflicts of in- terest and spirals of suspicion. Yet in each case, key leaders were aware that the revolution was unlikely to ? pread and that war would be expensive. As a result, both sides ultimately refrained from the large-scale use of force.
UncertaintyandMisinformation. Ifrevolutionsarebothhardtoexportand difficult to reverse, then why do states worry about either possibility? Our cases provide part of the answer: it will be extremely difficult for states to gauge their situation accurately after a revolution, because relations be- tween revolutionary states and other powers will be afflicted by very high levels of uncertainty and misinformation. The problem arises from several different aspects of the revolutionary process, so it will usually be difficult to overcome.
For example, it is hardly surprising that both sides have trouble estimat- ing the balance of military power, because the military capacity of the new state will rest on novel institutions whose effects can only be discovered through battlefield experience. Foreign powers will usually have good rea- sons to discount a revolutionary state's capabilities, if only because the armed forces of the old regime usually deteriorate in the short term. Yet in almost every case examined here, the revolutionary regime managed to use new myths, symbols, and institutions to create an unexpectedly formidable military machine. 6
The political consequences of a revolution are even harder to calculate in advance, because the political appeal of a revolutionary model is virtually impossible to gauge with confidence. Neither the revolutionaries nor their foreign opponents know if the revolution will attract adherents abroad or if foreign intervention will spark a counterrevolutionary upheaval. Faced
6 The United States and Mexico are partial exceptions to this claim, because neither revo- lution faced a large-scale foreign invasion. The rapid mobilization of U. S. naval power dur- ing the Quasi-War and the Constitutionalists' unexpected success in the civil war against Villa are consistent with this argument, however.
[3371
? Revolution and War
with this uncertainty, elites on both sides will tend to rely on ideology and recent experience and to assume that the initial revolutionary success was a harbinger of things to come. Even relatively modest signs of a sympathetic response elsewhere will be taken as evidence of a rising revolutionary tide, and testimony from foreign revolutionaries or counterrevolutionary exiles can reinforce these erroneous expectations.
Both sides overstate the likelihood of counterrevolution for much the same reason. The danger of an "aristocratic conspiracy" terrified the French from 1789 onward; the Bolsheviks feared their hold on power might lapse at any moment; the Islamic Republic has waged a brutal campaign against for- mer opponents as well. These fears reflect the intrinsic difficulty of accu- rately gauging the loyalty of the population at large. Although signs of dissent will always be present, neither the ruling authorities nor outside powers can know how strong the opposition really is.
To make matters worse, revolutions impede the acquisition of the infor- mation that might help correct these erroneous impressions. The revolution- ary states in France, Russia, and Iran were cut off from normal diplomatic contacts, intensifying their perceptions of threat and making subsequent ef- forts to reach a modus vivendi more difficult. Indeed, a striking feature of most of these cases is the extent to which states were forced. to conduct dipl9- macy through unofficial agents whose expertise and reliability usually left much to be desired.
Lastly, with all the uncertainty, other states will have trouble deciding how to respond to a revolution, and foreign powers will be hard-pressed to agree on a common course of action. 7 If the power of a revolutionary state is unclear, the danger of contagion uncertain, and the prospects for counter- revolution unknown, it will be difficult to obtain a consensus for interven- tion or abstention, and the other states will be more likely to respond in a haphazard and poorly coordinated manner. This problem arose in several of our cases: the European powers were often divided over how to respond to the revolution in France; the Entente could not adopt a unified policy to-
ward the Bolshevik or Turkish revolutions; Britain and the United States disagreed over the proper approach to take toward the revolutionary gov- ernments in Mexico and China; and there was little consensus on how to re- spond to Iran's revolution until its army had crossed into Iraqi soil. The result, unfortunately, may have been the worst of both worlds: other states did enough to antagonize the new regimes but not enough to eliminate them.
7 I identify some other reasons for this tendency in my chapter, "Collective Security and Revolutionary Change: Promoting Peace in the Former Soviet Empire," in Collective Secu- rity after the Cold War, ed. George W. Downs (Ann Arbor: University of Michigan Press, 1993).
? ? ? ? ? Conclusion
Realism and Revolution
The cases examined in this book both confirm the basic explanatory power of realism and suggest which strands of realist thought are most useful. In particular, realism is most powerful when it goes beyond a nar- row focus on the gross distribution of power and instead examines the bal-
ance of threats. Although the balance of power is hardly irrelevant, the behavior of states is also affected by how national leaders assess the inten- tions of other states and how they perceive the relative advantage to of- fense or defense. The offense-defense balance is not merely a function of
geography or military doctrine, however, but may also be affected by the potential appeal of particular ideas (such as a new revolutionary ideology). Realism gains even more explanatory power by incorporating the possibil- ity that states will misread these factors, and as we have seen repeatedly, such misperceptions are especially common after revolutions. 8 Beliefs about the possibility of revolutionary contagion-which ultimately rest on beliefs about the persuasive power of revolutionary ideas-have been crit- ical in shaping perceptions of the offense-defense balance and help explain why some revolutions led to war and others did not. In short, by incorpo- rating domestic politics and ideas into the anarchic setting depicted by re-
alism, we obtain a more complete picture of the forces that shape state behavior.
This argument points to another insight: realism may tell us more about international behavior in postrevolutionary periods than in more "normal" periods. Some realists depict international politics as a relentless struggle for survival whell"e security is extremely scarce and states must constantly strive for any advantage. 9 Proponents of this perspective are likely to view
revolutionary states as exceptions to realist logic, because their foreign pol- icy objectives are heavily affected by ideology and their leaders are pre- sumed to be less familiar with the subtleties of international diplomacy. Yet in the three main cases examined here, each side saw the other as an immi- nent and intense threat and discounted the possibility of a lasting peace. In other words, relations between these revolutionary states and other powers were virtually identical to the relationships depicted by the more extreme versions of realism. The key point, however, is that the level of conflict is not
8 Stephen Van Evera refers to this as "Type IV Realism. " It differs from other strands of re- alist theory by focusing on what Van Evera calls the "fine-grained structure of power" and by explicating the factors that shape how states perceive that structure. Causes ofWar, vol. 1: The Structure ofPower and the Roots ofWar (Ithaca: Cornell University Press, forthcoming), chap. 1. On the impact of ideas on the offense-defense balance, see George Quester, Offense and De-
fense in the International System (New York: Wiley, 1977), 67.
9 This view is most clearly expressed in the writings of John J. Mearsheimer: "Back to the
Future: Instability in Europe after the Cold War," International Security 15, no. 1 (1990); and "The False Promise of International Institutions," International Security 19, no. 3 (1994-95).
[3391
? Revolution and War
merely due to the state of anarchy and a particular distribution of capabili- ties (as a neorealist such as John Mearsheimer would have it) but is also heavily influenced by perceptions of intent and beliefs about the likelihood that the revolution will spread or be reversed.
With frhe passage of time, however, each side will acquire a more accurate estimate of the true threat that it faces. The security dilemma between them will ease and each will adopt a less vigilant posture. The relentless competi- tion depicted by some realists will abate, and prospects for cooperation will probably increase. By restricting themselves to examining the gross distri- bution of power, however, neorealists cannot explain why the level of con- flict varies even though the distribution of power is constant.
This impression was at best a caricature and at worst extremely mislead- ing. Although its leaders were willing to implement far-reaching and ill- conceived domestic programs such as the Great Leap Forward, China's foreign policy was acutely sensitive to external constraints and its use of
188 See Shu Guang, Deterrence and Strategic Culture, chaps. ? ; Gurtov and Hwang, China under Threat, chap. 3; Jan Kalicki, The Pattern ofSino-American Crises: Political-Military Interac- tions in the 1950s (Cambridge: Cambridge University Press, 1975), chaps. 3, 6, 8; and Allen S. Whiting, "New Light on Mao: Quemoy 1958, Mao's Miscalculations," China Quarterly 62 (1975)?
189 Foot, Wrong War, 27-28.
190 Quoted in Shu Guang, Deterrence and Strategic Culture, 239?
191 ForMao'sstatementsaboutnuclearweapons,seeAliceLangleyHsieh,ComunistChina's
Strategy in the Nuclear Era (Englewood Cliffs, N. J. : Prentice-Hall, 1962), 1, 52, 132; and Stuart Schram, Mao Tse-tung (Harmondsworth: Penguin, 1966), 291. On China's nuclear program, see John Wilson JLewis and Xue Litai, China Builds the Bomb (Stanford: Stanford University Press, 1988); and Avery Goldstein, "Understanding Nuclear Proliferation: Theoretical Expla- nation and China's National Experience," Security Studies 2, nos. 3-4 (1993).
192 See Chang, Friends and Enemies, chap. 8.
193 In 1966, Assistant Secretary of State William P. Bundy referred to the PRC as "a gov- ernment whose leadership is devoted to the promotion of communism by violent revolu- tion," and Secretary of State Dean Rusk told a congressional hearing that "a country whose behavior is as violent, irascible, unyielding, and hostile as that of Communist China is led by leaders whose view of the world and of life itself is unreal. " See Franz Schurmann and
Orville Schell, eds. , The China Reader, vol. 3: Communist China (New York: Vintage, 1967), 378-79- 508.
? ? The American, Mexican, Turkish, and Chinese Revolutions
force was cautious and restrained. 194 Mao was willing to raise tensions in the Taiwan Straits in order to convince the United States that the PRC could not be intimidated, but he went to some lengths to limit the risks and stayed well clear of the brink. 195 With the exception of Korea and Vietnam (where . Chinese strategic interests were directly involved), Beijing's support for "wars with national liberation" was more rhetorical than real, and Mao stressed that foreign revolutionaries must rely on their own efforts. 196 Simi- larly, Mao's seemingly provocative statements about nuclear weapons were intended primarily to bolster Chinese morale and to convince the United States that China could not be cowed by nuclear threats; significantly, Mao's private remarks were quite prudent, and his public statements became more
moderate as China's nuclear capability increased. 197
China's overriding concern for security was also evident in its sensitivity
to shifts in the global balance of power and its willingness to ignore ideo- logical principles in responding to them. 198 During the 1950s, the PRC saw the United States as its principal threat and chose to "lean to one side" with the Soviet Union. By the end of the decade, dissatisfaction with Soviet sup- port led Mao to adopt a neutral position between the two superpowers, while China expanded its ties with Western Europe and Japan and tried to unite the Third World in an "international united front. " The latter effort did not attract many adherents, however, and the onset of the Cultural Revolu- tion in 1966 initiated a brief period of self-imposed isolation. 199
194 See Allen S. Whiting. "The Use of Force in Foreign Policy by the Peoples' Republic of China," A n n a l s of t h e A m er i ca n A c a d e m y of P o l i t i c a l a n d Soci al S c i e n c e 4 0 2 ( 1 9 7 2 ) , 5 5 ? , an d Th e C h i n e s e C a l - culus ofDeterrence: India and Indochina (Ann Arbor: University of Michigan Press, 1975).
195 During the 1954 crisis, Mao prohibited "offensive actions against foreign warships and airplanes," and the Central Military Commission told Chinese troops that they could retali- ate to a U. S. attack only "if this involves no grave risks. " When the GMD evacuated the off- shore island of Dachen in February 1955, the Central Military Commission denied a local commander's request to occupy the island for fear that it would entangle them with U. S. forces. Mao acted cautiously in 1958 as well, rejecting a proposed air strike on Quemoy in Au- gust and ordering his artillery not to fire on U. S. ships. See Shu Guang. Deterrence and Strate- gic Culture, 198, 219, 236-37; and Chang. Friends and Enemies, 72.
197 In 1965, Mao claimed that his calling the atomic bomb a "paper tiger" was "just a figure of speech. " He suggested that a nuclear war "would be a catastrophe for the whole world" and stated, "If one must fight one should confine oneself to conventional weapons. " See Schurmann and Schell, China Reader, 364-65. See also Avery Goldstein, "Robust and Afford- able Security: Some Lesson from the Second-Ranking Powers During the Cold War," Journal ofStrategic Studies 15, no. 4 (1992), 492-{)6, 5oo-503; and Lewis and Xue, China Builds the Bomb, 34, 215-18, and app. A.
198 According to A. Doak Barnett, "when security interests have been at stake, Chinese Communist leaders have generally . . . given them priority over other interests, including ide- ological and economic ones. " China and the Major Powers in East Asia (Washington, D. C. : Brookings Institution, 1977), 254-55.
199 See Harry Harding. "China's Changing Roles in the Contemporary World," in Harding. China's Foreign Relations, 186-87; and Armstrong, Revolutionary Diplomacy, 77-90.
? 196 See Van Ness, Revolution and Chinese Foreign Policy, 72.
? ? ? Revolution and War
By the late 1960s, the growth of Soviet military power in the Far East and a series of border clashes between Soviet and Chinese troops convinced Mao that the main threat was "Soviet hegemonism. "200 Just as the PRC had once "leaned to one side" to balance the perceived threat from the United States, it now turned to the United States to balance the threat from the So- viet Union. The rapprochement became overt after President Richard Nixon's 1972 visit to Beijing, and given the aura of suspicion that had sur- rounded Sino-American relations since the late 1940s, it would be hard to find a more vivid example of unsentimental realpolitik. Moreover, this shift ? occurred in period when ideology was especially important in shaping pol- icy within China. Thus, although the Sino-Soviet alliance can be seen as the result of either balancing or ideological solidarity, the Sino-American rap- prochement represented strictly the former. Z01
China's willingness to subordinate ideological principles to its security in- terests was evident in other relationships as well. Despite their public com- mitment to "Asian solidarity" and "peaceful coexistence," recurring border disputes and mutual suspicions led to a major deterioration of Sino-Indian re- lations in the late 1950s. The two states eventually fought a brief border war in 1962, and China's forces acted both cautiously and defensively in repelling the Indian challenge. India's subsequent tilt toward the Soviet Union was matched by a de facto alliance between China and Pakistan, despite the lat- ter's nonsocialist character and its close ties to the United States. 202 Similarly, the U. S. withdrawal from Indochina in the early 1970s allowed the long-sup- pressed rivalry between China and Vietnam to reemerge; Vietnam strength- ened its alliance with the Soviet Union while China backed the Khmer Rouge in Cambodia, and China and Vietnam eventually fought a brief but intense border war following the Vietnamese invasion of Cambodia in 1979.
In addition to providing evidence of China's willingness to sacrifice prin- ciples to the requirements of balance-of-power politics, these developments can be viewed as part of the gradual process of learning and adaptation. Al- though the United States and China continued to exaggerate each other's hostility after the Korean War, they avoided a direct clash in subsequent crises in part because each had learned that such a war would be costly. 203
200 See Thomas W. Robinson, "The Sino-Soviet Border Conflict," in Stephen S. Kaplan, DiplomacyofPower:SovietArmedForcesasaPoliticalInstrument(Washington,D. C. :Broko ings Institution, 1g81).
202 See Anwar H. Syed, China and Pakistan: Diplomacy ofan Entente Cordiale (Amherst: Univer- sity of Massachusetts Press, 1974), chap. 4; and Ya'acov Vertzberger, The Enduring Entente: Sino- Pakistani Relations, 196o-t980, Washington Papers no. 95 (New York: Praeger, 1983), 15-24.
203 China reportedly suffered over 26o,ooo dead in Korea, while the United States lost around47,00. 0 AwarenessofwhatanotherclashmightcosthelpedkeeptheUnitedStatesout of Indochina in the 1950s, and the fear of Chinese intervention constrained U. S. behavior dur- ing the height of its subsequent involvement there.
201 I am indebted to Avery Goldstein for this insight.
? ? The American, Mexican, Turkish, and Chinese Revolutions
Over time, the PRC gradually evolved from a radical state to a reformist state to a conservative state, and although the CCP remains in power as of this writing, its policies bear little resemblance to the ideals of Marx or Mao, and its foreign policy aims are largely indistinguishable from the goals of other great powers.
The Chinese Revolution fostered intense perceptions of threat with other major powers, based on the shared belief that the other was inherently ag- gressive and that the threat could be reduced or eliminated through the use of force. Over time, however, the impact of these early views faded and the new regime and its neighbors gained a more accurate understanding of their respective interests and capabilities. Despite its revolutionary origins, revolutionary China generally behaved in a restrained and prudent manner after the Korean War. My theory thus receives further support.
Although the comparison is hardly novel, the similarities between the Chinese and Russian experiences are worth noting. Both regimes came to power after a world war, a situation that helped shape their early percep- tions and gave them a "breathing space" in which to consolidate their power. Both fought and won a civil war, faced pressure from hostile imperi- alist powers, and exaggerated their opponents' willingness to attack them directly. Having fought wars shortly after gaining power (though China's involvement in Korea was far more costly than Russia's war with Poland), both states gradually overcame their initial isolation and reemerged as ac- cepted (albeit suspect) players on the world stage. Finally, although both China and Rutssia were willing to compromise their ideological principles in order to preserve their security, neither abandoned these ideals completely, and they continued to interpret world events through the distorting prisms of a revolutionary worldview. This tendency helps explain why other states kept them at arm's length even when there were obvious incentives to forge a closer relationship.
CONCLUSION
Each of the revolutions examined in this chapter increased the intensity of security competition, and each of these states came close to war on one or more occasions. The conflicts arose for reasons that are consistent with the mechanisms identified in chapter 2 and bear more than a passing resem- blance to the dynamics observed in the French, Russian, and Iranian cases.
First, foreign reactions to each of these revolutions were heavily influ- enced by the potential impact on the balance of power. To France, the Amer- ican Revolution was an opportunity to weaken Britain; several foreign states saw the turmoil in Mexico as both a threat to their existing interests
? ? ? Revolution and War
and a chance to improve their positions at others' expense; Great Britain and Greece considered the revolution in Turkey a "window of opportunity" in the Near East; and the Chinese revolution was viewed as a major gain for the Soviet Union. In each case, foreign powers sought to take advantage of the power vacuum produced by the revolution, and their efforts led to con- flict with the new regime, other foreign powers, or both.
Second, after the revolution each of these states experienced spirals of hostility with other states that brought them to the brink of war at least once. The revolutionary state and other powers tended to suspect eaclh other's intentions, and each new regime was jealous of its sovereignty and obsessed with issues of security. Although each of these states did face real security threats after the revolution, the threats were usually exaggerated and, at times, wholly illusory. This paranoia was most pronounced in the case of China, but misunderstandings and exaggerations marred Anglo- Turkish relations during the Nationalists' bid for power, exacerbated U. S. relations with Britain and France in the 1790s, and disturbed revolutionary
Mexico's relations with the United States as well. As expected, lack of infor- mation and poor channels of communication generally made these prob- lems worse.
These states also exhibited the expected combination of vulnerability and! optimism, although the intensity varied greatly. U. S. leaders were deeply fearful of foreign interference yet confident that their new nation would! eventually control the entire continent, and their long-term optimism dis- couraged the use of force on several occasions. The Nationalists in Turkey and the Constitutionalists in Mexico were equally opposed to foreign inter- ference and willing to use force to prevent it, yet each ultimately preferred to negotiate rather than fight; moreover, foreign intervention was clearly discouraged by an awareness of what such a campaign might entait Thus, the predicted mixture of fear and overconfidence was muted in these three
. cases. It was clearly present in China, however, and helps explain Mao's de- cisi? n to intervene in Korea in 1950. Intervention risked U. S. retaliation and! jeopardized the reconstruction of his war-torn country, but Mao was con- vinced that the threat was too great to ignore and that his army could ac- complish nts mission at an acceptable cost.
Given the presence of these familiar sources of conflict, it is not surprising that each of these states came close to war. Yet as we have seen, the Ameri- can, Mexican, and Turkish revolutions did not go over the brink. The expla- nation lies in the strong condition of defense dominance that accompanied each; unlike in the four other cases examined in this book, both the internal character and external circumstances of the American, Mexican, and Turk- ish revolutions made the use of force less attractive. Although conflicts and crises did occur, both sides ultimately concluded that going to war would not be worth the cost and risk.
? ? The American, Mexican, Turkish, and Chinese Revolutions
This condition of defense dominance was attributable to three main fac- tors. First, the American, Mexican, and Turkish revolutions were all com- paratively moderate, relative to our other cases. In particular, none was based on a explicitly universalist ideology (the Founding Fathers being at best ambivalent about the ability of other societies to replicate the U. S. ex-
perience), and none of these states did much to export their principles to other countries. This greatly reduced the potential threat they posed and re- duced the need for other states to take countermeasures to contain them. Another consequence was to allow each of these states to reenter the exist- ing system of states more rapidly than postrevolutionary France, Russia, Iran, or China could do.
Second, each of these three states was both large and relatively weak-es- pecially in comparison with the other great powers-making a war by mis- calculation less likely. U. S. weakness helped limit pressures for war in the 1790s, as both Washington and Adams ultimately chose to negotiate with their foreign adversaries rather than risk a war for which they were clearly unprepared. Similarly, early in this century Mexican and U. S. leaders both understood that Mexico was by far the weaker of the two (and thus not much of a threat), though its sheer size would have made it difficult to oc- cupy and subdue. As a result, both sides had ample reason to act with fore- bearance. The Turks were willing to fight the Greeks, Armenians, or Georgians (the latter two being even weaker than they), but Kemal recog- nized that a war with the Entente would be counterproductive and possibly
fatal. By contrast, the combination of China's vast population, Soviet sup- port, and the proximity of Korea made it possible for Mao to contemplate the overt use of force, especially as he believed that the United States was preparing to attack as soon as it established its position in the Far East.
Third, these three revolutionary states were favored by geographic isola- tion and fortuitous timing. The United States lay an ocean away from the other great powers, and its potential opponents were preoccupied by events closer to home. Mexico enjoyed similar advantages; although the outbreak of World War I encouraged a certain amount of foreign meddling, the war in Europe distracted its potential opponents and played a key role in dis- suading the United States from more active interference. The Turks profited. from the revolution in Russia and the war-weariness afflicting the other Eu- ropean powers, which discouraged Western intervention and allowed the Turks to set the two sides against each other. Had the timing of these revo- lutions been different, it is easy to imagine a less favorable outcome.
Finally, the cases examined in this chapter suggest a reciprocal connection between revolution and war. Revolutionary change will make war more . likely, but the onset of war will shape the revolutionary process itself. In France, the outbreak of war in 1792 radicalized the revolution and led to the founding of the republic. Foreign intervention encouraged the Bolsheviks to
? ? Revolution and War
take more extreme measures, and the outbreak of war with Iraq strength- ened the radical elements in Iran as well. By contrast, the American, Mexi- can, and Turkish revolutions had more moderate outcomes in part because these states managed to avoid war until the revolution was essentially com- plete.
The American, Mexican, Turkish, and Chinese revolutions enhance our confidence in the propositions developed elsewhere in this book. That con- fidence is increased by the diversity of the cases and by the fact that even the apparent exceptions demonstrate that revolutions create strong pressures for war. In the next and final chapter, I consider whether more recent events fit this pattern as well and summarize the lessons that national leaders might draw from these results.
? ? ? ? ? ? Conclusion
"Revolutions . . . occurred and will always occur so long as human na- ture remains the same. "
-Thucydides
"In a revolution, as in a novel, the most difficult part to invent is the end. " -Alexis de Tocqueville
This boolk has explored some of the ways that revolutions affect interna- tional politics, focusing primarily on the relationship between revolution and war. I argued that revolutions alter the balance of threats between states, leading to more intense security competition and a heightened prob- ability of war. I tested and refined this argument by examining three major revolutions in detail-those of France, Russia, and Iran-as well as four ad- ditional cases where the fit between theory and reality was less obvious.
Four tasks remain. The first is to summarize and compare the results of the seven case studies, in order to highlight the principal theoretical conclu- sions we may infer from these events.
The second is to identify the policy implications of these results: when a revolution occurs, what precautions should other states take? What actions should they avoid? The third is to sketch what the theory tells us about the recent collapse of the Soviet empire and its effects on the likelihood of war. The final task is to consider the long- term relevance of this study: is mass revolution a fading phenomenon, or are the problems caused by past revolutions likely to occur in the future?
REVOLUTION, SECURITY COMPETITION, AND WAR
The cases examined in this book confirm that revolutions increase the in- tensity of security competition between states and raise the probability of war. Although war did not occur in every case, each regime came close to war soon after gaining power and each revolution fostered greater security competition among the other major powers. The occurrence of revolution
[331]
? ? Revolution and War
was hardly the only source of competition and conflict, but in each case the level of tension was higher than it would have been without the revolution. In the French case, the collapse of the old regime caused competition be- tween Austrfta, Prussia, and Russia to rise, triggered a spiral that placed France and Austria on a collision course, and brought the rivalry with En- gland to the point of open warfare in 1793. Although war might have oc-
curred even if the revolution had been averted, the internal turmoil in France was directly responsible for the war that did break out and for its rapid expansion.
Similarly, although the collapse of the Romanov dynasty temporarily lowered the level of conflict by taking Russia out of World War I, the estab- lishment of Soviet Russia ultimately increased it. The revolution encour- aged other states to compete over the lands of former tsarist empire and brought to power a regime that supported the violent overthrow of other states. The tsarist regime would never have acted as Soviet Russia did, and the other great powers would never have seen it as a subversive force. Again, it is safe to infer that the revolution was directly responsible for much of the enmity and insecurity that characterized relations between So- viet Russia and most other powers.
This pattern is even more apparent in the Iranian case. The revolution alarmed Iran's Arab neighbors, led directly to a protracted war with Iraq, and transformed U. S. -Iranian relations from close alliance to bitter rivalry. Relations with most other countries deteriorated as well, leaving the Islamic Republic largely isolated. Although the shah's ambitions had alarmed his neighbors on occasion, they never provoked the level of international en- . mity that the Islamic Republic has. Thus, Iran's present position is simply incomprehensible outside the context of the revolution.
The same effects were present in the other four cases, albeit to varying de- . grees. The United States experienced repeated conflicts with both its former British rulers and its putative French ally, coming close to war with each on one or more occasions. The revolution in Mexico exacerbated the competi- tion for influence between the United States, Britain, and Germany, led the United States to intervene in 1914 and again in 1916, and continued to mar relations between Mexico and the United States until the late 1930s. The Na- tionalist revolution in Turkey challenged European ambitions in the Near East, brought the new regime to the brink of war with Great Britain in 1922, and eventually forced the Entente to withdraw its troops and abandon the Treaty of Sevres. Needless to say, it is hard to imagine the Ottoman sultan achieving like results. Finally, the Communist triumph in China ended
decades of foreign interference, moved Beijing firmly into the socialist camp, and placed the new regime on the path to war in Korea. Once again, it is unlikely that Chiang Kai-shek and the GMD would have acted the same way or provoked similar responses.
? ? Conclusion
Hypothetical scenarios can never be tested, of course, but the possibility that these states would have experienced equally high levels of conflict in the absence of revolution seems remote. Although other causes of conflict were undoubtedly present, these cases show that revolutions have indepen- dent causal effects on the level of security competition and the probability of war.
Balance-of-Threat Theory
Why do revolutions make competition and war more likely? As described in detail in chapter 2, balance-of-threat theory suggests that revolutions cause security competition by altering the perceived level of threat between the revolutionary state and its main adversaries, on one hand, and by en- couraging both sides to believe that the use of force can overcome the threat at an acceptable cost, on the other.
The evidence strongly supports this general argument. The crises or wars that followed each of the revolutions examined here resulted from a combi- nation of opportunism and insecurity, based on misjudgments about the balance of power, overly malign perceptions of intent, and (in the worst cases) exaggerated beliefs about the likelihood of contagion or counterrevo- lution. Compounding the problem were uncertainty and misinformation, which reinforced each side's prior beliefs and made peaceful settlements more elusive.
TheBalanceofPower. IntheFrench,Russian,andIraniancases,therevo- lution's effect on the balance of power was a central cause of war. Yet the im- pact of a revolution on the balance of power does not cause war by itself. In particular, these effects cannot explain why some states try to exploit the op- portunities while others remain aloof, nor can they account for aggressive behavior on the part of the revolutionary state. For example, although Prus- sia saw French weakness after 1789 as a chance for expansion, most Euro- pean states welcomed the erosion of French power and did not use force to exploit it. Instead, the war began when, reacting to Austria's efforts to in- timidate the French, the Girondins convinced the Assembly to declare war
in April 1792. Similarly, Iran's weakness after the revolution does not fully explain why Iraq saw a military attack as desirable or necessary; the appar- ent collapse of Iranian power might just as easily been considered an op- portunity for Iraq to turn its attention to other problems. ' The four cases examined in chapter 6 reinforce this conclusion: although the American, Mexican, Turkish, and Chinese revolutions created significant power vacu-
1 Furthermore, focusing on power alone cannot explain why Iraq decided to go to war but ? Iran's other neighbors (the Soviet Union, Turkey, Afghanistan) did not.
? ? ? ? ? Revolution and War
urns and fostered greater security competition between other powers, only China was subsequently involved in war. If the belief that a revolution has weakened its victim is part of the link between revolution and war, it is hardly the whole story.
PerceptionsofIntent. Eachoftheserevolutionsproducedsharpdepartures from the foreign policies of the old regime, in tum creating sharp conflicts of interest with other states. Furthermore, both the revolutionary state and tlhe other great powers tended to exaggerate one another's hostility and aggres- siveness. One of the most characteristic features of postrevolutionary for- eign policy, this tendency is also one of the most pernicious.
These cases also confirm that spirals of suspicion can arise from several distinct sources. The most obvious source is ideology: if the worldview of a revolutionary movement stipulates that certain regimes are inherently hos- tile, the new regime is likely to interpret the behavior of foreign powers in the worst possible light. Asecond source is domestic politics. As the French and Iranian cases suggest, factions within a revolutionary movement may dramatize foreign dangers in order to consolidate their own positions. Spi- raling may also be fueled by testimony from emigres or foreign revolution- aries, whose desire for support gives them an obvious incentive to foster conflicts between the revolutionary state and other powers. This tactic played a modest role in several of these cases; French emigres echoed Euro- pean fears of the revolution in France, Russian exiles stiffened Allied resis- tance against normalizing relations with the Bolshevik regime, and Iranian exiles (including the shah) contributed to the deterioration of Iranian for- eign rellations. The activities of the "China lobby" and the misleading testi- mony provided by some U. S. diplomats after the American Revolution further illustrates this danger.
Offense, Defense, and the Export ofRevolution. The cases examined in this study also confirm that revolutions cause war by affecting perceptions of the offense-defense balance, conceived in both military and political terms. In the French, Russian, Iranian, and Chinese cases, decisions to go to war were encouraged by a combination of fear and overconfidence, usually based on each side's calculations of the likelihood that the revolution will spread or be reversed. Expecting that their example would be contagious, the revolutionary state was more confident, less willing to compromise, and prone to support revolutionary efforts abroad. At the same time, the general belief that the revolution might also be easily reversed made exporting the revolufrion seem necessary to the revolutionary state and helped convince its adversaries that they could eliminate the threat with little effort.
Ironically, history suggests that both these beliefs are usually misguided. Although each of these revolutions was accompanied by evidence of discon-
? ? Conclusion
tent in other societies, none of them spawned successful imitators during the decade after the seizure of power, and efforts to spread the revolution via propaganda or contagion only angered and alarmed other states. The Girondins' predictions of a universal crusade for liberty were disappointed; "Bolshevik" uprisings in Finland, Hungary, and Bavaria collapsed quickly, the Polish proletariat welcomed the Red Army with bayonets instead of flowers; and Soviet attempts to spark Communist revolutions in the Near East and China all failed. Efforts to export the Iranian Revolution have been equally abortive thus far, despite the universalist pretensions of Khomeini's Shiism and Iran's support for fundamentalist groups throughout the Muslim world. We should not be surprised at these results, however, because would- be propagators of revolution face several significant obstacles.
First, although a revolution often comes as a? surprise to virtually every- one (including the revolutionaries themselves), it also provides a timely warning to others. As a result, potential targets will be less prone to the mis- takes that let the old regime be toppled. Thus, the French example alerted the other European powers to keep a close watch on potential "Jacobins" and to make a number of modest reforms. These measures were universally successful except in areas conquered by French troops. Similarly, states fac- ing a Bolshevik challenge acted vigorously to suppress potential uprisings after 1917, just as the Persian Gulf states suppressed, coopted, or expelled anyone suspected of spreading Iran's revolutionary message. Louis XVI in Paris, Nicholas II in St. Petersburg, and Mohammad Reza Pahlavi in Tehran may have dithered their way to their own destruction, but their contempo- raries in other countries seem to have learned from their mistakes.
Second, potential victims of a spreading revolution also learn to balance against this frightening possibility. The growing danger from revolutionary France led to the formation of a large (if unruly) coalition by the summer of 1793, and the Entente maintained a common front against Bolshevism until the mid-192os. In the same way, the threat from Iran prompted greater co- operation between Iraq, Saudi Arabia, Kuwait, and the United States. 2
Third, the failure of revolutions to spread underscores the advantages that incumbent regimes ordinarily enjoy. Even states with severe internal problems usually retain some shreds of legitimacy, as well as a substantial' asymmetry of power over their internal rivals. Moreover, the combination of favorable circumstances and coincidences that make the first revolution possible are unlikely to occur elsewhere in precisely the same fashion. Thus, pro-French radkals in England were quickly overcome by prompt govern- ment action, the German Communists proved to be no match for the Reichs-
2 By contrast, because foreign powers were not especially worried that the U. S. , Mexican, or Turkish revolutions would spread, they did not band together in strong opposing coali- tions.
[335]
? ? ? Revolution and War
wehr in 1919, 1921, and 1923, and the Shiites in Iraq were crushed when their leaders tried to duplicate the Iranian experience. Given the barriers to a revolution's spreading beyond a single state's borders, the real mystery is why anybody believes that it wilV
The evidence also suggests that reversing a revolution is nearly as diffi- cult as spreading one. Revolutionary regimes survive because they are usu- ally adept at mobilizing military power; whatever their other failings, revolutionary movements are especially good at persuading people to run grave risks and make large sacrifices for the sake of an ideal. 4 Foreign inter- ference can facilitate this task by providing the legitimacy that a revolution- ary regime needs, and a foreign threat can make it easier for leaders to eliminate rivals in the name of "national unity. "
When foreign powers do not possess reliable sources of information and do not discount the testimony of emigres sufficiently, they are likely to end up backing far-fetched counterrevolutionary schemes that hold little chance of success-as illustrated by English support for the Quiberon expedition in 1795, British and French aid to the White armies in the Russian Civil War, Iraq's support for various Iranian exile groups in Baghdad, and covert U. S. efforts to support anti-Communist groups in China. Supporting counter- revolutionary forces may be a cost-effective means of pressuring a revolu- tionary regime, but it is unlikely to remove it from power and will almost
certainly fuel its perceptions of threat.
These obstacles do not mean that counterrevolutionary efforts never suc- ceed, but success will be more difficult than other states expect. Foreign in- tervention did reverse the Dutch revolt of 1787 and the Polish "revolt" of 1791, to cite but two examples, and both superpowers intervened to reverse unwanted upheavals in their own spheres of interest throughout the Cold War. 5 These examples suggest that outside intervention can work when
3 Exceptions arise when the new regime is able to defeat and occupy its adversaries and impose its system upon their populations by force. Thus, France's "sister republics" and the Soviet empire in Eastern Europe resulted from military expansion rather than the spread of revolutionary ideals. Examples of "spontaneous" Communist revolutions include those in the People's Republic of China, North Korea, Vietnam, Cuba, Ethiopia, Angola, and Nic- aragua, but these "triumphs" occurred decades after the 1917 revolution.
4 See Theda Skocpol, "Revolutions and Mass Military Mobilization," World Politics 40, no. ? 2 ( 11}88); and Ted Robert Gurr, "War, Revolution, and the Growth of the Coercive State," Com-
parative Political Studies 21, no. 1 (1988).
5 The Soviet Union intervened in East Germany in 1953, in Hungary in 1956, and in
Czechoslovakia in 1968. The Polish Army declared martial Jaw (with Soviet support) in 198o, and the Red Army invaded Afghanistan in 19Bo in an attempt to prop up a faltering Marxist regime there. The United States helped topple the Mossadegh regime in Iran in 1953 and the Arbenz regime in Guatemala in 1954 and played a subordinate role in removing the Allende regime in Chile in 1968. In 1983, a U. S. invasion ousted the New Jewel Movement in Grenada, and the U. S-backed contras forced the Sandinista government in Nicaragua to agree to new elections (in which they were voted out of office) after a protracted and bloody civil war.
? ? ? Conclusion
there is a large disparity in power between the two states involved. Even in such cases, however, the effort often turns out to be greater than the inter- vening power anticipated.
In short, the perceptions of a mutual offensive advantage that accompany most revolutions are especially dangerous, because the combination of in- security and overconfidence that leads to the use of force is usually ill- founded. In fact, revolutions are hard to export and difficult to reverse, and in most cases, both sides are more secure than they think and would be bet- ter off remaining at peace.
The tendency for revolutions to foster perceptions of offense dominance is not universal, and the absence of this factor is the main reason why the American, Turkish, and Mexican revolutions did not lead to war. These rev- olutions altered the balance of power and generated both real conflicts of in- terest and spirals of suspicion. Yet in each case, key leaders were aware that the revolution was unlikely to ? pread and that war would be expensive. As a result, both sides ultimately refrained from the large-scale use of force.
UncertaintyandMisinformation. Ifrevolutionsarebothhardtoexportand difficult to reverse, then why do states worry about either possibility? Our cases provide part of the answer: it will be extremely difficult for states to gauge their situation accurately after a revolution, because relations be- tween revolutionary states and other powers will be afflicted by very high levels of uncertainty and misinformation. The problem arises from several different aspects of the revolutionary process, so it will usually be difficult to overcome.
For example, it is hardly surprising that both sides have trouble estimat- ing the balance of military power, because the military capacity of the new state will rest on novel institutions whose effects can only be discovered through battlefield experience. Foreign powers will usually have good rea- sons to discount a revolutionary state's capabilities, if only because the armed forces of the old regime usually deteriorate in the short term. Yet in almost every case examined here, the revolutionary regime managed to use new myths, symbols, and institutions to create an unexpectedly formidable military machine. 6
The political consequences of a revolution are even harder to calculate in advance, because the political appeal of a revolutionary model is virtually impossible to gauge with confidence. Neither the revolutionaries nor their foreign opponents know if the revolution will attract adherents abroad or if foreign intervention will spark a counterrevolutionary upheaval. Faced
6 The United States and Mexico are partial exceptions to this claim, because neither revo- lution faced a large-scale foreign invasion. The rapid mobilization of U. S. naval power dur- ing the Quasi-War and the Constitutionalists' unexpected success in the civil war against Villa are consistent with this argument, however.
[3371
? Revolution and War
with this uncertainty, elites on both sides will tend to rely on ideology and recent experience and to assume that the initial revolutionary success was a harbinger of things to come. Even relatively modest signs of a sympathetic response elsewhere will be taken as evidence of a rising revolutionary tide, and testimony from foreign revolutionaries or counterrevolutionary exiles can reinforce these erroneous expectations.
Both sides overstate the likelihood of counterrevolution for much the same reason. The danger of an "aristocratic conspiracy" terrified the French from 1789 onward; the Bolsheviks feared their hold on power might lapse at any moment; the Islamic Republic has waged a brutal campaign against for- mer opponents as well. These fears reflect the intrinsic difficulty of accu- rately gauging the loyalty of the population at large. Although signs of dissent will always be present, neither the ruling authorities nor outside powers can know how strong the opposition really is.
To make matters worse, revolutions impede the acquisition of the infor- mation that might help correct these erroneous impressions. The revolution- ary states in France, Russia, and Iran were cut off from normal diplomatic contacts, intensifying their perceptions of threat and making subsequent ef- forts to reach a modus vivendi more difficult. Indeed, a striking feature of most of these cases is the extent to which states were forced. to conduct dipl9- macy through unofficial agents whose expertise and reliability usually left much to be desired.
Lastly, with all the uncertainty, other states will have trouble deciding how to respond to a revolution, and foreign powers will be hard-pressed to agree on a common course of action. 7 If the power of a revolutionary state is unclear, the danger of contagion uncertain, and the prospects for counter- revolution unknown, it will be difficult to obtain a consensus for interven- tion or abstention, and the other states will be more likely to respond in a haphazard and poorly coordinated manner. This problem arose in several of our cases: the European powers were often divided over how to respond to the revolution in France; the Entente could not adopt a unified policy to-
ward the Bolshevik or Turkish revolutions; Britain and the United States disagreed over the proper approach to take toward the revolutionary gov- ernments in Mexico and China; and there was little consensus on how to re- spond to Iran's revolution until its army had crossed into Iraqi soil. The result, unfortunately, may have been the worst of both worlds: other states did enough to antagonize the new regimes but not enough to eliminate them.
7 I identify some other reasons for this tendency in my chapter, "Collective Security and Revolutionary Change: Promoting Peace in the Former Soviet Empire," in Collective Secu- rity after the Cold War, ed. George W. Downs (Ann Arbor: University of Michigan Press, 1993).
? ? ? ? ? Conclusion
Realism and Revolution
The cases examined in this book both confirm the basic explanatory power of realism and suggest which strands of realist thought are most useful. In particular, realism is most powerful when it goes beyond a nar- row focus on the gross distribution of power and instead examines the bal-
ance of threats. Although the balance of power is hardly irrelevant, the behavior of states is also affected by how national leaders assess the inten- tions of other states and how they perceive the relative advantage to of- fense or defense. The offense-defense balance is not merely a function of
geography or military doctrine, however, but may also be affected by the potential appeal of particular ideas (such as a new revolutionary ideology). Realism gains even more explanatory power by incorporating the possibil- ity that states will misread these factors, and as we have seen repeatedly, such misperceptions are especially common after revolutions. 8 Beliefs about the possibility of revolutionary contagion-which ultimately rest on beliefs about the persuasive power of revolutionary ideas-have been crit- ical in shaping perceptions of the offense-defense balance and help explain why some revolutions led to war and others did not. In short, by incorpo- rating domestic politics and ideas into the anarchic setting depicted by re-
alism, we obtain a more complete picture of the forces that shape state behavior.
This argument points to another insight: realism may tell us more about international behavior in postrevolutionary periods than in more "normal" periods. Some realists depict international politics as a relentless struggle for survival whell"e security is extremely scarce and states must constantly strive for any advantage. 9 Proponents of this perspective are likely to view
revolutionary states as exceptions to realist logic, because their foreign pol- icy objectives are heavily affected by ideology and their leaders are pre- sumed to be less familiar with the subtleties of international diplomacy. Yet in the three main cases examined here, each side saw the other as an immi- nent and intense threat and discounted the possibility of a lasting peace. In other words, relations between these revolutionary states and other powers were virtually identical to the relationships depicted by the more extreme versions of realism. The key point, however, is that the level of conflict is not
8 Stephen Van Evera refers to this as "Type IV Realism. " It differs from other strands of re- alist theory by focusing on what Van Evera calls the "fine-grained structure of power" and by explicating the factors that shape how states perceive that structure. Causes ofWar, vol. 1: The Structure ofPower and the Roots ofWar (Ithaca: Cornell University Press, forthcoming), chap. 1. On the impact of ideas on the offense-defense balance, see George Quester, Offense and De-
fense in the International System (New York: Wiley, 1977), 67.
9 This view is most clearly expressed in the writings of John J. Mearsheimer: "Back to the
Future: Instability in Europe after the Cold War," International Security 15, no. 1 (1990); and "The False Promise of International Institutions," International Security 19, no. 3 (1994-95).
[3391
? Revolution and War
merely due to the state of anarchy and a particular distribution of capabili- ties (as a neorealist such as John Mearsheimer would have it) but is also heavily influenced by perceptions of intent and beliefs about the likelihood that the revolution will spread or be reversed.
With frhe passage of time, however, each side will acquire a more accurate estimate of the true threat that it faces. The security dilemma between them will ease and each will adopt a less vigilant posture. The relentless competi- tion depicted by some realists will abate, and prospects for cooperation will probably increase. By restricting themselves to examining the gross distri- bution of power, however, neorealists cannot explain why the level of con- flict varies even though the distribution of power is constant.
