Structural Sources of China's Territorial Compromises

Correspondence: China’s Territorial Compromises Correspondence Structural Sources of China’s Territorial Compromises Albert B. Wolf M. Taylor Fravel To the Editors: More than a decade ago, in a review of Jack Snyder’s Myths of Empire, Fareed Zakaria noted: “After over a decade of vigorous debates about realism, structural realism, neoliberal institutionalism, and hegemonic stability theory, political scientists are shifting their attention to the internal sources of foreign policy. Some even contend that realism’s dictum about the ‘primacy of foreign policy’ is wrong, and that the domestic politics of states are key to understanding world events. Innenpolitik is in.”1 As evidenced by Taylor Fravel’s “Regime Insecurity and International Cooperation: Explaining China’s Compromises in Territorial Disputes,” Innenpolitik is still in.2 In Fravel’s words, “Regime insecurity best explains China’s pattern of cooperation and delay in its territorial disputes” between 1949 and 2004 (p. 81). According to the regime insecurity hypothesis, “Internal conºict often creates conditions for cooperation, producing a ‘diversionary peace’ instead of war” (p. 49). Given the domestic vulnerability of the Chinese Communist regime during periods such as the revolt in Tibet and the upheaval that followed the Tiananmen crisis, Chinese leaders reached territorial compromises to extend their tenure in ofªce (p. 51). Although intuitively plausible, Fravel’s regime vulnerability argument is ºawed, for two reasons. First, his conªdence in the empirical accuracy of the domestic insecuritydiversionary peace theory rests on a realist straw man, arguing that China’s behavior is inconsistent with the predictions of realist theories because “China has not used its power advantages to bargain hard over contested land, especially with its weaker neighbors” (p. 46). When properly construed and understood, realist theories can explain territorial compromises as well as conºict. Realist theories suggest that strong states exploit the weak, but only under certain conditions because powerful states often face powerful incentives not to exploit their weaker neighbors: “As Stuart Kaufman suggests, the logic of self-help under anarchy ‘encourages strong powers to absorb Albert B. Wolf is a graduate student in the Department of Political Science at The Ohio State University. He wishes to thank Randall Schweller and Frances Wolf for their comments on earlier drafts. M. Taylor Fravel is Assistant Professor of Political Science and a member of the Security Studies Program at the Massachusetts Institute of Technology. He thanks Kevin Narizny and Benjamin Valentino for reading a draft of this response. 1. Fareed Zakaria, “Realism and Domestic Politics: A Review Essay,” International Security, Vol. 17, No. 1 (Summer 1992), p. 177. 2. M. Taylor Fravel, “Regime Insecurity and International Cooperation: Explaining China’s Compromises in Territorial Disputes,” International Security, Vol. 30, No. 2 (Fall 2005), pp. 55–57. Additional quotations appear in parentheses in the text. International Security, Vol. 31, No. 2 (Fall 2006), pp. 199–205 © 2006 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology. 199 International Security 31:2 200 weak ones when practical.’”3 Robert Gilpin shows that imperial expansion often provides diminishing returns to imperial powers.4 Kenneth Boulding ªnds that the farther away a nation is from its home territory and resource base, the weaker it becomes. As a result, its lines of communication grow more vulnerable, reducing its ability to project power into other primary ªelds of operations and leaving it open to attack.5 Second, Fravel places realist theory in a Catch-22. On the one hand, he contends that the territorial compromises China reached with its weaker neighbors are problematic for realism because China did not use its power advantages to exploit and bargain harder with its weaker neighbors. On the other hand, Fravel purports to show that China’s territorial compromises were the result of its weakness, including its inability to maintain “the control of territory claimed by the state,” “rebellions,” and “legitimacy crises” (pp. 50, 53). Does not the Melian dialogue, one of the foundational documents in the realist canon, state that “the strong do what they can and the weak suffer what they must?”6 Furthermore, how could China have possibly used its power advantages over its weaker neighbors if its domestic situation was as volatile as Fravel suggests? Particularly troubling about Fravel’s theory is its inability to explain China’s balancing behavior during periods of supposed regime vulnerability. According to the dictates of the regime vulnerability hypothesis, when faced with threats to the tenure of the regime, insecure states will engage in territorial compromises with external foes. Yet, despite the threat that the United States and its allies in East Asia posed to China during the early Cold War, the Chinese never reached any sort of arrangement with them. In fact, rather than compromising with the United States or its allies, China chose to balance, ªrst by forming an alliance with the Soviet Union and then by improving its own material resources.7 As Avery Goldstein notes, “Under Mao Zedong’s leadership, [China’s] Cold War approach was distinctively realist, sometimes to the point of caricature.”8 Fravel posits that “China’s behavior challenges existing arguments about the foreign policies of revolutionary states,” citing Stephen Walt’s Revolution and War as an example (p. 50 n. 16, pp. 68–69). In Revolution and War, Walt utilizes balance of threat theory to show that “revolutions intensify the level of security competition.”9 Balance of threat theory explains everything Fravel’s theory explains as well as a number of things that it 3. Quoted in David Kang, “Hierarchy and Stability in Asian International Relations,” in G. John Ikenberry and Michael Mastanduno, eds., International Relations Theory and the Asia-Paciªc (New York: Columbia University Press, 2003), p. 265. 4. Robert Gilpin, War and Change in International Politics (New York: Cambridge University Press, 1981), chaps. 1–2. 5. Kenneth Boulding, Conºict and Defense: A General Theory (New York: Harper and Brothers, 1962), pp. 79, 231. See also Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1987), p. 23. 6. Thucydides, The Peloponnesian War (New York: Modern Library, 1951), p. 351. 7. Thomas J. Christensen, Useful Adversaries: Grand Strategy, Domestic Mobilization, and Sino-American Conºict, 1947–1958 (Princeton, N.J.: Princeton University Press, 1996), chaps. 1–2, 5–6; and Dale C. Copeland, The Origins of Major War (Ithaca, N.Y.: Cornell University Press, 2000), pp. 159–161. 8. Avery Goldstein, “Structural Realism: Much (but Never All) of the Story,” in Andrew K. Hanami, ed., Perspectives on Structural Realism (New York: Palgrave Macmillan, 2003), p. 122. 9. Stephen M. Walt, “Rethinking Revolution and War: A Reply to Jack Goldstone and Kurt Dassel,” Security Studies, Vol. 6, No. 2 (Winter 1996/1997), p. 174. Correspondence: China’s Territorial Compromises 201 cannot. According to balance of threat theory, states balance against those states that are geographically proximate, possess offensive military capabilities, aggregate power, and have harmful or aggressive intentions.10 Since coming to power in 1949, the Chinese Communists have reached territorial compromises with their weaker, less threatening neighbors so that they could marshal their resources against their more powerful enemies, such as the United States, and their more aggressive neighbors, such as Taiwan. “Regime Insecurity and International Cooperation” builds on Steven David’s theory of omnibalancing to explain China’s territorial compromises (pp. 51–52). Omnibalancing theory hypothesizes that leaders form alliances as a means to balance against both internal and external threats. In “Explaining Third World Alignment,” David states, “The essence of balance of power theory is correct.”11 Fravel asserts that on many occasions, China accepted absolute gains (or relative losses) in its territorial compromises: “Moreover, [China] has offered substantial compromises in most of [its] settlements, usually receiving less than 50 percent of the contested land” (p. 46). China did not accept absolute gains (or relative losses) as often as Fravel suggests. On the surface, China’s territorial compromise with Burma in 1960 would appear to conªrm Fravel’s argument, but the empirical reality is more complex. In 1960 China did reach an agreement with the Burmese government, but this agreement included a clause that allowed the People’s Liberation Army to conduct raids against the Chinese Nationalist forces in Burma, and it compelled the Burmese army to undertake similar raids against the Nationalists.12 China may have received less than 50 percent of the land at stake, but it was Burma that accepted the sucker’s payoff.13 Moreover, contrary to the predictions of Fravel’s theory, China began to exploit its weaker neighbor in 1962 by supporting the Communist Party of Burma’s insurgency against the Burmese government, making the party China’s largest client after North Korea and North Vietnam.14 The Burma case provides an additional problem for Fravel’s hypothesis: negotiations to settle the Sino-Burmese border began in 1954, ªve years before the temporal introduction of the independent variable, in this case the Tibetan revolt (p. 64).15 The regime insecurity hypothesis is also consistent with many other outcomes, making it difªcult to falsify. For example, Fravel counts China’s resolution of the border with Burma as a victory for the regime vulnerability hypothesis. If, however, the Chinese Communists had reached a compromise with the Nationalists—elements of whom dominated the Burmese states of Kengtung, Kokang, and Manglun—this, too, could be counted as a victory for Fravel’s theory.16 Given that “neighboring states can provide a range of sup- 10. Walt, The Origins of Alliances, pp. 17, 18, 22–26. 11. Steven R. David, “Explaining Third World Alignment,” World Politics, Vol. 43, No. 2 (January 1991), p. 235. 12. John W. Garver, Protracted Contest: Sino-Indian Rivalry in the Twentieth Century (Seattle: University of Washington Press, 2001), p. 253. 13. Kenneth A. Oye, “Explaining Cooperation under Anarchy: Hypotheses and Strategies,” in Oye, ed., Cooperation under Anarchy (Princeton, N.J.: Princeton University Press, 1986), p. 7. 14. Garver, Protracted Contest, pp. 254–257. 15. Ibid., p. 253. 16. Mary Callahan, Making Enemies: War and State Building in Burma (Ithaca, N.Y.: Cornell University Press, 2003), p. 146. International Security 31:2 202 port for rebels or even seek to intervene in the conºict,” a compromise with the Nationalists would have bolstered the security of the Chinese Communist party-state (p. 53). Fravel never tells his readers why the Communists chose not to compromise with the Nationalists. Fravel’s work raises important questions about why some revolutionary and unstable regimes manage to survive, whereas others collapse. If Fravel is right about China, why did Mikhail Gorbachev’s retrenchment from Eastern Europe not provide for a “diversionary peace” for the Soviet Union? Would the Jacobins in France, or the Nationalists in China, have maintained their stay in ofªce if they had simply changed the borders of their respective countries and reached territorial compromises with their enemies? Although there may be a relationship between internal conºict and international cooperation, states’ territorial compromises are ªrmly rooted in changes in the balance of power. —Albert Wolf Columbus, Ohio M. Taylor Fravel Replies: In his spirited critique, Albert Wolf challenges my argument that regime insecurity best explains China’s compromises in its many territorial disputes.1 Wolf suggests instead that this behavior is “ªrmly rooted in changes in the balance of power.” I disagree. My claim that a strong state would be unlikely to offer signiªcant concessions in territorial disputes is hardly a realist straw man. As reºected in the literature on bargaining and war, structural realism would expect the division of an issue contested by two states to reºect relative military capabilities.2 Recent quantitative research demonstrating that militarily stronger states have been less likely to pursue compromise in territorial disputes is consistent with this expectation.3 Similarly, scholars working in the realist tradition as well as past and present observers have all expected China to bargain hard over disputed territory because of its military capabilities (p. 46).4 The Catch-22 alleged by Wolf is apparent, not real. Wolf mistakenly equates internal regime security with external military strength. Increased regime insecurity, however, does not necessarily reduce a state’s ability to project power, especially against a weaker neighbor. It can under extreme conditions when the functioning of a state largely collapses, such as during civil war. Under less severe conditions, however, a 1. M. Taylor Fravel, “Regime Insecurity and International Cooperation: Explaining China’s Compromises in Territorial Disputes,” International Security, Vol. 30, No. 2 (Fall 2005), pp. 46–83. Additional references appear in parentheses in the text. 2. James D. Fearon, “Rationalist Explanations for War,” International Organization, Vol. 49, No. 3 (Summer 1995), pp. 379–414. 3. Paul K. Huth and Todd L. Allee, The Democratic Peace and Territorial Conºict in the Twentieth Century (Cambridge: Cambridge University Press, 2002). 4. Guy Searls, “Communist China’s Border Policy: Dragon Throne Imperialism?” Current Scene, April 15, 1963, pp. 1–22. Correspondence: China’s Territorial Compromises 203 state can maintain its relative power advantages despite internal unrest. The Tibetan revolt, for instance, illustrates the lack of such a linear and inverse relationship between regime security and military power. Despite a real internal threat to its territorial integrity, China nevertheless dispatched troops to its border with India, resulting in clashes in August and October 1959—clashes, moreover, that suggested China might bargain hard over disputed land in the region. Wolf faults my theory for not explaining China’s alliance choices during periods of regime insecurity. Yet nowhere did I claim that regime insecurity offered a general theory of alignment. My purpose was more modest—namely, to explain why and when China pursued otherwise costly compromises in its territorial disputes. Moreover, logic does not require that a theory of territorial compromise must also explain alignment patterns, or vice versa. External balancing provides an explanation for why a state might offer territorial concessions, one to be tested against the historical record and alternative theories. Contrary to Wolf’s charge, my theory of regime insecurity is readily falsiªable. The theory would be invalid if China had compromised frequently in the absence of internal threats or, if during periods of regime insecurity, process-tracing revealed that these compromises were spurious. Wolf, however, does not offer any such evidence. Regarding Burma, the “empirical reality is more complex,” but not in ways that Wolf believes. Wolf overlooks my statement that regime insecurity does not explain China’s ªrst attempt to compromise with Burma in 1956 (p. 63). No theory accounts for every observation of the dependent variable, and my work is no exception. China’s additional concessions to Burma in January 1960, however, were compromises prompted by the regime insecurity following the Tibetan revolt (pp. 65–67). In addition, Wolf states that my theory is indeterminate because I do not explain “why the Communists chose not to compromise with the Nationalists” who had troops based in Burma. Yet just a basic understanding of the Chinese civil war makes this clear. Given the contest for legitimacy between the People’s Republic of China and the Republic of China to govern both the mainland and Taiwan, what could Beijing have possibly offered Taipei for the removal of its troops from a third country? As I noted, the zero-sum competition between the two made territorial compromise unlikely (p. 59). Furthermore, Wolf wrongly claims that Burma “accepted [a] sucker’s payoff” by agreeing to conduct military operations with China against the Nationalists. The Burmese government had unsuccessfully used diplomacy to achieve their repatriation, but it lacked the military capabilities to remove them through force, especially when Taiwan dispatched reinforcements in the late 1950s.5 China’s assistance in expelling the Nationalist troops was welcomed, not coerced.6 Moreover, my theory predicts that this 5. Victor S. Kaufman, “Trouble in the Golden Triangle: The United States, Taiwan, and the 93rd Nationalist Division,” China Quarterly, No. 166 (June 2001), pp. 440–456. 6. Yao Zhongming, “Zhou Enlai zongli jiejue ZhongMian bianjie wenti de guanghui yeji” [Premier Zhou Enlai’s glorious achievement in settling the Chinese-Burmese border problem], in Pei Jianzhang, ed., Yanjiu Zhou Enlai: Waijiao sixiang yu shijian [Studying Zhou Enlai: Diplomatic thought and practice] (Beijing: World Affairs Press, 1989), pp. 101–102. International Security 31:2 204 type of bargaining will occur: due to internal threats, China traded contested territory for Burmese assistance, such as the joint campaign against the Nationalists (p. 68). Again, as I noted, compromise through regime insecurity does not necessarily reºect benign intentions (p. 52). Likewise, China’s subsequent aid to the Communist Party of Burma fails to provide evidence against regime insecurity as an explanation for the Chinese leadership’s willingness to compromise with Burma in 1960. Wolf overlooks that China’s material support for the insurgency reached high levels only in the late 1960s, during the tumult of the Cultural Revolution. As hostile as China was, this support offers poor evidence against a decision taken almost a decade earlier. Additionally, Wolf misses the link between China’s aid and frontier security, as Burma’s Communist Party occupied the areas adjacent to Yunnan Province vacated by the Nationalist troops when they ºed to Laos in 1961.7 Ultimately, however, Wolf’s commentary falls short because the balance of threat explanation that he champions fails to capture the variation in China’s compromises. For Wolf, China’s behavior in the early Cold War reºects broader efforts to balance against the U.S. threat. Until the escalation of the Vietnam War in 1965, the United States was probably most threatening to China in the early 1950s, ªrst when U.S. troops crossed the 38th parallel on the Korean Peninsula and then when it formed alliances in East Asia, including the 1954 treaty with Taiwan. China did seek to balance U.S. inºuence during this period through the consolidation of its alliance with Moscow and “Bandung diplomacy,” but, puzzling for Wolf, pursued territorial compromise just twice. Yet if threat explains China’s behavior, China should have compromised frequently in the early 1950s, not later. In Wolf’s view, for instance, China ought to have compromised with Pakistan eight years earlier than it did, perhaps to discourage Islamabad from joining the U.S.-led Southeast Asia Treaty Organization established in 1954. The early 1960s nevertheless offers an inclusive test of my argument because regime insecurity and external threat point to the same outcome of compromise. Other tests are needed where the two theories provide divergent predictions of the same phenomena. The 1970s offers one such test. Then, the level of threat was high, as the Soviet Union had tripled the number of divisions deployed along China’s border (later including SS-20 missiles) and used treaties with India and Vietnam to squeeze China diplomatically.8 Although Beijing countered Moscow through rapprochement with the United States, it did not offer new territorial concessions to any of its neighbors. This is a striking failure for the balance of threat explanation because such compromises might have been just the tool with which to drive a wedge between India, Vietnam, and the Soviet Union. By contrast, regime insecurity accounts for the lack of territorial concessions due to the absence of ethnic unrest in the frontiers and relative political stability at the core. The 1990s provides another useful test of my argument. Here, the collapse of the 7. Bertil Lintner, Burma in Revolt: Opium and Insurgency since 1948, 2d ed. (Bangkok: Silkworm, 1999). 8. National Intelligence Council, “The Sino-Soviet Relationship: Military Aspects,” NIE 11-13-73 (Washington, D.C.: National Intelligence Council, 1973), p. 5. Correspondence: China’s Territorial Compromises 205 Soviet Union produced China’s least threatening external security environment since 1949 (pp. 74–75). If the balance of threat explanation is correct, China should have rarely if at all pursued territorial compromise because there was no aggression to counter. Yet China offered many concessions from 1990 to 2002, moves that track with the political instability after Tiananmen and unrest in Xinjiang (pp. 74–80). Balance of threat as an explanation for China’s behavior thus ºunks a “most likely” test in the 1970s and fails to pass an albeit tougher “least likely” test in the 1990s. Regime insecurity can account for the variation in China’s willingness to compromise in these two periods, but balance of threat cannot. My article sought to explain why and when territorial compromise occurs. Wolf fails to explain China’s compromises with any precision in the early Cold War and cannot explain China’s behavior in the 1970s and the 1990s. Innenpolitik remains in— in insecurity. —M. Taylor Fravel Cambridge, Massachusetts