SURRENDER? WHAT SURRENDER? YANXISHAN'S RECONSOLIDA TION OF POWER IN TAIWAN JUL Y-A UGUST, 1945 BY J A M E S A L E X A N D E R M I T C H E L L B F A , T H E A T R E , T H E UNIVERSITY OF REGINA, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF HISTORY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA AUGUST 1995 (c) James Alexander Mitchell, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ft\ <^^f The University of British Columbia Vancouver, Canada DE-6 (2/88) 11 ABSTRACT In August 1945 North China found itself in a situation where 'defeated' Japanese troops were in possession of territory which was contested by several 'victors', including both rival Chinese and foreign military forces. Not only did this complicate the process of surrender, but it provided the opportunity and conditions for the resumption of civil war. The placement of Japanese troops in places throughout the country put them in a good position to counter CCP advances until such time that G M D troops could arrive to take the surrender. In Shanxi, warlord governor Yan Xishan - Jiang Jieshi's commander of the 2nd War Zone - had been living in quasi-exile in southwestern Shanxi. He took this formula one step further by negotiating a set of conditions, some of which had been in place before the surrender, under which the Japanese would 'surrender' only to his own troops, and in fact to join him to fight against the communists who surrounded the cities and rail lines. During the summer of 1945, in anticipation of surrender, Yan moved closer to Taiyuan, the capital, and began to negotiate his return there with the local Japanese commander, along with formal acceptance of the surrender. When the war ended, he was already moving his armies and himself toward the major cities, especially the Fen River basin around Taiyuan. The 8th Route Army in Shanxi held the majority of the province, for its own part, but their guerrilla strategy kept them away from the cities until after the surrender. Yan's familiarity with the Japanese allowed him a measure of flexibility throughout the war which facilitated collaboration. By 1945 a virtual ceasefire existed between them, and Yan moved his troops easily through their lines to attack the communists. In addition, both Yan and General Sumita in Taiyuan recognized that without each other they both faced certain defeat, and began preparing for the surrender in Shanxi. Power was to be transferred under this scheme to Yan alone. When the end of the war did come, then, Yan shuttled toward Taiyuan where, on August 30, he arrived by armoured car, protected by his own and Japanese troops. With Japanese soldiers still in positions along the rail lines and in many towns, He was not only able to protect himself as he returned to Taiyuan, but denied the communists the opportunity to expand from the countryside. There presently exists very little in English on this topic. Recent source material in Chinese, however, has made it possible to look in detail into the circumstances of these events, and confirms that the surrender in August 1945 represented neither the end of war nor the end of Japanese intervention in China. iii TABLE OF CONTENTS page ABSTRACT ii TABLEOF CONTENTS Ui LIST OF FIGURES AND TABLES IV ACKNOWLEDGMENTS V NOTEON SOURCES.. '. VI 1) INTRODUCTION: THE END OF THE WAR IN CONTEXT / 2) SETTING THE STAGE: YAN, THE JAPANESE AND THECCPIN SHANXI a) Yan Xishan's Provincial Legacy. 8 b) The Anti-Japanese War in Shanxi 12 3) PREPARATIONS FOR SURRENDER, JULY 1945 18 4) NEW UNIFORMS, OLD SOLDIERS: CHANGING OF THE GUARD a) Surrender? What Surrender?. 23 b) Toward Taiyuan. 30 5) CONCLUDING REMARKS. 39 SOURCES CITED. 43 IV LIST OF FIGURES AND TABLES PAGE # FIGURE 1 DISTRIBUTION OF POWER IN NORTH CHINA, 1945. 2 FIGURE2 YANXlSHAN, a LATE 1930'S. 9 FIGURE 3 SHANXI PROVINCE 13 FIGURE 4 YAN'S RETURN ROUTE 31 FIGURE 5 YAN, IN FULL UNIFORM: TAIYUAN, 1945. .'. 34 FIGURE 6 CARTOON, JIEFANG RIBAO, OCT. 7,1945. 40 TABLE 1 YAN'S PRINCIPAL COMMANDERS AND OFFICIALS. 20 TABLE 2 8M ROUTE ARMY COMMANDERS IN SHANXI. 22 TABLE 3 JAPANESE PERSONNEL IN SHANXI. 24 TABLE 4 YAN'S PROVINCIAL DEFENSE ARMY: A UGUST30,1945. 36 ACKNOWLEDGMENTS During the course of this research, I have had to make a difficult transition from being a generalist to acquiring a sense of being a 'specialist'. One of the steps in this process, inevitably, has been learning to tap into the knowledge and experience of the specialists working around me. I have had the extremely good fortune, both in Canada and in Taiwan, to discuss my work and its various aspects with many people, whose knowledge and insight has both clarified my own understanding and allowed me to enjoy the research and writing. I am extremely grateful to Dr. Lary, my supervisor, for giving me both a sense of discipline and the freedom to conduct research on my own terms. Her encouragement to look further into the issues of regionalism and imperialism has helped tremendously in my understanding of revolutionary China. My own emphasis on Chinese sources, as well, comes in part from her belief I have come to share, that one cannot do justice to research on cultures other than our own without adequate knowledge of the operative language in that culture. In addition, I am indebted to Drs. Glen Peterson and Diane Newell, from the Department of History at UBC, as well as Dr. Rene Goldman from the Asian Studies Department, for their comments and criticism at various stages. I would also like to thank Drs. Chen Yongfa and Zhang Yufa, both of the Academica Sinica in Taibei, for their early comments and especially for steering me away from my original thesis topic. The staff at the U B C Asian Library - S. Y. Tse, Tsai Yuh-jen and Hsiao Mei-chih - have been indispensable in guiding me through both research methodology and language difficulties. I am especially grateful to Mr. Gonnami, who taught me how to source Japanese names properly, and without whom I would still be trying. Several colleagues and friends have also been more than generous with their time and attention in working out the problems which have come up along the way. Many thanks to: Ian Petrie, Dan Schnick, Stewart Muir, and my father Ken for editing my mistakes; and Y u L i , Y i n Wenji Direne Liu and Chihiro Otsuka for getting me through the sources. NOTE ON SOURCES In researching mainland source materials, one is conspicuously aware of the consistency of the Chinese sources - not only in reference to this period in Shanxi - to the point of near-repetition in certain episodesA; this certainly lends itself to the belief that they represent an 'orthodox' version of what occurred at the end of the anti-Japanese War. In fact, several of the sources on this subject are people who were at the time officials serving in Yan's or puppet adrninistration, and their testimony must be seen in this light. This 'version', however, is strengthened by two factors: first, the record in English is too general to be able to provide a counterpoint; secondly, those details which exist - particularly those regarding the nature of collaborative activity before 1945 - would appear to support the Chinese narratives. The only other major sources of historical narrative on the period are written in Japanese and, again, consist mainly of memoirs and such by former Japanese soldiers. Barring the possibility of reading such material, the identities of those in Shanxi have been corroborated through biographical reference material, such as the Japan Armed Forces Directory. In addition, the Shanxi wen shi zi liao includes, in one of its volumes, a Chinese translation of an excerpt from a book written by Jono Hiroshi, a significant figure in the "stay behind" movement after the surrender. He was captured by the communists at the end of the civil war, and spent fifteen years in Chinese prisons (at one point Puyi occupied the cell next to him) - before returning to Japan in 1964. Even his account, as such, does not conflict with PRC histories. The challenge, then, is for future research to provide a broader context. There is likely to be further evidence in Shanxi, as well as in private archives in Taiwan, which would do much to clarify what has been sketched out here using sources available in the public record. 'Good examples are the meeting in Xiaoyi in August and Yan's later arrival in Taiyuan by armoured train. 1 1. I N T R O D U C T I O N : T H E E N D O F T H E W A R IN C O N T E X T "The actual question of whether or not Yen Hsi-shan is guilty of what would in western eyes be considered traitorous dealings with the enemy is probably riot of great importance. I am personally prepared - on his record and otherwise almost unexplainable fact of his continued existence - at least to accept the common Chinese description of him as a bamian linglong (an eight-faceted glittering gem) - two faces for the Central Government, two faces for the Japanese, two faces for the Communists, and two more for the people." John Service, Yanan 19441 When the Anti-Japanese War came to an official end in August 19452, Japanese positions in North China were in many places still secure; that is, at the time there was little chance of an immediate military defeat by Chinese or Allied forces. Still, the various Chinese military forces - principally those of the Guomindang (GMD) and the Chinese Communists (CCP), but as we shall see with the case of Yan Xishan in Shanxi, not without other sub-groups - were quick to begin flexing their military muscles in anticipation of taking over this territory, regardless of the formal command structure under which the Third United Front co-ordinated its 'national' war resources. Both the United States and the Soviet Union, as well, claimed mandates for accepting the 'Cited in U.S. Senate Judiciary Committee, Amerasia Papers: A Clue to Understanding the Catastrophe of China ed. by Robert Kubek. {Washington, D C , 1970) p. 771. This collection of documents was produced as part of an investigation by the senate committee into charges brought against John Service and other former China field officers, during the McArthey hearings. Several of these documents contain reports of parts thereof which were deleted from diplomatic records. This publication should not be confused with a shorter summary of a similar name, written by John Service himself. ^On August 10, Japan "announced the acceptance of the proclamation of the Potsdam Conference demanding the unconditional surrender to China, the United States, Britain and Soviet Russia. Upon receipt of Japan's declaration of surrender, the national Military Council of the Chinese Government immediately instructed all armed forces in the country to wait for orders pertaining to the acceptance of surrender in accordance with Allied agreements." History of the Sino-Japanese War (1937-1945). (Taipei: Chung Wu Publishing Co., 1971), p.547. 2 FIGURE l: DISTRIBUTION OF POWER IN NORTH CHINA, SUMMER 19453 3 All maps drawn by the author. Information for Figure 1, as well as Figure 3 (p. 13) was drawn from the following sources: Suzanne Pepper, Cambridge History p. 708; Van Slyke, Lyman P., ed. The Chinese Communist Movement: A Report of the U.S. War Department. July I94i (Stanford: Stanford University Press, 1968), p. xii; Wei Hongyu, ed., Huabei Kane Ri Gendudi Jishi (Records of the Anti-Japanese Base Areas in North China) (Tainjin: People's Publishing, 1986), frontpiece. 3 surrender either 'on China's behalf or jointly where they were able to4: the Soviets declared war on Japan on August 9, sweeping through Manchuria and sending the Japanese into rapid retreat. U.S. forces at the same time began to prepare for the airlifting of G M D troops to North China to 'accept the surrender' in the major cities. The Eighth Route Army, throughout North China, had been operating in a 'no-man's zone' between the Japanese and their own border regions, and began now to move toward major cities such as Beijing, Tianjin and Zhangjiagou. In other places throughout the informal Japanese Empire - the 'East Asia Co-Prosperity Sphere' - In Vietnam, for example - Japanese forces also stood by, representing a potential 'third force' which could be used by the 'recognized' regime to prevent communists from gaining ground. In the former, ironically, Chinese Nationalist troops were also moved into North Vietnam, as British troops were deployed in the south, to help restore the French and disarm the communist-dominated Vietminh. The placement of military forces at the end of hostilities was thus of great influence in determining the conditions and extent of surrender in the field. No foreign army, however, can survive in a totally hostile environment. The very fact of military occupation of territory anywhere, after all, is made possible only by the participation - or at least the acquiescence - of local officials and other influential groups. The problems associated with collaboration have inevitably, in the course of China's own history, been those of time frame and longevity of the occupying regime. The dynastic cycle has been punctuated with transitions between regional or factional power holders, at which time the 'winner' has been determined primarily by permanence. When the Japanese invaded China proper in 1937, the majority of people had no idea how long they would remain occupied: Taiwan, after all, had already been a formal part of Japan's empire since 1895. There was thus the possibility that they would be living 4See for example Suzanne Pepper, "The KMT-CCP Conflict 1945-1949", in Denis Twitchett, John K. Fairbank, eds., The Cambridge History of China, vol. 12, part 2 (Cambridge: Cambridge University Press, 1986) pp. 726-727. 4 under occupation for a very long time. And, as Fu Poshek wrote recently, "Patriotic defiance brought risk of death; compromise seemed to many inevitable, yet morally questionable"5. In other words, one's existence at such times may dictate behavior which one would not contemplate under less adverse conditions. Those who collaborated with the Japanese, such as Wang Jingwei and Yan Xishan, were motivated by much more than the treacherous schemes with which he is credited by Chinese critics. At the end of the war, the issue of who the Japanese were to surrender to was simple enough in Jiangnan and the south, where the CCP was isolated in small pockets and thus had less influence. In the north, however, it was immediately complicated by separate and competing demands by both CCP and G M D leaders that they alone should accept the surrender. In places, then, tracking the actual 'end of the war' is problematic. First, there was a continuation of hostilities among various groups, including the Japanese themselves,6 as they jockeyed for position in anticipation of a post-war struggle for redistribution of power. In addition, the 'puppet' troops and officials who had served the Japanese became, in some cases, responsible for taking the surrender of their former occupiers. The surrender itself, announced formally on August 14th by Emperor Hirohito, brought the war to an end but intensified the sense of potential luan (disorder) which hung over China, poised for the consummation of a civil war which had been brewing for almost twenty years. The Japanese presence, from Manchuria to Guangdong, only added to this sense of chaos. Neither the pronouncements of the Japanese Emperor nor those of General Okamura Yasuji, chief commander of the Imperial forces in the China theatre, were sufficient in many places to ensure an orderly, consistent or even peaceful process 5Fu Poshek. Passivity. Resistance and Collaboration: Intellectual Choices in Occupied Shanghai. 1937-1945 (Stanford: Stanford University Press, 1993), p. 156. 6The Japanese - like Jiang Jieshi - were fiercely anti-communist, and "undismayed by the surrender at Tokyo, fought back with an intensity they had not shown for years.The [puppets] who had befriended the Japanese became staunch supporters of the Guomindang, ran up the national flag, and denounced the communists as disloyal". Theodore White, Thunder Out of China (New York: William Sloane Associates, 1946), p. 272. 5 of power transfer. In major cities such as Shanghai, Beijing and Tianjin, American marines and G M D soldiers frequently encountered armed Japanese troops7, who generally felt that they were not surrendering to the Chinese, but to the Americans. As it was, there were relatively few 'incidents' in these GMD/U.S.-controlled zones. In areas where the CCP held control, however - particularly in the 'border regions' of North China, which crossed into Japanese-held territory - very little went smoothly. Marshal Zhu De, speaking from Yanan on August 10, ordered communist forces throughout the border regions and elsewhere in China to take the surrender wherever they were able, despite orders from Jiang Jieshi to the contrary. The eight-year Anti-Japanese War in China was waged over vast areas of the country, and there exists an equally vast amount of source materials and written histories. With regard to the prosecution of the war itself, however, the centres of military and political activity have been subject to such voluminous research that many areas on the periphery have been left relatively obscure. We are thus deluged with histories of the 'major' events on the Yunnan/Burma front, in Manchuria and East China, as well as in Chongqing and Yanan. When the war ended, the 'race for North China' between the G M D and the CCP centred on Henan Hebei and Shandong, where Japanese front lines were located. Shanxi Province Was not, as such, in the main arena of such attention. Certainly there were locations along its perimeter that were both contested and reported on both during and at the end of the war8. There has been little mention in English, however, concerriing what occurred in Taiyuan - the capital, located near the geographical centre of the province - between the announcement of Japan's intention to surrender in August good example of this was reported by Tillman Durdin in The New York Times (NYT) on September 10, 1945. ^Throughout the summer there was heavy fighting between Yan's armies in the Shangdang region around Changzhi; as well as in Yanbei and toward Datong For summaries of this activity see Minpuo Junshi Shi Lueqiao (An Outline Military History of the Republican Period) Volume 3, Part 2 (Beijing: Zhonghua Publishing House, 1991), pp.869-897. 6 and the reconsolidation of Yan Xishan's power there in September. Donald Gillin's 1967 biography of Yan contains approximately one page of commentary on both Yan and conditions in Shanxi during this crucial period9. A later article by Gillin, concerning the post-war presence of Japanese troops in China 1 0, suggests that Yan was helped by the Japanese. He provides, however, few details of such an arrangement. While compiling sources for this project, it became apparent that virtually nothing exists in English that details the nature of surrender and transfer of power involving Yan - the commander of the Nationalist 2nd War Zone, responsible for accepting the surrender on behalf of the central government - and General Sumita Raishiro, the commander of Japanese forces in Shanxi. In fact, the process by which this occurred - between August 11 and the beginning of September - was made possible through a system of diplomacy and secrecy which had been cultivated between Yan, the Japanese military in Shanxi, puppet intermediaries and Yan's own senior military and civil personnel. According to available Chinese sources11, on August 23 Yan embarked on the final leg of a much-anticipated return to Taiyuan from his forced exile in Jixian by armoured train, provided and guarded by the Japanese. At the time this occurred, the entire question of the official surrender, the disarming and the repatriation of Japanese troops was neither clearly articulated nor clearly uniformly carried out.12 In fact, the level of complexity with which Yan's reconsolidation of military power in Taiyuan was carried out, between July and August 1945, has been either underestimated or ignored in ^Donald Gillin, Warlord Yen Hsi-shan in Shansi Province 1911-1949 (Princeton: Princeton University Press, 1967), p. 285. 1 0 Donald Gillin, "Staying On: Japanese Soldiers and Civilians in China, 1945-1949", in The Journal of Asian Studies. Volume 32, Number 3 (1983), pp. 497-518. ' ^This is referred to in a variety of sources, a good new one being the Yan Xishan Pingzhuan (A Critical Biography of Yan Xishan) (Zhangjiagou: CCP Central Party School Publishing House, 1991), p.387. l^The situation was indeed confusing at times. As one Marine Lieutenant in North China related in the Stars and Stripes at the time,"As an officer I am supposed to tell them [why they were there - i.e. to disarm the Japanese], but you can't tell a man that he's here to disarm Japanese when he's guarding the same railroad with Japanese." Cited in Dick Wilson, "Leathernecks in North China". Bulletin of Concerned Asian Scholars. Vol. 4, (summer 1972), p 36. Henry Lieberman reported in The New York Times, on October 29, 1945, that "two Japanese brigades...were fighting shoulder to shoulder with Government troops on a 'sixty-mile front". 7 virtually all the secondary literature available in English on this topic. It is not the case that no-one knows what happened in Shanxi during this period: evidence exists which describes various military, economic and governmental aspects of the existing relations between Yan's officials and the occupying Japanese regime. To the present, however, there has been no attempt to synthesize the various into a coherent narrative. There are thus two contentions in this paper. First, on a historiographical level, there is a demonstrated need to look again at these events of fifty years ago in the light of more recent source material which, at least in the case of Shanxi, seem to have been almost untouched by scholars writing in English. Some Chinese sources have begun to refer to such histories, but only beginning in the 1990's, and only Mainland scholars. Some of this new source material takes the form of written memoirs, reminiscences, accounts of battles, telegrams and the like, a good example of which is the Shanxi Wenshi Ziliao periodical, discontinued in 1965 but revived in 1979. Beginning in the 1980's, as well, a wealth of biographical material has been produced on Yan, in addition to all the major CCP figures. They represent, for one, an attempt by the PRC government to record the experience of the revolution while it was still possible to do so - i.e., while the participants were still living. Emerging from the dark cloud of the Cultural Revolution, the Chinese government was anxious to establish its legitimacy in front of a younger audience which had no revolutionary memories of its own. Not surprisingly, most of this material is highly critical of Yan, the G M D and the United States, and vulnerable to charges that they are simply a reconstruction of the 'winner's version' (see 'note on sources'). Until other evidence becomes available to provide further corroboration, they represent the extent of present understanding of the events under scrutiny. The second point is that the official announcement of surrender by the Japanese on August 10 represented neither the end of the war nor the end of Japanese intervention in China. This in itself is not unique to Shanxi: there were other places where the 8 Japanese continued to hold their positions, such as the major cities on the east coast, in Shandong and in Jiangnan, where Japanese troops were often deployed in conjunction with GMLVU.S. forces to ensure 'order' until actual transfer of power became possible -that is, to ensure that the CCP did not gain territory.13 The difference in the case of Shanxi, however, is not only that it occurred autonomously (though with the knowledge) of the 'national' chain of command, but that it also took place within the context of an arrangement made between Yan himself and the Japanese commander that would both place Yan back at the top of the provincial hierarchy, and allow them to jointly maintain a de facto state of war against the communists surrounding them. 2. S E T T I N G T H E S T A G E : Y A N , T H E J A P A N E S E A N D T H E C C P IN S H A N X I A)YAN XISHAN'S PROVINCIAL L E G A C Y A tenacious survivor of the Warlord period, Yan has been described as a master of strategic alliances. Bom into a banking family from the Wutai Region northeastern Shanxi, he journeyed to Japan in 1904 as a young man to study at the Imperial Military Academy in Tokyo; there, like so many other students from China, he was soon enrolled among the ranks of Sun Yat-sen's Tongmeng Hui (which later became the GMD). Upon his return to Shanxi in 1909, he was awarded a second-class juren degree by the Qing governor and a military posting in the New Shanxi Army. When the news of the Xinhai Revolution reached military leaders in Taiyuan, they seized military control quickly, and 1 3 For an excellent description and analysis of the use of Japanese troops see Dick Wilson, "Leathernecks", pp.33-37. 9 Yan was himself appointed military governor by the new national regime - a post which he held almost continuously until his final defeat in 1949 at the hands of the communists. With the disintegration of Yuan Shikai's regime in 1916, Yan Xishan was thrown, like the other major regional military leaders, into a chaotic cycle of uneasy, temporary alliances and warfare that has become known as the Warlord Era. He was in fact a warlord, perhaps the most persistent and successful. It is important to point out however that his circumstances, indeed his style of rule, differed significantly from his peers. The natural geography of Shanxi, for one, afforded Yan a measure of security due to its isolation that was perhaps paralleled only in the case of Sichuan; unlike the latter, however, Shanxi's close proximity to Peking and other centres of political power, as well as its historically strategic military routes, gave Yan a degree of leverage and flexibility which guaranteed him a position of power which extended beyond his own territory throughout his rule there. FIGURE 2: YAN XISHAN, a LATE 1930^ 1 4 Neither this photo nor the one on page 34 is given a credit - fairly typical of Mainland sources - in the Yan Xishan Pinzzhuan. where they are both featured in the frontispiece. 10 The intricacies of warlord politics are beyond the scope of this paper, and have been well documented elsewhere.15 There are, however, a number of important and inextricably related elements of Yan's political style, important in his later career, which emerged during that period and which must be made clear. First, he identified himself so closely with Shanxi that his authority there remained unchallenged until the Japanese invasion. Secondly, his education in Japan left him with a deep sense of admiration and respect for the Japanese - their discipline and loyalty, as well as their modern industrialization and militarization programmes - that profoundly influenced his relations with them even after war broke out between them in 1937. Thirdly, his virulent anti-communism made him a natural ally of the Japanese, and this became the main justification for their wartime co-operation. Like other warlords, Yan was concerned not only for his own survival in power, but also the role he and his province would play upon eventual reunification. Until such cohesion became feasible, he preferred to remain at arm's length from anything resembling hasty efforts to establish a 'national' polity which, after the launch of the Northern Expedition in 1926, came increasingly under the mandate of Jiang Jieshi. In the meantime, he "tried to remain aloof from the struggle for power and concerned himself almost exclusively with the problem of modernizing Shansi and...developing its resources" which by the 1930's had gained the reputation of "model province."16 For all his de facto autonomy, however, it is a telling sign of the limitations of regionalism in China that at no time during his rule was Shanxi considered anything but a 'province' by Yan or anyone else. As one work on regionalism of the period points out, the regionalists would apply themselves practically, in the regions they controlled, with the aim of serving the nation. Regionalism was thus elevated from a faute-de-mieux to a positive policy, justified by the high 1 •> Authoritative overviews can be found in Lucian Pye, Warlord Politics: Conflict and Coalition in the Modernization of Republican China (New York: Praeger, 1971); and Ch'i Hsi-sheng, Warlord Politics in China. 1916-1928 (Stanford: Stanford University Press, 1976). 16Gillin(1967), pp. 21-22. 11 nationalist cause which it served. The idea of levels of loyalty was spelled out, in such a way that regionalism and nationalism could co-exist, as different, but interdependent levels in a pyramid of loyalties.17 There is thus no inherent contradiction in Yan's behavior, in which one's allegiances can be seen as a series of concentric circles. Yan's autonomy from the G M D government was further extended by the invasion of the Japanese in 1937, and his political and military relations with the latter must be observed with this in mind. 1 8 His military command of the 2nd War Zone (Shanxi) was, indeed, part of the overall command structure as laid out by the central government, but the same could be said of the 8th Route Army and the border regions. If Yan was a Nationalist, then he was also a collaborator, a modemizer and a dictator. However, it is of much greater utility to view his 'loyalty' - and that of many others - rather as being determined by a shifting set of circumstances, in which ideological commitment is necessarily secondary to political and military survival. As a regional leader during a period of weak central authority, moreover, Yan's primary sense of loyalty was to Shanxi, and his permanent identification with his home province is evident. After all, he remained there throughout the war: instead of joining his contemporaries in Chongqing where he could expect to live in the luxury accorded such a leader, he endured the hardships of a spartan life in exile in Southern Shanxi. There is thus no need make the logical leap required to claim that Yan was a 'stooge' for the enemy: he and Japanese military leaders in Shanxi worked together primarily to rid themselves of a common enemy, despite the rhetoric of "preserving [Japanese] power".19 17Diana Lary, Region and nation: The Kwangsi Clique in Chinese Politics (Cambridge: Cambridge University Press, 1974), pp. 18-19. 1 8For a recent analysis of Yan's 'collaborative' behavior, see Ma Zhendu, "Huabei Difang Shili Pai Kang Ri Taidu zhi Bijiao Yanjiu" (Comparative Research on Anti-Japanese Positions by Ruling Cliques in North China). Minguo Dangan (Republican Archives, Nanjing,) 1993 No.2. (Quarterly), pp. 98-100, 102-103. Yan's attitudes are compared with other local power holders in North China at the time of the Anti-Japanese War. negotiations, both sides regularly made mention of the importance of continued co-operation in business after the war's end. See Huang Yougang, Kang Ri Zhanzheng Shiqi de 'Heping' Yundons ( The 'PeaceMovement'During the Anti-Japanese War) (Beijing: PLA Publications, 1988), pp.316-323. 12 B) THE ANTI-JAPANESE WAR IN SHANXI In the autumn of 1937, following Japan's invasion of Shanxi, Yan and his government retreated to the southwest corner of the province, and settled in the village of Kenanpo ( ^ i t $ ), Ji County - facing the Yellow River and Yanan to the west, and protected from the other directions only by isolation and rugged terrain. This remained his base of operations until close to the end of the Anti-Japanese War. A regional government and plan of action was soon organized, in Yan-like fashion: a *big government' approach to the distribution of essential resources, but accompanied by strict authoritarian political rule, much as had been the case before but on a smaller scale - and easier to control2 0. Toward the end of the war, however, his fortunes were threatened by renewed attacks by the 8th Route Army 2 1 - but also by hard times and, as some reports had it, even famine. The fact that this was not the case in communist-held areas to the north certainly suggests that i f resources were scarce in Yan's territory, it was due to poor distribution. By the spring of 1945, he was thus in a weak position - holding an estimated total of only 24 of Shanxi's 105 xian, while his military strength "did not exceed 100,000 men"2 2, which may well account for the renewed attention he began paying to potential allies in the months leading up to the war's end. At the outset of the Anti-Japanese war, the Second War Zone command under Yan co-operated with elements of the Eighth Route Army and the Shanxi "New Army" 2 3 2 0 A concise explanation of Yan's administrative policies in Jixian can be found in "Yen Hsi-shan's Political Position in North China", Office of Strategic Services (OSS) State Department Intelligence and Research Reports: Japan and its Occupied Territories During WWII. (Washington D.C.: Department of State, declassified 1968), p. 2 - 3. (on microform) 21zhu De began ordering counterattacks in January 1945. Zhongguo Renmin Jiefang Jun Da Ship. ( A Chronology of the People's Liberation Army) (Beijing: People's Military Affairs Publishing, 1983), p. 216. ^OSS Documents. J945. p 1 (territorial claims), p. 6 (troops). It is difficult to glean actual figures from Mainland sources. 23The "New Army" grew out of the Shanxi"Dare-to-Die" Corps, organized in 1937 by young militants including then-student Bo Yibo - later Shanxi governor under the people's Republic - in the wake of Japan's FIGURE 3: SHANXI PROVINCE24 invasion. It was reorganized when the invasion extended to Shanxi, with an estimated 15,000 irregulars. Gillin (1983),p.231-233, 249, 262; and Amerasia Papers . p.768. 24Information for Figures 3, as well as Figure 4 (p. 31) based on a map in Gillin (1967), p. 7. 14 in their strategic moves against the enemy. By the end of 1939, conditions of mistrust and charges of double-dealing led to clashes between them2 5; from that time onward they remained separated by an ever-shifting front, leaving Yan's forces surrounded on all sides by potentially hostile armies. The Japanese in Shanxi, as in other places in North China, generally controlled only the cities and large towns, which were connected only by the skeletal pattern of provincial transportation and communication routes. Except for their periodic raids, intended to collect food and other resources - as well as to terrorize the local population into submission and 'pacify' the communist resistance26 - the occupiers kept to their fortified positions. This left the countryside largely in the hands of local militia and 8th Route Army regulars, often working behind the main lines of Japanese blockades. The CCP structured its main areas of such control into three "border regions", all of which extended beyond Shanxi's own borders. To the Northwest, the Jin-Sui Border Region (# IE) connected the other regions to the main base area at Yanan, and expanded greatly through the war. The Jin-Cha-Ji Region ( f | | jf MM. BE X centred around the Wutai Mountains, was more active in Hebei due to heavy blockading in Shanxi. In the southeastern quarter. The Jin-Ji-Yu Border Region (^MWSK ), the communists used the Taihang and Taiyue mountain ranges as base areas to protect them from Japanese attacks. Although there were communist guerrillas operating close to urban areas, throughout the war there was very little large-scale military activity by the 8th Route in the Fen River Valley around Taiyuan; deterred by Shanxi's mountainous terrain, as well ^There are varying accounts of this breakdown in relations. The "New Army", for one, denied throughout the war that they were 'part of the communists, and kept their own designation even though they appeared to be working together with the 8th Route Army. Yan claims they disobeyed his orders, while Bo cites Yan's collaboration with the Japanese against them. See Amerasia Papers, p.768; and Bo Yibo, "Jielu Yan Xishan Tongdi Panguo Neimu" (Exposing Yan's Secret Contacts with the Enemy, first published August 1944). Bo Yibo Wenxuan {Selected Works of Bo Yibo, Beijing: People's Publishing, 1992), pp. 34-38. 26for reference to Japanese campaigns, see Jie fang Jun Da Shift. pp.216-219. 15 as the placement of Japanese blockhouses and other defenses, concentrated troop movements were rendered practically impossible.27 Thus patrols and units working in the countryside were forced, like elsewhere in North China, to fight in smaller guerrilla units, and to travel close to enemy areas only under the cover of night.28 Regular attacks by Japanese 'pacification units', were followed up by periodic 'mop-up campaigns' into the interior. While it is true that the 8th Route Army was extremely active in 1945, launching a campaign in January 1945 to "expand the base areas"29, their major regions of deployment were all, except for the easternmost part of Jin-Sui, well away from the urban centres of the province. While the Anti-Japanese War continued, moreover, this remained the policy of the 8th Route Army central command.30 In addition to these 'declared' holders of power, many counties were run by puppet administrations, under the nominal control of the Japanese - who they supplied with grain and other resources - but not without Yan's influence. As the war progressed, in fact, the distinctions between Yan's and Japanese puppet officials became less and less clear. The appointment of officials serving in Japanese-held areas, for example, was reportedly still conducted from Jixian, with the knowledge - and even request - of the Japanese.31 Su Tiren, who had before the war been part of Yan's administration in Taiyuan, served as 27Extensive deforestation over the centuries laid bare the loess soil covering much of the interior in North China, causing such erosion that ravines and gorges cut deeply into the hillsides; movement between base areas was often too difficult with anything except light equipment, and it was necessary to descend into the valleys and the relatively even ground in order to travel long distances and through mountain passes.. It was along these routes which the Japanese placed defenses and garrisoned blockhouses, nearby the towns and villages they occupied. Michael Lindsay, together with his family, gives a first-hand account of this, in The Unknown War: North China. 1937-1945 (London: Bergstrom & Boyle Books, 1975), pp. 39-44, 73-75.. 29'Jiefang Jun Da Ship, p. 216. 3 0This policy of limited warfare changed on August 10th when Zhu De began ordering all-out attacks on cities. He pulled back from this twelve days later when their experience in cities such as Zhangjiagou and Beijing proved too costly, but in Shanxi the main army groups were either transferred out, such as that of Liu Bocheng, or fighting peripheral contests in Yanbei and Changzhi and Fushan. For more on Zhu's policies, see Huabei. p.531-532. 3 1 Amerasia Papers, p. 1428; Yan Xishan Tongzhi Shanxi Shishi (Historical Evidence on Yan Xishan's Rule in Shanxi) (Taiyuan: People's Publishing, 1984), p. 329-330. 16 puppet Governor until he was promoted by the Japanese to work in Beijing in 1943.32 Wang Xiang, also recommended by Yan, served as "Puppet Japanese" Governor during the war, distinct from the former, further blurring the official division of loyalties. These cadres, operating in conjunction with the Japanese military, were not only able to furnish Yan with intelligence, but also acted as his intermediaries. As 1945 progressed, these officials were instrumental in preparing for the coming transition of power. It is evident from existing sources that through much of the war, in fact, Yan directed many of these officials as they negotiated with the Japanese toward furmering the existing state of co-operation between them. General Sumita, the commander of the North China Imperial Army in Shanxi province, got along well with the ever-diplomatic Yan, although usually through intermediaries. His approach, consistently, was to push for an anti-communist alliance, in an arrangement which became known as gongtong fangong (&m& or "cooperating to oppose the communists."33 In practice, this meant allowing Yan's troops to cross Japanese lines in action against the 8th Route Army, as well as maintaining a general ceasefire between themselves. In fact, he retained representatives from both the Japanese forces and the communists during his stay in Jixian County, but he maintained little contact with the Yanan representative after 193934. Evidence now suggests that intermediaries of Yan Xishan and General Sumita had actually been involved in negotiations off and on since as early as this, some of which resulted in active military co-operation against communist areas in the province.35 3^Niu Xintian, "Shanxi Riwei Zhengquan de Jianli he Fumie" (The Rise and fall of Japanese Puppet Political Power in Shanxi), Shanxi Wenshi Ziliao. {Shanxi Historical and Cultural Materials) ( Taiyuan: Shansi Historical Cultural Materials Research Committee) vol. 41 (1985), p.161. There was also Zhao Rui, stationed in Linfen, and various other officials in Fushan, Xixian, Fenyang, Lishi and other major towns. H In fact, Yan used this slogan throughout the war: in 1940, he reportedly pursued a policy of "Asian Alliance, co-operate against the communists, equal foreign policy, autonomy of internal affairs". Heping Yundong. p.305. 3 4 A good witness is Cao Yanxing, who was stationed in the CCP's '2nd War Zone General Affairs Office and remained there throughout the war, until the spring of 1945. His commentary is found in Shanxi Gemin Huiyi Lu {Memoirs of Shanxi Revolutionaries) { Taiyuan: Shanxi People's Publishing, 1985), pp.39-59. 35 "Riben Waijiao Dangan Zhong Guan Yan Xishan Touxiang Huodong de Cailiao" (Materials from the Japanese Foreign Ministry Concerning Yan Xishan and the Surrender), Shanxi Wen shi zhi liao , vol. 6 17 Yan himself recalled two direct meetings with the Japanese: one in 1940, and another in 1942 with Iwamatsu Yoshio (g- i£ gi ) - Sumita's predecessor - in Lirifen; although Yan denied that any deal was in the making, available evidence indicates that nothing could be further from the truth. Yan reportedly assured Iwamatsu on this occasion "that he would aid them in their fight against communism in China i f they withdrew a large part of their forces in Shansi and helped bim strengthen his own army by giving him food, weapons and $CH15 million worth of specie."36 That deal never materialized, and in fact resulted in several months of fighting during the winter and spring of 1942. According to several Chinese sources, there were many other cases throughout the war which proved more successful.37. According to an American intelligence report in July 1945, relations appeared to be on firm ground once more: ...it is reported that since July 1944 [Yan's] forces have not taken a single Japanese prisoner or Japanese document. Generally, the agreements reached between the Japanese and Yen are believed to give Yen permission to operate against the communists in certain areas, including the area east of the Tung-Pu railway. They also assure Yen that he will be apprised of the circumstances of Japanese withdrawals so that he can move in before the communists, and include the understanding that each will avoid hostilities:38 As the coming of defeat drove the Japanese toward accommodation, Yan himself desperately needed their military manpower to re-establish control of his old seat of power39. He and his envoys began to lobby actively for an agreement that would use Japanese military power to finally destroy the communists before they pushed him out (1963), pp. 10-15. It consists of transcripts of telegrams between Yan and the Japanese commander in Taiyuan dating from 1941; Heping Yundong takes these activities back to 1937 (p. 291). 3 6 Gill in , p.280; Harrison Forman, Report from Red China (New York: Henry Colt & Co., 1945), p. 32. 37 The Fenyang Agreement is an interesting example. This deal was eventually abrogated as well: negotiations had been conducted in the autumn of 1941, when Japan was at the height of its power. On Dec. 8, 1941, when Yan heard that the United States had declared war after Pearl Harbour, he began to vacillate on bis position, and the negotiations were soon called off. Yan believed from that time onward that Japanese surrender was inevitable, and thus relations with them were conducted accordingly. Heping Yundong. p.307-312; Amerasia Papers, p. 1428. 3 8 "Yen Hsi-shan's Political Situation in North China" OSS Documents. (1945), p.5. 39 .Amerasia Papers, pp. 1427-1430. 18 altogether. The actual size of his own forces at the time is difficult to even guess at due to the vagueness of sources, but even at an estimated 100,00040 he was dwarfed by the 8th Route Army. Responding to the CCP's counterattacks, Yan called for a new plan in February to "co-ordinate with the Japanese to take back the cities in Shanxi."4 1 To facilitate this, he allegedly dispatched "Red Gang" (hong bang, a secret society) members under his control - Wang Qingguo, his nephew among them - to contact local organizations to "mobilize anti-communist power"42 in Wutai, Jiaocheng, Datong, Yanbei, Changzhi and even Taiyuan. These groups worked with the consent of Japanese and puppet garrisons, who were themselves terrified of the communist guerrillas in the countryside which surrounded them. 3. P R E P A R A T I O N S F O R S U R R E N D E R , J U L Y 1945 B y the summer of 1945, losses in the Pacific War had seriously damaged Japan's capacity to make war; it was obvious to all, not least to Yan (who had predicted such an end in December 1941 that its surrender was in the making). On the military front, he continued to deploy troops straight through Japanese lines and strongpoints, such as in the Fendong region (east of Linfen toward Fushan), where this had been permitted by agreement since early 194443. In Central Shanxi, between Jixian and Taiyuan, puppet and Japanese-held towns became the targets of Yan's plan to 'retake the cities'. Cavalry 4 0 U.S. Department of State, Foreign Relations of the United States (FPUS). Diplomatic papers, vol. 7: China (Washington, D C . GPO, annual), p. 202 ^Pingzhuan, p. 381. 4 2 Yan Tongzhi. p. 328-330. It is claimed that this 'secret society' influence was exercised from early on in the Anti-Japanese War. A group of people were sent, for example, to Wutai, Yan's home district, in order to "gain a foothold", even though there was a large CCP presence there. ^Zhao Rui, "Yan Xishan Tong Di Panguo Zuixing Jiyao" (Records of Yan Xixhan's Traitorous Communication with National Enemies), Wenshi Ziliao Xuanji (Beijing: Wenshi Ziliao Publishing), vol. 29 (1962), p..233; Pingzhuan, p. 378-380; Heping Yundong. p. 321. 19 commander Wen Huaiguang, for example, was dispatched to see Qixian's puppet garrison cornmander, Zhao Rui. According to Zhao, Yan's orders were as follows: First, when Japan surrenders the newly-organized army needs to watch its rear positions, and must especially not abandon protected areas and allow the 8th Route Army to occupy defended territory. Second, all counties in Central Shanxi - especially [those on] the transportation route between Xiaoyi and Taiyuan - must be defended so that the when [Yan] returns to Taiyuan, everything wil l be safe".44 Zhao also indicated that when Yan was safely back in the capital, he planned to reorganize his own and the puppet forces into a new "provincial Defense Army", and that the two of them would essentially be awarded promotions: Zhao Rui himself, as the case turned out, was made commander of the '2nd Army' under this scheme, deployed in the Shangdang region around Changzhi. On the diplomatic front, he named one of his most trusted commanders, Zhao Chengshou, to be his Director of Political Development', negotiating the co-ordination of military and administrative resources with Sumita in the eventuality of surrender. He was kept busy shuttling to the cities around the capital meeting with chiefs-of-staff and puppet commanders, laying out the procedure - called "gc Mc H " (jieshou zhengquan, or 'taking over political power') - to be implemented upon the official end of hostilities.45 In mid-July Yan and his staff moved north - staying clear of the railway lines - to Xixian (f>H f | ) , approximately 80 kilometers northeast of Kenanpo and perhaps a third the distance to Taiyuan4 6. That he felt confident enough to make such a move is evidence of the virtual ceasefire that extended to all groups except the communists. According to one biographical source, Yan immediately began receiving reports from Sumita in Taiyuan, who seemed anxious to make new arrangements: ^Wen Huaiguang, commander of the First Cavalry in Jixian; cited in Shanxi Wang Yan Xishan (Shanxi Ruler Yan Xishan) (Henan: People's Publishing, 1990), p.223-234. 4 5 Yan Tongzhi. pp. 329-330. 4 6 Yan TongzhL p.332. 20 The Commander of the Japanese Army stationed in Shanxi [Sumita] believed that very few of the present provincial [officials] were sufficiently competent, and wished that a number of new personnel could be recommended as soon as possible, and that Yan come to Taiyuan to fight against the communist resistance47. Yan sent immediately a team of his staff to Taiyuan to negotiate the transfer of new officials to puppet-run county seats. He instructed them also "to prepare for the appointment of a new [transition] governor."48 Yan himself wished to choose Liang Shangchun, a long-time staffer and advisor to Su Tiren, but Liang himself deferred in favour of his nephew, Liang Yanwu. TABLE 1: YAN'XlSHAN'S (2ND WAR AREA) PRINCIPAL MILITARY COMMANDERS, AUG. 1945 NAME Gu Zongfen DESIGNATION ft Chief-of-Staff, 2nd War Area AREA Yang Aiyuan mm M 6th Army Group: 19th, 23rd Corps Shi Zebo 19th Corps (6th Army Group) Fushan Zhao Chengshou 7th Army Group: 33rd, 34th Corps Xixian SunChugi % $ 8th Army Group: 43rd, 61st Corps Jixian Liang Peihuang mm !§[ 61 st Corps (8th Army Group) Fushan Wang Jingguo 3 E 5 t i § 13th Army Group: 83rd Corps, 1st Cavalry Jixian Sun Fulin I t 83rd Corps (13th Army Group) Xiangning The quiet, personal diplomacy conducted through Yan's various envoys and representatives bore quick results. A further agreement, according to the Shanxi Da Shiji. was reached in Fenyang on July 23, under which Japanese military responsibilities were to be handed over to Yan's control step by step, and certain troops were to be used to help Yan's "Bandit Suppression'. At the same time, Yan acted to 4 7 Yan Tongzhi. p.330 48Jbid, P-331 21 protect Japanese business interests in Shanxi ... and [puppet] troops assumed defense of enemy positions in Lishi, Zhongyang, Xiaoyi and Fenyang"49. A l l of these towns are located along the major transportation and communications routes extending southwest from Taiyuan (see map), and were thus capable of controlling access to the capital. According to a Jiefang Ribao report, this was enough to put Yan's security worries at ease, and on August 2 he and his staff moved further north to Xiaoyi County, accompanied by his 82nd Army 5 0 (he always travelled under heavy guard, and went to extremes to ensure his own safety51). From his close proximity to the capital now, he was in a much better position to communicate with the centre of power, to make further co-operation possible with the Japanese: Yan was deeply aware that, after the Japanese surrender, i f he wanted to recover his control over Shanxi, it was simply not enough for him to rely on his own power. Because of this, he came up with a plan under which the strength of remaining Japanese soldiers would be used to achieve his goal. 5 2 The 8th Route Army was not unaware of Yan's movements, and equally appreciated the importance of securing the capital. On all fronts, army and guerrilla units continued to attack train lines, destroy blockhouses and occupy towns where they had enough strength; still, penetrating the main defense lines required both a larger concentration of troops and, significantly, a good reason to proceed past them. Each of Shanxi's three base areas included territory outside the province, and the focus of military activities was thus scattered over a wide area, from Suiyuan (now Inner Mongolia) to Henan, which followed a rural, guerrilla strategy. CCP leaders, like everyone else, did 4 9 Shanxi Da Ship, p.275. ^Jiefang Ribao (Liberation Daily, Yanan) Aug. 11/45, p. 2: "Yan announces defection to the Enemy". The army group under question is more likely to be the 83rd Corps, under the 13th Army Group commanded by Yan's brother-in-law, Wang Qingguo. For further documentation of Yan's movements, see Yan Tongzhi. p.334-335. * ^Amerasia Papers, p. 1430. 52Pingzhuan. p. 385. 22 not expect the war to end suddenly, and until it did, the emphasis on small-scale guerrilla activity remained in place. TABLE 2: 8TH ROUTE ARMY COMMAND STRUCTURE, SHANXI BORDER REGIONS Liu Bocheng Jin-Ji-Yu (TaihangMts) iEm Lu Zhengcao Jin-Sui (Yanmen) n i t He Long Jin Sui (Luliang) mmm Nie Rongzhen Jin-Cha-Ji (WutaiMts) In the summer of 1945 Nie Rongzhen was kept busy from Jin-Cha-Ji headquarters in the Wutai region, as attacks were sustained in southern Chahar (now Inner Mongolia), Hebei and Yanmen (northern Shanxi). To the south, in Jin-Ji-Yu, troops descended from the Taiyue Mountains for guerrilla raids on Yangquan, Yushe and Zaoquan, and harassing Japanese and puppet positions as close to Taiyuan as Taigu 5 3. In the Taihang region, communist troops under Liu Bocheng had been under attack by Yan's 61st Army since the spring of 1944, and constant shifts between advance and retreat had not ceased since then.54 It was in the Jin-Sui region, however, where the 8th Route Army succeeded in operating closest to Taiyuan; during the final summer of the anti-Japanese war, guerrillas were able to attack roads leading from Taiyuan to both Lishi and Fenyang, as well as southward from Yanbei. Harrison Forman, who spent several weeks in the region, reported aggressive action against Japanese lines, the destruction of pillboxes and ^Minguo Junshi. p. 871 5 4 Ibid, pp. 1427-1429; Pingzhuan. p. 378-80. 23 garrisons, and ambushes on patrols and convoys in the 8th Sub-region, west of Taiyuan5 5. Their actions, however, like in other border regions, were not generally offensive in nature, maintaining hit-and-run attacks on the front and behind enemy lines but not gathering a large single force together. 4. N E W UNIFORMS, O L D SOLDIERS: C H A N G I N G O F T H E G U A R D A) SURRENDER? WHAT SURRENDER? A t the beginning of August,56 Takahashi Tadashi - Chief of Staff for the North China Army central command in Beijing - flew to Taiyuan, where he was joined by Yamaoka Michitake, his counterpart in the Japanese First Army stationed in Shanxi. 5 7 From there they were driven to Yaopu Village, in Xiaoyi County - close to Fenyang, south of Taiyuan. The purpose of the visit was to meet with Yan Xishan, who had arrived there three or four days earlier. Also in attendance was Su Tiren, the wartime puppet governor of Shanxi, Liang Yanwu - his newly-appointed successor - his uncle Liang Shangchun, and other senior staff members.58 Yan had met with senior Japanese officials stationed in Shanxi during the war 5 9 to discuss Japanese peace proposals, but was held back by his aversion to signed treaties with the 'enemy' which would not only render him unable to defend himself against charges of collaboration, but would cause him tremendous loss of face in his own home 55Harrison Forman, pp. 200 - 242. 5 6 All sources are similarly vague about the exact date of this meeting. One would assume, given the nature of their discussion, that it was either before the first atomic bomb was dropped, or at least before the news of its occurrence had reached the parties concerned. ^ Yan Tongzhi. p. 333. 5 8Zhao Rui, Xuanii. vol. 29, p.239. 5 9 When Yan met with Sumita, a photograph was apparently taken of them which, somewhere along the line, reached Jiang Jieshi's desk. When Jiang demanded an answer, Yan "replied that the picture could not be of him because he had worn nothing but his military uniform since the 'war of resistance'. Amerasia Papers (1970), pp. 1428-1429. 24 province. He had no qualms about the substance of co-operation, but would not jeopardize his command of the 2nd War Zone. And while Chief of Staff Yamaoka had also met with representatives of Yan before, the participation of a high official from Beijing such as Takahashi - not mentioned in English sources - seems to suggest that the Japanese were again prepared to regard Yan seriously, i f not with caution.60 TABLE 3: JAPANESE PERSONNEL IN SHANXI NAME \l ffl B W Sumita, R(a)ishiro Jono, Hiroshi Takahashi, Tadashi ill fss} jll Yamaoka Michitake }pj ^ f^S Kawamoto Daisaku Iwamatsu Yoshio POSITION(S) • Commander, 1st Army (Taiyuan), from Nov/44 • pre-surrender: various civil & military posts -post: organised "Stay Behind" Campaign • Chief of Staff, North China Army(Beijing) from Oct/44 Chief of Staff, First Army (Taiyuan) from Dec/44 Manager, Shanxi Industrial Company, 1944 - Former chief-of-staff, Kwantung Army Commander, 1st Army (in Taiyuan), until Aug/42 The meeting revolved around two main issues. The first was the question of imminent surrender61, and Yan's possible role in it. The other was a query regarding Japanese willingness to agree to Yan's latest proposal which went by the slogan 73 (qi cun wu li, or "co-opt existing military power"), under which Yan would have access to Japanese military resources and personnel, with the aim of launching an anti-communist offensive upon the official cessation of mutual hostilities. There are several versions of this meeting in Chinese sources, some of which vary in exact wording but most of which are identical in tone. In the following excerpt from the Yan Xishan Pingzhuan there is a noticeable politeness in the wording which seems to capture the mood Yan was able to assume. Takahashi first explained why b U Shanxi Wang, p.227-228. 6 1The date of this meeting is vague, but based on the way it is reported appears to place it at least before the knowledge of the first bomb in Hiroshima became known to them. 25 he had come: the Japanese forces in North China would consider surrendering to Yan alone, denying the CCP and even the allies the honour of the formal occasion: Takahashi: The situation in North Asia is about to change: the Japanese are at the point of stopping the war, and announcing a surrender; it is hoped that your excellency can get to Beijing by tomorrow morning, to participate in the North China Political Affairs Council62. Yan: If Japan is about to announce their surrender, (then you) must ask them first to notify me, so that I can make preparations; according to what the Japanese side has indicated, this should be beneficial to both parties. Takahashi: This is a matter of course. Before I receive such orders, they will first notify me; it is also hoped that your excellency can make communication with [the GMD government], on behalf of the Japanese rnilitary, congratulating the government and his excellency [Jiang Jieshi]. Japan's intention is to leave aside the Soviet, American and British allied countries, and to surrender directly to China. If this is successful, this will be, for Your Excellency Yan and His Excellency Jiang, equally advantageous, to the greatest possible degree. Yan: ...I am the most senior of the War area commanders, so it doesn't matter if Japan surrenders early or later on. Both ways there is power in areas under my control - which, incidentally, are being reorganized. Because of this, you really should direct the surrender of the North China Army to me alone. In my capacity as the commanding officer of this war area, specific changes are being made to numbers of designations to be granted Chinese troops. Later on we will file a report for the Chongqing government. I think that Mr. Jiang could not but agree!63 Takahashi agreed that Yan's scheme was amenable, but that he had came under orders from his superiors to negotiate only in principle. He thanked Yan and promised to contact his superiors in Tokyo when he got back to Beijing, for a final decision on the issue. If all went well, he said, he would come again to negotiate the specific details of a counterproposal. With the news of the bomb in Hiroshima, followed by the Nagasaki attack on August 8, the speed and volatility of events in this 'transitional period' became acute. Yan Xishan did not go to Peking, nor did he take the surrender of the Japanese army in all of China. As one source described the fate of these plans, it was "a wonderful thought, but 6 2 This is presumably a puppet organization, several of which the Japanese had attempted - in vain - to entice Yan to join during the war. ^Pinszhuan. p. 383-385. 26 all to nought", as the surrender was made to the Allies. 6 4 The 'qi wu curt W plan, however, was not shelved in Shanxi, and the reshuffle of local leaders continued as his military establishment began to "launder", in effect, many former puppets back onto the official payroll. 6 5 Yan's presence in Xiaoyi, less than 150 kilometers from Taiyuan, certainly made communication to and from Taiyuan relatively unproblematic, especially as it was mainly puppet-led troops which now held control of the major intersections along the way (or were at least in contact with the Japanese garrisons66). On the evening of August 9th, when Japan officially announced its willingness to surrender, Yan was already meeting with his core military staff - likely including Wang Jingguo, Zhao Rui and his Chief-of-StafF Guo Zongfen - to lay out plans for securing the capital6 7. The results of this were quickly disseminated the following day, as a former local official recounts: [We] received an urgent telegram from Yan, ordering the 8th administrative group head Zhang Yishan, as well as his subordinates Guo Qinghua and Zhang Yizhong, to instruct the local leaders in the nine counties under their jurisdiction [Taigu, Yuci, Yangqu, Wenshui and Taiyuan among them] to proceed to Xiaoyi...immediately for an emergency meeting. When the meeting began, Zhang announced that Japan had decided to surrender; the heads of the nine county workgroups soon decided that the county governments should be changed, and to produce a revised list of county leaders. They were also to seize the county seals, consolidate all arms, and to mobilize all necessary resources in order to quickly prepare for taking over power in all these places.68 The group was to co-ordinate these civil efforts with the advance military forces being sent ahead to clear the way, which received their marching orders the same day to ^Heping Yundong. p. 323. 6 5 Zhao Rui writes that he himself - the then-commander of the puppet-Japanese "Bao An" or "Peace Maintenance" Army - received a new title in Yan's "newly-organized 2nd War Zone"; similarly, Li Baosen, a puppet officer in Manchuria, became commander of the "2nd Bandit-suppression Corps." Zhao Rui, Xuanji. vol. 29 (1962), p. 237. ^^These included Taigu, Qixian, Wenshui, Lishi, Zhongyang and other county seats southwest of Taiyuan which were to be given over to Yan under the second 'Fenyang Agreement' (see note 30). ^Shanxi Dashiji. (Taiyuan: People's Publishing, 1987), p.276.' ^ 8 Gu Xiangting, "Yan Xishan Jieshou Taigu Xian Zhengquan Jishi" (Recollections of Yan Xishan taking over Political Power in Taigu County), Wenshi Ziliao. vol. 26 (1983), pp. 133-134. At the time, Gu was a member of Yan's 8th district administrative group, stationed in Xiaoyi City. 27 proceed toward Taiyuan, along two routes, led by Chu Xichun and Peng Yubin, the 7th and 8th Army Group vice-commanders respectively. Each passed through the major county seats close to Taiyuan, and several were already occupied by puppet troops. Guo's party of officials was to accompany Chu's forces, with Zhang's group joining Peng along the other route. On August 10th they were given the order by Yan to advance, slowly and deliberately so as not to cause friction of any kind. As Guo relates, everything went relatively smoothly until his party reached Taigu (the family home of H. H. Kong, Jiang Jieshi's brother-in-law and leading G M D banking figure). Here there was some resistance by the local Japanese commander, who did not appear quite ready to surrender. Huo Aiting, the work team leader assigned to Taigu, had approached the local puppet leader on the 16th - as he had been instructed - and promptly seized the county seals, wrote posters announcing the change and had them duly pasted up throughout the town. Soon after, the local Japanese commander rode up on his horse, incensed: he claimed that when the [Chinese] brigade commander had come to tell him the group would be coming to take over power, there had been no mention of an unconditional surrender. But the first point listed on the posters was that there had been an unconditional surrender. They had heard nothing of the sort, and "being Imperial soldiers...they were incensed"69. Before riding back, he threatened to set his men to pillaging the town unless they were taken down, and that i f they began firing, he (Huo, that is) would have to take responsibility. The solution, as it turned out, was to go out during the middle of the night and, under the privacy of an enforced curfew, and take the offensive posters down themselves. If anyone asked about them, they would say that there were "8th Route Army Guerrillas in the city, and the posters had been removed by them".70 Such antics - and the refreshing candor of the former official in this case - well illustrate the established order 6 9 G u Xiangting, Wenshi Ziliao. vol. 26 (1983), p.135. 70Ibid., p. 135. 28 of relations and protocol which existed between local officials who served under the Japanese. Communicating from Chongqing on August 9th, Jiang Jieshi ordered communist armies throughout China to halt, and to allow Nationalist forces to take the Japanese surrender where they were. As the commander of the 2nd War Zone, the '18th Army Group' [8th Route Army] came under Yan Xishan's command, with Zhu De acting as 'vice-commander'. At this point, however, Zhu was far more focused on his duties with the 8th Route Army. The same day he began issuing a series of seven commands71, the first of which ordered cornmunist troops "to send orders to nearby Japanese and puppet troops, stationed in cities and along lines of communication, to surrender all their arms within a stated time limit. Those who refused were to be destroyed at once."72 Considering the size of Yan's own core military strength - roughly estimated at 100,000 troops 7 3 - it appears ironic that he should be made responsible for preventing Communist troops from doing anything. Zhu De, his nominal subordinate, had at his disposal many times more military power that Yan throughout North China, 7 4 and ordered an all-out attack on cities throughout North China. Many of the major offenses to which these forces were being deployed were, as before, outside Shanxi. In Jin-Cha-Ji, the targets were Beijing, Tianjin and Zhangjiagou. Lu Zhencao in Yanbei - and much of what had been known as the "New Army" - proceeded to march northward to Manchuria, leaving He Long in charge of troops in Jin-Sui. He ordered attacks on three fronts on August 12, and by the 16th, confirmed attacks are reported in Lishi, Fenyang, Pingyao I^Jiefang Jun Da Shin.. p.219; Agnes Smedley. The Great Road: The Life and Times of Chit Te (New York: Monthly Review Press, 1956), pp. 417-418.. ^^Smedley. The Great Road, p. 17 7 3££f/£(1945), p. 202. 74 A figure of 910, 000 is given, for regular troops only, inMinguo Junshi. p. 813; similarly in Lyman Van Slyke, "The Chinese Communist Movement During the Sino-Japanese War 1937-1945". Cambridge History, vol. 12, part 2 , p. 708. 29 and Wenshui, but their own commitments to the front in Suiyuan and elsewhere75 kept them well away from Taiyuan itself. The CCP complained loudly over Yan's alleged involvement with the Japanese: The New York Times reported their claims that "the Japanese objective in the deal was to assure maintenance of an island of Japanese influence in North China even in defeat"76. In lieu of evidence that such a plan was actually made or sanctioned by the leadership in Tokyo, it seems more likely that someone such as Sumita in Taiyuan, with Okamura's knowledge, would have come up with it. In the same way that Yan himself was commander of the 2nd War Zone - while obviously setting bis own agenda - Sumita also played a double role, staying as he did in Shanxi as one of Yan's 'advisors' until at least 1947. In an interesting aside to the CCP's charges against Yan, Michael Lindsay writes that there was, indeed, hard evidence to prove the claims about activities, but that the Xinhua News Agency in Yanan stuck to its habit of jargonistic invective, with the observation that "only the communists who understood dialectical materialism could believe something because of evidence...it was, therefore, pointless to set out evidence in newspaper articles."77 This is evident from reading the Jiefanz Ribao^ in which the accusatory rhetoric is unrestrained, and where detailed information on Yan's activities is difficult to locate. According to Yan's former Chief of Staff Deng Zongfen7 8, before Zhao Chengshou left Fushan for Beijing, he dispatched Shi Zebo's 19th Army Group to take the surrender from the Japanese garrison there, which was already supplemented by Chinese soldiers. Shi's forces now began to advance toward Changzhi, taking the surrender in towns along the way, and meeting fierce resistance from communist troops. From mid-75M/wgwo Junshi. p. 877. 16NYT. August 12, 1945, p. 31 ^Michael Lindsay, The Unknown War. p.88. 7 8Guo Zongfen. Wenshi Ziliao vol.8 (1963), pp. 16-17. 30 August, and continuing well into the fall, the Battle of Shangdang prevented communist forces in the southeast from moving in on central Shanxi around Taiyuan, as had been planned.79 B) TOWARD TAIYUAN W h e n Yan ordered Chu Xichun and Peng Yubin to begin their two-stage advance toward Taiyuan on the 10th,80 Chu's forces swept northward first, on either side of the Fen River, stopping at the county seats to inform those in charge that work teams would be arriving shortly. By the 15th, an advance party had arrived at the town of Xiaodian, a short distance from the capital. They made contact with the Japanese the next day to parley, and the next morning Chu sent a couple of regiments around the city to the outlying towns of Huanghoyuan - 'Empress Gardens' - on the northern outskirts of the city, as well as Huangzhai, and then divided to move south from there to garrison the suburb of Xincheng, where many of the Japanese had made their homes since 1937. When on Aug. 14 Japan officially announced its unconditional surrender,81 Yan sent an urgent telegram to Zhao Chengshou, urging him to hurry to Linfen, where he was to "take the train quickly toward Taiyuan, before any shots were fired, to parley with the Japanese about the question of "accepting [the surrender of] Taiyuan"82. Zhao arrived there on the afternoon of the 17th, after which - according to his own testimony - he paid a visit to Sumita's headquarters. His orders were to "make the Japanese understand that in military terms there was no desire for conflict, only to prevent the communists from occupying Taiyuan."83; they soon reached what is vaguely referred to as a "so-called 79J bid., p.; Yan Tongzhi, p. 342. 8 0 Pingzhuan. p.386. 8 1 Shanxi Wang, p.229. ^Shanxi Wang, p.229; Shanxi Da Shiii. p. 277; Pingzhuan. p.386.. ^Yan Tongzhi. p.336. 31 'settlement' of the issue,"84 according to which the Japanese throughout the province were, first, "to wait for orders where they were, until Yan's armies came to accept the surrender, and second, not to allow defended territory to fall into the hands of the communists"85 Both sides would co-operate to block the communists from accepting the surrender in Taiyuan, but the Japanese agreed to "take primary responsibility"; in addition, i f they wished to make any troop transfers, they would first have to request by telegram for Yan's approval. Characteristically, he seized the weakness of the Japanese position - now that the surrender was official - and began to speak with a more forceful voice. ^Pingzhuan p. 386. 8 5Zhao Rui, WenshiZiliaoXucmji, vol. 26 (1983), p.242-243. Zhao Rui's testimony is quoted heavily in much of the biographical materials. 32 By mid-August, American military sources reported not only that Yan was on the move, but that there was "increasing friction between Communists and Yen's forces, and probability of serious clash for Taiyuan is in offing."8 6 The New York Times makes a vague reference to a reported attack by communists at the airport near Taiyuan.8 7 There is no mention of this in the Chinese press; they do, however, contain reports of clashes around Taigu, Fenyang and Linfen. The 8th Route Army held no permanent territory in these places; thus, when the central government reported that "Chinese troops" had by August 21 taken several major points near Taiyuan, including Taigu, Fenyang and Taiyuan Junction,88 one must assume that they refer to Yan's spearhead forces under Peng Yubin and Chu Xichun. Their purpose there was not, however, to 'take the surrender', which was being handled by other military and civil officials bringing up the rear, but to secure safe passage for Yan's return. Now that this protective corridor was in place along his way, Yan planned to move again, from Xiaoyi - where his headquarters had been located for a month now -along the rail line eastward to Jiexiu, which sat on the main rail line running north. Sections of the railway between Jiexiu and Pingyao, however, had been destroyed by guerrillas, and it was not until the 23rd that he arrived in Jiexiu. From here he sent a telegram to Sumita requesting that preparations be made for the formal surrender. Sumita responded by sending Yamaoka to meet Yan in Jiexiu, along with an armoured train, car and a contingent of "more than 500 soldiers" as an escort.89 The following morning he boarded the train with Yamaoka to begin the northward journey, but they only got as far as Y i A n before they were stopped. During the night, communist guerrillas had come and ^FRUS (1945), p.535 8 7 A2T, August 21 1945, p.2. 8 8/&/ H A S Mtifc W A 0IB 1927-1982 (A Chronological History of the People's Liberation Army). Beijing: People's Military Affairs Publishing, 1983. Zhongguo Renmin Jiefang Jun Zhanvi tf3 S A S ^ K I I R S (PLA Military Campaigns). Beijing: PLA Publishing, 1987. Zhongguo Renmin JiefansJun Lishi Cidian ^ \&$£ffl(.W-M. !$l $Q (PLA Historical Dictionary). Beijing: People's Military Affairs Publishing, 1990. (Japanese) Nihon Riku-Kaigun sogo jiten (Japan Armed Forces Directory) Tokyo: Tokyo Daigaku Shuppankai, 1991. E N G L I S H S O U R C E S Barnett, Doak A. China on the Eve of Communist Takeover. Seattle: University of Washington Press, 1963. Ch'i, Hsi-sheng. Warlord Politics in China. 1916-1928. Stanford: Stanford University Press, 1976. Deane, Hugh. Good Deeds and Gunboats: Two Centuries of American-Chinese Encounters. China Books and Periodicals, Inc., 1990. 45 Forman, Harrison. Report from Red China. New York: Henry Colt & Company, 1945. Fu, Poshek. Passivity, Resistance and Collaboration: Intellectual Choices in Occupied Shanghai. 1937-1945. Stanford: Stanford University Press, 1993. Gillin, Donald. Warlord Yen Hsi-shan in Shansi Province 1911-1949. Princeton: Princeton University Press, 1967. "Staying On: Japanese Soldiers and Civilians in China, 1945-1949". The Journal of Asian Studies, vol. 32, no. 3, May 1983., pp. 497-518. Hersey, John. "Letter from Peiping", The New Yorker, Feb. 1946. History of the Sino-Japanese War (1937-1945). Taipei: Chung Wu Pubhshing Co., 1985. Lary, Diana. Region and Nation: The Kwangsi Clique in Chinese Politics Cambridge: Cambridge University Press, 1974. Lindsay, Michael. The Unknown War: North China 1937-1945. London: Bergstrom & Boyle Ltd., 1975. Nakamura, Yakafusa, "Japan's Thrust into North China, 1933-1938", in Iriye, Akira, ed.. The Chinese and the Japanese. Princeton" Princeton University Press, 1980. The New York Times., 1945. Stories filed by Henry Lieberman, Tillman Durdin. Pepper, Suzanne. "The KMT-CCP Conflict 1945-1949". Twitchett, Denis, Fairbank, John K., eds., The Cambridge History of China. Vol. 13: Republican China. 1912-1949. part 2 pp. 723-788. Cambridge: Cambridge University Press, 1986. Pye, Lucian. Warlord Politics: Conflict and Coalition in the Modernization of Republican China. New York: Praeger, 1971. Smedley, Agnes. The Great Road: The Life and Time of Chu T.e. New York: Monthly Review Press, 1956. United States Department of State. Foreign Relations of the United States (FRUS): Diplomatic Papers, vol. 6: China, 1944; and vol. 7: China, 1945. Washington: GPO, 1967 United States Office of War Information, Yan Xishan's Political Position in North China. Office of Strategic Services (OSS). Documents: 1945 . (on microform) United States Senate, Judiciary Committee. Amerasia Papers: A Clue to Catastrophe of China, for Subcommittee to Investigate Admimsttation of Internal Security Act and Other Security Laws. Washington: ConMnittee print, January 1970. Prepared by Anthony Kubek. 2 Volumes, 1819 pages. Van Slyke, Lyman P. "The Chinese Communist Movement During the Sino-Japanese War 1937-1945". Cambridge History, vol. 12, part 2 , pp.609-722. 46 5 ed. The Chinese Communist Movement: A Report of the U.S. War Department, July, 1945. Stanford: Stanford University Press, 1968. White, Theodore. Thunder Out of China. New York: William Sloane Associates, 1946. Wilson, Dick. "Leathernecks in North China, 1945". Bulletin of Concerned Asian Scholars, vol. 4, 1972, pp. 33-37.