5—
Centralization and the Provinces under the Dictatorship of Yuan Shikai

A significant feature of the provincial regimes that emerged from the 1911 Revolution was the extent of their autonomy from central government control. In contrast to the centralized bureaucratic empire it replaced, the new Chinese Republic was, in effect, a confederation of semi-autonomous provinces. The increased provincial autonomy of the early Republic was not simply the result of a breakdown of central authority. Rather, it represented a new emphasis by politicized elites on the province as an alternate arena for the pursuit of broader national goals. The proponents of this new provincialism thus did not see the autonomy of postrevolutionary provincial regimes as dangerous political fragmentation. Rather, this provincialism was founded on an implicit federalist assumption that the creation of strong self-governing provinces would strengthen the nation as a whole.

The federalist justification for the new powers claimed by provincial regimes at the beginning of the Republic was not, however, universally accepted. A competing view, already evident in the late Qing program of official reform, argued that national strength could only be assured by increased administrative centralization. Although such centralizing efforts were temporarily derailed by the revolution, the supporters of this view still believed that the autonomy of the provincial regimes could only result in national weakness. Yuan Shikai had been a leading figure in the official reform movement, and after assuming the presidency of the Republic he remained a proponent of the need for more, not less, centralized power. The de facto autonomy of the provincial regimes therefore presented Yuan with a serious political challenge and set the stage for a conflict that would reshape the Republican polity. In mid 1913 Yuan successfully overcame a "Second

Revolution" mounted by his opponents in the provinces. He used this victory to establish a centralizing dictatorship that sought to halt the Republican experiment with provincial self-government. The provincial regimes of Hunan and Hubei both fell victim to Yuan's centralizing drive, scuttling the provincialist alternative they offered.

The Second Revolution was important not only because of the political issues it resolved but because they were resolved by recourse to military power. Many factors contributed to Yuan's victory, including a fairly strong aversion among the elite to a renewal of military conflict. Nonetheless, once military conflict broke out, superior military power determined the winner. Likewise, the extension of Yuan's centralizing dictatorship would have been impossible without the military power he wielded. The Second Revolution therefore marked a crucial step in the militarization of politics—that is, the application of military power to achieve political ends.

Although the establishment of Yuan's dictatorship generally enhanced the political importance of military power, some qualifications must be made about the limits Yuan himself sought to put on military power and its use. Yuan's intent was not to institute military rule, let alone to encourage warlordism. Ultimately, his centralizing efforts were aimed at the reduction of both military and provincial autonomy. As Ernest Young has pointed out, the emergence of warlordism was not so much an "intended consequence" of Yuan's policies as the result of their failure.^[1] There was no necessary contradiction in the use of military power to create a political structure that would seek to restrict the political power of the military. Yuan could have found a model in the pattern of Chinese dynastic history, where military conquest and unification preceded reestablishment of political legitimacy and the reassertion of civilian rule. While using military force to destroy the civilian alternative offered by the provincialist regimes, Yuan sought to install an equally civilian-oriented central bureaucratic administration.

An examination of the extension of Yuan's control over Hubei and Hunan shows that he had considerable success in his objectives. With the extension of central military power into the provinces, civil and military administration was largely resubordinated to central bureaucratic control. At the same time, a variety of measures served to restrain the autonomy of the military governors and military forces sent to these provinces as the agents of central power. Having reinforced the structure of central control, though, Yuan failed to rebuild a civil base of authority to support it. By relying on coercion

to enforce his political will, Yuan was unable to forgo the military foundation of his political authority. This failure was never more evident than in his attempt to make himself emperor. Although the goal of the monarchist movement was to provide Yuan's regime with greater legitimacy, it could only be implemented by coercion.

The Second Revolution

The dispute between the adherents of centralized government and of provincial autonomy was only one of a number of interrelated constitutional issues at stake in the Second Revolution. The existence of political controversies, no matter how bitter, does not however explain why military force was ultimately used to resolve them. The context of military conflict was the lack of a clear consensus about processes or institutions able to legitimate the resolution of political conflicts. On one side there was Yuan Shikai's rejection of the participatory politics that had legitimated revolutionary regimes in 1911. On the other side there were revolutionaries who could find no common ground with Yuan once his authoritarian tendencies became clear. At the same time, years of revolutionary experience had inured revolutionary activists to the purposeful application of military power. Those who saw Yuan's activities as a betrayal of the revolution had little difficulty supporting any means, including military force, to carry out a "second" revolution against him. Unfortunately for the revolutionary cause, though, there was no broader elite consensus that the political situation in 1913 required a military solution. Without this broader support, the Second Revolution could not repeat the success of 1911. Meanwhile, the suppression of this revolt gave Yuan an opportunity to apply his own military solution to the early Republic's political controversies.

Although the issues fought over in the Second Revolution can be defined entirely in constitutional terms, they were also interwoven with the fundamental practical question of politics—who exactly would wield power in the Republic. This issue was particularly fueled by the dissatisfaction of revolutionaries over their underrepresentation in the postrevolutionary political structure. During the revolution, concern for the preservation of order, and to prevent foreign intervention, led revolutionaries to accept political compromises with more conservative reformers among the gentry, or even with imperial officials, as long as they accepted the overthrow of the Qing dynasty. Thus, just as revolutionaries in Hubei and Hunan had been willing to accept Li Yuanhong and Tan Yankai's leadership in revolutionary

provincial regimes, the presidency of the Republic had been offered to Yuan Shikai in exchange for his support of the call for the abdication of the Manchu emperor. After the establishment of the Republic, though, considerable discontent festered within revolutionary ranks over the preponderance of national and provincial posts in non-revolutionary hands. The potential for conflict with Yuan Shikai thus grew as revolutionary leaders sought to ensure that more of the political rewards of the revolution went to those who had promoted it and who would safeguard its objectives.

While their presence in Republican administration remained weak, revolutionaries saw new hope for an expansion of their political influence in electoral politics. This was particularly evident in early 1913, when the revolutionary-based Nationalist Party won the largest representation in the National Assembly, as well as in a number of provincial assemblies. Still unresolved though was the exact division of authority between the executive and legislative branches that would determine the power political parties would exert through these assemblies. The settlement of this issue awaited the drafting of a permanent constitution, slated to take place after the 1913 elections. Whereas Yuan naturally favored a strong presidency, Song Jiaoren, the organizational leader of the Nationalist Party, campaigned for the creation of a parliamentary system that would reduce the powers of the presidency in favor of a premier directly responsible to the National Assembly's majority party. Insofar as the assembly was to draft this constitution, the Nationalist Party's electoral victory in 1913 opened the way for the transfer of national political power into revolutionary hands. If the premiership envisioned by Song Jiaoren were established, Song, as legislative leader of the leading party, would no doubt replace Yuan as the chief executive of the central government.^[2]

An equally important issue awaiting more precise constitutional definition was the question of the proper relationship between central and provincial authority. Yuan and the revolutionaries were not so clearly in opposite camps on this issue. Many revolutionary leaders agreed with Yuan that a strong central government was a prerequisite for national power. In September 1912, Sun Yat-sen and Huang Xing met with Yuan and acknowledged the need for centralized administrative control over the army, foreign affairs, finances, the judiciary, and transportation and communication systems.^[3] Song Jiaoren was also a consistent advocate of centralized national government, albeit under a strong premier rather than a strong president. As long as Yuan remained in control of the central government, though, most revolution-

ary leaders were reluctant to see the extension of central power into those provinces where they retained significant political influence. Neither, as a matter of practical politics, could the Nationalist Party risk its local political bases by taking a strong stand in favor of central power. The party's platform thus stressed support for local self-government alongside a much vaguer call for political unity.^[4] The issue of centralization also became enmeshed in the debate over the relative powers of the executive and legislative branches of the government. On the one hand, Yuan Shikai saw a strong executive as essential for effective central government. On the other hand, most national and provincial assemblymen were inclined to defend the political interests of their provinces from an aggrandizing central administration.

The 1913 electoral success of the Nationalist Party thus threatened Yuan on several fronts. He faced the possibility of a strengthened National Assembly hostile to his centralization plans, a yet-to-be-drafted constitution that might undercut the executive powers of his office, and the potential loss of his own political power to a premier representing the revolutionary party. Presented with these challenges, Yuan went on the offensive. His first action was to remove his most immediate political adversary, Song Jiaoren. On March 20, 1913, Song was assassinated in Shanghai by men later proven to have been agents of Yuan's government. This was a chilling indication that Yuan would not shrink from the use of force to obtain his political objectives. Indeed, Yuan was now prepared to use the military power of the central government against both his political enemies and the recalcitrant provinces.

One obstacle that remained in the path of an effective mobilization of central military power was the central government's continuing financial weakness, largely created by the loss of tax revenues from the provinces. In April 1913, Yuan found a temporary remedy to this problem by negotiating a large government "reorganization loan" from a foreign consortium.^[5] In obtaining this loan, Yuan ignored vehement nationalist opposition that saw the terms of the loan, permitting foreign supervision of China's salt administration, as a betrayal of Chinese sovereignty. At the same time, Yuan showed his disdain for legal procedures by concluding the loan without the approval of the National Assembly, as explicitly required by the provisional constitution. Of course, the willingness of the foreign powers to approve this loan also contributed to Yuan's ability to ignore these political and legal restraints. Charles Tilly has noted that foreign

military aid strengthens the capacity of Third World armies for independent political action by freeing them from the need to reach compromises with social or political forces in their own societies.^[6] Yuan's ability to apply the reorganization loan to meet central military expenses similarly encouraged his proclivity to resort to a military solution to his political problems.

With his finances assured, Yuan initiated an attack on provincial autonomy by ordering the replacement of the military governors of Jiangxi, Anhui, and Guangdong provinces. By this means, Yuan moved to reassert central appointment powers over provincial officials without waiting for constitutional authorization for this prerogative. The replacement of these three governors in particular also served a political purpose. All three were revolutionary activists who had gained these posts during the 1911 Revolution, and their armies remained the strongest revolutionary military bases. Thus the extension of central authority into these provinces would also diminish revolutionary power. Ignoring the debate over the legality of his orders, Yuan showed his determination to enforce his will by advancing loyal Beiyang Army units into Jiangxi.

For many revolutionary leaders, including Sun Yat-sen, the assassination of Song Jiaoren was sufficient reason to seek Yuan's removal from office. At the same time, Yuan's defiance of the National Assembly and the provisional constitution over the reorganization loan convinced them that he could not be stopped by legal means. By April 1913, some revolutionaries had therefore begun to plot military uprisings to overthrow Yuan. Other revolutionaries, represented by Huang Xing, recognized the weakness of revolutionary military forces and hoped to delay open conflict. Yuan's removal of revolutionary military governors and the advance of Beiyang troops into Jiangxi forced the issue by threatening the revolutionaries' only strong military bases.

On July 12, 1913, Jiangxi declared its independence, officially initiating the Second Revolution. In the next few weeks, six more provinces would follow Jiangxi's lead. However, the assessment of revolutionary weakness was proven correct when Beiyang armies loyal to Yuan inflicted a series of defeats on revolutionary forces. Although revolutionaries at Nanjing held out until September 4 and Sichuan did not cancel its independence until September 12, most serious military resistance to Yuan collapsed by mid August.

The weakness of the revolutionary position in the Second Revolution is evident in the response to this crisis in Hubei and Hunan. As the birthplace of the 1911 Revolution, Hubei might have been ex-

pected to be at the forefront of the anti-Yuan struggle. Instead, Li Yuanhong opposed the Second Revolution and effectively suppressed any action in its support. Hunan under Tan Yankai's reluctant leadership joined the Second Revolution, but could offer very little to prevent its defeat. Although ending up on different sides, the situations in both provinces reflected the weakness of the political consensus behind the Second Revolution.

Given his reluctant role in the 1911 Revolution, Li's opposition to the Second Revolution was hardly surprising. He did, however, attempt to take a conciliating position between Yuan and his opponents. Making the preservation of order his main goal, Li advocated the resolution of all political conflicts through peaceful and legal means. Thus, while decrying Song Jiaoren's assassination, Li urged that justice be sought through the courts. Likewise, while favoring Yuan's reorganization loan, Li proposed that the political conflict over it be settled peacefully within the National Assembly.^[7] On a number of occasions, Li personally mediated between Yuan and revolutionary provinces to reduce the tensions that had arisen from these and other issues. Even on the eve of hostilities, Li offered to act as a mediator in the dispute over Yuan's removal of revolutionary governors.^[8] Thus up to the outbreak of the revolt, Li's activities were in effect an extension of the mediating political role he played in Hubei into the national political arena.

There were limits though to Li's willingness to make compromises. First, he adamantly refused to support any call for Yuan's overthrow, asserting that his leadership remained essential for the maintenance of order and national unity. In May 1913, amid rising political emotions, Li banned all anti-Yuan protest meetings in the name of preserving public order. Second, Li consistently denounced any use of military force by revolutionaries seeking to redress their grievances against Yuan. Thus Li not only rejected revolutionary overtures for him to lead the military struggle against Yuan but also reportedly turned down appeals to remain neutral in the event of an outbreak of hostilities.^[9] Finally, despite his position as head of Hubei's semiautonomous provincialist regime, Li generally agreed with Yuan on the necessity of a strong central government. Li's attitude in this regard was perhaps related to his concern, as a military man, for national defense. Indeed, other professional military men in similar positions, such as Yunnan's military governor, Cai E, expressed similar concerns for a powerful central government.^[10] At any rate, when other provinces began to declare their independence, Li announced that he would

obey the orders of the central government and support the suppression of rebelling provinces.^[11] When it became necessary to take a stand, Li came down firmly on Yuan's side.

The efforts of revolutionary leaders to gain Li's support for their struggle against Yuan were no doubt hampered by the antagonistic relationship that had developed between Li and revolutionary activists in his own province. Hubei revolutionaries had already instigated several military plots directed against both Li and Yuan in late 1912. Song's assassination encouraged them to redouble their efforts, and in April 1913 a number of revolutionary military officers, including the commander of the 8th Division, Ji Yulin, were implicated in an uprising plot.^[12] In the succeeding months, Wuhan stayed in a permanent state of tension as Li's detectives thwarted similar plots by searching out and executing suspected plotters. The outbreak of the Second Revolution simply added further impetus to an ongoing revolutionary struggle.^[13]

The main revolutionary strategy in Hubei during the Second Revolution was to subvert the Hubei army in order to repeat the success of the 1911 Revolution with a military coup. However, conditions for such a coup were no longer favorable. Revolutionaries still had some influence within the Hubei army, but troop disbandment and the suppression of previous plots had long since broken up any effective revolutionary organization within it. With insufficient time to rebuild their organizations in the army's ranks, many revolutionaries focused their efforts on gaining the support of officers who might bring their entire units over to the struggle. However, as in the case of the 1911 Revolution, most officers, even if sympathetic, were unwilling to risk their positions unless there was some assurance of success. Success was even less likely than in 1911 because Li, taking advantage of his accumulated experience in suppressing revolutionary plots, had built up elaborate precautions against military uprisings. Furthermore, as noted in Chapter 4, the political instability of the Hubei army had led Li to request military assistance from Yuan, so northern troops loyal to Yuan were also on guard in Hubei against revolutionary attempts. As a result, revolutionary plotters had little leeway for action. Although a number of minor and short-lived uprisings did occur in outlying counties, Hubei under Li Yuanhong remained firmly in Yuan's camp.^[14]

Unlike their comrades in Hubei, Hunan's revolutionaries were able to bring their province over to the Second Revolution. The assassination of Song Jiaoren, a Hunan native, raised particularly strong anti-Yuan tempers in the province. After this point, Tan Yankai came

under increasing pressure from national revolutionary leaders and local activists to break with Yuan's government. As befitting his position as the head of the Hunan Nationalist Party, Tan generally supported the party's national leadership in its disputes with Yuan Shikai. Given Tan's decision to tie his political fortunes to the Nationalist Party, the pressure for independence was not easy to resist. Indeed, at one point Tan reportedly complained that he had become "nothing more than a monkey obeying his trainer's commands."^[15] Nonetheless, Tan opposed an open break with Yuan and forestalled Hunan's declaration of independence for many weeks.

Tan's most persuasive argument against independence was based on military considerations. The near-complete disbandment of the Hunan army, and the province's empty treasury, left Hunan particularly vulnerable to Yuan's armies. Tan himself seemed less concerned about the insufficiency of Hunan's military preparations than he was desirous of avoiding hostilities altogether. He was reluctant to have all he had accomplished in disbanding Hunan's forces undone by the new round of military recruitment that a struggle with Yuan would require. Only under extreme revolutionary pressure did Tan finally agree to begin rebuilding the Hunan army.^[16] Nonetheless, even with the outbreak of military conflict, Tan continued to hope for a peaceful settlement. In a meeting called immediately after Jiangxi's declaration of independence, Tan not only expressed pragmatic concern about Hunan's military weakness but warned of the internal disorder and opportunity for foreign intervention that might result from a civil war. It is a credit to Tan's political skills that even at this late date he persuaded Hunan's revolutionaries to agree to one last attempt at a mediated settlement. It was quickly apparent, however, that no settlement would be possible. On July 25, Tan finally succumbed to revolutionary pressure and agreed to a declaration of independence.^[17]

Tan's agreement to declare independence clearly went against his better judgment. Only days before the actual declaration, he told a party of Hubei revolutionaries who had fled to Hunan that he had no confidence in Hunan's ability to resist Yuan.^[18] After the declaration, he frankly told an American missionary that he had been forced into this action against his own will.^[19] In a later defense of Tan to Yuan, Li Yuanhong claimed that Tan had secretly disclosed his determination to commit suicide before agreeing to independence. Li reported that he had instead urged Tan to "appear to go along [with the Second Revolution], while secretly planning its suppression" if the pressure for independence became too great.^[20] Some Chinese historians have taken

this account as proof of Tan's secret desire to destroy revolutionary power, whereas others question its veracity, insofar as Tan never confirmed this story in his own defense.^[21] In the end, there is no reason to believe that Tan's reluctance to declare independence was based on anything more than a realistic assessment of the chances of winning a war with Yuan and fear of the general disorder such a war might bring. At the same time Tan evidently felt he could not afford to alienate revolutionary opinion in Hunan and perhaps provoke an uprising. In the short run, accepting independence may have seemed the only way to forestall discord within Hunan itself.

Tan's doubts about Hunan's chance of succeeding in the Second Revolution proved well founded. Hunan's hurriedly recruited troops were too few and too poorly trained to make a real impact on the war. Hunan also suffered a serious military setback even before the outbreak of war when the provincial arsenal was destroyed by a fire set on July 7 by Yuan's agents. Most of the weapons needed to arm new forces were lost in this fire.^[22] After Hunan's declaration of independence, Zhao Hengti's Guangxi brigade advanced cautiously across the Hubei border, but it only engaged in a few minor skirmishes with northern troops.^[23] In the end, the collapse of the Second Revolution in other provinces placed Hunan in an impossible position. On the very day that Hunan declared independence, the Jiangxi army suffered a major defeat, from which it never recovered. By the end of July, Huang Xing, who had taken command of the revolutionary forces at Nanjing, admitted defeat and fled to Shanghai. On August 6, Anhui's military governor also abandoned his post, and the province revoked its independence.^[24] Cheng Qian, an ardent supporter of the Second Revolution who had been appointed head of Hunan's Department of Military Affairs to oversee the rebuilding of the province's military forces, frankly reported to Tan that there was no real hope of a successful outcome, and on August 13 Tan rescinded Hunan's independence and recalled Zhao's forces from the Hubei border.^[25]

The failure of the Second Revolution reflected the inability of China's revolutionaries to build the same broad political consensus against Yuan that had been so effective against the Manchu dynasty. One Hubei activist later acknowledged that "at the time, Yuan's evil tracks were still not fully evident and all the people in the nation were very tired of disorder. This was greatly different from the revolutionary conditions that had existed before the 1911 Revolution."^[26] Constitutional conflicts over central and provincial authority, or over executive and legislative powers, were less easily defined in nationalist terms

than the anti-Manchuism of 1911. Indeed, in contrast to the idealism of the 1911 Revolution, many saw the Second Revolution as merely a conflict among self-interested politicians. On the eve of Hunan's declaration of independence, an English-language  newspaper  noted that

although the atmosphere in Changsha is electric, it is of a heavy dull nature in contrast to the current of contagious enthusiasm which marked the revolution. No desire for war is seen among the soldiers; the man on the street has no grievance against the government and is mystified as to the reasons for this new outbreak. Yuan Shih-k'ai [Shikai] cannot be called popular in Hunan, yet there seems no desire to oppose him except in the hearts of political job hunters and other self seeking "patriots." . . . Many in high quarters look with an ill-concealed concern upon the present situation and the people frankly state that they have known no real reason for revolt.^[27]

The absence of a consensus in support of the Second Revolution can be seen in the stands taken by merchant organizations in both Hunan and Hubei against revolutionary calls for provincial independence. Citing the preservation of order as their primary concern, these groups were not yet willing to consider war as the only solution to the problems confronting the Republic.^[28] In this instance, the proponents of the Second Revolution, not Yuan, appeared to be the party that most threatened the nation with disorder and disunity. Li's opposition to the Second Revolution, and Tan's reluctance to support it, were actually more in touch with this sentiment.

In the absence of widespread political support for their cause, the success or failure of the revolutionary struggle against Yuan Shikai came down to a question of military power. Here Yuan had an obvious advantage. The political effects of troop disbandments in Hubei and Hunan quickly became clear. Whereas Hubei had provided the military vanguard for the 1911 Revolution, in 1913, with the assistance of Yuan's northern troops, Li Yuanhong had no trouble suppressing revolutionary attempts to organize uprisings among the remnants of the provincial armies. Likewise, Hunan, though a major supplier of revolutionary troops in 1911–12, was unable to bring sufficient military force to bear to make an effective contribution to the Second Revolution.

Their failure in Hubei and Hunan shows the dilemma revolutionary activists faced in mid 1913. While believing that Yuan had left them no recourse but military force to stop the arbitrary consolidation of his political power, they were fully aware of the weakness of their military position. The defeat of the Second Revolution shattered any

hope of replicating their 1911 victory and paved the way for Yuan to implement the policies their revolt had sought to block.

The willingness of both Yuan Shikai and the revolutionary party to use military force to resolve the political disputes between them was an ill omen for the new Republic. Following so quickly after the 1911 Revolution, the Second Revolution heightened the importance of military power as a political force. At the same time, the rise of provincial military resistance to his central authority confirmed Yuan's views on the disruptive political effects of provincial self-government. Yuan thus emerged from the Second Revolution even more determined to consolidate the power of the central government by destroying the autonomy of the provincial regimes.

Centralization and the Demise of the Provincial Regimes

Although Li and Tan ultimately stood in opposite camps during the Second Revolution, until the actual outbreak of conflict their political positions were not that far apart. Both men were strongly motivated by a concern for order, which led them to advocate peaceful mediation and political compromise. Although perhaps somewhat naive about the possibility of an accommodation with Yuan, their attempts to avoid military confrontation were very much in step with the general elite response to the Second Revolution. With the end of the war, their common political viewpoints, as well as the common interests of their provincial regimes, drew Li and Tan together again as natural allies. Their actions showed that they still hoped to reach a political settlement with Yuan that would permit the continuation of some degree of provincial self-government. As it turned out, the revolutionaries had been more correct in assessing Yuan's true intentions. Indeed, after his victory Yuan felt even less compelled to make compromises. In a calculated manner, Yuan initiated measures to install a centralized dictatorship and in the process cause the demise of most autonomous provincial regimes, including those of Hunan and Hubei.

One of Yuan's primary postwar objectives was to strengthen his executive authority as president. Up to this point Yuan had been "provisional" president, with elections for a "regular" president slated to take place after the drafting of the permanent constitution. Even before the Second Revolution ended, Yuan pressured the National Assembly to agree to earlier presidential elections. With the revolt's failure, the assembly agreed to Yuan's proposal and set October 6 as the date of the election. On this day, unwilling to leave anything to

chance, Yuan organized a "citizens' corps" that kept the assembly captive in its chambers until his reelection was assured.^[29] Yuan's desire to have the National Assembly renew his presidency reveals his recognition of the importance of representative assemblies in providing political legitimacy. His use of coercion to achieve his ends in the assembly, though, showed his mistrust of the participatory politics that was the real base of this legitimacy. Through this coerced election, then, Yuan strengthened the importance of force as an alternate determinant of political authority.

Having used the National Assembly to consolidate his position, Yuan set out to destroy it. His assault on the assembly began indirectly as an attack on the Nationalist Party. Despite the complicity of many of its leaders in the Second Revolution, the Nationalist Party represented a political coalition broader than the original revolutionary Tongmenghui. Thus the party was far from united behind the rebellion. Intent on fostering this division, Yuan made no attempt to suppress the party while the rebellion still continued. The majority of Nationalist Party assemblymen therefore dissociated themselves from the Second Revolution and remained in their positions.

Once Yuan was confirmed as president, however, he was ready for stronger action. In early November 1913, he ordered a nationwide ban on the Nationalist Party and the expulsion of all national and provincial assemblymen who had ever held membership in it. So many national assemblymen and their alternates were connected to the Nationalist Party that the National Assembly could thereafter no longer form a functioning quorum, and Yuan's final dissolution of the assembly on January 10, 1914, was something of an anticlimax. Yuan easily discarded a draft constitution prepared under the assembly's auspices that would have severely limited the powers of the presidency. In its place, he later directed the preparation of another constitution that would confirm his dominant position in the central government.^[30]

The consolidation of Yuan's executive authority in the central government was accompanied by an extension of central authority into the provinces. Indeed, the two processes were interconnected. By eliminating the National Assembly, Yuan removed the one voice in the central government most likely to defend provincial interests. Likewise, the constitution produced under his direction asserted the principle of centralized administration over that of local self-government. This simply confirmed centralizing measures that Yuan had already initiated. This process began immediately in provinces, such as Jiangxi, Jiangsu, and Anhui, that had been militarily defeated and occupied

by Beiyang Army units. Centrally appointed provincial officials were quickly installed, and provincial assemblies that had supported declarations of independence were dissolved. Yuan was more cautious about pressing his objectives too quickly in areas beyond the immediate reach of his army. For a time, this caution seemed to hold out some hope for the survival of Hunan's provincial regime.

The main difference between Hunan and other defeated provinces was that Tan had annulled his province's declaration of independence before any northern troops crossed its borders. With no occupying army on Hunan soil to enforce his will, Yuan initially accepted Hunan's submission without insisting on Tan's removal. Yuan was also restrained by Li Yuanhong's personal pleading in Tan's favor. As noted above, Li claimed that Tan had been coerced into declaring independence against his will. In recognition of this, and in the interests of preserving order in Hunan, Li argued for the retention of Tan in his post.^[31] For a short period these conditions appeared to provide Tan with some leeway for independent action. Thus, while on the one hand Tan placated Yuan by suppressing revolutionary plots seeking to force a second declaration of independence, on the other hand he also aided in the escape of revolutionary leaders whom Yuan had ordered arrested.^[32] Tan also sought a settlement with Yuan that would preserve at least some elements of provincial self-government. Thus while Tan agreed to the dissolution of the Hunan Provincial Assembly, he worked to negotiate conditions for its revival. He even appeared to obtain Yuan's acquiescence to reconvening a rump assembly.^[33] Realizing that the absence of northern troops was crucial to his political flexibility, Tan also used Li as a mediator to delay the advance of Beiyang forces into his province. Even though Yuan insisted on sending some northern troops into Hunan as a sign of his authority, a settlement seemed possible when he initially limited the advance of his troops to Yuezhou on Hunan's northern border. Indeed, the occupation of Yuezhou by northern army and navy forces did not occur until late September, over a month after Hunan's formal submission.^[34]

As it turned out, Yuan had been simply waiting until his position was consolidated elsewhere before turning his full attention to Hunan. Although giving the appearance of a compromise, the military occupation of Yuezhou seriously weakened Hunan's position. Located at the point where Hunan's main river, the Xiang, flows into the Yangzi River, Yuezhou was Hunan's main transportation and commercial gateway. With the northern occupation of this city, Hunan not only yielded its most defensible point but gave Yuan the ability to place an

economic stranglehold on the province. In early October, Yuan appointed Tang Xiangming, the commander of northern naval forces at Yuezhou, to a special inspector's post (chabanshi ) over Hunan and ordered him to proceed to Changsha with a small troop detachment to "meet" with Tan. As soon as Tang reached Changsha, Tan was ordered to Beijing to "await punishment." Given the larger northern forces waiting at Yuezhou, Tan had little alternative but to obey. Once Tan was gone, Tang was formally appointed military governor in his place.^[35]

Because of his background and connections, Tang Xiangming's appointment as military governor served to soften the entry of northern power into Hunan. First, Tang was a naval officer, educated in England and France, and so strictly speaking not a member of Yuan's "Beiyang faction." Tang also had a revolutionary reputation for bringing the Yangzi naval fleet at Wuhan over to the revolutionary side in 1911. More important, Tang was the younger brother of Hubei's leading politician, Tang Hualong. As head of Hubei's late Qing Provincial Assembly, Tang Hualong had developed close ties with his Hunan counterpart, Tan Yankai, in the constitutionalist movement. After his short service in the Hubei revolutionary government in 1911, Tang Hualong had moved into national politics and was elected speaker of the lower house of the National Assembly.^[36] Thus because of his own background and his elder brother's political influence, Tang Xiangming's appointment was more easily accepted in Hunan than one of Yuan's Beiyang generals. Likewise, because of Tang's Hubei background, Li Yuanhong also approved the appointment, even though he opposed Tan's removal.^[37] Nonetheless, there was no question as to where Tang's loyalties lay. During the Second Revolution, Tang gained Yuan's appreciation by actively leading naval forces in support of Yuan's army along the Yangzi River.^[38] Tang's assumption of the military governor's post marked the end of Hunan's provincialist regime and paved the way for the extension of Yuan's control over the province.

Once Yuan had dealt with the provinces that had opposed him in the Second Revolution, he turned his centralizing efforts against the provincial regimes of putative allies such as Li Yuanhong. Despite Li's professed loyalty to the central government, Hubei's independence of central control was only slightly less than Hunan's. Yuan also had political reasons to be concerned about Li's position in Hubei. Li's prestige as the first revolutionary military governor and vice president of the Republic could as easily be turned against Yuan as used in his

favor. Indeed, rather than showing complete subservience to Yuan, Li continued trying to carve out a special role for himself as a mediator between Yuan and his political opponents. Li's attempt to intervene on Tan Yankai's behalf was only one example of his political maneuvering. Li also took an ambiguous position on Yuan's attempt to destroy the powers of representative assemblies. In November 1913, Li willingly carried out Yuan's orders banning the Nationalist Party in Hubei and expelling its members from both provincial and county assemblies. Li was less willing, though, to accept the permanent dissolution of representative institutions. By December, Li was actively involved in negotiations for the revival of national and provincial assemblies.^[39] By continuing to act as a political mediator, Li might emerge as a rallying point for those opposing Yuan's policies, so from Yuan's perspective he had to be removed.

Yuan had tried to entice Li away from Hubei on a number of occasions. Nonetheless, even the quite reasonable argument that Li's position as vice president required his presence in Beijing had been insufficient to achieve this end. By late 1913, the general consolidation of Yuan's power and the occupation of Hubei by Beiyang forces seriously reduced Li's maneuverability. Finally in early December, Yuan sent his minister of war, Duan Qirui, to Wuhan to "invite" Li personally to proceed to Beijing. Li could no longer refuse. On December 9, he boarded a train for Beijing, leaving his chief of staff as acting military governor. Even before Li's train reached Beijing, central orders named Duan as acting military governor in Li's place.^[40] In February, Duan Qirui returned to his post in Beijing and was replaced by another loyal member of Yuan's Beiyang Army faction, Duan Zhigui.^[41] While the immediate effect of these appointments was to bring Hubei within the orbit of Yuan's power, they also initiated a rule of Beiyang generals in Hubei that did not end until 1926.

The removals of Tan Yankai and Li Yuanhong were parts of a more general program of administrative centralization under Yuan that ultimately placed the majority of China's provinces under Beijing's dominance and significantly increased central influence over the others. Given the semi-autonomy of most provinces following the 1911 Revolution, this was a considerable achievement. The success of this centralization, however, also meant the demise of the distinct provincialist regimes that had emerged from the 1911 Revolution. Yuan's determination to undercut the political bases of these regimes was especially apparent in early 1914, when he finally ordered the nationwide dissolution of all remaining provincial and county assemblies.

Yuan thus destroyed the institutions that had legitimated the provincial regimes, as well as the channel they had provided for expanded political participation. The authority of the provincial governments that followed had no similar political foundation. Instead, they were extensions of central military and bureaucratic power.

The Realignment of Central and Provincial Military Power

The extension of central authority over the Yangzi River provinces was largely made possible by the southward expansion of the Beiyang Army and other northern units loyal to Yuan during the Second Revolution. Nonetheless, the remaining provincial military forces still posed some threat to the consolidation of central power. Thus one important task of the new centrally imposed provincial regimes was a realignment of central and provincial military power that would ensure provincial subordination to the center. This required both the continued strategic dominance of northern forces and the reduction or marginalization of those provincial forces whose loyalty to the center could not be trusted. The rearrangement of military power in Hubei and Hunan after the Second Revolution clearly shows the achievement of these objectives.

As noted in Chapter 4, the buildup of northern troops in Hubei began, with Li Yuanhong's permission, before the outbreak of the Second Revolution. By the end of the war, between thirty and forty thousand northern troops were deployed within Hubei's borders. Eventually many of these troops would be transferred either eastward into Jiangxi or southward into Hunan. After this, the Beiyang 2d Division under Wang Zhanyuan was left as the main northern army of occupation in central Hubei.^[42] In extending his military control over the Yangzi provinces, Yuan had to spread his armies quite thin. To remedy this. Yuan generally encouraged the expansion of occupying Beiyang armies. Thus Wang added a "supplemental brigade" (buchonglü ) to his 2d Division. In 1914, this brigade was expanded into an independent mixed brigade, ultimately given the designation of the National 6th Mixed Brigade, under the command of one of Wang Zhanyuan's 2d Division brigade commanders, Wang Jinjing.^[43] Yuan relied on these two forces then to maintain his control over Hubei.

The extension of Yuan's military power into Hunan began with the occupation of Yuezhou by the Beiyang 3d Division under Cao Kun and the 39th Brigade of the Fengtian 20th Division under Wu Xiangzhen in late September 1913. Yuan initially appointed Wu Xiangzhen

as Yuezhou garrison commander (zhenshoushi ), but after Tang Xiangming's advance to Changsha, parts of Wu's army were shifted to central Hunan and Wu's title was revised to Chang-Yue (Changsha-Yuezhou) garrison commander. Later in 1914, Cao Kun was also given the special title of commander-in-chief of upper Yangzi River defenses (changjiang shangyou jingbei zongsiling ). While his headquarters remained at Yuezhou, Cao's forces were spread out over a number of strategic points both in Hunan and along the Yangzi River in Hubei.^[44]

Yuan's use of Beiyang and other northern troops as instruments of central power was, of course, only possible because they were willing to subordinate themselves to his authority. Any one commander's subordination to Yuan may have been influenced by a number of factors, ranging from close personal or professional ties to a patriotic faith in Yuan's ability to provide national unity. Nonetheless, Yuan's victory in the Second Revolution also gave those officers and commanders who had shown their loyalty increased opportunities for career advancement. Conversely, insubordination would be quickly dealt with by other commanders seeking Yuan's favor. Whatever the case, Yuan's control over these commanders was evident in his ability to determine their posts and to deploy their forces wherever a show of central power was necessary. In 1913, for example, Wu Xiangzhen's "Fengtian" brigade was shifted first to Hubei, then to Yuezhou, and then to Changsha. In 1914 Wu's assignment was changed again to South Hunan garrison commander. Finally, in late 1915, Yuan transferred Wu out of Hunan to a new post as South Sichuan garrison commander.^[45] Likewise, Wang Zhanyuan's 2d Division, although assigned to Hubei, was sent into Henan in 1914 to help fight the bandit White Wolf.^[46] Although Cao Kun's 3d Division was concentrated along the Yangzi River, parts of it were also sent further afield as the need arose. For example, one of his subordinate brigade commanders was detached to replace Wu Xiangzhen as South Hunan garrison commander.^[47] Such deployments reflected a very high degree of responsiveness to central military command.

Whereas the northern forces transferred into Hubei and Hunan provided a strong base of central military power, they were not the only military forces in the two provinces. Despite the large-scale troop disbandments that had taken place before the Second Revolution, considerable provincial forces still existed. Although weakened by their fragmented organization, taken together these provincial troops outnumbered Yuan's armies. To redress this imbalance, and further consolidate northern military power, one objective of the new military

governors in Hubei and Hunan was the further reduction of provincial forces.

As noted in Chapter 4, Li Yuanhong continued his program of disbandment during and after the Second Revolution. Thus even before his removal, Li proposed the contraction of Hubei's undermanned three divisions and two independent brigades into one division and two mixed brigades.^[48] Upon assuming the military governorship of Hubei, Duan Qirui reportedly favored the total disbandment of the Hubei army and its eventual replacement with a new provincial army to be recruited from northern provinces. Although he never fully implemented this plan, the troops discharged by Duan during his short term of office further reduced the size of the Hubei army.^[49] Finally, in February 1914, the regular army's remaining troops were combined into a single division, the Hubei 1st Division, under the command of Shi Xingchuan. At first, this division was composed of three infantry brigades, but in May 1914 Shi was ordered to bring his division into line with national army regulations by the further elimination of one brigade.^[50] After this, no further disbandment occurred. However, to limit its power, most of the 1st Division was deliberately scattered among small garrisons in outlying towns.^[51]

In contrast to the contraction of the regular Hubei army, no attempt was made to disband or reduce Li Tiancai's Jiangnan 1st Division. No doubt Yuan Shikai, like Li Yuanhong before him, saw the non-Hubei composition of this force as a factor in its favor. Moreover, Li Tiancai and his division had proven their loyalty during the Second Revolution by aiding in the northern attack on Jiangxi and the suppression of revolutionary uprisings in Hubei.^[52] Yuan thus not only allowed Li to keep his division intact but rewarded him with several special appointments. In April 1914, Li was made Jingzhou garrison commander, headquartered at the site of the late Qing Manchu garrison. In January 1915, his appointment was changed to Xiang-Yun (Xiangyang-Yunyang) garrison commander, guarding the upper reaches of the Han River.^[53] In 1915, Li's division was also honored with an upgrade in its designation from a local (Jiangnan) force to a "national" unit, the 11th Division (revised in 1916 to the National 9th Division). This placed Li's army directly under the Ministry of War, which theoretically meant better and more reliable pay.^[54] By patronizing Li's division. Yuan sought to take advantage of its position as a "guest" army on Hubei soil to make it a supplemental agent for central military control.

The former Hubei 2d Division commander, Du Xijun, also retained

his post as Hankou garrison commander. It was no coincidence that Du was a native, not of Hubei, but of Zhili Province. Du's Hubei troops also presented little threat because they had been reduced to a small police force, only a battalion in size. It is of significance that a 1915 plan to increase Du's troops proposed recruiting soldiers not from Hubei but from Henan Province.^[55]

The consolidation of central military control over Hubei marked the culmination of the disbandment process that had begun under Li Yuanhong's influence. The regular Hubei army, which had grown to over eight divisions after the 1911 Revolution, retained only one division. At the same time, non-Hubei "guest" armies had progressively supplanted Hubei troops as the province's dominant military forces. One contemporary account notes that "Hubei men from this time forward lost [the opportunity] for a military livelihood."^[56] The broader purpose served by this change, though, was to divorce military forces in Hubei from any connection to provincial interests. The new configuration of military power in the province thus reduced the potential for military-supported resistance to Yuan's centralization program.

A similar reduction of provincial military forces took place in Hunan after the Second Revolution. This process again had some help from Tan Yankai before his departure. As previously noted, there had been a hurried attempt to rebuild the regular Hunan army before the outbreak of the revolt. A number of new units were thus recruited under the command of anti-Yuan military officers. Anticipating that war might increase local disorder, Guard Corps District commanders were also ordered to expand their troops to ten battalions per district.^[57] After the cancellation of Hunan's independence, Tan ordered the demobilization of all new units as a sign of good faith to Yuan. After its return to Changsha from the Hubei border, Zhao Hengti's Guangxi brigade was again the only regular army unit left in the province.^[58] In the interim before Tang Xiangming's arrival in Changsha, a number of local forces were also eliminated. For example, Zhao forcibly disbanded the Capital Guard Corps after an uprising in August 1913.^[59] Following the northern occupation of Yuezhou, 3d District Guard Corps units stationed in that city were also disbanded, while their commander, Chen Fuchu, was arrested and sent to Beijing to stand trial for his support of the Second Revolution.^[60]

After Tang Xiangming took office as Hunan military governor, he secured his position by eliminating all the provincial or local forces in Changsha and its surrounding counties. Because of its participation in

the Second Revolution, Zhao Hengti's Guangxi brigade was completely disbanded, and Zhao was sent to Beijing to stand trial.^[61] Tang also disbanded other small units, such as guard forces, left behind by Tan Yankai in Changsha. By the end of December 1913, Tang had demobilized the last remaining battalions of the 3d District Guard Corps stationed at Xiangyin between Changsha and Yuezhou. From December 1913 to April 1914, the 1st District Guard Corps, garrisoned east and south of Changsha, was also disbanded.^[62] At this point, then, the only troops left in the strategic and economic heart of the province around Changsha was the Fengtian 39th Brigade, commanded by Wu Xiangzhen, thus assuring northern military dominance.

Even with the establishment of northern military control over the center of the province, a large number of provincial Guard Corps and Green Standard troops remained in western and southern Hunan. Most of these forces were subsumed under a new garrison commander system, shown in Table 10.^[63] In most cases, these garrison commander appointments simply recognized the original commanders of these forces. Thus Tian Yingzhao's appointment as West Hunan garrison commander confirmed his authority over the Green Standard army that he had first received when Tan Yankai made him West Hunan zhentai in mid 1913.^[64] The 4th District Guard Corps commander Wang Zhenya continued to control his original forces as Chang-Li (Changde-Lizhou) garrison commander. The 2d District Guard Corps

commander Zhao Chunting was made South Hunan garrison commander. Zhao, however, was also given control over the 6th District Guard Corps.^[65] Less straightforward was the appointment of Tao Zhongxun as West Hunan vice garrison commander. Tao had originally been commander of the 3d District Guard Corps at Yuezhou. Before the outbreak of the Second Revolution, Tao had been ordered to exchange positions with Chen Fuchu, commander of the 5th District Guard Corps at Hongjiang in southwest Hunan. After a short time, he also left this post. Tao's garrison commander appointment placed him back in control of the 5th District's troops.^[66]

To see the purpose of these appointments, it is important to understand the place of garrison commands in the military structure of power created by Yuan Shikai in this period. This is particularly true in contrast to the position of garrison commanders in later Republican history. Eventually, like military governor , the title garrison commander became almost synonymous with warlord . In the warlord period, the granting of such titles often simply confirmed the positions of powerful local or regional military commanders who were already beyond central control.^[67] The reality of what garrison commanders would ultimately become should not, however, distort an understanding of these commands in this period. In most cases, Yuan was not simply giving central recognition to unassailable local military powers. Yuan had no trouble shifting the appointments of these commanders when he desired, even the stronger ones like Wu Xiangzhen or Li Tiancai. Indeed, such transfers worked to prevent the development of settled territorial bases or local ties. Furthermore, the garrison commands in this period were still merely military posts, without any authority over civil administration. Their commanders were thus given less chance than later warlords to build independent power bases. More important, Yuan was able to use these appointments for centralizing purposes.

The garrison command system was initially conceptualized as part of a centralizing scheme to eliminate the provinces as administrative units. Replacing the provinces would be smaller units, the circuit (dao ) on the civil side and the garrison command on the military side. The goal of this administrative reorganization was to remove the threat to the center represented by the concentration of civil and military power at the provincial level, especially as seen in the postrevolutionary military governorships. At the same time, circuit and garrison command boundaries would not be contiguous, thereby lessening the possibility of future military interference in civil administration. A

tentative implementation of this plan can be seen in several southern provinces, though not in Hunan or Hubei, where Yuan initially replaced military governors with several garrison commanders.^[68] In the end, though, the proposed elimination of provinces was never carried out. While still in the process of consolidating his position, Yuan could not afford to risk the instability that might accompany such a drastic administrative change. Nonetheless, the appointment of garrison commanders still achieved some of the centralizing goals that had originally been intended for them. First, Yuan used garrison command appointments, along with other titles, to reward loyal commanders, as well as to assert his authority over them. The appointments given Li Tiancai in Hubei fit well into this category. Second, even if the military governors were not eliminated, the garrison commanders could still function as potential central checks on them. Certainly, Wu Xiangzhen's various garrison commander appointments in Hunan served this purpose in relation to Tang Xiangming.

The appointments of garrison commanders from among Hunan's local commanders also served Yuan's purposes. The political objective behind these appointments can be seen in their timing. These appointments were first issued by Yuan on August 9, before Hunan annulled its independence.^[69] Tian Yingzhao, Zhao Chunting, and Tao Zhongxun had all been open in their opposition to Hunan's participation in the Second Revolution.^[70] This was no doubt the original reason for Tao's transfer from his original post at strategic Yuezhou to more remote Hongjiang.^[71] Although available records are silent on Wang Zhengya's position on the Second Revolution, his conservative background suggests that he was probably not a strong supporter. Yuan's appointment of these four men to garrison command posts thus appears to have been an attempt to ensure, or reward, their loyalty, and to create internal divisions in Hunan's military. Because the initiative for these promotions came from the central government, not from the province, a stronger central claim was staked over Hunan's military structure. For the same reason, the men who benefited from these appointments could be expected to show their appreciation through loyalty to Yuan. Finally, the confirmation of preexisting commands seen in these appointments does not support the conclusion that these commanders were too strongly placed to be removed. Rather, their comparative weakness is shown in Tang Xiangming's ability to carry out a fairly successful disbandment of a good number of their troops.

In terms of maintaining a balance between northern and provincial forces, Tang was far from comfortable with the number of Hunan

troops left under arms within the garrison command system. At the time of his arrival in Hunan, well over twenty thousand troops probably remained of Hunan's original Guard Corps and Green Standard forces.^[72] In late 1913, Tang ordered the disbandment of all excess troops from Guard Corps units beyond their original regulation sizes.^[73] During this time, as noted above, he also initiated the complete disbandment of 1st and 3d District Guard Corps remnants. In April 1914 Tang proposed a comprehensive plan to discharge most of the remaining Guard Corps troops controlled by the Hunan garrison commanders, leaving only a small number to serve as police under local magistrates.^[74] Although this program was never carried out in its entirety, it did lead to a fairly drastic reduction of Guard Corps troops.

One of Tang's main targets was the Chang-Li garrison commander, Wang Zhengya. Beginning with a five-battalion base, Wang had eventually expanded his 4th District Guard Corps to nine battalions. After a series of reductions ordered by Tang, by mid 1914 Wang's army retained only four companies, the equivalent of one battalion. Equally significant was the transfer of Wang's headquarters from Changde to Lizhou. Located close to the point where the Yuan river ran into Dongting Lake, and thus to the Yangzi River, Changde was the commercial gateway to western Hunan. With Wang's removal, northern troops took over the defense of this strategic city.^[75]

The second main target in Tang's disbandment program was the South Hunan garrison commander, Zhao Chunting. Neither Tang nor Yuan had much confidence in Zhao, and they distrusted Zhao's troops even more because of their participation in the 1911 Revolution. The ultimate goal in Zhao's case, then, was to make the reduction of his army a prelude to the complete abolition of his post.^[76] Tang first weakened Zhao's command by removing the 6th District Guard Corps from his direct control and placing it under a Hubei protégé, Wang Yunting.^[77] By mid 1914, Zhao was ordered to carry out a four-fifths reduction of his remaining troops, leaving only two battalions.^[78] After some foot-dragging, Zhao began to carry out these orders. Just then several units gathered for demobilization at the city of Chenxian mutinied. Although the mutineers at first appeared only opposed to disbandment, revolutionary activists soon persuaded them to declare their mutiny an uprising against Yuan Shikai. Soon other troops of Zhao's joined the rebellion, and they were reinforced by new troops recruited by revolutionaries.^[79]

Although provoked by the act of disbandment itself, the Chenxian mutiny was exactly the type of political threat in Hunan's military

forces that Tang's program of disbandment sought to eliminate. Thus, both Tang and Yuan took the uprising very seriously. Northern troops from Cao Kun's 3d Division and Wu Xiangzhen's 39th Brigade were quickly sent south to keep the rebellion from spreading. By early August, the rebellion was suppressed. Naturally, with the defeat of the rebellion, the disbandment of Zhao's troops was easily completed. In September 1914, Zhao himself was removed from office and Wu Xiangzhen was appointed South Hunan garrison commander.^[80] One effect of the rebellion, therefore, was to draw northern military power from central to southern Hunan.

All Guard Corps forces were not, however, removed from southern Hunan. The 6th District Guard Corps troops at Yongzhou under the control of Wang Yunting proved their value in campaigns against local bandits and in the suppression of the Chenxian mutineers. Although these troops had originally all been slated for disbandment, Tang limited their reduction from ten to six battalions. Yongzhou, located on the main pass controlling Hunan's southern trade with Guangdong, was another strategically important city, and Tang relied on Wang's loyalty to keep control over this area and saw to it that he was rewarded for his efforts. In early August 1915, arguing that the extreme southern counties were too distant to be patrolled effectively by regular troops, Tang received permission for Wang's appointment as Lingling (an alternate name for Yongzhou) garrison commander.^[81]

Tang was generally less concerned about the local forces stationed in western Hunan. Although initially slated for extensive disbandment, the 5th District Guard Corps forces under Tao Zhongxun's West Hunan vice garrison command were left with five battalions, about one-half their original number.^[82] There is no sign that Tang made any effort to tamper with the Green Standard forces under Tian Yingzhao. In general, western Hunan was a poor, mountainous region with very limited arable land and difficult lines of transportation. As a result, provincial authorities in this period seldom put a high priority on western Hunan's military arrangements—a factor that influenced the survival of Green Standard forces there in the first place. Tang seemed to carry on this policy of neglect toward western Hunan. At the same time, with their headquarters at Hongjiang and Fenghuang on Hunan's remote western border, neither Tao nor Tian presented much threat to Tang's Changsha government.

Less than a year after his assumption of Hunan's military governorship, Tang Xiangming had accomplished a considerable realign-

ment of military power in Hunan. After an extensive program of disbandment, the only Hunan troops left were much reduced forces along the province's periphery. The heartland of the province, meanwhile, was well under the control of northern troops. As in the case of Hubei, then, the extension of Yuan Shikai's authority into Hunan was buttressed by reducing the strength of provincial forces in relation to occupying northern armies. It is significant, though, that no effort was made in either Hubei or Hunan to eliminate all local or provincial forces in favor of central armies. This may have been a matter of temporary expediency, so that the northern forces loyal to Yuan would not be stretched too thin by their sudden expansion into central and southern China. At the same time, this situation suggests a continuing tendency to see an advantage in the counterbalancing effects of the organizational fragmentation of the military.

Military Governors and Central Control

Once Yuan abandoned the idea of eliminating provinces altogether, the success of his centralizing plans depended not only on the extension of central military power but on the subordination of military governors to his authority. One factor aiding this was that the new governors of Hubei and Hunan, unlike Li Yuanhong and Tan Yankai, only gained their positions as a result of Yuan's favor. Of equal significance was Yuan's success in limiting the personal military power of his appointees. The new military governors were, of course, military men. Tang Xiangming was a naval officer, and both Duan Qirui and Duan Zhigui had served as Beiyang Army commanders. Upon assuming their posts, however, none of these men had direct personal command over significant military forces; rather, they had general jurisdiction over military forces in their provinces. The most important of these forces were "national" armies that owed their ultimate loyalty, not to the military governors, but to the central government. Thus, although the military governors could deploy these forces as agents of the central government, they were less able to use them as power bases against the central government. Just as these military governors' central appointments distinguished them from their immediate predecessors, their weaker relationship to military power distinguished them from their warlord successors.

The military governors appointed by Yuan were not indifferent to their own power interests. In assuming their offices, Yuan's military governors became heirs to an established institutional tradition of strong provincial executives. These new governors were hardly reluc-

tant to take advantage of this tradition to expand their powers at every opportunity. The result was a continuing tension between the aggrandizing efforts of the military governors and Yuan's centralizing goals. This tension was mainly visible in two areas. The first was in the realm of military power. Both Duan Zhigui in Hubei and Tang Xiangming in Hunan were acutely aware of the increased power they could wield by establishing their own military bases. Although he made some concessions in this regard, there was a limit to how much military power Yuan would permit them. The second area of tension was civil administration. Here Yuan's efforts to enforce a separation of civil and military powers ran into the military governors' attempts to maintain the supremacy of their offices over both areas.

Seeking to increase his own military power, Duan Zhigui made several attempts to bring the existing forces in Hubei more under his direct control or to create new forces to the same end. In mid 1914, Duan complained to the Ministry of War that the disbandment of Hubei troops had put the province in a weak military position. He also expressed concern that national armies in Hubei could be called away at any time. He therefore proposed that a supplemental brigade within the 2d Division be transformed into a "Hubei" mixed brigade under provincial authority.^[83] Although this request was denied, Duan was allowed to recruit a new independent Hubei regiment. Duan increased his personal control over this regiment by placing it under the command of a young protégé, Lü Jinshan, a Zhili native and graduate of Japan's Army Officers' Academy. Significantly, Lü recruited this new "Hubei" unit in Henan.^[84] This repeated the pattern seen before of using non-native soldiers to assure greater political reliability. Because of their lack of local ties, these soldiers could be expected to be more loyal to the military governors or commanders who supported them. Not satisfied with this one regiment, in early 1915 Duan gained permission to establish a second regiment, recruited from Jiangsu and Anhui. These two regiments were then combined to form a new Hubei 3d Infantry Brigade, with Lü Jinshan as brigade commander.^[85] Duan also made further efforts to have Wang Zhanyuan's 2d Division and Li Tiancai's Jiangnan Division officially reclassified as Hubei forces. Yuan would not, however, agree to the provincialization of central military forces. Indeed, not only was Duan's proposal to remove the 2d Division's national designation denied, but the Jiangnan Division was reclassified as a national unit.^[86] Thus the creation of the Hubei brigade did not upset the balance of military power in the province in Duan's favor.

In Hunan, Tang Xiangming also tried to create military forces that would be more directly under his personal control. In early 1915, Tang received permission from Yuan to recruit a new regular Hunan army, and by the middle of this year he had established the Hunan 1st Brigade. As might be expected, this brigade was again an "outside" force recruited in Hubei, Tang's home province. Eventually the brigade was further expanded into a mixed brigade, with three infantry regiments as well as artillery, engineering, and cavalry components.^[87] In order to build a strong relationship with this force, Tang frequently visited its soldiers in their barracks and kept them well and regularly paid.^[88] This was a particularly good example of the increasing use of personalism to assert the primacy of one loyalty (to Tang) over another (to the central government). Nonetheless, as in the case of Duan Zhigui in Hubei, the establishment of this brigade did not raise Tang to a predominant military position over other military forces in Hunan.

Beyond his attention to the military balance of power in the provinces, Yuan sought to limit the authority of the military governors by appointing civil governors to each province. In Hubei, Yuan simply followed the precedent already established by Li Yuanhong. When Li was removed from his post, the current civil governor, Li's protégacute; Rao Hanxiang, gave up his post to follow Li to Beijing. Yuan immediately replaced him with Lü Diaoyuan, an Anhui jinshi whose prior official career had culminated in a circuit intendancy. Lü was also related to Yuan by marriage.^[89] Similarly, in Hunan, Tang Xiangming's appointment as military governor was accompanied by orders naming another former Qing official, Wang Hu, as civil governor, instituting this post in Hunan for the first time.^[90]

Yuan took a number of measures to emphasize the separation of powers he sought through the appointment of civil governors and in so doing revealed his intention of establishing civilian primacy. First, in mid 1914 the civil governor's title was changed from minzhengzhang , literally "head of civil government," to a vaguer and hence actually more encompassing term, xun'anshi . At the same time, the original broader term for military governor, dudu , was abolished and replaced by the more specific military title of "general" (jiangjun ) or, in the case of Hubei and other provinces that had had governors-general in the late Qing era, "high general" (shangjiangjun ). The administrative system behind these new titles purposefully emphasized the power of the civil governors at the expense of the military governors. The civil governor was given official precedence in his rank. A

clear delineation of civil and military affairs was ordered, and the military governor was strictly enjoined from interference in civil administration. The civil governor was even granted a degree of military power by being assigned authority over police, local self-defense forces, and any remaining old-style troops. The central appointment of civil governors was thus clearly intended to undermine the strong military governorships that had come to represent the growth of provincial autonomy since the 1911 Revolution.^[91]

The reversal of the established prerogatives of the military governors was not, however, a process that could be achieved by simple fiat. Yuan's own recognition of the need to tread carefully was revealed in the selection of civil governors believed amenable to the military governors. For example, besides his relationship with Yuan, Lü Diaoyuan was Duan Qirui's close friend and a fellow Anhui provincial, and Duan personally recommended him for the Hubei civil post.^[92] Although Hubei's next military governor, Duan Zhigui, was also from Anhui, he had no close ties to Lü. The two men's relationship quickly soured. Therefore, in October 1914, Lü was replaced with Duan Shuyun, a Jiangsu jinshi with a record of previous provincial and central posts. Not coincidentally, Duan Shuyun had a close acquaintance with Duan Zhigui, solidified by a clansmen's pledge.^[93] Wang Hu's appointment as Hunan civil governor was primarily owing to the influence of Xiong Xiling, a prominent politician from western Hunan who was Yuan's premier at this time. This appointment did not, however, meet with Tang Xiangming's approval. As a result, Wang delayed proceeding to Hunan and finally resigned in April 1914.^[94] In July 1914, Yuan appointed Liu Xinyuan, the former Hubei civil governor, to take Wang's place.^[95] As an important member of the Hubei gentry and a long-time associate of Tang Xiangming's brother, Liu was an appointee Tang could hardly refuse. The influence of such considerations in the selection of civil governors shows that they did not enter their offices with the primacy Yuan was attempting to assign to them.

In the end, the authority of civil governors over provincial administration remained dependent to a large degree on the acquiescence and cooperation of military governors. The separation of military and civil administration came closest to actualization during Duan Qirui's short term as Hubei military governor. Duan, after all, had only taken the Hubei post as a temporary assignment before returning to Beijing as minister of war. He could therefore be expected to promote Yuan's centralizing efforts. Duan drew a clear line between areas of military

and civil authority, and there was reportedly very little communication between Duan's office and civil departments. Only when Lü Diaoyuan met with special difficulties would Duan take a hand in assisting him, and then only "from behind the scenes." Nonetheless, Duan's role here appears very similar to Li Yuanhong's. Good intentions aside, Duan still acted as the Hubei government's final arbiter and thus was recognized as the greater authority. As in Li's case, this allowed Duan to exert considerable influence when he desired, as in the selection of civil personnel. Therefore, although some progress was made in building the civil governor's position under Duan Qirui, a complete separation of powers was not achieved.^[96]

Neither Duan Zhigui nor Tang Xiangming was as committed to the principle of the separation of military and civil administration. These two men, unlike Duan Qirui, saw the province as the arena in which to increase their political influence. Thus they were less willing to yield power to civil governors. Tang Xiangming had been appointed acting Hunan civil governor pending Wang Hu's arrival. Wang's timidity essentially left Hunan's civil administration in Tang's hands.^[97] By the time Liu Xinyuan took office, Tang had had time to consolidate his own ascendancy in the Hunan government. In Hubei, Duan Zhigui's arrival reversed the steady accumulation of power Lü Diaoyuan had enjoyed under Duan Qirui. Regulations and policies that by right should have been initiated by the civil governor were soon being drafted in Duan Zhigui's office for Lü's co-signature. Duan Zhigui also increased his influence over important civil appointments.^[98] Neither the mid-1914 attempt to reemphasize the separation of military and civil powers nor the appointment of the more amenable Duan Shuyun appears to have had much effect. After a special visit to Yuan in early 1915 to plead Hubei's strategic importance, Duan Zhigui received permission to continue to "assist" the civil governor in all important matters.^[99] Most contemporary accounts leave little doubt that the military governor continued to be the dominant partner in the provincial governments of both Hubei and Hunan.

Yuan Shikai's appointment of civil governors at the least asserted the principle of the separation of military and civil administration. His efforts to this end were not a complete failure. The civil governors were never totally subordinated to the military governors. They continued to be centrally appointed, ultimately responsible to the central government, and active in implementing central policies. Nonetheless, the establishment of civilian primacy in provincial administration remained one of Yuan's goals rather than one of his accomplishments.

In attempting to control his military governors, Yuan faced a basic dilemma. On the one hand, overly strong military governors could be potential threats to central power. On the other hand, Yuan relied on strong military governors to ensure that his will was enforced in the provinces. Ultimately, the goal of civilian primacy gave way to this later concern. By the same token, Yuan's failure to achieve this particular goal did not necessarily mean failure for his broader program of administrative centralization. Despite the tension between Yuan and his military governors, considerable progress toward increased central control over provincial administration was evident in a number of areas.

Administrative Centralization in the Provinces

The replacement of provincial governors with central appointees and the elimination of provincial assemblies were essential steps toward the reintegration of provincial governments under central bureaucratic control. To complete the destruction of the postrevolutionary provincial regimes, Yuan revived the late Qing law of avoidance preventing officials from holding posts in their native provinces. Provincially oriented politicization in the late Qing period had led to a rejection of this rule during the 1911 Revolution and its replacement by an opposite tendency—the monopolization of provincial and local offices by the native gentry of the province. This was one feature of the Hubei and Hunan regimes that gave them their distinct provincialist character. Some claimed, though, that rejection of the law of avoidance had brought about the very corruption that the law had originally sought to prevent. According to one report in early 1914, the domination of Hubei posts by local gentry had led to a situation where "officials and gentry colluded together to devour public funds and exploit the common people."^[100] Whether accurate or not, such charges served to justify administrative purges of the previous provincial regimes. In both Hunan and Hubei, one of the first duties of the new centrally appointed governors was the wholesale replacement of local officeholders.

It is debatable whether the restoration of the law of avoidance had any real effect on reducing corruption or improving the quality of officials. Indeed, the rapid turnover of officials provided an opportunity for a new favoritism by provincial administrators. The only difference was that this favoritism benefited men from the native provinces of officials, not from the provinces where they held their offices. In Hunan, the most influential man under Tang Xiangming in the selection of

magistrates and financial officials was Hu Ruilin, head of both the Department of Finance and the Department of Internal Affairs. Hu owed his own position to his personal ties to Tang Xiangming. Hu was a Hubei native, a long associate of Tang Hualong's, and was even related to the Tang family by marriage. Provincial connections in turn figured prominently in the nominations Hu made for Hunan government posts. Almost 80 percent of these nominees were Hubei natives, many of whom turned out to be very poorly qualified for their posts.^[101] In Hubei, Duan Qirui, Duan Zhigui, and Lü Diaoyuan were all Anhui men. Not unexpectedly their fellow provincials claimed a large number of Hubei posts.^[102] Because a preference was shown for supposedly better qualified late Qing officials, the removal of less experienced native officeholders was justified as an improvement in administration. According to one account, though, the effect of these changes was simply to "drive the tiger away from the front door, while letting the wolf in at the back."^[103]

Although Yuan was generally willing to accede to the recommendations of top provincial officials with respect to lower-level posts, he was not content to leave personnel selection entirely in their hands. He therefore made serious efforts to institute new central standards to improve the quality of officeholders. Among the measures introduced was the establishment of a new national examination system for magistrates.^[104] Such measures were not without results. For example, in early 1914 over thirty unqualified Hubei magistrates were removed from their posts and sent to Beijing to participate in the new examinations.^[105] When Duan Shuyun assumed office as civil governor in October 1914, he was forced to turn away many former subordinates who came to him seeking offices. Although this was partially owing to conflicts with Duan Zhigui and other officials, stricter central qualifications for officeholders also limited Duan Shuyun's ability to recommend his own favorites for office.^[106] Through such means the central government began to tighten its authority over the appointment process.

In a number of cases, Yuan was also able to exert even more direct control over key provincial posts below the governors, which helped central influence penetrate provincial government even more deeply. One example occurred in Hubei, where the central Ministry of Finance made its own candidate concurrently head of the province's National Tax Office and Department of Finance. This man assumed sole authority over Hubei tax collections and the selection of local

financial personnel. The civil governor's power in these areas was thereby reduced to a supervisory role.^[107]

The assertion of central appointment powers was not an end in itself. Rather, it was a necessary prerequisite to make the provincial governments more responsive to central administrative direction. The progress made toward this goal could be seen in increased central control over provincial fiscal affairs in both Hubei and Hunan. The underlying objective here was to revive the flow of provincial funds to the central government, which had largely halted since the 1911 Revolution. To this end, the central government began to set provincial budgets and to project large surplus revenues in them.^[108] Both Hubei and Hunan then came under relentless pressure to remit these surpluses to the central government. At one point, the annual remittances targeted for Hubei and Hunan were over nine million and four million taels respectively.^[109]

To enable the provinces to meet its demands, the central government again intervened in provincial finances from two directions. First, the government ordered, and the provinces implemented, the restoration of all Qing taxes that had been abolished after the 1911 Revolution and imposed a wide range of new taxes.^[110] Second, the provinces were forced, again often under specific Ministry of Finance orders, to cut back provincial government expenses.^[111] This demand necessitated the retrenchment of provincial reform programs. Education, a primary concern of gentry reformers, was particularly hard hit in both Hunan and Hubei by this program of fiscal austerity. In Hubei the Department of Finance seized provincial and local educational funds to finance other expenses after centrally ordered cutbacks. As a result, the Department of Education closed all county technical and middle schools and consolidated all primary schools into one per county.^[112] In Hunan, Tang Xiangming also reduced educational funds and closed schools, halting educational development plans drawn up after the 1911 Revolution. These measures, according to one Hunan educator, effectively shattered the province's educational dreams.^[113]

The constant pressure on Hubei and Hunan by the central government for more funds suggests that they never fully met its demands. Both Hunan and Hubei continued to plead special financial difficulties, and apparently the provinces had some leeway to negotiate the level of their contributions. Nonetheless, this does not obscure the essential fact that both provinces were actively forwarding a significant amount of their revenues to the central government on a regular basis.^[114]

Yuan's ability to tap into provincial finances was an important reflection of the increased subordination of provincial administration to the central government.

The attempt to make provinces more responsive to central control in some ways mirrored the two-sided approach in the military arena that combined the imposition of central-government troops with the undercutting of provincial forces. On one side, the "provincialist" bases of previous administrations were undermined by the removal of independent military governors, the dissolution of local and provincial assemblies, and the restoration of the law of avoidance. From the other side, Yuan asserted the power of central appointment over top provincial officials and increased central-government influence in selecting lower-level personnel. The new provincial administrators were more dependent on the central government, and hence, as seen in the case of provincial finances, much more responsive to its direction. Although the goals of administrative centralization were far from completely attained, there had been a definite shift in the balance of power between the center and the provinces in favor of the center.

Political Authority and Coercion under the Dictatorship

The process of administrative centralization carried out by Yuan Shikai in the provinces after the Second Revolution was accompanied by an equally dramatic change in the nature of the political authority behind provincial government. The legitimacy of the previous provincial regimes in Hunan and Hubei had been grounded in the expanded elite politics that had also provided the consensus for the overthrow of the Qing dynasty. To some degree, Yuan's victory in the Second Revolution was also facilitated by a political consensus, at least among non-revolutionary elements of the elite, that Yuan's leadership provided the nation's best hope for order and unity. There is no evidence, though, that this consensus extended to Yuan's subsequent destruction of provincial self-government and representative institutions. In eradicating these manifestations of elite political participation, Yuan showed that he had no intention of making the success of his centralizing program dependent on elite consent. Instead, under Yuan's dictatorship, provincial administrations again derived their authority bureaucratically from the central government in a manner reminiscent of the imperial period. These administrations were thus designed not to incorporate political participation but to extract political obedience.

As a consequence, Yuan's authority depended increasingly on his ability to ensure the acceptance of his programs through coercion.

Political coercion in the Chinese Republic did not, of course, begin with Yuan Shikai. Postrevolutionary elite provincial regimes regularly suppressed any threat to their rule from below. In a number of provinces, factional political struggles also resulted in heavy-handed acts of political suppression. Thus, in the last half of his term in Hubei, Li Yuanhong increasingly inclined toward authoritarian measures to deal with challenges to his rule. In general, though, government in the early Republic had a consensual authority that was not based on coercion. Ultimately, the survival of this polity depended on ensuring that consensus not force remained the foundation of political authority. This battle was lost with the imposition of Yuan's dictatorship. Once he had turned his back on the consensus-building potential of participatory politics, coercion became the primary means for Yuan to impose his political will on the provinces. The foundation for this political coercion was laid by the expansion of Yuan's military power during the Second Revolution. Once this military power was in place, Yuan ensured the acceptance of his objectives by extensive political suppression and even terror.^[115]

The ostensible purpose of political suppression carried out after the Second Revolution was to prevent any further uprisings by revolutionaries, officially identified in this period as luandang , or "disorderly parties." In both Hunan and Hubei, alarms over real or suspected revolutionary plots justified the frequent imposition of martial law, until martial law almost became a normal state.^[116] Walled cities, in particular the capitals of Wuchang and Changsha, were heavily guarded and patrolled by armed troops. Soldiers manned checkpoints at city gates and wharves to search for suspicious persons, weapons, or revolutionary literature. To guard against the movements of revolutionary organizers, stringent residency requirements were enforced, neighborhood mutual responsibility systems were activated, and hotels and schools were closely watched. Even personal letters and telegrams were regularly scrutinized for suspicious wording that might indicate revolutionary activity.^[117]

Political suppression in practice was not just aimed at the capture of revolutionaries but at the more general repression of all forms of political dissent. This broader goal was most clearly discernible in the censorship of  newspapers . On his arrival in Hunan, Tang Xiangming closed all Nationalist Party papers and forced all others to accommo-

date themselves to official views. Any offense was sufficient to cause the closing of a  newspaper  and the arrest of editors or reporters. For example, the Hunan gongbao was banned for half a year for questioning the appropriateness of one political execution and then closed for two months more for simply reporting the 1914 Chenxian mutiny.^[118] By 1916 Hunan  newspapers  had to submit to prior censorship and often appeared with blank columns where articles had been excised.^[119] In Hubei, many  newspapers  were published within the Hankou foreign concessions and so had slightly more freedom from arbitrary censorship. Even so, many  newspapers  were closed down for political offenses, either by direct government order if within the Chinese city or by negotiation with foreign authorities if within the concessions.^[120]

One important manifestation of heightened political suppression in this period was a proliferation of government spies and detectives. One special investigative bureau established by Tang Xiangming reportedly employed over four thousand detectives. Detective units were also attached to county government offices, police organizations, and military commands.^[121] In Hunan, there was even a special unit of women spies used to seek out female revolutionaries.^[122] Undercover detectives loitered at teahouses and hotels, patrolled the streets, and took passage on steamers, always on the lookout for suspicious persons or activities.^[123] Generous rewards were offered for the capture of revolutionaries or the discovery of revolutionary organs. According to one itemized account, in the last two months of 1913 alone, Tang Xiangming paid out over thirteen thousand yuan in such rewards. As a further inducement, detectives were also sometimes required to fill quotas of suspects to retain their jobs.^[124]

All these measures resulted in a steady stream of politically motivated arrests and executions. By all accounts, the number of executions carried out in Hunan was particularly high. One investigation completed after Tang's fall from power compiled the names of over sixteen thousand people who were killed during his rule. Besides these victims, countless other unidentified people were also reported to have lost their lives.^[125] The scale of executions in Hunan earned Tang the popular nickname of "Butcher Tang."^[126] Shanghai  newspaper  accounts for the period also report a very high frequency of arrests and executions in Hubei, even though no exact figures are given. The U.S. consul at Hankou noted "reports received from our many American mission stations scattered throughout the central Yangzi region that the local officials are, with but few exceptions, vigilant in apprehending suspicious characters, and numerous executions are reported."^[127]

The sheer number of executions led to a new level of brutality. One account from Hunan reported that overworked executioners virtually hacked prisoners to death with blades dulled from use.^[128] An English-language  newspaper  in Hankou made an unfavorable contrast between the brutal executions of 1914 and the orderly and cleanly administered beheadings of the late Qing era.^[129]

The detective system, with its rewards and quotas, encouraged not only arrests on very little evidence but a great number of false accusations based on guilt by association, fabricated evidence, and coerced confessions. One example occurred in Hubei's Yangxin County, where detectives attached to the local garrison sought extra merit by accusing innocent townspeople of revolutionary involvement. Broken by torture, these people not only confessed to revolutionary plotting but implicated their friends and relatives. In a short time, over three hundred people had been arrested, and many were executed. Soon the entire town was hiding behind locked doors while merchants paid out bribes to detectives to avoid false accusations. This reign of terror only halted when the local battalion commander finally intervened to demand more exact proofs of guilt.^[130] The use of torture to extract confessions was not an uncommon occurrence. The head of the Hunan Office of Military Law acquired the sobriquet "Living King of Hell" for his creative torture devices and his seeming pleasure in their application.^[131] Opportunities for corruption also existed everywhere, since detectives used searches to extort funds from a terrified citizenry. One account described Wuhan detectives as "beasts of prey" who "squandered money like dirt, and were luxuriously clothed and fed." Any opposition to their exactions brought accusations of luandang involvement, against which there was very little defense.^[132]

Without a doubt, this political suppression achieved some of its desired effect. In late 1914, the U.S. consul at Hankou wrote that "the general consensus of opinion throughout the territory is that conditions are peaceful, more so, in most cases, than at any other time since the revolution." Nonetheless, he also noted that, despite these generally peaceful conditions, executions of political suspects continued unabated.^[133] There is a suggestion here that the level of suppression far exceeded any real revolutionary activity. Yet the continued terror did serve a purpose. When the slightest expression of dissatisfaction with the government could be severely punished as luandang rumormongering, any political discussion became dangerous. The execution of a middle-school teacher in Hunan for expressing general concern over national affairs in the classroom was not an exceptional case.^[134]

A 1914 account of conditions in Hunan noted an atmosphere that chilled even trivial political conversation: "Detectives are everywhere, and the people are silent as cicadas in winter. On guard against each other, they dare not speak about current affairs. There have already been numerous cases where people have been arrested over some ambiguous idle talk."^[135] Another account of the effects of political suppression in Hunan frankly described the province as a "world of terror."^[136]

A significant feature of the political terror in this period is that even the generally privileged gentry received no immunity. In the eyes of Hunan's elite, the most shocking action taken by Tang Xiangming was his arrest of many gentry members of Tan Yankai's administration. These men had not believed themselves in danger, and had remained at their posts to smooth the transition to Tang's regime. Subsequently, several of these highly respected men were executed for little more than having carried out their normal administrative duties during Hunan's short period of independence.^[137] These executions were an object lesson that social status alone would offer no protection in the case of political offenses. Indeed, the literate and politically conscious gentry were particularly targeted by censorship and the suppression of political dissent.^[138] By all indications, previously politicized elites were generally cowed by the extension of political terror into their ranks. Political suppression, then, may have been at least superficially successful in strengthening bureaucratic administrative control by silencing potential political opposition. It did so, though, at the cost of alienating much of whatever support Yuan may have still had among the nation's elite.

Yuan's Monarchist Venture

The political risk engendered by the broad application of terror might have proved worth taking if it had resulted in centralized bureaucratic efficiency capable of fulfilling national aspirations. Instead, as Ernest Young has shown, the achievements of Yuan's dictatorship fell far short of its promises.^[139] Domestically, financial retrenchment may have restored some degree of fiscal responsibility, but only at the cost of higher taxes and the stagnation of reform programs. In foreign affairs, the centralization of administrative power had little effect on Yuan's ability to resist imperialist pressure. Indeed, in mid 1915 Yuan was forced to accept almost all the points of a Japanese ultimatum, known as the "Twenty-one Demands," for new economic and political concessions. In the midst of these troubling de-

velopments, Yuan took the controversial step of replacing his presidential chair with a monarch's throne. Yuan's decision to make himself emperor has often been explained in terms of his own personal ambition. While ambition, or perhaps hubris, was no doubt an influential factor, Young has shown how Yuan's monarchist venture also reflected his determination to pursue his original political objectives in the face of mounting problems. "Unwilling fully to face his own failures of leadership and blind to the fundamental flaws in his strategy of bureaucratic centralization, Yuan opted for monarchy as an accommodation to popular psychology and as a means of gaining public order and greater power for the central government," Young observes.^[140]

The main justification advanced for Yuan's assumption of imperial status was that it would strengthen the authority, and hence the effectiveness, of the central government. In theory, the government would draw new strength from the symbolic power the monarchy still held for China's inarticulate masses. Paradoxically, while Yuan cloaked his ascension to the throne as a response to the will of the masses, the method chosen to express this will was organized elite support. The monarchist movement was initiated in mid August 1915 with the formation of the Peace Planning Society (Chou'anhui) by six leading gentry figures. The society was purportedly organized to discuss the comparative advantages of different political systems. All provinces were called upon to send representatives to Beijing to join in this discussion. It soon became apparent, though, that the real purpose of the society was to prepare the way for the restoration of the monarchy with Yuan as emperor. Once Yuan's true intentions became known, provincial administrations threw themselves into the task of providing the expressions of support required to justify this objective.

The orchestration of monarchist plans in the provinces revealed how effectively Yuan's control was exerted on and through the provincial administrations. As soon as Yuan's desires became clear, provincial instruments of terror were turned against any opposition to Yuan's imperial goal. In Hunan,  newspapers  opposing the monarchy were suppressed, letters and telegrams were searched for criticism of the proposed change of government, and martial law was imposed to tighten precautions against any possible disorder.^[141] As a result of the political tension caused by the news of the Peace Planning Society's activities, martial law was also declared in Hubei. The targets of this precaution were not, of course, the society's supporters, but its opponents. Special police searches were ordered to find and destroy litera-

ture opposing the society on the grounds that these materials contributed to disorder.^[142] In Wuhan, casual criticism of monarchist plans was sufficient reason for arrest by omnipresent detectives. An anomalous situation thus arose: it had become subversive under the Republic, not to speak out in favor of a monarchy, but to talk about preserving the Republic.^[143]

Complementing the suppression of opposition to the monarchy in the provinces was the organization of "popular will" in its support. Documents published after Yuan's fall proved the extent to which the provincial response, including wires of support from provincial officials in the name of influential gentry and merchants, was coordinated in detail by Yuan's supporters in Beijing.^[144] An account of the engineering of the monarchist movement in Hunan is given in the memoirs of a member of Tang Xiangming's staff, Tu Zhuju. Tu and some forty other specially selected bureaucrats were ensconced in a well-provisioned office behind the military governor's yamen as a secret center for the organization of Hunan's pro-monarchy movement. Generous funding (estimated by Tu at a million yuan) was provided by the Peace Planning Society in Beijing, and no expense was spared. The office received exact instructions from Beijing on the measures to be taken to show a crescendo of support for Yuan's ascension to the throne, including the fabrication of "popular will." This office, then, not only organized but actually became the main source of monarchist support in Hunan.^[145]

The culmination of the monarchist movement in the provinces was the calling of "citizens' conventions" for the ostensible purpose of making recommendations on the political system most suited to China's needs. In fact, the conventions, held in Hunan on October 28 and in Hubei on November 1, were ceremonial functions, where assembled gentry electors were simply asked to deny or affirm their preference for the monarchical system. Electors marked their ballots in the presence of "supervisors" and provincial officials, leaving little doubt as to the outcome. In Hunan and Hubei, as in other provinces, the conventions voted unanimously for the establishment of a monarchy, and with equal unanimity petitioned Yuan to ascend the throne as emperor.^[146]

The carefully constructed outpouring of elite approval for the monarchy was a hollow exercise. Almost all independent accounts of the monarchist movement in Hunan and Hubei reveal a general lack of enthusiasm that, if not leading to open opposition, expressed itself in widespread apathy. Various reports noted that public opinion in Wuhan retained its republican sympathies and was generally cold to

the idea of a revival of the monarchy. The Peace Planning Society failed to gain a real foothold in Hubei, and the only active support for the monarchy reportedly came from officials. In both Hunan and Hubei, Chambers of Commerce and other merchant organizations were viewed as being pressured against their will to participate in Peace Planning Society discussions.^[147] The U.S. consul in Changsha noted widespread opposition or indifference to the change in government reflected in the reluctance of many qualified voters to participate in the citizens' conventions. A missionary in Xiangtan noted that the city "did its duty in supporting the change of Government, but as one Chinese said, 'That is what they were instructed from Peking [Beijing] to do, so what else could be done?' As another said, 'There is none in China who want Yuan as Emperor.'" Reports cited from other cities expressed a similar lack of enthusiasm for the monarchist cause.^[148]

Naturally, when Yuan ascended the throne, suppression of the monarchy's opponents intensified. A Western missionary in Hubei in early 1916 reported:

At present a lot of extra spies have been put on to arrest any person who says a word against the present change of government. A good many have been arrested for saying very little, now everybody has got frightened [sic ] so no one dares to say anything, as spies seem to be everywhere, and even those who are not spies are paid for any information given, so many seek to earn a little extra in this way.^[149]

Eventually, to forestall even the slightest potential for political criticism, a ban was issued in both Hunan and Hubei on any discussion of national affairs (guoshi ).^[150] Thus, to ensure expression only of support for his ascension to the throne, Yuan attempted to reverse the whole process of participatory nationalism that had politicized China's elite since the late Qing period. In the end, political participation in the monarchist movement, with the exception of a few sincere proponents, was less a reflection of genuine support than a tribute to the bureaucratic efficiency and coercive powers of local and provincial administration under Yuan's control.

Muted beneath the fanfare of officially engineered popular support for the monarchical system, though, there remained a more substantial elite concern for the continued preservation of order. This concern appeared as a recurring theme in the political turmoil of the early Republic. Elite assessment of the best means to maintain order had played an important role in the consensus behind the 1911 Revolution, in the compromise that made Yuan president, and in the failure of the

Second Revolution. A belief in Yuan's ability to keep order appears to have provided some support for his rule even as disillusionment over his policies increased. In late 1914, for example, the U.S. consul at Hankou recorded a consensus in missionary reports from throughout the central Yangzi area that "in many quarters a general dislike and distrust for Yuan Shih Kai [sic ] is noted, although everywhere there seems to be a great respect for his ability to maintain peace and order."^[151]

Ultimately the preservation of order, not the pros and cons of the monarchical system, would remain the overriding concern of a significant section of the population. According to one report, "the people of Wuhan have really not had any kind of reaction to the issue of the nation's political system. However, in most hearts there is a hope for peace."^[152] Even strong anti-monarchical feelings were often reported as being tempered by a broad popular desire for peace and order.^[153] For some, this desire would outweigh their opposition to Yuan's imperial plans. For example, the opposition of many Hankou merchants to an armed struggle against the monarchist movement was interpreted as revealing the priority they placed on peace over their republican preferences.^[154] For others, however, Yuan bore the onus of provoking disorder through his bid for the throne. Early in the monarchist movement, one courageous member of the Hubei elite attempted to organize a group of Wuhan gentry and merchants to protest pro-monarchy telegrams issued by the Hubei government in their names. Appealing to Yuan to ban the Peace Planning Society, this group placed the blame for creating instability and inciting revolutionary opposition directly at the feet of the monarchist movement.^[155] Once armed resistance to the monarchy finally broke out, many more saw the key to the preservation of order not in the continuation of Yuan's rule but in his removal.

In the end, Yuan's attempted restoration of the monarchy had an effect opposite to its intent. The flawed assumption behind the monarchist movement was that the imperial symbols of the past, combined eclectically with the formalities of elite elections, could provide a new and stronger basis for central authority. Unfortunately for Yuan, the traditional appeal of the imperial throne had already lost much of its potency, and sham elections alienated rather than consolidated elite support. Indeed, by provoking a new round of political instability, Yuan's attempt to make himself emperor undermined the foundation of order that had provided the strongest reason for elite acquiescence to his rule. The result of the monarchist venture, then, was to weaken,

not strengthen, both Yuan's personal authority and the legitimacy of the central regime he had attempted to create.

Conclusion

As evidenced in the cases of Hubei and Hunan, the original direction taken by Yuan Shikai in the establishment of his dictatorship did not necessarily suggest an advance into warlordism. A concerted effort at bureaucratic centralization reversed the tendencies toward political fragmentation that had been represented by postrevolutionary provincial autonomy. Central powers of appointment over provincial governments were no longer simply claimed but enforced, and provincial administrations as a whole were made more responsive to central control. Measured against the previous provincial regimes, Yuan made considerable progress toward a restoration of central power in the provinces. His promotion of civilian primacy in provincial administration was somewhat less successful. In both Hunan and Hubei, the military governor remained the dominant figure in provincial government. Nonetheless, civil governors were appointed and the goal of civilian rule upheld. If given sufficient time, increased central bureaucratic control in the provinces might have enabled the civil governors to win their contest with military governors for the control of provincial administration. After all, at this point the military powers of the military governors in Hunan and Hubei were constrained by the comparatively small number of troops under their personal command and by the primary loyalty of other key military commanders garrisoned in their provinces to Yuan and the central government. Clearly the policies and political structure of Yuan's centralizing dictatorship showed a sensitivity to the danger of militarism and were designed to avoid it.

Despite all his efforts to reinforce centralized and civilian government, in one crucial area Yuan fostered a countervailing tendency toward military rule. This was in his reliance on military power and coercion to enforce his political authority. Yuan could not have been unaware of the lesson traditionally drawn from Chinese dynastic history about the intrinsic instability of authority based solely on force. A general could conquer from horseback, but he had to dismount to rule. Unfortunately for Yuan's political aspirations, he was never able to dismount. Military power enabled Yuan to establish his dictatorship, but to maintain it he found no alternative to coercion.

Yuan's ascension of the imperial throne was a belated recognition of the need to shore up the foundations of his authority by some

means other than military force. Unfortunately for Yuan, imperial symbols had lost most of their efficacy. Ironically, the monarchist movement reflected both the achievements and the weaknesses of Yuan's rule. The initially successful orchestration of the monarchist movement in the provinces clearly reflected the consolidation of central administrative control under Yuan's dictatorship. At the same time, the means by which the movement was pursued revealed the extent to which Yuan's central authority continued to rely almost solely on bureaucratic obedience enforced by coercion. Whatever political consensus supported Yuan at the beginning of his rule had evaporated by the time he attempted to make himself emperor. Instead of building new bases of support for his authority, Yuan's betrayal of the Republic created a rallying point for his opponents and initiated a civil war that not only brought an end to his imperial dreams but led to reversal of the centralizing gains of his dictatorship.