I. SEMI-SOVEREIGNTY: EXCEPTIONS, ABERRATIONS, AND VIOLATIONS?

Semi-sovereignty occupies an uneasy place within international relations. Although there existed (and continue to exist) a bewildering range of semi-sovereign international entities which are not fully sovereign but which are not entirely lacking in sovereignty, they have often been dismissed as little more than aberrations, minor exceptions to the spread of sovereign states as the default type of international polity.Footnote ² Moreover, the diversity and the fleeting existence of many semi-sovereigns make their systematic study difficult. Semi-sovereignty included a wide variety of international relationships and entities, including dominions (Australia), protectorates (Egypt), vassals (Tibet), neutralized states (Switzerland), certain federal units, and mandatory territories and trusteeships, each with its own peculiarities. Such categories were, moreover, constantly in flux: for instance, dominions emerged in the middle of the nineteenth century, acquired unquestioned international personhood after World War I, formal equality with Britain in 1931 through the Statute of Westminster, and disappeared after World War II, all within the space of a century.

However, far from being a purely antiquarian exercise, the study of forms of semi-sovereignty such as suzerainty has the potential to yield important insights for our understanding of the historical development of the international order and of international law. At the most basic level, semi-sovereignty comprises some of the most explicit manifestations of hierarchy within International Relations, a subject which, after a long period of neglect, has returned to the forefront of the discipline in recent years.Footnote ³ Hence, the rise and fall of semi-sovereignty speaks directly to the changing place of hierarchy within the international order. Moreover, the persistence, and indeed growth, of semi-sovereignty (albeit in ever-evolving forms) during the nineteenth and early twentieth centuries challenges the traditional English School of International Relations’ narrative surrounding the expansion of international society, according to which a “ready-made” grundnorm of Westphalian sovereignty, developed and perfected in Europe, was exported throughout the non-Western world, rapidly becoming the default mode of political authority to the exclusion of more “traditional” forms of authority.Footnote ⁴ Indeed, although semi-sovereignty tended to be superseded by sovereign nation-states within Europe during the period, European expansion into the non-Western world was in large part conducted through the extensive use of semi-sovereignty, relationships which were sometimes later converted into direct colonial (and therefore fully sovereign) rule, but which often continued in their semi-sovereign form until decolonization.

But if semi-sovereignty is treated (unfairly) as an oddity within the Western scholarly literature, most Chinese academics reject even the possibility of semi-sovereignty. Modern Chinese literature on sovereignty emphasizes the indivisibility of sovereignty, which is assumed to be one of its fundamental attributes; semi-sovereignty is rarely discussed, but is sometimes described as a sort of colonialist plot.Footnote ⁵ The Chinese emphasis on the indivisibility of sovereignty is unsurprising: modern Chinese historiography adheres to the narrative arc of the “Century of Humiliation”, during which China first lost a large part of its sovereignty during the Qing dynasty to foreign powers—both through the loss of territory and the loss of certain types of authority within its remaining territory—before a gradual process of recovery of the lost sovereignty, resulting in the abrogation of the so-called “Unequal Treaties” and the restoration of China to full sovereignty.Footnote ⁶ Given the centrality of this narrative to China's national self-conception, it is unsurprising that the divisibility of sovereignty, invoking as it does the central trauma of modern Chinese history, should prove to be an unpalatable idea. The PRC's continuing championing of the indivisible, maximalist understanding of sovereignty internationally, leading some to describe it as the “vicar of the high church of Westphalia”,Footnote ⁷ can also be understood as the result of this particular set of historical experiences, which have served to elevate sovereignty to a quasi-mystical status in contemporary Chinese political thought. Finally, compounding the difficulty is the fact that semi-sovereignty has modern political and legal implications for China, which is involved in several territorial disputes where questions of historical title feature prominently. China has consistently maintained, for instance, that it has possessed full sovereignty over Tibet since antiquity, even though the claim is arguably anachronistic, involving, as it does, a notion which was not introduced to China until the nineteenth century. To acknowledge that China's title over Tibet was at any point anything less than full sovereignty would therefore pose serious problems for China's position in that regard.

This paper will examine the struggle over the legal status of Tibet and Mongolia between China, on the one hand, and Britain and Russia on the other, during the Republican era (1911–49). In short, the dispute was whether they were under the suzerainty of China—a type of semi-sovereignty and the term favoured by the Western powers—or under its sovereignty. Modern Chinese historiography has categorically repudiated Chinese suzerainty in those territories, describing it as “the product of imperialist aggression”,Footnote ⁸ a notion “cooked up by old and new imperialists out of their crave to wrest Tibet [and Mongolia] from China”,Footnote ⁹ and “the sword [used by] Western powers to try to separate Tibet and Mongolia”,Footnote ¹⁰ a view which has also found favour among some Western scholars.Footnote ¹¹

However, this paper will argue that suzerainty, although initially imposed upon China by Western powers for reasons of political expediency, was in fact not only accepted, but also instrumentalized by Republican China.Footnote ¹² By doing this, it not only ensured the continuation of nominal Chinese title over these territories, as well as preventing their de jure independence, but it also formed the legal basis for the re-absorption of both territories later. China did so by negotiating the parameters of suzerainty with Western powers, and by using suzerainty as a tool to exclude foreign intervention in these regions. Thus, at a time when China was unable to exercise any form of effective territory over its frontier regions, it was able to secure its future sovereignty over Tibet and Mongolia by accepting legal limitations over its sovereignty over these regions, which was paradoxically more effective at maintaining Chinese sovereignty in the long run. In other words, though it was outwardly an outside imposition, semi-sovereignty in the form of suzerainty was in fact indispensable to China's legal construction of its territory, and to the maintenance of Chinese sovereignty over its borderlands. More broadly, this paper will argue that suzerainty, despite the best efforts of international lawyers to define it, in fact had no fixed meaning, but was a legal empty vessel, which could be used to denote a range of hierarchical relationships that could range from de facto independence for the vassal to de facto political hegemony for the suzerain. That empty vessel, in turn, was often deployed to script the legal identities of non-Western entities, but could lead to unexpected results, as it did in both Tibet and Mongolia, in ways which illuminate the relationship between international law and imperialism during the long nineteenth century.

II. SUZERAINTY, ANCIENT AND MODERN: FROM FEUDAL TO INTERNATIONAL LAW

Suzerainty has its origins in French feudal law, where it denoted “the relation between the feudal lord and his vassal”.Footnote ¹³ In a suzerain relationship, the superior lord, or suzerain, received the homage and certain feudal services of the inferior lord, or vassal, who in turn was afforded the protection of the suzerain lord.Footnote ¹⁴ The exact nature of such services varied considerably in practice, and often depended on local customs or specific legal documents creating a suzerain relationship. Though the etymological evidence is imperfect, it seems that suzerainty as a word post-dates the emergence of the feudal system and was not in widespread use until the fifteenth century, mostly in modern-day France.Footnote ¹⁵

By the beginning of the nineteenth century, suzerainty was far down the path of obsolescence. Although feudalism had effectively ended in much of Europe by the sixteenth century, many of its institutional features remained in being, including numerous suzerain relationships. The upheavals brought by the French Revolution and the Napoleonic Wars led to a culling of suzerain relationships, as they were supplanted by sovereign ones. For instance, in the 1814 Treaty of Paris, France was made to renounce all her “suzerainty …” beyond her borders as set by the Treaty.Footnote ¹⁶ At the 1815 Congress of Vienna, Austrian suzerainty over Lusatia and the Kingdom of the Two Sicilies’ suzerainty over Elba and Malta were all converted into sovereign titles. Prussia and Saxony renounced all their feudal rights beyond their newly fixed borders in each other's territories. Even where suzerain rights were explicitly preserved (those of Prussia over Wernigerode and the Two Sicilies over Piombino), they were explicitly coupled with full sovereignty, which superseded any lesser feudal title.Footnote ¹⁷ Prussia continued to claim a vague suzerainty over Neuchâtel until 1857, but when it abandoned this increasingly unreal claim via treaty, it did not mention suzerainty, only “sovereign rights”.Footnote ¹⁸ Thus, after 1815 suzerainty essentially became extinct in Western Europe as a type of legal relationship.

Yet just as what could be described as “traditional” suzerainty, anchored in feudal law, was disappearing from Western Europe, the concept found a new lease of life, first in what is now Eastern Europe but was then the western frontier of the decaying Ottoman Empire. In 1827, Britain, France, and Russia agreed to give to the Ottoman Porte suzerainty over Greece: the Greeks were to pay tribute to the Porte, but were to acquire internal autonomy as well.Footnote ¹⁹ To this the Porte agreed in the 1829 Treaty of Adrianople, which extended Ottoman suzerainty to Moldova and Wallachia.Footnote ²⁰ Thereafter there was a steady increase in the number of suzerainties around the world. Germany and Belgium concluded many treaties of suzerainty with local chiefs in Africa, while Britain used the term to define its relations with the South African Republic and the Indian princely states. In 1907 Britain and Russia acknowledged Chinese suzerainty over Tibet, and in 1915 Russia and China recognized the latter's suzerainty over Outer Mongolia. All in all, it was a remarkable and unexpected revival for a concept whose use was rare even in the Middle Ages.Footnote ²¹

But what exactly did suzerainty mean in the post-feudal, extra-European context? This point puzzled publicists and statesmen alike, and it was never to be entirely resolved. As Lassa Oppenheim put it, with considerable understatement, “the relations between a suzerain and its vassal state create much difficulty in the science of the Law of Nations”.Footnote ²² Robert Treat Crane illustrated the situation more dramatically in 1907, when he compared no fewer than eleven definitions given by leading legal scholars, all of which were contradictory to some extent. A vassal state was alternately described as a non-sovereign, a semi-sovereign, an unacknowledged semi-sovereign, a semi-sovereign on the way to becoming fully sovereign, a synonym to a protectorate, or of a status completely unrelated to sovereignty.Footnote ²³ Befitting the variegated definitions of suzerainty, vassals occupied a wide range of positions within international society. At the one extreme, entities such as Bulgaria and Egypt, in addition to internal autonomy, exercised almost all of the powers associated with external sovereignty, such as the despatch of diplomatic representatives and the conclusion of international treaties.Footnote ²⁴ At the other end of the spectrum, the Indian Princely states had no separate international legal personhood at all.Footnote ²⁵ Suzerains also differed widely in the extent of their power over the internal affairs of their vassals: in British India the suzerain could depose vassal princes at will, whereas in Mongolia the right was explicitly forbidden to China.

This wide discrepancy in vassal states’ international status was justified by publicists’ insistence that each suzerain relationship was sui generis. The way to determine the contours of each suzerain relationship, according to them, was by reference to the specific treaty that created it. The treaty specified the rights and duties of each party, and on that basis could the international status of a vassal be determined.Footnote ²⁶ Traditional incidents of medieval suzerainty, such as tributes and protection, were not essential to the new type of suzerainty. Thus, suzerainty was in practice a sort of legal empty vessel, which would be used to denote a range of hierarchical relationships that could range from de facto independence for the vassal to de facto political hegemony for the suzerain. That empty vessel, in turn, was often deployed to script the legal identities of non-Western entities. In fact, suzerain relationships were just as often the result of external imposition as they were the result of a willing claim to authority by the suzerain power. Such scripting inevitably had the effect of weakening the political control over the vassal that the suzerain previously had, causing some writers to describe suzerainty, in practice, as a kind of halfway house for territories on their way to independence.Footnote ²⁷ For instance, Ottoman suzerainty over Greece and the Danubian Principalities was the result of the Porte's defeat at the hands of Russia in the 1828–29 Russo-Turkish War: the concept had not existed in the Ottoman Empire before then.Footnote ²⁸ The imposition of nominal suzerainty was used to deprive the Sultan of any real control over the territories in question, without, however, precipitating the immediate independence of these territories. This was well understood at the time: indeed, Lord Ellenborough, the British Lord Chancellor, described suzerainty as “anything from the palfrey of the King of Naples up to real dominion”,Footnote ²⁹ and said that Ottoman suzerainty over the Principalities was deliberately conceived of by the Western powers which had proposed the arrangement as a way to strip it of any influence over these territories.Footnote ³⁰ A few writers, like the Scottish lawyer Sir John Macdonell, even went so far as to state that “[suzerainty's] very indefiniteness [is] its recommendation”.Footnote ³¹

Most, however, were unwilling to go so far: nineteenth-century international lawyers thought of their discipline in scientific terms—the “science of international law” being the title of a famous lecture by Lassa Oppenheim—and necessary to this project was the aspiration of divorcing international law from politics, a project shared, but never fulfilled, by international lawyers to the present day.Footnote ³² To acknowledge the fundamentally political nature of what purported to be a legal concept, or indeed the many ways in which international law was inextricably embedded to the advance of imperialism, would have been problematic for the discipline's legitimacy.Footnote ³³ Illustrative of this approach was the work of the British publicist Thomas Baty: in the midst of the Second Boer War, when the meaning of suzerainty had become a matter of public controversy, Baty assiduously but fruitlessly surveyed legal authority on suzerainty all the way to the mid-eighteenth century. Coming to no very clear conclusion, for the various authorities were largely contradictory, he nevertheless concluded that:

According to Baty, suzerainty had to have a true meaning because reality had to match theory, for otherwise Johann Jakob Moser (who was writing under very different circumstances in 1777) would be proven wrong. And suzerainty had to have a concrete meaning because otherwise the nineteenth century's new suzerain relations would be meaningless. This neatly encapsulated the circularity of the efforts of publicists at fixing the meaning of suzerainty: because the concept existed it must have had a meaning, even if no one could ever identify it. Hence, the cottage industry of public international law writing on suzerainty continued unabated.

This approach presented several advantages from the Western point of view. It allowed for the creation of de facto independent states on their peripheries, without, however, giving the latter full international legal personality. In this manner, the Family of Nations would not be unduly burdened with new non-Western entrants who could claim the full rights attached to international sovereignty. But treaties of suzerainty usually allowed just enough external autonomy to the vassal for it to conclude commercial treaties and the like, enabling Western countries to obtain commercial and strategic advantages in these areas. Finally, as for the suzerain power, the preservation of a nominal title through suzerainty made their loss of power more digestible, as they could still claim nominal authority over the territory in question. Indeed, the short-lived British suzerainty over the South African Republic [ZAR] between 1881 and 1884—a rare example of a claim to suzerainty that was made willingly by the suzerain power—was established by the British government precisely to allow it to maintain the illusion of political control over the ZAR, whereas such control proved in fact to be essentially non-existent.Footnote ³⁵

Conceptually, it might be helpful to think of suzerainty as a close relative of the protectorate, which has been the subject of much more modern academic analysis. Just as the protectorate was a legal technique which enabled “European states [to exercise] extensive control over non-European states while not officially assuming sovereignty over those states”,Footnote ³⁶ suzerainty can be thought of as a legal technique designed to achieve similar benefits, but without requiring European states to assume any responsibility, nominal or otherwise, over the targeted territory—effectively a sort of extreme indirect rule. The reluctance to establish a protectorate, or indeed to directly annex the territory concerned, could have many reasons. In the case of the Danuban Principalities it was to avoid disturbing the European balance of power; in the case of Tibet the British had determined it to be too costly.Footnote ³⁷ But whatever the reason, the device of suzerainty enabled a similar result to be achieved, without the costs associated with direct or indirect rule.

In sum, although widely used in nineteenth-century international law, suzerainty was not so much a well-defined legal concept as an imperial legal technique widely employed to script the political and legal identities of territories on the periphery of non-Western imperial state systems. The result was a series of entities which existed in a sort of legal twilight zone: part sovereign, part dependency, in a state which was beyond explanation within the existing frameworks of international law. The next section will explore the impact of the conception on China, and its intersection with China's attempts at defining the extent and nature of its authority over its frontier territories.

III. THE (MIS)USES OF SUZERAINTY IN CHINA

Suzerainty is often used to describe China's historical relationships within countries and territories within the Chinese world order, both in the scholarly literature and elsewhere.Footnote ³⁸ Indeed, suzerainty is sometimes used synonymously with the Chinese tributary system. However, this approach, while convenient, trades nuance for clarity. As Feng Zhang puts it, such a usage “brings with it historical baggage associated with the term … that might contravene historical realities in East Asia”.Footnote ³⁹ Indeed, the notion of suzerainty had no ready equivalent within either the Chinese language or system of international relations, and there was no vocabulary for it in Chinese until the late nineteenth century. In fact, W.A.P. Martin's seminal translation of Henry Wheaton's treatise on international law, which introduced international law to East Asia for the first time, did not mention suzerainty at all.Footnote ⁴⁰ The first Chinese word for suzerainty was shangbangzhiquan [上邦之权], meaning roughly “the rights of the upper [country]”, which was used in the 1894 Anglo-Chinese Convention on Burma.Footnote ⁴¹ By 1902, Japanese legal textbooks were introducing a new series of translations of Western legal vocabulary to China, and shangbangzhiquan was progressively superseded by zongzhuquan [宗主权], which remains in use today.Footnote ⁴² Zhuquan was the Chinese word for sovereignty, and zongzhuquan, taken as a whole, meant something like “ancestral sovereignty”. This linguistic shift from shangbangzhiquan to the sovereignty-based zongzhuquan is suggestive of the extent to which the notion of sovereignty had by then been understood in China as the default Western mode of political authority vis-à-vis which all other forms of authority are defined.

What is generally now described as suzerainty in the Chinese context is more usually known as the Chinese tributary system, a clumsy form of words which belies the fact that the concept was of post-World War II Western coinage: there was no Chinese name for the system, and indeed it was not conceived of as a system by its practitioners. An “open access [and] inherently elastic” system, China did not generally directly administer internal affairs of its tributaries, nor did it constrain their external relations, although it reserved the right to intervene militarily in them.Footnote ⁴³ Under the prevailing Western legal view, however, China's level of authority over those regions fitted uneasily with Western understandings of sovereignty and of authority. Legally, sovereignty was inextricably tied to the notion of effective territorial control, one of the determining factors in determining whether a state had a sovereign title over a given territory. This arose from the doctrine whereby sovereign states had not only international rights, but also international obligations. A sovereign who exercised little control over a territory could not, the logic went, fulfil his international duties within that territory, notably that of protecting the rights of other states, and could therefore not claim a complete sovereign title. As Max Huber famously put it in the seminal Island of Palmas case:

Hence, as Western contacts with China increased, and as it became necessary to define the boundaries between China and Asian territories under Western influence or control, suzerainty became an increasingly common way in the West to denote China's relationships with territories within its periphery during the second part of the nineteenth century. Moreover, the focus on effective territorial control had important political advantages, as many non-Western forms of political authority failed the test of effective territorial control and could be described by Western powers as non-sovereign.

The case of Tibet is illustrative in that regard. Early official British travellers to Tibet referred to China's authority there as one of sovereignty. From the 1880s onwards, however, official British documents began to define China's historical authority there in terms of suzerainty, without there having been any alteration in the material circumstances of China's presence there.Footnote ⁴⁵ The British did recognize that there was a degree of incommensurability between the Chinese and Western understandings of political authority in these parts. For instance, Sir Charles Bell, the British political officer for Tibet, remarked that “it might be argued that … [Tibet] was undoubtedly under the suzerainty of China. But Asia does not think along European lines. The Tibetan Government maintains that the Dalai Lama is the spiritual guide and the Chinese Emperor his lay supporter.”Footnote ⁴⁶ Nevertheless, having decided against ruling Tibet directly or to establish a protectorate over it, the British considered that an autonomous Tibet under nominal Chinese control was the best way of securing India's northern flank. Hence, it began to describe China's authority in Tibet in terms of suzerainty, and the position was maintained until 2008.Footnote ⁴⁷ Such a description was not the result of dispassionate legal inquiry of the actual relationship between the two, but rather the result of political expediency. The claim that the relationship was one of suzerainty by the British, in other words, came after it was decided that Tibet should be autonomous from China for strategic reasons, suzerainty being merely the legal instrument chosen to implement that objective.

The differences between the European and Chinese approaches is neatly illustrated by an exchange between French prime minister Jules Ferry and Chinese diplomat Zeng Jize over China's legal position in Vietnam during the Sino-French War, the French having insisted that China merely possessed nominal suzerain rights in Vietnam at some point in the past, given that it never interfered with Vietnamese internal affairs:

Zeng's sharp remark notwithstanding, France won the war and made China renounce all rights in Vietnam. In relation to this episode, Lloyd Eastman notes that, although the idea “that the Chinese had not comprehended in the mid-1870's that the tributary relationship had no status in Western international law” seems incredible, in Chinese officials’ understanding at the time, a state could be both a tributary of China and independent. Indeed that was the answer Chinese officials gave to puzzled Western diplomats who queried Korea's international status.Footnote ⁴⁹ But to be independent and dependent at once, although perfectly compatible in the Chinese worldview, was hardly one that could be slotted into modern Western legal understandings of the time. Nevertheless, Chinese diplomats began to realize the importance of the precise legal vocabulary that was used to describe China's position in its frontier regions. During the Sino-French War, Zeng used souveraineté and suzerainté interchangeably in his communications with the French government.Footnote ⁵⁰ But later Chinese diplomats became much more careful in their choice of words, carefully rejecting, in their conversations with Western interlocutors, suggestions that China's authority over Tibet and Mongolia was suzerain rather than sovereign. As the diplomat Tang Shaoyi wrote to his superiors in 1905:

It was within this context that, in 1911, the Xinhai Revolution precipitated a reckoning on China's periphery. Taking advantage of the chaos in China, Mongolia and Tibet both declared their independence towards the end of 1911, expelled the Chinese ambans (residents) and garrisons, and established independent governments. Both Mongolia and Tibet justified their independence on the grounds that the territories’ attachment to China was a personal one between the Manchu emperors and the territory, and that the territories were not part of China proper. With the dissolution of the Qing dynasty, so did the personal union between them and China also dissolve.Footnote ⁵² Both governments sent out requests for recognition of their statehood to the Great Powers, requests that were not acceded to. Faute de mieux, Mongolia and Tibet concluded a mutual treaty in early 1913 in which they acknowledged each other's independence.Footnote ⁵³

Meanwhile, the new Chinese government took the opposite position: not only did it contend that both Mongolia and Tibet continued to be parts of China, it also proclaimed that they were integral parts of the national territory that should be politically undifferentiated from the rest of the country. The 1912 provisional constitution of China declared that “[t]he territory of the Chinese Republic consists of twenty-two provinces, Inner and Outer Mongolia, Tibet and Qinghai”,Footnote ⁵⁴ laying the legal framework for China's territorial claims (as well as indicating, through their absence, which of its traditional territorial claims it had forfeited for the time being). It was no longer considered adequate to declare tautologically that “the Great Qing Emperor rules over the Great Qing Empire”, as the 1908 draft Qing constitution did.Footnote ⁵⁵ And to underline the new status of Tibet and Mongolia, Yuan Shikai, the new Chinese president, decreed that they would henceforth be treated in the same manner as the 内地 (neidi, interior) provinces, and expressions such as lifan (vassal) and zhimin (colonize) would be dropped from official documents.Footnote ⁵⁶ All this was part of a new vision of the Chinese nation—the idea of “Five Races Under One Union”, under which Mongolians and Tibetans were both integral to the Chinese nation.Footnote ⁵⁷

Yet the shift from the traditional form of Chinese authority to a national one on its periphery was more theoretical than real. The Chinese government had limited resources to reassert its authority over Tibet and Mongolia, or to establish a new system of civil administration. Hence, the Peking government simultaneously reaffirmed the traditional bounds between the Qing throne and the outlying territories. Yuan's administration issued decrees reconfirming all the titles and privileges granted by Chinese emperors to high-ranking Mongols and Tibetans, bestowed higher ones on loyal aristocrats, and generally positioned himself and the Republic as the direct successors of the Qing,Footnote ⁵⁸ in a manner that implied the seamless continuation of the traditional tributary system.

There was a fundamental paradox in Republican China's initial claims of legal authority over the whole territory of the former Qing Empire. It claimed a modern, national form of authority over the entirety of the Qing territory, yet what bound some of these territories to China was based on an entirely different logic of governance. It was, moreover, a logic which ill-fitted the strictures of international law to which China, in its claim to become a normal member of the Family of Nations, was bound to follow. Very soon, this contradiction would come to the forefront, as China's neighbouring countries challenged its claims over its boundary regions.

IV. THE NEGOTIATION OF CHINESE SUZERAINTY: MONGOLIA, 1911–15

By the time of its unilateral declaration of independence in 1911, Mongolia had been politically bound to China since the seventeenth century. It was governed separately from China proper, but Qing officials governed Mongolia directly, unlike the situation in more distant territories. For much of modern history the nature of China's authority over Mongolia was not seriously questioned in the West.Footnote ⁵⁹ Russia, the other power which, by virtue of its geography, had a special interest in the region, recognized China's authority over Mongolia in 1727–28, and reaffirmed it in 1881, in return for preferential commercial treatment for Russian merchants.Footnote ⁶⁰ China's authority over Mongolia was not seriously challenged by the Russians for the next century: a standard nineteenth-century account of the region by a Russian diplomat, for instance, described Mongolia as “composed of several principalities, which recognize the sovereignty of the emperor of China”.Footnote ⁶¹ All the while, however, Russian strategists hoped that Mongolia (and Manchuria) would become independent if the Qing Empire were to collapse, and in 1907 secured from Japan recognition of a Russian sphere of influence in Outer Mongolia in a secret treaty.Footnote ⁶²

Only four years after Russia obtained the guarantee of a free hand from Japan, on 1 December 1911 Outer Mongolia proclaimed its independence from China. On 29 December the eighth Jetsun DampaFootnote ⁶³ was enthroned as the Bogd Khan, or Emperor, a title that had previously been assumed by successive Qing emperors.Footnote ⁶⁴ The Chinese amban and his troops were disarmed and expelled in short order afterwards. As geography dictated, the new Mongolian government sought Russian support for its independence. However, the Russian government was unwilling to countenance Mongolian independence, but wanted to keep Mongolia autonomous as a buffer to China. It therefore set on the formula of “Chinese suzerainty and Mongolian autonomy”, the same formula Britain had decided upon in Tibet. Taking advantage of China's weak position, on 5 November 1913 China was induced to acknowledge the formula, with the full details of its suzerainty to be fleshed out at a later date in tripartite talks with Mongolia and Russia.Footnote ⁶⁵ The formula satisfied neither the Mongolians, bent on independence, nor the Chinese, who sought sovereignty (although they extracted from the Russians recognition that Mongolia was part of China's territory). Hence, when the tripartite talks began at Kyakhta in September 1914, both set out to put their own spin on what Chinese suzerainty would mean in practice: the Mongolians in order to make it nominal, the Chinese to gut it of any meaning.

Negotiations began in earnest on 23 September, when the Chinese representative introduced a proposed outline for the draft tripartite treaty, which contained four demands: (1) Mongolian recognition of the Sino-Russian Declaration of 1913; (2) the rescinding of Mongolian independence; (3) the abandonment of the title of emperor by the Jetsun Dampa; and (4) the abandonment of the new Mongolian regnal calendar in favour of the Chinese republican calendar.Footnote ⁶⁶ He insisted no other issue could be negotiated until Mongolia agreed to the four demands, a line which he held for most of the conference.Footnote ⁶⁷ In reply, the Mongolian delegate categorically refused to abandon either the Jetsun Dampa's imperial title or the new Mongolian calendar. His government was willing, he said, to accept at the outermost the requirement that it should not make treaties or alliances that would harm China or its territorial integrity, but he otherwise maintained that “no [other] suzerain obligation can be recognized”.Footnote ⁶⁸ For the most part, however, he claimed (perhaps genuinely, given Mongolia's relative isolation) not to understand what suzerainty and autonomy meant, pushing China to produce increasingly detailed outlines of the rights it was claiming.

In the ensuing negotiations, the Chinese side proceeded on two very different tracks. On the one hand, it effectively wielded modern concepts of international law to its advantage. It pointed out that Mongolia's independence “had no reality in international affairs”, given that no other country had recognized it.Footnote ⁶⁹ The Chinese side also put much insistence on the preservation of China's territorial integrity, a relatively new concept at the time (it was not to become a mainstay of international relations until the establishment of the League of Nations).Footnote ⁷⁰ At the same time, however, the Chinese negotiators also framed their arguments in terms of Chinese cultural exceptionalism. Thus, when the Russian delegate cited the precedents of Egypt and of Morocco for the proposition that royal titles can subsist under a suzerain, the Chinese delegate brushed them aside by claiming that “under Oriental customs”, suzerainty could not countenance Jetsun Dampa holding the title of Bogd Khan under the Grand President of China, as Bogd Khan was previously used in Mongolia to refer to Qing emperors, and “there cannot be two masters within one territory”.Footnote ⁷¹ The same went for the issue of the calendar: according to the Chinese delegate, “among those places which exist in Chinese culture, if there is a country it necessarily has [its own] regnal year … an autonomous region cannot have a regnal year”.Footnote ⁷²

Moreover, Chinese negotiators were bent on perpetuating the traditional forms of authority formerly exercised by the Qing dynasty in Mongolia. In contrast to the creation of properly republican ceremonials within “China proper” which was occurring at the same time, the Chinese negotiators insisted on a clause that all traditional ceremonies between Mongolia and China continue as before.Footnote ⁷³ All Mongolian titles, including that of Jetsun Dampa, were to be conferred by the Grand President, who thus would have assumed for himself, in relation to Mongolia, the status of his imperial predecessors. All official appointments were to be confirmed, as before, by Chinese authorities. The Chinese draft treaty also contained a residual clause that provided that “all the matters in Sino-Mongolian relations not covered by the present treaty are to continue as under the old system”.Footnote ⁷⁴ Finally, there was much skirmishing over terminology. The Chinese insisted that Mongolia was to be treated as a difan (地方), or locality, as opposed to a guo (国), or country. Instead of a government (政府), it was only to possess an administration (官吏), and so on.Footnote ⁷⁵ Mongolia, by contrast, insisted on calling itself a guo or indeed, somewhat grandly, as a 帝国 (empire).Footnote ⁷⁶

The Chinese delegation also proposed that lawsuits between Chinese and Mongolians should be tried under Chinese law by Chinese officials. In effect, China was seeking extra-territoriality in Mongolia for Chinese citizens, which sat oddly with its claim of Mongolia as being merely an autonomous part of Chinese territory over which it had sovereignty. This demand was justified by the Chinese delegate on grounds of Mongolian officials’ mistreatment of Chinese (i.e. non-Mongolians) subjects in legal disputes.Footnote ⁷⁷ The arguments made by China—that the local justice system inadequately protected its subjects—would have been familiar to Western diplomats seeking extra-territoriality in China, but with the added twist that China was claiming the right in relation to one of its own territories. Russia, playing off this apparent contradiction, argued that, as Mongolia was Chinese territory, Chinese subjects there ought to be tried by local law, which is ipso facto part of Chinese law, and accused the Chinese government of inventing new doctrines within international law.Footnote ⁷⁸ Could a country in fact assert extra-territoriality within its own territory? And if it did so, could it still claim to be sovereign over it? Those questions were never satisfactorily answered, but illustrate the fact that Chinese elites were just as likely to condemn foreign practices that impinged on its sovereign authority as they were to use the same techniques against weaker counterparts.

Those debates were based on a more fundamental dispute as to the meaning of suzerainty. Whereas the Chinese delegate defined suzerainty as the voluntary cession, on the part of the suzerain, of part of its sovereignty over a given territory, the Russian negotiator defined it as complete autonomy and self-government for the vassal, subject to reserved powers for the suzerain. Such fine distinctions arguably did not matter from a practical point of view, but it was evidence of China's determination to exclude the Mongolians from the definition of their status. In fact, the Chinese were unwilling even to allow Mongolia a place at the table. In its view, it was not up to Mongolia “to determine the limits of suzerainty as a precondition for accepting suzerainty”: since suzerainty was the partial relinquishment of sovereignty by the sovereign power, the sovereign alone decided what should be relinquished.Footnote ⁷⁹ Hence, not only was the struggle over what was suzerainty, it was also about who was entitled to define it.

More fundamentally, China's negotiating positions reflected a government torn between national and imperial logics of governance; it aspired to the former, but the continuation of the latter which was for the moment the most practicable. It simultaneously sought to define Mongolia as an undifferentiated part of the Chinese guotu, while maintaining its distinctiveness by seeking the wholesale maintenance of pre-republican modes of political control and extra-territoriality. Such an approach was bound to lead to absurdities, such as the claim to extra-territoriality in what was purported to be an integral part of China, or the preservation of the neidi/waidi distinction even though Mongolia was supposed to be undifferentiated guotu. All this was bounded by Western norms concerning sovereignty and suzerainty, within which China had to make its case. On the other hand, the Chinese approach was consistent insofar as all of its demands were designed to create or to reinforce effective Chinese control over Mongolia's internal administration: the specific logic of governance did not matter so much as long as its control over Mongolia was complete and effective, since only in that manner could it regain internationally recognized rights over it. That the result consisted of an internally incoherent series of negotiating points does not detract from its fundamental purposefulness.

The final tripartite treaty, signed on 7 June 1915, embodied these contradictions.Footnote ⁸⁰ In the treaty, Mongolia was simultaneously under the suzerainty of China, part of Chinese territory, and autonomous from China.Footnote ⁸¹ Its head was allowed to keep the title of Bogd Khan, but the imperial title was to be conferred by the republican Chinese president (surely the only recorded instance of a president creating an emperor).Footnote ⁸² The Chinese calendar was to be used, but alongside the Mongolian one.Footnote ⁸³ Mongolia was forbidden from concluding “treaties concerning political and territorial questions”, but could conclude commercial treaties.Footnote ⁸⁴ Neither Russia nor China was allowed to interfere in Mongolia's internal administration, but both could establish extra-territorial courts and maintain troops, and China was allowed to maintain officials throughout Mongolia.Footnote ⁸⁵ Effectively, Mongolia was at once Chinese and Mongolian, independent and dependent. The paradoxical situation would only be resolved in the most traditional of manners: taking advantage of the Russian Civil War, in 1919 China invaded Mongolia, forced its leaders to “petition” the Chinese government to surrender its autonomy, and the Bogd Khan was made to kowtow to a portrait of the Chinese president, bringing to an end (for the time being) to the concept of an independent Mongolian state.Footnote ⁸⁶

V. WHAT IS TIBET? SUZERAINTY AND THE LEGAL POSITIONING OF TIBET, 1914–51

It was not until 2008 that the British foreign secretary abandoned his government's previous recognition of Chinese suzerainty over Tibet, in favour of the recognition of Chinese sovereignty. By that time, the origins of that suzerainty had long been forgotten, yet to different people the term had radically different meanings. To the Chinese government, suzerainty was yet another Western colonial imposition, part of the long century of humiliation, and its rejection by the British was merely in conformity with international recognition of Tibet as a part of China;Footnote ⁸⁷ to the Tibetans in exile, suzerainty meant acknowledgment that Tibet was not a part of China, and the British government's volte-face was decried as an act of betrayal.Footnote ⁸⁸ As always, the truth was far more complicated, and evolved in ways which those involved in the shaping of Tibet's future could not have foreseen, and which have been much misunderstood since.

As described earlier, the legal status of Tibet vis-à-vis China was not one which had a clear answer in the Western legal framework due to the perceived incommensurability between traditional Chinese and Western doctrines of political authority. The British tried to convince the Chinese to accept suzerainty over Tibet in the 1906 Anglo-Chinese Convention, but were rebuffed by the Chinese envoy, who insisted on Chinese sovereignty instead.Footnote ⁸⁹ The final document said nothing about Tibet's political status, but was essentially negative in nature—Britain promised not to interfere in Tibet beyond certain limits, and the Chinese undertook not to allow any foreign power (i.e. Russia) to interfere in Tibet. Formal recognition of Chinese suzerainty came in the most roundabout of ways the following year, when Britain and Russia, in that year's Anglo-Russian Entente, “recognizing the suzerain rights of China in Thibet”, agreed to only negotiate with Tibet through China “[i]n conformity with the admitted principle of the suzerainty of China over Thibet”.Footnote ⁹⁰ Not being a party, neither China nor Tibet was legally bound by that “recognition”, but the position taken by the great powers in the region could not help but be influential for the future of both China and Tibet.

Having realized the weakness of their position, China sent a military expedition to establish direct rule over Tibet in 1910, and the Dalai Lama fled to India. Chinese direct rule was short-lived; in the chaos that followed the Xinhai revolution in 1911, the Tibetans drove out the Chinese garrison, and in 1913 the Dalai Lama returned, proclaiming the end of the patron-priest relationship (the term he used to characterize the historical relationship between the Qing crown and Tibet) due to the collapse of the Manchus. It was in fact a declaration of independence, which was confirmed the same year in a treaty with Mongolia in which each recognized the other's independence. In any event, it was the only formal international recognition each nascent state would ever receive.

The British Indian government decided to solve the question of Tibet's status once and for all. In Delhi it was feared that the Tibetans might seek assistance against China from Russia, thus undermining Britain's attempts to exclude Russia from the region.Footnote ⁹¹ Pressure was applied on both Tibet and China, and a tripartite conference, at which Tibet and China had equal status, was convened at Simla in 1913.Footnote ⁹² The Tibetan envoy asked for the repudiation of the 1906 Anglo-Chinese Convention and recognition of Tibetan independence and the exclusion of Chinese troops and officials, as well as the recognition of the Dalai Lama as head of Buddhism in China. The new Tibetan state would include Qinghai and portions of Sichuan and Gansu—in effect Greater Tibet—and China would be required to pay compensation to Tibet for damage caused during the recent military clashes.Footnote ⁹³ The Tibetan envoy justified his demands on grounds that the pre-existing relationship between China and Tibet was one of “the disciple towards the teacher”. In summary, he said that “Tibet and China have never been under each other and will never associate with each other in the future”.Footnote ⁹⁴ As in Mongolia, it was a clear demand for independence, albeit on even less favourable terms than those of the Bogd Khan, without even symbolic retained rights for China. In support of their demands, the Tibetan side came prepared with a large quantity of historical documents aimed at establishing both its historical independence and its title over Greater Tibet.Footnote ⁹⁵

By contrast, the Chinese delegation came unprepared for the Tibetan case, and could muster no documents apart from a list of MPs purported to have been elected for Tibet in the Chinese parliament. It asserted instead authority over Outer Tibet on the basis of “what is called in International Law ‘effective occupation’”, but was at times reduced to quoting from British travel books to prove its case.Footnote ⁹⁶ In the main, the Chinese demands were a statement that Tibet forms “an integral part of the territory of the Republic of China”, and that the rights associated with that fact should be respected by Britain (although the exact nature of the rights was not specified). China would not convert Tibet into a province, and Britain would not annex it. Tibetan foreign affairs were to be conducted through China, except for commercial intercourse with Britain. Finally, Tibet would be partitioned into two parts, the boundary of which approximates to the restrictive definition of Outer Tibet.Footnote ⁹⁷

Finally, the British presented their draft. It also split Tibet into two parts, albeit on terms more favourable for Outer Tibet. Article 2 defined Tibet as “a State under the suzerainty, but not the sovereignty, of China”.Footnote ⁹⁸ As suzerain, China would not be deemed a “foreign power” under the 1904 Anglo-Tibetan Convention, and the Chinese representative in Inner Tibet would be entitled to an escort of 100 men. But overall, the draft entirely decoupled suzerainty from sovereignty. China retained no external or international rights in relation to Outer Tibet, and the latter could only conduct relations with foreign countries and with China through the intermediary of the UK. Neither was China given any right in relation to Tibet's internal affairs.Footnote ⁹⁹ China could not convert Tibet into a province, represent it in the Chinese parliament, or keep any official or “colonist” inside Outer Tibet (Inner Tibet, which had a large non-Tibetan population, was excluded from these provisions).Footnote ¹⁰⁰ On those terms, Chinese suzerainty would be entirely deprived of any significance, and Tibet would become, de facto but not de jure, independent.

The Chinese negotiator, Ivan Chen, protested to the Indian foreign office, and in April, after exhaustive talks, managed to wrestle concessions from the British.Footnote ¹⁰¹ The “but not the sovereignty of China” formula in the original draft was dropped, so that the exact relationship between sovereignty and suzerainty was obscured once again. The British side also agreed to recognize that “Tibet forms part of Chinese territory”, but only as a separate note and not within the main treaty.Footnote ¹⁰² Chen was, however, rebuffed when he asked for an agreement defining the “political, not geographical” limits of Chinese suzerainty in Tibet. The British side countered that, since the term was vague and there existed differences of opinion among publicists, the term could not be defined in the treaty, which naturally raises the question of why the term was chosen in the first place.Footnote ¹⁰³ Any discussion aimed at defining suzerainty would be tantamount to winning more concessions for the Chinese side, given that the alternative was to admit that suzerainty meant nothing in practice.

The final draft was presented by Sir Henry McMahon, the British plenipotentiary, in April. Chen was at first reluctant to sign, and it was only when McMahon threatened to drop British recognition of Chinese suzerainty from the treaty (tantamount to an acknowledgement of Tibetan independence) that he relented, and reluctantly initialled the treaty, pending confirmation from his government.Footnote ¹⁰⁴ Two days later, Chen informed the other delegates that the Beijing government had repudiated his acceptance of the terms. Contrary to later Chinese claims, the reason for the rejection was based not on the rejection of Chinese suzerainty over Tibet, which at any rate must have been known in Beijing for some time, but was because the proposed frontier between Inner and Outer Tibet was not acceptable to the Chinese government.Footnote ¹⁰⁵ The British and Tibetan representatives then proceeded to ratify the convention regardless, with the provision that the Chinese government would be disbarred from its rights under it until it ratified it.Footnote ¹⁰⁶ The point, at least legally, became moot when the British government, in turn, rejected the convention as being incompatible with the 1907 Anglo-Russian Convention, which forbade either power to deal directly with the Tibetan government “in conformity with the admitted principle of suzerainty of China over Tibet”.Footnote ¹⁰⁷

Chinese historiography has tended to portray China's actions at Simla as a forceful rejection of British designs over Tibet in general, and of suzerainty in particular.Footnote ¹⁰⁸ However, although the proposed boundary was unacceptable to the Chinese government, Chen had in fact scored an important success. The affirmation that Tibet was Chinese territory, coupled with the deliberate ambiguity over the nature of sovereignty, meant that the prospect of Tibetan independence with British support was averted for the time being, in addition to providing a basis for later Chinese claims over Tibet, which would have been more difficult to maintain had China accepted either that suzerainty was not sovereignty, or that Tibet was not part of its territory.

Thus, the suzerainty that the British proposed at Simla was, in fact, quite acceptable to China for the time being. The best evidence for the Chinese government's shift of stance comes from an August 1915 proposal by China to evacuate Chamdo (then under Chinese military occupation) in exchange for Tibetan acknowledgement of Chinese suzerainty in a tripartite treaty (at Simla, China had insisted on retaining Chamdo).Footnote ¹⁰⁹ In other words, China was proposing to cede territory to Outer Tibet in exchange for the acknowledgement of its suzerainty that had been agreed upon at Simla. V.K. Wellington Koo, who conveyed the message to the British ambassador to China, made favourable references to the recently concluded Russo-Sino-Mongolian Kyakhta agreement, which he said China wished to emulate in relation to Tibet, because suzerainty allowed China to “save face”.Footnote ¹¹⁰ The favourable reference to Kyakhta, which then and now was considered by almost all outside observers as a Russian diplomatic success, is also telling. This fitted with a broader pattern whereby the Nationalist government was happy to tolerate a great deal of ambiguity on its boundaries in order to focus on other priorities, whilst simultaneously appeasing domestic audiences.Footnote ¹¹¹

In the end, the British did not pursue the proposal because they sought further concessions as to the border between Inner and Outer Tibet, but the Chinese government's willingness to go so far as to trade territory in exchange for a recognition of its suzerainty demonstrates a significant shift in Chinese attitudes towards suzerainty, in contrast with the late Qing insistence on sovereignty. The Chinese government later reaffirmed their acceptance of suzerainty on a number of occasions, most notably in a 1921 political declaration, in which it asserted it wanted “nothing more than the re-establishment of Chinese suzerainty over Tibet, with recognition of the autonomy of the territory immediately under the control of the Lhassa government”, a position the Chinese government seems to have maintained until the 1930s.Footnote ¹¹²

Many on the British side found the importance that China now attached to its newly declared suzerainty puzzling, since from their point of view Chinese suzerainty in Tibet was effectively meaningless. Jordan said as much to his Chinese interlocutors, whom he thought were sacrificing “the substance over the shadow” by focusing so much on their suzerainty.Footnote ¹¹³ In fact, the British Indian government had always regarded Chinese suzerainty in Tibet as something of a fiction. As early as 1903, when China was still resisting the British claim of suzerainty, the viceroy Lord Curzon reported to London that “[w]e regard the so-called suzerainty of China over Tibet as a constitutional fiction—a political affectation which has only been maintained because of its convenience to both parties”.Footnote ¹¹⁴ Whitehall, on the other hand, was more circumspect, and had rebuked Curzon for his stated view in 1903, as the issue involved not only Indian, but also wider British interests in China.Footnote ¹¹⁵

However nominal the British intended China's suzerainty to be, they quickly discovered that suzerainty did in fact limit Britain's scope of action in Tibet, if only by dint of the term's uncertainty. In 1919, the Tibetan government requested arms from the British; as the Chinese and Tibetan armies had clashed in 1917, and China had only recently forced Mongolia to surrender its autonomy, the Tibetan government was naturally anxious to strengthen its defences. The government in Delhi was strongly in favour of the sale, but in London the Foreign Office dithered because Britain was a signatory of the 1919 Saint-Germain Convention, which prohibited international arms sales except to the high Contracting Parties.Footnote ¹¹⁶ China was a signatory, but Tibet, with its dubious international status, was not. Obviously, sale through China would be impossible. Hence, the exact nature of Chinese suzerainty, which the British at Simla had categorically refused to clarify, came back to haunt them.

The Indian government and some in the Peking legation argued that the status of Tibet under Chinese suzerainty was really similar to that of a British self-governing dominion, another semi-sovereign status which had recently entered the international vocabulary. Just as they could ratify the Convention on their own, nothing precluded Tibet from doing so without reference to China.Footnote ¹¹⁷ At any rate, they argued, there was no effective government in China as there was in Tibet, and it would be just to help an organized political community to defend itself against an unorganized one.Footnote ¹¹⁸ And China's refusal to ratify at Simla deprived it of any suzerain rights in any case. In the final formulation, the Indian government proposed that Britain should deal with Tibet as a “self-governing dominion of Chinese Commonwealth, standing in same relation to China as Canada does to Great Britain”.Footnote ¹¹⁹ On the other hand, the bulk of Foreign Office opinion was against the sale.Footnote ¹²⁰ Remembering that the notes attached to the Simla Convention did in fact recognize Tibet as part of Chinese territory, an official minuted that “[w]e should find it difficult to argue that Tibet is not subject to Chinese suzerainty—and if so, is Tibet not part of ‘China’?”Footnote ¹²¹ But to that question not even the Foreign Office legal advisor could provide a definitive answer.Footnote ¹²² Whitehall also argued that suzerains, lacking in international personality, could not enter into binding international agreements.Footnote ¹²³ In the end, weapons were supplied in order to induce China to reopen negotiations over Tibet after it was remembered that the government had not yet ratified the treaty.

But the episode alerted the British government to the fact that suzerainty was not, in fact, meaningless, even if they did not know, any more than anyone else, what it meant. The Foreign Office tried to press the Chinese government into renewing talks, but the latter dithered and the tripartite talks were never renewed.Footnote ¹²⁴ Given that the talks were not going to lead to British acknowledgement of Chinese sovereignty over Tibet, the Chinese government had little incentive to disturb the status quo, which served China's interests, at least until Tibet could be re-absorbed like Mongolia.

Thereafter there were periodic British proposals to scrap the recognition of Chinese suzerainty because it hampered Britain's room for manoeuvre in Tibet—a 1943 proposal made it as far as the War Cabinet, but was never acted upon.Footnote ¹²⁵ Nor did the ever-lurking idea of dominion status find favour; suzerainty was already obscure enough without introducing another concept without equivalent in Asia.Footnote ¹²⁶ Meanwhile, Tibet, even though it was de facto independent, laboured under the handicap of being a vassal, and was not generally recognized as a sovereign country, though it did maintain relations with a few foreign countries.Footnote ¹²⁷ After World War II, China renewed anew its claims to Tibet and, as is well known, after the settlement of the Chinese Civil War the new Communist regime annexed Tibet in 1951, thus bringing an end to its ostensible suzerainty, although the British maintained the pretence of Chinese suzerainty until 2008, by which time the origins of the concept had become even more nebulous than they had been hitherto.