EXPLAINING PEACEFUL CHANGE WITHOUT DEMOCRACY: THE CASE OF THE SINO-SOVIET SECURITY COMMUNITY

: Some scholars have claimed that democratic regime type needs to be treated as a necessary precondition for the formation of a pluralistic security community. This essay argues that one should not overestimate the explanatory power of linking the democratic peace proposition to the study of security communities. Democratic values, norms, institutions, and practices may certainly facilitate the formation of a security community, but it is by no means the only or even most plausible path to assure dependable expectations of peaceful change. While a number of authors have of late made similar claims, what is not settled is why non-democracies can form security communities. The findings in this essay advance scholarship on this issue by showing that the same causal logics commonly attributed exclusively to democratic security community formation are also present in the formation of non-democratic security communities. The study adds empirical evidence to this argument by developing a historical case study of the Sino-Soviet relationship. In sum, the findings demonstrate that (1) democracy is not a necessary (though facilitating) precondition for the development of a pluralistic security community and (2) a pluralistic security community may form between autocratic regimes based on the causal logical nexus of non-democratic norm externalization, ideological coherence, a common Other (normative logic) and autocratic domestic institutional constraints (institutional logic).


Introduction
W hy can non-democracies form security communities? Are the same causal logics commonly associated with democratic security communities also present in non-democratic ones? What does this say about democratic peace theory? 8

MGIMO REVIEW OF INTERNATIONAL RELATIONS • 2 • 2019
In autocratic political systems, the apparent lack of liberal norms, institutional self-restraint and transparency that could otherwise check the freedom of action of political decision makers, prevent the marginalization of domestic opposition and minorities, inhibit the closure of elites and allow outside observers to 'see into' such states makes a compelling argument why mutual trust and dependable expectations of peaceful change, essential for the formation of a security community, may not easily develop among non-democratic countries [9; 26; 43; 74; 76; 77; 94]. On the other hand, some scholars have argued that there may exist an 'autocratic peace' or 'dictatorial peace' that complements the proposition of the 'democratic peace' [11; 12; 34; 48; 73; 102]. According to this view, autocracies do share institutional constraints, identities, and normative values that may create a 'separate peace' among these similar regime types [17, p. 500; 66; 69, p. 20].

The Argument
This essay states that democracy is not a necessary precondition for the formation of a pluralistic security community. Conversely, the article argues that there is the complementary existence of autocratic pluralistic security communities. 1 This is certainly not a novel argument. However, while a number of authors have recently made similar claims, what is not settled is why non-democracies can form security communities [1; 5; 10; 44; 47; 61]. As Bennett notes, 'although the apparent existence of this alternative peace has been noted, theoretical arguments for why it should exist, and empirical analyses that reveal more detail of it, have been underdeveloped' [11, p. 313]. The findings in this essay advance scholarship on this issue by showing that the same causal logics commonly attributed exclusively to democratic security community formation are also present in the formation of non-democratic security communities. A security community is considered to be a group which has become integrated, where integration is defined as the attainment of a sense of community, accompanied by formal or informal institutions or practices, sufficiently strong and widespread to assure peaceful change among members of a group with 'reasonable' certainty over a 'long' period of time (italics and quotation marks in the original, see [18, p. 33]).
Peaceful change is defined as 'the resolution of social problems (…) without resort to large-scale physical force' . In a pluralistic security community, members 'retain their legal independence' as sovereign states [19, p. 5-6]. This study does not argue that autocratic security communities are necessarily more successful and longer lasting than democratic ones. In fact, the historical record implies that they are probably not. Also, I do not pretend that democratic institutions and practices do not matter at all in the formation of stable peace. My proposition simply states that we should not bind our-are defined as "regimes in which the party has some influence over policy, controls most access to political power and government jobs, and has functioning local-level organizations" [30, p. 20]. Finally, the room for analyzing the case study is inevitably restricted. Due to limited space, I will thus provide an abbreviated survey of the main events, processes, and dynamics involved in the formation and disintegration of the Sino-Soviet security community. For the purpose of analytical clarity and space, it is impossible to narrate the Sino-Soviet relationship here in its entire complexity. Potentially missing data will be controlled by the method of process tracing.
The essay is divided into four parts. First, I will give a brief overview on some of the arguments raised by the democratic peace theory that are relevant to the study of security communities. Second, I will review what we theoretically know about the formation of security communities among non-democracies. While there are a number of possible factors typically associated with democratic peace, I will subsequently focus on two causal logics that are invoked by scholars who argue that democracy is a necessary precondition for the formation of security communities. In the third section, I will empirically test these causal logics to find out if they were also present in the Sino-Soviet case. Finally, the essay will revisit the propositions raised above and review what the findings say about democratic peace theory.

The Democratic Puzzle and Security Communities
Many scholars have explained stable peace among nations by focusing on domestic preferences and regime type. The following overview does not pretend to be a comprehensive summary of their assumptions and findings. Rather, it seeks to explain some of the arguments that have been raised with regard to the study of pluralistic security communities. As early as 1795, Immanuel Kant proposed the possibility of a 'pacific federation' or 'pacific union' among liberal republics. The democratic peace theory basically argues that democracies do not go to war with each other, though they certainly fight wars against non-democracies [7; 8; 15; 24; 25; 52; 65; 76; 77; 84; 85; 91]. This empirical phenomenon known as the 'democratic puzzle' has been typically resolved following two broad lines of argumentation. First, the political structure argument states that leaders of democratic governments are more constrained in their policy behavior than leaders of other forms of government. The democratic process makes it relatively easy for domestic oppositions to mobilize against the use of force against outsiders. This argument is based on the Kantian premise that, because democratic states require (at least in the long-run) the consent of its citizens to go to war, violent interstate conflict will be less likely as the citizens themselves will have to bear the costs. The second argument focuses on the distinct political culture of democracies based on individual rights and liberties as well as norms and preferences of orderly and peaceful conflict resolution within a domestic society that are externalized and exported into its international environment [23; 59; 62; 89]. In sum, propositions of democratic peace are based on institutional, informational, normative, and preference-based arguments [41]. According to Lektzian and Souva [51,p. 23], however, only shared preferences may also lead to peace among autocratic dyads.
Whenever states and societies share the perception that they are liberal democracies, they are able to distinguish between likeminded democratic 'doves' and nondemocratic 'non-doves' in the international sphere [13]. As a result, liberal democratic states may form democratic zones of peace by building mutual trust and a democratic identity and, at the same time, separating themselves from the war-prone world of non-democracies [103]. Some proponents of the democratic peace theory have thus linked their explanations directly to the Deutschian concept of security communities by arguing that through social learning and the process of mutual recognition as liberal democracies these states develop collective democratic identities that play a constitutive and disciplining role in the formation of democratic pluralistic security communities [9; 38; 77; 93; 94; 97; 105]. In sum, based on these findings some of these authors have argued that democracy is essentially 'the basic requirement for integrated political security communities as defined by Deutsch' [26, p.  However, the claim that democratic regime type needs to be treated as a necessary precondition for the formation of a security community must be handled with caution on both theoretical and empirical grounds. To begin with, the original concept of security community does not support this claim. Deutsch concludes that a consensus on 'main values' is sufficient (and necessary) in maintaining stable peace. He does not link this condition to any specific type of state organizing principle [19, p. 123]. To restrict the concept of security community to liberal democratic types unduly limits its intellectual utility and depth. For example, the democratic peace theory cannot explain the potential or actual development of pluralistic security communities among primarily non-democratic countries raised by some scholars such as the Association of Southeast Asian Nations (ASEAN), the Economic Community of West African States (ECO-WAS) or the countries of the Southern African Development Community (SADC) [1; 37; 44; 58; 64; 87; 88; 98]. One might agree or disagree with their assumptions but if we limit ourselves to the study of democratic pluralistic security communities, we will not be able to debate this matter at all [63]. Moreover, we may possibly exclude large parts of the (non-Western) world from our analytical focus.
Second, it should be clear that a pluralistic security community encompasses a much broader concept than the democratic peace proposition [94, p. 155; 44]. They may share some overlapping theoretical and empirical features in the sense that they both help to consolidate and preserve stable peace. Security communities are the most advanced type of stable peace. A security community involves not just the absence of war but also the absence of 'significant organized preparations for war' [20, p. 99]. Hence, the threshold for achieving stable peace in the case of a pluralistic security community is significantly higher.
Finally, the notion that democracies recognize each other as members of a security community through shared democratic institutions and norms raises the question of who gets to decide. How do countries with substantially different political and economic institutions, and different cultural values (e.g. France and the USA) agree on 'we-ness' based solely on a democratic identity [28]? In sum, democratic institutions and norms may certainly facilitate the formation of a security community but it is by no means the only or even most plausible path.

Non-democratic security communities
What do we know about non-democratic security communities? While there can be little doubt about the existence of non-democratic security communities, little is known about the causal processes involved in their development [11, p. 313]. The main theoretical disagreement revolves around the question whether 'democratic peace' and 'autocratic peace' are distinct phenomena and hence should be studied separately or whether there exists a single theory that links both types of empirical observations. Proponents of the latter view argue that peaceful dyads result not from democratic regime type but from political and economic institutional similarity [92; 102], 'joint coherence' [11], regime stability [39], economic interdependence [82] or societal and cultural commonality [47]. What these authors fail to address, however, is the fact that 'sameness' alone can hardly explain the formation of a security community in the absence of a common Other [4, p. 56; 68, p. 77].
The most explicit and knowledgeable attempts to theoretically explain the formation of security communities among non-democratic states have been advanced by constructivist scholars. Perhaps most importantly, Adler argues that since security communities are socially constructed, high levels of mutual trust and a collective identity may also develop among illiberal regimes [2]. This may occur in two ways. First, non-democratic states may adopt 'selected' liberal practices from democratic international institutions. Second, illiberal ideologies may produce a set of common norms and ideas. Wendt [101], Acharya [1] and Pecency [69] suggest a similar argument though (as Adler) fail to specify exactly how this may occur. Adler stresses the practical basis of security communities, which leaves open the possibility of security community formation among non-democratic regimes. Since he treats peaceful practices of self-restraint and non-violent behavior as ontologically prior to liberal values and institutions, Adler [5, p. 206] concludes that 'community of practices' (such as security communities) may also spread to non-democratic regions. Kupchan [47, p. 55] makes a similar point by claiming that non-democracies often include elements of domestic constraint that enable these states to form zones of peace: 'it is the exercise of strategic restraint, not regime type per se, that is a necessary condition for stable peace' . Of course, such theoretical claims are rebutted by democratic peace theorists who claim that autocratic leaders and non-democratic societies may never develop a collective identity and level of mutual trust necessary to form a security community since there is nothing in their domestic values and norms that would prescribe practices of nonviolent behavior, responsiveness or compromise [77, p. 501-505; 85, p. 35].
Hence, in order to answer the question whether some of the peace-causing attributes commonly associated with democracies also play a role in the formation of non-democratic security communities, it is necessary to focus next on the causal mechanisms typically associated with the development of democratic security communities. After briefly laying out these variables I will subsequently look at the Sino-Soviet case in order to empirically test their validity with regard to an autocratic security community. In doing so, I seek to demonstrate that the same causal logics commonly attributed exclusively to democratic security communities are also present in the formation of non-democratic security communities. Scholars who claim that democracy is a necessary precondition for the development of a pluralistic security community commonly base their argument on two causal explanations that can be summarized as the normative logic and the institutional logic [80]. These causal logics have not remained uncontested [25; 46; 81; 82; 89]. Still, for the purpose of this essay we may accept their underlying assumptions in order to show that the same causal logics also apply to autocratic security communities.

Normative logic
According to the normative logic the domestic structure of democracies consists of a set of liberal norms and ideas that is deeply rooted in political culture and institutionalized in the political system [77]. These norms (individual freedom, human rights, minority protection, and equal participation) prescribe a certain type of domestic behavior that is based on non-violent conflict resolution and compromise [22; 23; 85]. Political leaders internalize these norms through social learning and subsequently externalize them in their dealings with other states. On a transnational level, these norms are further externalized through non-governmental groups as well as interactions and cultural exchange among individuals [16, p. 4; 60, p. 69]. This transnational identity corresponds with Deutsch's definition of a security community as "a group of people which has become integrated" (italics added) [19, p. 5]. In addition, these liberal norms enable democracies to set themselves apart from their nondemocratic Other. In sum, liberal norms guide interaction at the domestic and international level, provide a framework of shared and collective meanings, and regulate and reproduce a peaceful social order based on mutual trust and a collective democratic identity. In theory, such a normative framework could also exist among nondemocracies either through diffusion of 'selected' liberal practices of self-restraint or a shared normative commitment to an illiberal ideology [2; 69]. However, there are two important caveats. First, we would hardly expect to find people-to-people ties among autocratic states that are not government-mandated. Second, shared ideologies may also legitimate violent behavior to enforce ideological purity both domestic and abroad. As I probe the normative logic in the Sino-Soviet case, we need to bear these caveats in mind.

Institutional Logic
According to the institutional logic democratic institutions and processes hold political leaders accountable to a large and heterogeneous domestic coalition of societal groups and individuals. In addition, the political system is characterized by a set of checks and balances between the executive, legislative, and judicial branch. Due to theses vertical and horizontal mechanisms that enable audiences to monitor and sanction political action, democratic leaders will only engage in large-scale violent conflicts if there is broad and sustainable domestic support [49; 67; 85]. In addition to these audience costs, the transparent character of democratic institutions and processes provides good information to outsiders about the intentions and resolve of democratic leaders [14; 93]. Aware of these institutional constraints, democratic leaders know that their democratic counterparts are equally unlikely to engage in war and, in case of a conflict of interest between them, will instead prefer a negotiated settlement. This institutional logic correlates with the security community concept in the sense that these institutional constraints reflect the practical intersubjective knowledge of a 'civic culture' among democracies [2; 6]. As Starr points out: 'the social integration process provides decision-makers with overwhelming information which allows them to have full confidence in how they separate states (…) This is the definition of the security community' [93, p. 211]. In theory, autocratic leaders (especially in single-party regimes) may face similar institutional constraints as they are held accountable by party members and local party officials and need to constantly provide information to these groups [17; 69; 100]. As caveats, we need to keep in mind that autocratic leaders usually face a smaller 'selectorate' than their democratic counterparts and that the information channeled to their audiences is often ideologically filtered.
In sum, scholars who reject the possibility of non-democratic security communities argue that both normative and institutional logics necessary to establish dependable expectations of peaceful change are limited to democratic states due to the assumption that autocratic regimes face fewer if any audience costs, the lack of transparency, and the absence of domestic norms of non-violent behavior. However, as I indicated above, autocratic regimes may, in theory, be capable of forming pluralistic security communities based on the same logical nexus of non-democratic norm externalization and domestic institutional constraints. Having laid out the theoretical foundations for such a possibility, I will now empirically investigate these causal logics in the Sino-Soviet case.

Ascendancy and Decay
The Sino-Soviet rapprochement has typically been understood as a classical alliance against an outside threat. Indeed, the Sino-Soviet security community was ini-tially founded as an anti-systemic alliance directed against the United States and Japan within the evolving bipolar international order of the Cold War (104, p. 165). However, the Sino-Soviet relationship ran much deeper than one might expect. Both sides demonstrated a high degree of confidence in each other's intentions, practiced reciprocal restraint in their behavior and accepted mutual vulnerabilities [47, p. 160]. In fact, it may be argued that the level of security integration at certain points even eclipsed the level of integration between Western democracies at the time, for example the level of nuclear technology transfer.
While the relationship between Russia and China can be traced back to the beginning of direct relations in 1689, closer contact during Russian imperialism in the 19th century, and to its earlier common history as parts of the Mongolian Empire, the roots of the security community between the Soviet Union and Communist China developed immediately after the 1917 October Revolution and the Communist seizure of power in Russia [96, p. 325]. In 1921, with substantial help from the Soviet-dominated Comintern, the Chinese Communist Party (CCP) was founded. However, when the rivaling anticommunist Guomindang (GMD) came to power in China in 1927, the CCP under Mao Zedong became domestically marginalized and its members persecuted while the Soviet Union under Stalin sided with the GMD by sending some advisors and limited material support. In 1945, the Soviet Union even concluded a treaty of alliance with the Guomindang, which left the CCP comrades irritated and isolated. Indeed, it appears strange that Stalin did not offer exclusive and unlimited support to the CCP. Yet for Stalin, ideological considerations at this point were no match to his geopolitical concerns about Japanese expansionism and the war against Nazi Germany [72, p. 4]. The treaty with the GMD also ensured recognition of Soviet territorial interests in the region by the US and Great Britain at the Yalta Conference.
Of course, all of this changed when the CCP emerged victorious from the Chinese Civil War in 1949 while, simultaneously, tensions between the Soviet Union and the United States came to dominate the international arena. In fact, the new Chinese leader Mao Zedong displayed a great desire to enter into alliance with the Soviet Union but was turned down several times by Stalin who, initially, had no intentions to jeopardize Soviet territorial gains by replacing the existing Sino-Soviet Treaty with the GMD [104, p. 9]. It was not until 1950 when, for the first time, Stalin even agreed to meet Mao in Moscow to discuss negotiations of a new treaty. The Moscow Summit produced the Sino-Soviet Treaty of Friendship, Alliance and Mutual Assistance. The central focus of the treaty was to prevent a 'repetition of aggression (…) on the part of Japan or any other State which should unite with Japan'; the latter part implying mutual protection against a potential military conflict with the United States. In addition, the Soviet Union provided substantial military and economic aid including $300 million in credit loans and the development of the Chinese air force and long-range artillery. Moreover, Stalin recognized Chinese sovereignty rights by promising the withdrawal of Soviet troops from Port Arthur, to yield Soviet privileges at Port Dalny, and to surrender control over the strategically important Changchun railroad, all by 1952.
At the same time, Stalin imposed significant demands on the Chinese such as Soviet participation in joint-stock enterprises as well as Soviet influence over Outer Mongolia and the Chinese provinces of Manchuria and Xinjing including the right to transport Soviet troops across Northeast China [104, p. 11]. Despite the obvious asymmetrical nature of the Sino-Soviet relationship with China being the junior partner, Mao did get most of what he wanted in terms of bilateral security and economic assistance to consolidate the nascent People's Republic of China, improve the country's international position, and strengthen Mao's standing at home [72, p. 9]. The Korean War put the treaty into practice sooner than expected as China and the Soviet Union, despite minor disagreements over negotiations of a cease-fire on the Korean peninsula and a quarrel over Soviet air cover, cooperated essentially harmoniously in military strategy planning and consultation. It was only for Stalin's continued humiliations, for example by making the Chinese pay for Soviet military equipment used during the Korean War, that put a little damp on an otherwise successful community-building experience.
After Stalin's death in March 1953, the Sino-Soviet relationship transformed significantly. If we need to set a 'take-off point' 3 for the beginning of the Sino-Soviet security community, 1953 would be the starting year. Up to this point, the Sino-Soviet relationship represented a state of rapprochement by developing mutual expectations of peaceful coexistence. However, the lack of mutual trust and responsiveness kept both countries locked in the security dilemma. In a complete reversal of Stalin's personal engagement with China, the new Soviet leader, Nikita Khrushchev, used his first visit abroad to travel to Beijing symbolizing the increased importance that he assigned to the Sino-Soviet relationship. At the meeting in Beijing in September/October 1954, Khrushchev extended economic cooperation to a level that made some members of the Politburo wonder whether the Soviet Union could even afford such massive spending. In fact, Soviet economic assistance to China added up to the largest foreign aid program in the socialist camp ever. In addition to new credit loans, the Soviets passed on sophisticated technological know-how almost free of charge and aided in the construction of several hundreds of enterprises in China [107, p. 202 In addition to military and economic cooperation, social and cultural integration between the PRC and the Soviet Union also intensified during this period. The Chinese suggested to establish 'postal, telegraph, railway, and air services with the USSR (…) and also (…) to set up a joint Soviet-Chinese air company' [104, p. 301-313]. The number of Soviet experts and advisors, mostly engineers and teachers, sent to China increased from about 400 in early 1954 to multiple thousands in the following years. 4 Conversely, many Chinese workers, scholars, and students came to the Soviet Union for education and research. Soviet books and pamphlets were translated into Mandarin while Soviet films penetrated even the most remote areas of the PRC [104, p. 17]. To be sure, most of these transactions travelled one-way but were met by an enthusiastic welcome from the Chinese people. Both China and the Soviet Union engaged in a multilateral community-building dialogue by developing common and consensual mechanisms to communicate and coordinate their interests on important international issues. Common normative ideas of communism served as the core of a Sino-Soviet collective identity which were frequently reproduced and reestablished in communication and speech acts: 'We trust you people, because you are from a socialist country, and you are sons and daughters of Lenin' [107, p. 209]. Based on his 'lean-to-one-side'policy of 1949, Mao initially completely aligned Chinese foreign and domestic policies with the Soviet Union [21 p. 12; 42, p. 247]. In doing so, both sides were able to identify with each other against the 'Western imperialists' and to perceive their needs and interests as mutually indivisible: ' Attacking China means attacking the Soviet Union' [35, p. 151]. These identity narratives produced a sense of community and were certainly not limited to intergovernmental and interparty relations but comprised all levels of Chinese and Soviet society as people on both sides developed a feeling of mutual sympathy and formed an 'indestructible friendship' [32, p. 57; 45, p. 136].
The first cracks in the relationship became visible after Khrushchev had delivered his so called 'secret speech' at the Twentieth CPSU Congress in Moscow in February 1956 denouncing Stalin's crimes and policies. While Mao was irritated that Khrushchev had not consulted him prior to the speech and certainly did not display much personal amity for the new Soviet leader, his main disapproval of the speech resulted from the fact that Khrushchev's criticism of Stalin potentially undermined Mao's undisputed ideological authority at home [106]. The subsequent ideological debate between Chinese 'dogmatism' and Soviet 'revisionism' gradually came to dominate and haunt the Sino-Soviet relationship in the years to follow. Another point of contention was Khrushchev's policy of peaceful coexistence with the United States and its Western allies, which Mao viewed as a sign of weakness in the face of continuing 'imperialist aggression' as well as a potential US-Soviet alliance to isolate the PRC. Despite these simmering disputes, however, Sino-Soviet cooperation actually progressed in the meantime with the 1957 agreement on further exchange of nuclear technology aimed at enabling China to produce its own nuclear weapons by 1960 and extended agreements on economic and educational cooperation in the same year.
A serious bilateral crisis occurred when the Soviets pushed for even closer military cooperation by proposing to the Chinese a joint Sino-Soviet nuclear submarine fleet and the construction of long-wave radio stations on Chinese territory jointly operated and paid for by the Soviet Union. Mao interpreted the Soviet proposal as a deliberate attempt to undermine Chinese sovereignty. Conversely, Soviet leaders started to slow down the transfer of nuclear technology worrying about Mao's unorthodox views on nuclear war: 'I believe that the atomic bomb is not more dangerous than a large sword. If half of humanity is killed during this war, it will not matter' [104, p. 158

Normative Logic
The Sino-Soviet relationship not only shared a common ideology but was also built on a common definition of the Other represented by the 'capitalist imperialists' in general and the United States and Japan in particular. Both parties based their relationship on a common understanding about the future political order (Communism) and ideology (Marxism-Leninism) as well as their confrontational dealings with external and domestic threats to that order. This shared and internalized normative structure essentially constituted the collective identity necessary for developing and maintaining the security community. Precisely, the Sino-Soviet security community was founded on two interconnected normative structures that led to the formation of mutual trust and a collective identity.
First, both sides shared a common understanding about the future political order. The Soviet Union and the People's Republic of China both sought to build a communist society as the final stage of human history based on the theoretical writings of Marx and Lenin. This included the exclusive control of the Communist party at the expense of all other political movements and the abolition of social class and individual property. In sum, Marxism-Leninism provided the constitutive normative framework for the Sino-Soviet security community. In this context, one may counter that such ideological visions may be manipulated for short-term political gains and thus may be used as an instrumental device. In some cases, this may certainly be true. However, given the historical evidence, we may confidently claim that both Soviet and Chinese leaders acted on the genuine belief in the superiority and realization of communist societies [56, p. 24; 57, p. 9].
From this first normative structure follows a second one. As both actors shared common beliefs about the future of political order they also established a common understanding of how to deal with any external and internal threats to that order. While the marginalization of domestic resistance has already been mentioned, the interaction with external threats deserves some further attention here. The Soviet Union and the PRC were both committed to world revolution. This objective included the promotion of the communist political order in other countries, which, ultimately, meant the overthrow of the existing 'imperialist international system' by expropriating the capitalist classes worldwide and place the productive capacities of society into collective ownership. Naturally, the United States as the dominant power and ideological leader of the 'imperialist bloc' played a central role in Soviet and Chinese dealings with their external environment [57, p. 10]. As a joint declaration in late 1954 states, '(the PRC and the USSR) note the full coincidence of their views both on the all-round cooperation developing between their countries and on international affairs' [32, p. 289]. Some scholars have claimed that the opposition to the 'Free World' was not sufficient to guarantee peace between the Soviet Union and the PRC [74, p. 38]. However, this assumption is flawed as it is based exclusively on the empirical evidence of the violent border clashes of the late 1960s and completely ignores the period of peaceful integration that took place during the 1950s. In sum, the Soviet Union and the PRC developed a normative community and a collective non-democratic identity based on the normative understanding and externalization that socialist states must remain united against a capitalist Other and assist each other by practicing self-restraint and solidarity.

Institutional Logic
Beginning in 1953, there were substantial changes in political and military planning that allowed for transparency and a relatively free flow of information and technology. The highly advanced level of military cooperation included the transfer of state-of-the-art weapon technology (e.g. fighter jets and missiles) that even the Soviet army had hardly tested before. Mutual trust even expanded into the most sensitive area of military know-how and capabilities with the establishment of joint nuclear research and the construction of a nuclear reactor in China. Apparently, the Soviet Union did not regard a potential Chinese nuclear bomb as a threat but, instead, extended its nuclear umbrella over the PRC [57, p. 35]. Likewise, the Chinese did not view the presence of thousands of Soviet specialists and advisors in its ministries, military bases, and party organizations as problematic. Scholars who claim a lack of transparency among autocratic states must acknowledge that even the United States was never willing to provide the same level of direct assistance and information-sharing to its European allies in the way the Soviet Union did in China [35, p. 158]. Needless to say, during the period between 1953 and 1960, according to the historical evidence, neither China nor the Soviet Union prepared for war against each other. Moreover, Mao and Khrushchev faced significant domestic audience costs. Under their leadership, both the Soviet Union and the PRC were stable single-party authoritarian regimes. Hence, their leaders had much less control over the bureaucratic apparatus or individual officials than for example the personalist regimes of Communist leaders Kim Ill-Sung in North Korea or Enver Hoxha in Albania. The presence of intraparty competitive elections for certain offices and the relative autonomy of top officials and party leaders significantly increased audience costs for Khrushchev and Mao [30; 100].
Hence, both leaders being conscious of their domestic institutional constraints practiced considerable self-restraint in their dealings with each other in order to consolidate their domestic power base and to mobilize the people and resources for the transformation of domestic society and in support of Sino-Soviet friendship. In this sense, the lines between government-mandated societal integration and independent transnational ties increasingly blurred to a point when they became virtually undistinguishable [45]. Both societies constantly reproduced the practical intersubjective knowledge of a distinctly 'proletarian culture' , which enabled them to generate mutual trust and responsiveness necessary to form a non-democratic security community. To be sure, these institutional constraints may not reflect the density and depth of transnational networks and societal transactions in civic cultures but they were nevertheless present in the Sino-Soviet case enabling both states to read each other's peaceful intentions and to recognize each other's dovishness.

Conclusion
This essay has shown two things. First, it cautions scholars not to overestimate the explanatory power of the democratic peace proposition when studying security communities. In contrast to assumptions made in previous works, democracy is not a necessary precondition for the formation of a pluralistic security community. Democratic values, norms, institutions, and practices may certainly facilitate the formation of a security community but it is by no means the only or even most plausible path. While a number of authors have recently made similar claims, what is not settled, however, is why non-democracies can form security communities. The findings in this essay advance scholarship on this issue by showing that the same causal logics commonly attributed exclusively to democratic security community formation are also present in the formation of non-democratic security communities. Again, I do not argue that autocratic security communities are necessarily more successful and longer lasting than democratic ones. In fact, the case of the Sino-Soviet security community implies that they are probably not. Yet, its very existence, if only temporary, and more importantly, the corresponding causal mechanisms involved in its formation need to be recognized. Hence, scholars should not limit themselves to studying democratic norms and institutions when studying security communities. Instead, one should look at democracy as a facilitating rather than a causal factor. In sum, the case of the Sino-Soviet security community has demonstrated the need to rethink our understanding of the dynamics and processes of that particular relationship as well as of the study of security communities more generally.
Second, the findings in this essay also have significant implications for democratic peace theory. While there can be little doubt about the existence of an autocratic peace, scholars differ over its causal processes. The main theoretical disagreement revolves around the question whether 'democratic peace' and 'autocratic peace' are distinct phenomena and hence should be studied separately or whether there exists one theory that links both types of empirical observations. While it is theoretically possible that a democratic peace and an autocratic peace could result from separate causal processes, the findings at hand tend to support the latter view as the causal logics of security community can be equally (though to different degrees) applied to both democratic and non-democratic dyads. However, as this essay has also shown, it is not 'sameness' or similarity per se that enables non-democracies to form pluralistic security communities. The presence of a common Other and ideological coherence are at least equally important. The significance of these factors became visible in the Sino-Soviet case when domestic de-Stalinization and peaceful coexistence with the United States in the Soviet Union increasingly conflicted with re-Stalinization and anti-imperialism in China ultimately leading to the breakup of the security community. Also, the findings demonstrate the importance of individual 'peacemakers' [3, p. 168] or 'peace promoters' [27, p. 280] in the process of security community formation as, in this case, Khrushchev and Mao.
In the end, this essay represents only a first attempt to systematically investigate the causal logics involved in the formation of non-democratic security communities. As I hope to have convincingly shown, the combined findings demonstrate the causal logics involved in the formation of a single-party autocratic security community. Further research needs to be conducted to find out if there are more empirical cases of peaceful change without democracy. Specifically, we need detailed empirical case studies on the formation of security communities among other types of autocracies (personalist, military), the impact of alternative non-democratic norms and ideas, and the role of public opinion and societal as well as transnational actors in creating peaceful change among autocracies.