Towards a global security studies: what can looking at China tell us about the concept of security?

Existing scholarship has demonstrated that theorising about security is Eurocentric. This leaves us with a partial account of the concept of security, which is presented as universal. This in turn generates explanatory problems because we are only seeing part of the picture. Yet there have been few attempts to move beyond critiques of Eurocentrism to examine the concept of security ‘elsewhere’. This paper takes China as its starting point, asking: what can looking at China tell us about security? In answering this question, the paper makes two contributions. First, it presents new empirical findings, building a conceptual history of security in China. Drawing on 140 key texts dating 1926–2022, the paper traces the emergence of the concept of security in China and its evolution through three explicit security concepts. Drawing on postcolonial insights it demonstrates that these concepts are hybrid, evolving out of multiple domestic and international influences. They have similarities as well as differences with the Eurocentric concept that dominates International Security Studies (ISS) and produce a discrete approach towards security that has been overlooked in a discipline that uses ‘Europe to explain Asia’. Second, considering these insights, the paper demonstrates that the universal concept of security that underpins theorising in ISS is partial and misleading. Differences in security concepts matter for theorising security and for understanding security policy. Consequently, I argue that we need to provincialize the concept of security: a truly global security studies is of necessity a provincial one attuned to difference and similarity.

Security has become a central concept for understanding the contemporary international system. Scholars of International Relations have theorised its importance since the inception of the field: the idea of 'national security' lies at the heart of the discipline (Neocleous, 2000: 8). Existing scholarship has demonstrated that theorising about security is Eurocentric: it 'derives its core categories and assumptions about world politics from a particular understanding of European experience' (Barkawi and Laffey, 2006: 330). It privileges some experiences over others and considers them both as universal and as 'fact' (Sabaratnam, 2020: 10). In the words of Barkawi and Laffey (2006), 'conventional security studies . . . is a product of Western power ' (p. 352). This causes three problems. First, it leaves us with a partial account of the history and concept of security, which is presented as universal. Second, this then generates explanatory problems because we are only seeing part of the picture: the field 'mistakes "Western" experiences for the universal' and so fails to see different insecurities experienced elsewhere (Bilgin, 2010: 619). Indeed, the field 'provides few categories for making sense of the historical experiences of . . . most of the world's population' (Barkawi and Laffey, 2006: 332) -and for how those experiences have in turn shaped the world as we know it. This limits the discipline's ability to understand and explain international politics. Third, presenting particular experiences as universal has broader political implications: it reflects and reproduces existing power (im)balances. This much we know. However, there have been few attempts to move beyond critiques of Eurocentrism to examine the concept of security 'elsewhere' and to consider what these insights reveal about the future direction for the study of security. That is the focus of this paper.
Taking this problem seriously requires a research design that centres an in-depth case study. The concept of security has never been neutral: it has always been laden with history and culture (see Neocleous, 2006). This paper introduces China as an alternative starting point, to retrace security and its emergence in a different geographical context, in order to see what that can tell us about the concept itself. Since taking power in 2013, president Xi Jinping has placed security at the heart of the political project. Announcing a new 'total concept of national security' with 'Chinese characteristics' in response to 'unprecedented challenges' facing China today (Renmin Ribao, 2014), his security doctrine is enshrined in the Chinese constitution, illustrating its importance. Security matters in contemporary Chinese politics, but China fits neither the profile of the EuroAmerican experience that forms the basis of existing theory, nor the developing/Third-World/postcolonial experience that forms the basis of postcolonial work in International Security Studies (ISS). 1 This makes China a particularly salient alternative starting point. 2 The paper traces the emergence and evolution of the concept of security in China over the past century, writing a counter-history of security. In the process, I ask: what can looking at China tell us about security?
In answering this question, the paper makes two contributions. First, it presents new empirical findings, building a conceptual history of security in China. Because political science has tended to favour grand narratives, sweeping assumptions have often gained more attention than nuanced empirical analyses of Chinese history (Chong, 2014: 941-942). The paper uses conceptual history as method, 3 an approach that aims to show 'the political importance of conceptual change', contributing to 'historical thinking about politics . . . [and] to the activity of political theorising' (Farr, 1989: 37-38). The analysis draws on a close reading of 140 key texts dating between 1926 and 2022, centring key speeches and texts by top leaders, significant policy documents and legislation. Texts were selected on two criteria: their relevance to understanding the concept of security and their importance (see Appendix). Analysing political discourse, the paper traces first the emergence of the concept of security and later its evolution and transformation through three explicit security concepts. Drawing on postcolonial insights, it demonstrates that China's security concepts are hybrid, evolving out of multiple domestic and international influences. They have similarities with the security concept/s found in ISS, but they are not the same. Chinese elites actively and explicitly pursue conceptual innovation, transforming the concept to suit the changing needs and interests of the Partystate. They also contain Chinese influences and ideas that originate outside of the Eurocentric concept. When it comes to security, China has drawn on institutional templates from the Soviet Union and conceptual vocabularies from Europe and the United States, but neither are pure replicas. The analysis shows two core differences that endure: (1) the referent object of national security is not the state, but the party-state and (2) threats are not (primarily) external to the state. These differences are not merely of theoretical importance, they also shape policy: producing a discrete approach towards security that has been overlooked in a discipline that uses 'Europe to explain Asia' (Kang and Lin, 2019: 399).
Second, considering these insights, the paper demonstrates that the universal concept of security that underpins theorising in ISS is partial and misleading. By building an alternative conceptual history, the paper shows the importance of context and domestic factors in shaping a state's concept of security. This shows that we need a provincial approach to analysis for the future of IR. This matters because only by pursuing such provincial analysis can we understand the meaning of security, but also how security policy develops. The concept of security does not simply describe an independently existing reality, it is also 'a medium for defining the possibility of politics' (Lund Petersen, 2011: 696). I understand security here as a political discourse that legitimates the exercise of state power. Whether or not something is considered 'security' and what security is taken to mean shapes resource allocation, policy direction, and political action. Therefore, understanding security matters. Consequently, I argue that we need to provincialize the concept of security. The paper therefore sets a new agenda for security studies, arguing that a truly global security studies is of necessity a provincial one attuned to difference and similarity.
The paper proceeds in three parts: (1) outlines the Eurocentric limits of International Security Studies, (2) presents the conceptual history, and (3) returns to the research question to draw some broader lessons for understanding the concept of security.

The Eurocentric limits of International Security Studies
Theorising about international politics is Eurocentric: it views the whole world through EuroAmerican history and experience, and presents the Western world as more developed, civilised, and/or morally superior (see Hobson, 2012). We know that ISS suffers from the same problems (Barkawi and Laffey, 2006) 4 : less attention has been paid to how Eurocentrism has shaped our understanding of the concept of security specifically. That is my focus here. EuroAmerican experience has been the foundation for theorising a concept of security that is presumed to be universal. While I cannot provide a complete picture in the space available nor do full justice to the nuances or divergences within theoretical schools, I hope to illustrate here in broad strokes the Eurocentric concept of security. There are also epistemological limits: the Eurocentric concept of security is slippery and exceptions abound because no real singular Europe underpins Eurocentric thought. It is shaped by an imaginary Europe and a story of European history that is often factually incorrect (Osiander, 2001). This does not make it less influential. To borrow a phrase from Chakrabarty (2007), something about this concept '. . . remains deeply embedded in cliched and shorthand forms in some everyday habits of thought' (p. 4) and continues to shape scholarship on security politics all over the world.
ISS tells a story about the world where the concept of security is both self-evident and universal. 5 The international is characterised by anarchy because no authority exists above the state. Consequently, states face an 'ever present possibility' of military threats from other states and 'each state must guarantee its own survival since no other actor will provide its security' (Mearsheimer, 1990: 12). Thus security is about protecting the state from external (military) threats, a self-perpetuating process necessitated by anarchy. This understanding of national security was dominant for the duration of the Cold War 6 and remains embedded in everyday habits of thought. 7 There are two key things to note here: (1) security involves protecting the state and (2) threats are external and usually military. The concept of security is neutral and universal -it just 'is', because anarchy is. But this story hides the origin of both the concept and the discipline itself. ISS and the concept of security are shaped by the US post-war era and Cold War context in which they emerged (Neocleous, 2006). Scholars preoccupied with looking at how Western governments dealt with war and the central strategic balance of the superpower rivalry (Buzan and Hansen, 2009: 2) largely overlooked the experiences and issues faced by the rest of the world (Acharya, 1997: 300). This produced a narrow concept of national security based on EuroAmerican experience, yet considered a universal model driving state behaviour around the world.
With the end of Cold War, the concept expanded to include non-military threats, but most of the discipline retains 'the predominant national security frame' (Buzan and Hansen, 2009: 2). Later and particularly after 9/11, the concept opens to encompass internal threats (Bigo, 2002: 63), and as the 2000s progress, the incorporation of 'risk' further breaks down some foundational categories like inside/outside, war/crime, and military/police (Lund Petersen, 2011: 695) in at least parts of the discipline. But this often repeated 8 tale of the evolution of security also primarily reflects EuroAmerican experience. It remains a predominantly 'Western subject, largely done in North America, Europe, and Australia' (Buzan and Hansen, 2009: 1), and principally reflects changes in thinking and (geo)political context in these spaces. But these shifts are read as being of universal significance for understanding the concept of security everywhere. In contrast, many developing states continue to adhere to a more narrow military concept (Brauch, 2008: 29).
International Security Studies continues to centre analysis on a concept of security that is shaped by EuroAmerican theoretical origins, historical experience, and contemporary politics. Presenting a provincial analysis as universal produces a misleading account of the central concept underpinning the discipline. It also limits our empirical understanding of much of the world. Critiques of mainstream ISS have made some inroads, some of which need elaboration. Starting in the late 1980s, scholars in Third World Studies argued that developing states face a wider range of -often internal -threats (Ayoob, 1995: 4; see also Thomas, 1987). 9 A broader and deeper critique came from critical security studies, aiming to open up the concept to new threats and alternative referents. However, these critiques centred on the concept of security in International Security Studies, leaving the international largely unchallenged (Bilgin, 2023: 3). Critical security studies broadly continues to assume that security has a 'universal logic' (Browning and McDonald, 2013), while centring theories based on EuroAmerican experience and/or political theory. The field fails to fully interrogate the historical legacy of Eurocentrism or to move forward by taking experiences of (in)security elsewhere seriously on their own terms. This is clearest in the case of critical and human security approaches which centre security-as-emancipation, where 'the agent of emancipation is almost invariably the West' (Barkawi and Laffey, 2006: 350).
Securitization theory attempted to reconcile more radical calls for expanding security with the mainstream's desire for a narrow concept. It contributed a broadening of security beyond the military into five sectors, together with a redefinition of security as a speech-act. The approach simultaneously opened up the concept and locked it down. The broadening sectoral approach has had limited uptake, while the redefinition of security as securitization has produced a burgeoning field of study. But although anything can (hypothetically) be securitized, the Copenhagen School's definition of securitization as a speech-act relies on a narrow Eurocentric concept of security that is closely related to that of mainstream ISS (see also Lund Petersen, 2011: 710). They argue that because of how 'security' is used in the [EuroAmerican] field of practice, it 'has to be read through the lens of national security' -it cannot escape its historical connotations (Waever, 1995: 49). It is closely related to 'power politics' and a 'traditional [EuroAmerican] militarypolitical understanding' (Buzan et al., 1998: 21): as a result, only particular speech-acts qualify as securitising acts. Thus, the theory 'limit[s] the meaning of security to very specific usages by particular actors' (Nyman, 2016: 825). 10 Foucauldian security studies associates security closely with the rise of the liberal state (Neocleous, 2007). This has brought valuable insights for understanding the evolution and practice of biopower and governmentality, but its theorisation of security remains deeply rooted in Western historical experience. Indeed, the point of reference remains the liberal state. 11 Further, 'Foucault's origin story for biopower remains sanitised of colonial domination and violence' (Howell and Richter-Montpetit, 2018: 5). It rests on an idea of the 'human' that does not acknowledge that some have been considered less-than-human, resulting in a sanitisation of state violence experienced by racialised subjects (Howell and Richter-Montpetit, 2018: 5-7). Last, the poststructuralist project in IR has tended towards an 'idealisation of the marginalised' that does not engage with their actual position, voice, or agency (Sajed, 2012: 143). Rooted in a refusal to speak for the other, it has tended to elide non-Western ideas, views, and subjectivities (Sajed, 2012: 162), focusing instead on critique of Western practice.
Postcolonial critiques have demonstrated that ISS is Eurocentric (Barkawi and Laffey, 2006) but have struggled to make further headway. They focus primarily on abstract 'macro-/meta-theoretical analysis' (Hönke and Müller, 2012: 384) and critique. The few works that do look beyond 'the West' variously emphasise the need to look at security in postcolonial states, the Third World, or the Global South (Barkawi and Laffey, 2006;Bilgin, 2010;Hönke and Müller, 2012). China again makes an interesting case in this context (see footnote 1). Eurocentrism in ISS has primarily been addressed by scholars working in war studies or on violence (Barkawi and Laffey, Jabri, Howell and Richter-MontPetit) and thus tells us little about the impact of Eurocentrism on the concept of security specifically 12 (Barkawi and Laffey, 2006: 349). This is a problem since widened security agendas span a range of non-traditional security issues. In sum, while postcolonial security studies provides a clear diagnosis of the problem, it has told us less about what a global security studies might actually look like: this is what this paper attempts to provide.

Provincializing the concept of security
We cannot ignore or escape Eurocentrism, since it has shaped thinking about security all around the world. But an alternative history of the concept that starts with a recognition of hybridity and interconnection can be open to both similarity and difference.
Bhabha shows how Eurocentrism has blinded us to the hybrid constitution of the international, masking the productive power of colonialism. Colonial power produced similarity in colonised spaces, creating hybrids that mimic the centre, but what may appear the same -mimicry -has both camouflaged and produced difference (Bhabha, 1984: 131). It is 'almost the same, but not quite' (Bhabha, 1984: 126). This can in turn be a source of power for the subjugated population (Bhabha, 1984: 126). For Bhabha (1994), hybridities have emerged through moments of transformation, so we need to explore the 'in-between space', 'the cutting edge of translation and negotiation' (p. 38). This necessitates paying attention to both difference and similarity: the hybrid is by its very nature in-between (Bhabha, 1994: 219). Ling (2002: 18) introduced these ideas into IR, arguing that mimicry can be a survival strategy, a process of postcolonial learning, which produces similarity. She uses this to analyse the impacts of East-West colonial encounters in East Asia, resulting in a 'mutually produced hybrid' where 'each derives of and from the other' (Ling, 2002: 75). As Chakrabarty (2007) shows, European thought is 'both indispensable and inadequate' for understanding the experiences of non-Western spaces. Our task is not to reject European thought, but to explore how it has influenced, engaged with and transformed (in) non-Western spaces, and how it can be 'renewed from and for the margins', to provincialize Europe (Chakrabarty, 2007: 16). Thus I bring together Bhabha's theorising of hybridity and Ling's theory of postcolonial learning to understand the emergence and evolution of the concept of security in China. Chakrabarty's call to provincialize Europe provides the basis for moving forward.
Hybridity, mimicry, and postcolonial learning have obscured difference in security concepts, masking Eurocentrism in the discipline. At the same time, Bilgin demonstrates that it is not enough to 'add the non-West and stir' -we have to understand the effects of historical Eurocentrism, both on the discipline and on how security is practised today in different parts of the world: many non-Western elites have embraced the 'standard' notion of security and utilized it in building national security states. What security studies had on offer (a state-focused approach to world politics and 'national security' as the language of state action) also served the interests of non-Western elites busy with state-building. (Bilgin, 2010: 618) Policymakers and scholars in non-core places 'were no mere vessels but also merchants of the increasing production and consumption of 'standard' notions of security' (Bilgin, 2010: 618). In response, Bilgin suggests analysing the production of similarity. In the end, we need to understand 'the "non-West" as the constitutive outside of ISS . ..those ideas, practices, and institutions that are typically ascribed to the "West" have been coconstituted by both' (Bilgin, 2023: 7). Here I build on Bilgin, to show that Chinese elites are not simply pushing standard (EuroAmerican) concepts of security -they are innovating, adapting the concept to suit their own agendas. This in turn shows both the agency of elites in non-core states and the need for analysis attuned to both difference and similarity in security concepts.
Before moving on to the empirical analysis, let us consider what we already know about security in China.

Security in China: what do we know, what do not we know?
Much of IR and ISS scholarship produced both in the West and within China takes the Eurocentric concept of security as its starting point. It focuses primarily on external threats, reflecting and often drawing explicitly on the American realist tradition of thinking about security. Nathan and Scobell's China's Search for Security is a good example here: it specifically aims to understand Chinese foreign policy, focusing on border challenges, external threats and relations with neighbouring states and the United States (Nathan and Scobell, 2012: 18-19). The Routledge Handbook of Chinese Security considers how Chinese history shapes contemporary policy on security, but in empirical terms similarly places the focus on 'China's relations with the great powers, regional security, and China's involvement with collective security organisations' (Dittmer and Yu, 2015). Scholars within China often similarly stress the role of geopolitics and external threat/s in discussions about security (Kong, 2010;Liu, 2004). I am not suggesting such analyses of Chinese security politics are incorrect, but they present a partial picture.
Scholars working within critical security studies have developed in-depth investigations of the concept of security but have mostly neglected China. A few exceptions are worthy of note. Beeson (2014: 2-3) points to broader concepts of security across Asia, going beyond external military threat. Radtke's comparative analysis of the origins of East Asian thinking on security argues that it can be found in the concept of 'disorder', or luan, which predates contemporary discourses on (in)security (Radtke, 2008: 204). Zhang Yongjin argues that internal threat and instability are central to China's understanding of security (Zhang, 2001: 252). Vuori's analysis of Chinese national security discourse suggests the term itself is imported from the West and adopted in the reform period of the 1980s, noting a shift in the discourse from 'counter-revolution' towards 'national security' (Vuori, 2014: 56). While broadly focused on securitization, he has also analysed the construction of internal threats as security concerns, such as the Falun Gong, indicating that a focus on external threat may not suffice for understanding China's concept of security. You Ji draws on securitization theory to point to an expansion of security under Xi Jinping, suggesting that 'securitizing everything' has become the norm (You, 2016: 180). Here there are similarities with Western developments post-9/11 but no discussion of where these developments originate or what the differences might be.
Meanwhile, critical security studies has been largely ignored in China (Yu, 2014: 24-25). Scholarship on security within China has some key differences worth mentioning. Political sensitivities leave scholars (understandably) reluctant to debate or criticise official state policy (Zeng et al., 2015: 253). In practice, 'security concepts that circulate in the academic world tend to reflect, interpret, and. . .comment on those articulated by the state' (Liu, 2012: 73). The Chinese academic debate on security is often either policycentred (driven by leaders' agendas) (eg Liu, 2019;Ma, 2001) or event driven (Hu, 2014), and usually inward-facing. Some pieces broadly summarise new policy initiatives (Gao, 2015), while others give policy advice ('my country should . . .') (Hu, 2020;Luo and Yuan, 2005). Thus in the early 2000s, we see papers developing a broadened concept of security, (Liu et al., 2002;Ma, 2001) while post-2014, we see suggestions for developing Xi's national security concept (Gao, 2015;Liu, 2019;Sun, 2019).
Chinese language scholarship has also included some thoughtful attempts to characterise China's changing thinking on security. In the early years of the PRC until the 1990s, Hu Hongbin argues that thinking on security was dominated by military and defence, since a Marxist understanding of the contradiction between imperialist and communist countries made war seem inevitable (Hu, 2014: 151). Liu Yuejin points to the (relative) novelty of 'national security' language in formal Chinese political discourse compared with many Western states. In his analysis, the term first appears in Chinese top political discourse in 1983, while the broader discourse only starts to shift from 'war and peace' to 'security and threat' in the 1990s (Liu, 2014: 26). Ling and Yang examine the evolution of Chinese thinking on security since the founding of the PRC. They recognise geographical variation in security concepts, pointing to key differences in the status and strength of a state, ideology and political system, culture, history, and geography. They suggest China's security concept has gone through three stages of evolution. First, from 1949-1978, a traditional concept dominates to emphasise political security and military methods in the face of internal and external military threats. Then from 1978 to 2012 a non-traditional security concept emerges, stressing 'economic security' (Ling and Yang, 2019: 9). The third phase emerges under Xi Jinping, with a new concept centring 'political security', defined as maintaining the status and power of the CCP and the CCP regime (Ling and Yang, 2019: 18).
Eurocentrism has shaped our understanding of security, but existing literature about China's concept of security gives us the beginnings of a different picture, suggesting that China may have a broader understanding of threat, pointing to the role of disorder and internal instability.

A brief conceptual history of security in China
Approaches to conceptual history have much in common: the very idea that concepts matter and that they change suggests that they are not neutral labels for transhistorical phenomena but political constructs. They also place importance on language. In IR, conceptual history is mostly associated with Skinner, though some studies inspired by Foucault touch similar ground (Leira, 2016: 29). 13 Here I use Walters' (2012) approach to tracing 'lines of descent': following 'the [multiple] pathways by which something significant and valued in the present came to take the form that it has' (p. 118), which draws on Foucault. This involves decomposing 'what otherwise appears integral and complete . . . to reveal that a final product is actually a hodgepodge of bits and pieces, each of which has its own history' (Walters, 2012: 118). Tracing lines of descent can be used to understand the history of a concept by developing 'detailed histories revealing when, where, under what circumstances . . . [it] came into being' (Walters, 2012: 118) and paying attention to 'subtle shifts in political logics ' (p. 122). This is a broader historical method that can also be used to understand phenomena other than concepts.
I also draw on Farr and Skinner's explication of concepts and conceptual history specifically. Understanding the histories of concepts is important because the concepts embedded within a political discourse constitute 'the beliefs and practices of political agents' (Ball et al., 1989: 1;Farr, 1989: 38). Crucially, conceptual change is a form of political innovation: 'to understand conceptual change is in large part to understand political change, and vice versa' (Farr, 1989: 24-25). Conceptual histories can help us understand why a particular concept emerges by showing what can be done with it that could not be done without it (Skinner, 2002: 178). Concepts do not simply represent a world that exists out there, they constitute meaning and political practice, shaping our political reality (Lund Petersen, 2011: 709). Conceptual histories tend to be interested in three things: where do particular concepts come from, when and how do they change, and what are the effects? Context is crucial, since concepts are situated in time and place (Skinner, 2002: 177).
My empirical focus is Chinese political discourse about security: how do political elites define and articulate the project of security and how has that changed over time? What logics does it rely on, how is it organised, what practices does it enable, and what are the political implications? What type of security is at stake here? Studying China, political discourse can tell us a lot about how particular leaders understand security: it is performed for an audience and has effects. Here I follow Sorace: . . . in both traditional and modern Chinese political theories and practices of statecraft, language has played a central role in the articulation and maintenance of political order. For the Chinese state, official discourse and terminology are not merely descriptive; they are also meant to be exemplary and normative, authoritative and binding. (2017: 7) To reflect this, Sorace (2017) takes Communist Party ideas and discourse as 'central to how that organisation formulates policies, defines legitimacy, and exerts its power' (p. 6). How party elites articulate security matters. Indeed, 'the Party extends vast resources on developing, maintaining, and advancing its discursive imaginary' (Sorace, 2017: 149). What matters is not so much whether people actually believe the ideology and ideas articulated, but that regardless of what people believe, ideology shapes how they behave and what they can say on an everyday basis (Sorace, 2017: 10). Here I focus on elite discourse about security, for an exploratory study examining how ordinary Chinese citizens experience state security, see Nyman (2021).
For purposes of transparency, I have included some discussion of choices made regarding translation. 14 I have written elsewhere about the concept of security in Mandarin and linguistic translation (Nyman, 2023a). In Mandarin, the word for security (安全, anquan) also means safe, or secure. In everyday use, it is often associated with crime or safety standards (such as in labour or manufacturing processes). The more specific national security (国家安全, guojia anquan) is the more commonly used term for discussing state-level policy, but does not appear in policy documents until the late 1970s. 15 The concept of security is also translated across time and space, and in the process, it transforms (Gulsah Capan and Grasten, 2021). To understand hybridity and adaptation in the Chinese case, it is worth noting that when concepts are imported into China they are often 'Sinified': 'given a specific meaning and understanding that work in (and arguably for) the Chinese political context' (Zeng et al., 2015: 246). This is also a form of conceptual innovation and is reflected in Xi Jinping's (post-2013) emphasis on security 'with Chinese characteristics'. The importance of adaptation is stressed through emphasis on the uniqueness of China's context and history in both policy discourse and in the academic debate (eg Yin, 2013).
I found three explicitly articulated concepts of security in China (to date), with the first emerging in the 1990s ('new security concept'), the second in the early 2000s ('comprehensive national security'), and the third under Xi Jinping ('total national security'). I begin by tracing their pre-history to understand their roots (Farr, 1989: 38) and then move on to unpack the three concepts in detail.

The pre-history of security in China: from order to economic growth
The contemporary concept of national security was imported into the Chinese context alongside other concepts like 'nation-state' and territorial/Westphalian notions of sovereignty (Shih, 1998;Vuori, 2014: 53), but it does not arrive in a vacuum. It takes on traces of past thinking as well as being 'Sinicized' to suit the contemporary political agenda of the leadership at the time. Here I identify four consecutive legitimating ideas -order, revolution, survival, and development -that drove policy in China before the emergence of explicit security language, performing a similar function to security today. Traces of each of these can be found in the contemporary concept and practice of security.
In ancient China, the question of order was the central problematique (Zhang, 2014: 174), making 'disorder' or chaos (luan, 乱) the central threat (Radtke, 2008: 204). This was underpinned by a different understanding of political legitimacy, based not on sovereignty within geographical boundaries (Shih, 1998: 125). Here there was no absolute outside, 'only relative degrees of promixity to the center' (Hevia, 1995: 23). Chinese philosophy stressed coexistence and relationalism, and thus 'maintaining harmony' was the central objective (Yu and Xie, 2015: 21;Yu, 2014: 29). Security did not centre on protecting the inside from an anarchic outside, instead, order and stability came from the centre. This challenges the foundational inside/outside distinction that underpins the Eurocentric concept of security.
The rise and eventual ascendancy of the Chinese Communist Party (CCP) turned this on its head. The CCP's driving goal was revolution and disorder was necessary for the formation of the new regime. Mao does not distinguish between external/internal threats, but between those who are for or against the revolution: 'Who are our enemies? Who are our friends? This is a question of the first importance for the revolution' (Mao, 1926). The two chief enemies are 'imperialism and feudalism, the bourgeoisie of the imperialist countries and the landlord class of our country' (Mao, 1939). This reflects a different notion of political community from the liberal-democratic states that form the basis of security theory: here boundaries are drawn on ideological rather than territorial lines, which shapes both the conception of threat and the referent object of security. Enemies and threats are defined in ideological terms, and in practice, internal threats often took precedence. The CCP reformulated the category of "the people" so that 'anyone who endorsed the socialist path belonged to the people, and anyone who opposed it became an "enemy of the people'" (Guan, 2019: 176). Centring "the people" also disperses agency and responsibility for spreading revolution, enabling mass mobilisation (Mao, 1934). This was fundamental to Mao's vision of the 'people's nation': he later called for 'public security committees' to assist the government in 'eliminating counter-revolutionaries, guarding against traitors and spies and safeguarding our national and public security' (Mao, 1951).
As the PRC became more established, the underpinning goal becomes the survival of the new republic. We see a territorialisation of ideological boundaries as China abandons international revolution to focus on surviving as a state in the international system, while continuing to eliminate ideological opposition domestically (Mao, 1955(Mao, , 1957. Here there are parallels with Western thinking, emphasising the balance of power and territorial integrity, but without reference to the term security. The new state and its 'democratic dictatorship' exists to suppress the 'internal enemy' and to protect the state from subversion and aggression by 'external enemies' (Mao, 1957). Internal and external enemies are often blurred, as domestic supporters of alternative politics become one with 'imperialist forces'. The key threats are more commonly grouped as 'imperialism, feudalism and bureaucrat-capitalism' (Mao, 1955), with the country divided into 'the people' and 'the enemies of the people' (Mao, 1957). These threats also legitimate a strong central state (Mao, 1957). There is explicit attention to adaptation and context, with Mao noting the need to integrate 'the universal truth of Marxism-Leninism with the concrete practice of the Chinese revolution' (Mao, 1956). This can also be seen in the institutional architecture: while parts of the structure of the Chinese security system are inherited from the Soviet Union, they operate differently, drawing on Chinese historical practice (Guo, 2012: 11).
The next big shift in thinking comes after Mao's death in 1976. As Deng Xiaoping takes over, he transforms China with Reform and Opening. This changes the narrative about China's international role and understandings of security and threat. Deng did not see war as imminent (Deng, 1977). Instead, China's security now depends on continued economic development, and this becomes the core legitimating idea and driver of policy. This still requires keeping an eye on 'destabilising factors' domestically (Deng, 1980b) and modernising the armed forces (Deng, 1981), but the legitimacy of the regime is now tied to its ability to continue to produce economic success. Economic success demands internal stability, legitimating a strengthening of the state (Deng, 1979b) and zero tolerance for 'proponents of erroneous tendencies' (Deng, 1979a). China cannot 'afford any more disorder or unrest. . .the key to China's development is political stability' (Deng, 1987). There is still talk of 'hostile forces' but these are now mostly domestic: 'counterrevolutionaries, anti-Party and anti-socialist elements and criminals' (Deng, 1980a). The priority is to protect stability and political unity, 'a sound, secure public order' (Deng, 1980b). Here earlier ideas about the ideological boundaries of political community are explicitly tied to order, suggesting political unity is central for security.
Explicit references to 'security' are rare but I have found a couple of early mentions. 16 In 1978, Deng argues that without economic development, 'our socialist political and economic system cannot be fully consolidated, and there can be no sure guarantee for the country's security (Deng, 1978). This is the first explicit reference to national security I encountered (original phrasing: '我们国家的安全/women guojia de anquan', 'our country's security' (Deng, 1994: 86)) -linked directly to economic development. A more familiar use appears in 1981, when Deng refers to 'intensified rivalry between the superpowers', as posing 'a serious threat to world peace and to our own national security' (Deng, 1981). Here the exact phrasing is '我国的安全/wo guo de anquan', our country's security (Deng, 1994: 395). Both of these uses of the concept security include the possessive particle 'de' between the words country and security, operating like the added possessive 's in the translation our country's security, indicating that security here is one abstract noun among many attached to the country (comparable with our country's economy, our country's people, our country's development). As the security discourse becomes more established, the possessive particle is dropped, indicating a formalisation of the concept. 1983 sees the establishment of the Ministry of State Security, which focuses primarily on intelligence and anti-espionage activities. This spurs increased use in the term national security in Chinese academic research, where at this point it had a narrow meaning, emphasising sovereignty, military security, and national defence.

National security emerges: three concepts
The CCP has advanced three different security concepts, and with each of these has made very clear that this is purposeful conceptual innovation. While the term national security is likely imported from Western discourse since it emerges later and without linguistic precedent, it is defined and used differently in China. The language around each concept stresses its 'Chineseness': that it has Chinese roots and/or is responding to specifically Chinese conditions and challenges.
The new security concept (1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000). 1989 sparked a series of changes in how the Chinese leadership conceptualised security. The Tiananmen protests represent a major challenge to party authority and the military crackdown leads to worsening relations with the outside world. At the same time, other Communist states are falling and no one knows quite what the post-bipolar world order will look like. China needs stability for economic development, which now underpins CCP rule. This is when security starts to emerge as an explicit political discourse.
Domestically, stability becomes the core term driving and legitimising policy, stressing both political and social stability. This echoes some of the earlier language about order: 'China must avoid chaos', and 'stability overrides everything' (Deng, 1989d(Deng, , 1989b. 'Social stability' requires both military defence and economic development (PRC Central Government, 1996), evoking discourse from the previous decade. This discourse is used to justify the Tiananmen crackdown and becomes deeply entrenched throughout the 1990s, legitimating an expansion of the domestic security apparatus to eliminate 'unrest' (Deng, 1989c). Wang and Minzner point to growing 'stability maintenance' (维稳, weiwen) policies in the 1990s, one of the results of which is a 'securitization of local governance in China'. This policy sees a wide range of previously unconnected areas, from ordinary civil disputes to environmental accidents, 'sucked into the weiwen vortex' (Wang and Minzner, 2015: 351).
Distinct from this internal emphasis on stability, a separate security discourse emerges centred on foreign policy. At the end of the 1990s, President Jiang Zemin launches a 'new security concept' (新安全观, xin anquan guan), formally embracing security as a political concept for the first time. The new security concept emphasises peaceful coexistence and cooperation with other states (PRC Central Government, 1998) to build 'common security' (Jiang, 2000). It also recognises emerging non-traditional security threats for the first time (PRC Central Government, 1998). Unlike the stability discourse, this concept has a dual audience and is also promoted externally in venues like the United Nations, supporting China's attempts to counter international concern about its rise. The concept represents the beginning of China's many attempts to shape international security discourse. This can also be seen in the title of the concept dropping the word 'national/ state', guojia: this is about shared security, not the interests of any individual nation. There are clear parallels here with attempts to shift towards broader notions of security in international discourse during the 1990s, including the UN's own human security concept.
Over the course of the 1990s these two discourses start to merge, reconnecting internal and external elements and integrating stability and security. Leaders reformulate security strategy to reflect the relationship between external security threats and domestic political stability. Referent objects of security become broader, and we see the concept stretched to permit more expansive action. Economic development and stability must be protected, but so must political independence and sovereignty. This is well illustrated in the 1993 State Security Law, said to be formulated for the purpose of safeguarding State security, protecting the State power of the people's democratic dictatorship and the socialist system, and ensuring the smooth progress of reform, opening-up, and the socialist modernization drive (PRC Central Government, 1993). In the beginning of the period, the key threat emphasised is domestic 'turmoil', or 'counter-revolutionary rebellion' (Deng, 1989a). By the late 1990s we see references to broader 'factors of instability' including international threats, from 'hegemonism' to military alliances, an unfair economic order, and non-traditional threats from terrorism to arms proliferation to smuggling, pollution, and refugees (PRC Central Government, 1998), reflecting wider global trends. Domestically, the 10th Five Year Plan also stresses social stability (PRC Central Government, 2001).
All of this enables a range of policies, including renewed mass mobilisation, with regular citizens as well as formal organisations from the armed forces and state organs to private and public enterprises tasked with a responsibility to 'safeguard the security, honour and interests of the State' (PRC Central Government, 1993). We also see a strengthening of national defence and the People's Armed Police in the face of internal and external threats (PRC Central Government, 2000).
Comprehensive national security (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012). The first decade of the new millennium is an era of complex new threats. The US-led Global War on Terror justifies ever-increasing security measures and in 2002-2003, the SARS epidemic sparks further expansion of the concept of security in China. Insecurity is said to be increasing (PRC Central Government, 2004), and in 2008 the Sichuan earthquake rocks confidence in party management. Responding to these challenges, the Hu Jintao/Wen Jiabao administration centres the concept of 'harmonious society', with clear parallels to earlier focus on stability. The period sees the emergence of a 'hybrid socialist-neoliberal form of political rationality' (Sigley, 2006) as governing and economic structures evolve. In an age of partial privatisation, maintaining security becomes a key function of the state.
The goal is now explicitly to protect 'national security', though this is sometimes paired with other terms, such as 'harmonious society', 'unity', or 'social stability'. The concept evolves and expands from the foreign policy-focused 'new security concept' to cover a much wider range of threats, captured by the new term 'comprehensive national security' (综合国家安全, zonghe guojia anquan). China now seeks a more 'comprehensive national security in the political, economic, military and social areas' (PRC Central Government, 2004) and non-traditional security threats grow in importance (FMPRC, 2002). This discursive shift has a few key features worth noting. First, the return of 'national'/guojia-reverting from common security, the new concept explicitly focuses on national/state security. Second, 'national security' is now one phrase, with the earlier possessive particle dropped, suggesting this is now a more established concept. Finally, the addition of the term 'comprehensive', zonghe, connotes a concept that is made up of several elements that are integrated and synthesised. Defence white papers still promote the 'new security concept' but expand it beyond foreign policy: it begins to blur lines between internal/external security and traditional/non-traditional threats (PRC Central Government, 2006b. The language of national security begins to creep into national Five Year Plans (PRC Central Government, 2006a: chapter 41 section 4).
The concept of security starts to become stretched to include a much wider range of threats: 'terrorist, separatist and extremist forces' emerge as a major focus (PRC Central Government, 2002), while SARS brings a focus on health and human security (see annual Defence White Papers from 2004 onwards). There are also a number of ongoing international challenges (PRC Central Government, 2004). Traditional/non-traditional security threats and domestic/international threats, are officially 'interwoven and interactive', and China faces 'strategic maneuvers and containment from the outside while having to face disruption and sabotage by separatist and hostile forces from the inside' (PRC Central Government, 2008). There may be a wider range of threats, but the thing to be protected remains the party-state (PRC Central Government, 2006a). The expanding security concept is mobilised to enable new policies that protect the party-state, entrenching party power. Enhanced public security is necessary for a 'stable social environment' to maintain national security and social stability (PRC Central Government, 2006a). During the SARS epidemic, the administration use emergency measures to centralise political power (Thornton, 2009: 40), echoing Mao-era measures including calling for a 'People's war' against SARS (Hanson, 2008; see also Gao, 2003). The scope of defence expands, with the military tasked with protecting against non-traditional threats at home (including 'stability maintenance') and abroad (PRC Central Government, 2010). The People's Armed Police also play a growing role against domestic threats (PRC Central Government, 2009: article 1).
The War on Terror is worth discussing in more detail, as it illustrates the messiness and complexity of hybridity and co-production well and shows the CCP's willingness to draw on security discourses emanating from elsewhere when it suits the interests of the Party. The CCP has faced ongoing challenges from domestic independence movements, with unrest in Xinjiang in particular growing in the 1990s and producing a crackdown in the form of an anti-crime campaign targeting 'splittism' (分裂主义, fenlie zhuyi) and 'unlawful religious activities'. After the 9/11 terror attacks, the Chinese state subsumes its struggle with domestic separatists into the global war on terror and 'Xinjiang separatists were renamed as terrorists virtually overnight', in turn providing post hoc legitimacy for China's crackdown in Xinjiang (see PRC Central Government, 2002;Vuori, in press: 150-154). China also retroactively re-labels the events of the 1990s as terrorism (see Pokalova, 2013: 288-289). The realignment of China's domestic discourse with the global War on Terror legitimised these policies both internationally and domestically. At the same time, China's understanding of terrorism remains conceptually distinct from that of the US, linking 'terrorism, separatism, and extremism' as 'three evils' to be combatted (Li, 2019: 312). China is not simply mimicking Western discourse, but innovating to suit the domestic agenda: using it to strengthen the domestic security apparatus while also legitimising the US-led war on terror, shaping global practice. Another example from the same period is human security. In the early 2000s, China adopts and adapts the UN's concept of human security, but in the Chinese version of the concept the state remains central as 'the key guarantor of human security, not a threat to it' (Breslin, 2015).
Total national security (2013-). The consolidation and expansion of the concept of security reaches new heights under Xi Jinping. When he comes into power in 2013, he explicitly places security at the heart of the political project. He creates a new top-level National Security Commission (NSC) to coordinate policy, and at the opening meeting, Xi-also head of the new Commission-announces a new national security concept: 总体国家安 全/zongti guojia anquan. This is often translated as 'comprehensive national security', but while accurate this translation can be both confusing and misleading, conflating it with the earlier security concept. The word translated as 'comprehensive' here is zongti, rather than zonghe which was used in the previous concept: these have overlaps in meaning but are not the same. Zongti connotes something that comprises the whole, or total, and could equally be translated as whole system, holistic, or total national security. 17 Here I translate Xi's concept as total national security to reflect this, and to distinguish it from the previous concept. In his speech, Xi declares that 'China's unique conditions and historical experience' require a different understanding of national security from the West, calling for an approach based on 'security with Chinese characteristics' (Renmin Ribao, 2014). In its original iteration the new concept has 11 listed official aspects, with political security as the foundation and then spanning issues as wide-ranging as cyber security, cultural security, and environmental security. At the time of writing, the concept has expanded to 16 areas. We are told that Xi's concept is both new and uniquely Chinese, '. . . permeated with the genes of Chinese traditional culture' (Xi, 2018). In practice, it is a hybrid concept, reflecting the expansion of security concepts in Western discourse in the 1990s and post-9/11 and building on conceptual developments within China in the 1990s and early 2000s.
Later we see the new concept enshrined in the constitution as one of the principles of 'Xi Jinping Thought' (Xinhua, 2018) and entrenched in a series of new security laws. Xi's vision of total national security is vague and open-ended, defined in his new National Security Law as . . . the relative absence of international or domestic threats to the state's power to govern, sovereignty, unity and territorial integrity, the welfare of the people, sustainable economic and social development, and other major national interests, and the ability to ensure a continued state of security. (PRC Central Government, 2015b: article 2) Liu Yuejin, a key thinker on the new concept, argues that the leadership headed by Xi Jinping pays more attention to national security than any previous administration (Liu, 2017), and use of the term in key speeches increases exponentially (Xu, 2022) (see also Five Year Plans of the period). 18 Under Xi Jinping, security becomes the central organising political discourse. Where before the concept had some (albeit permeable) boundaries, under Xi Jinping we see a form of 'hypersecuritization' where security discourse proliferates and the notions of threat and risk expand and extend to a point where anything, anywhere, at any time could be considered threatening.
The logic underpinning Xi's vision of total national security has three key features that explicitly distinguish it from previous Chinese concepts. First, while there are 16 aspects of security these are not equal: political security is the foundation for national security, and sits 'at the core of the national security system' (Chen, 2019). This is really about the security of the CCP as the ruling party and the system they have built (Gao, 2020), and in practice blurs the Western concepts of state security and regime security (see You, 2016: 179). Here there are parallels with earlier delineation of political community in ideological terms. Political security is national security, because only those who support the political project of the CCP are part of the political community. This also explains the ongoing domestic focus (see Chen, 2019;Gao, 2020). Second, the concept itself is based on intertwined relational concepts that cannot be analytically separated, including 'external and internal security', 'traditional and non-traditional security' (Gao, 2020;PRC Central Government, 2015b:: article 8). In the same way, security and economic development are intertwined (Gao, 2020;Xi, 2018), threats and risks become inseparable (Xi, 2019;Xinhua, 2015b). Third, we see a shift towards a proactive logic of security that centres risk prevention (PRC Central Government, 2015b: article 9), with relevant departments given the task to 'prevent problems before they happen' (Gao, 2020). The concept creates a permanent sense of crisis that requires constant vigilance, preparation and planning for foreseeable and unforeseeable risks and threats. It emphasises 'continuous response' (Xi, 2018): rather than countering specific threats that emerge at particular moments, it creates an institutionalised permanent securitization of everything that shapes and permeates everyday life. The concept pushes officials to be proactive about spotting threats, replacing older 'stability maintenance'/weiwen with new discourse centred on fangkong, prevention and control (Greitens, 2022: 4). This contrasts the risk management system of the previous decade, which centred reactive emergency response (see Zhang and Zhong, 2010).
The logic of risk produces an ever-expanding list of threats, mirroring the speculative security culture centred on scripting disaster that emerged in post-9/11 Europe and the US (de Goede et al., 2014). Traditional military threats remain, with insecure borders, terrorism, separatism, fears of 'anti-China forces' pushing a 'color revolution' (PRC Central Government, 2015a), while non-traditional security issues such as 'cultural security, economic security, and network security also pose new challenges.'.. (Xi, 2017). Widely different risks intertwine, raising the potential for 'risk chain reactions' with 'the transformation of economic and financial risks into political and social risks, the risks of cyberspace into real social risks, the risks of external struggles into internal security and stability risks.'.. (Chen, 2019).
Security becomes a form of governance infiltrating every level of policy-making. The new concept has provided a framework for assessing threats, creating National Security Commissions at every level from national down to village committees, entrenching the practice and centralising power and control. It has also resulted in an expansive new legal framework covering a wide range of issues and areas. The clearest illustration of this shift is the replacement of the 1993 State Security Law, which focused primarily on espionage, with the 2015 National Security Law, which effectively covers almost every aspect of public life: from politics, defence, finance, environment, cyberspace to culture and religion (Xinhua, 2015a). Alongside centralisation of power we also see increasing mass mobilisation and participation. Responsibility over security work is legally obligated for regular citizens as well as 'all State organs, armed forces, political parties and public groups, and all enterprises and organizations' (PRC Central Government, 2014). Cadres at all levels are obligated to study national security, while new security volunteer groups have formed to patrol and monitor citizen behaviour. By 2017, such volunteers numbered 850,000 in Beijing alone (Guo, 2017). Party officials in charge of national security are embedded in nominally private corporations. Meanwhile, Xi has strengthened CCP control over the military, and the armed forces are increasingly tasked with duties that go far beyond their traditional roles to 'firmly maintain social stability' and uphold the CCP's ruling position (PRC Central Government, 2015a). It is worth noting here that the armed forces, unlike in most political systems, are under the jurisdiction of the party, not the state. The fundamental principle underpinning national security work is to 'uphold the party's absolute leadership ' (Xi, 2018).
Understanding conceptual change as political change, these shifts enable growing authoritarianism, shoring up and increasing the power of the state, legitimating pre-emptive exercise of state power. They consolidate the position of Xi Jinping, who after constitutional changes faces no term limits. The language of national security is also used to legitimate these policies internationally, a form of postcolonial mimicry where what emerges is 'the same but not quite': most notably in ongoing justifications for extra-legal incarceration of ethnic minorities in Xinjiang in the name of national security. We can also see the shift illustrated in China's COVID-zero policy, which relied on a sense of crisis to extend the infrastructure of control. Many states have enforced lockdowns and travel-bans, but China's reliance on a combination of traditional forms of monitoring through mobilising neighbourhood party committees to police local residents, with hightech methods such as QR health codes to monitor movement, created a more total lockdown and a demonstration of party power that cannot be compared to the lockdowns seen in most other states. There are also significant differences with China's response to SARS. This also mobilised a crisis mentality and response, but technological developments and an enhanced infrastructure of control put in place during Xi Jinping's decade in power in the name of total security have enabled a much stronger response. Xi's concept and the practices that it enables differ both from the Eurocentric concept and from previous Chinese concepts.

Almost the same, but not quite: similarity and difference
China's security concept has multiple lines of descent: the term is imported from the West and in early periods we see elements of mimicry and postcolonial learning, but as security is 'vernacularised' (Tickner, 2003: 306) it acquires both old and new meaning. Today, China's security concept retains the seeds of its multiple origins: a fear of disorder, a Maoist focus on the party-state, Western cold-war thinking, and more recent broadening to take in new threats in both Chinese and Western theorising and practice. Mimicry hides difference: at first glance, Xi's national security discourse appears similar to the concept presented in ISS. The term 'national security' is invoked to enable and legitimate the exercise of state power and stresses a heightened vulnerability and a growing sense of threat that is often seen in other states' discourse today. But taking postcolonial insights seriously reveals hybridity, a complex interplay of similarity and difference. Some shifts in China's concept of security occur in parallel with similar changes elsewhere: a broadening of the notion of threat in the 1990s and early 2000s, and later the growing entrenchment of risk analysis in security thinking. Yet the discipline's understanding of these conceptual shifts is based on EuroAmerican experience, and the failure to consider how experience in other states shapes the evolution of the concept of security overlooks other triggers for change. For example, in China SARS was perhaps the biggest catalyst producing an expanded notion of threat.
Other aspects show significant difference. Chinese elites are not simply merchants of EuroAmerican notions of national security, but actively innovating to support their own interests. They pursue purposeful and explicit conceptual innovation, drawing on China's particular branch of Maoist-Marxist-Leninism that stresses continuous theoretical development and adaptation to Chinese conditions. They continually insist on the uniqueness of China's experience while leaving the precise nature of that uniqueness unclear. This in turn serves to support nationalist discourse and the idea of 'socialism with Chinese characteristics', both sources of authority for the party, signalling that 'Western institutions, definitions, and practices cannot be imported into China because they will fail to take root' (Smith, 2019: 66). This allows rejection or revision of things which do not 'fit' -or serve the interests of the party-state. Security 'with Chinese characteristics' has two core differences to the Eurocentric concept of security that survive throughout the period studied here.
First, although China uses the term 'national security' and at times talks about the survival of the state, the underpinning referent object of security is not the state. Instead, the thing to be protected is the party-state. Liberal-democratic states that form the basis of theorising about security are based on a clear separation between the state and the government: national security is about securing the state rather than any particular government. In contrast, China is a party-state, with the CCP controlling every branch of government, the armed forces, and the judiciary. National security here means political security, maintaining the status and power of the CCP and the system they have built. The basic unit is not the territorial state and all those who reside within it, but the party-state and its supporters. Thus we are talking about party or regime survival rather than the survival of a state. 19 This has a profound effect on security policy, enabling a wider and more authoritarian range of policies and control, and shaping a more expansive understanding of threat: which includes citizens who do not support the party-state, in turn enabling the persecution of dissidents. Even where a more 'conventional' understanding of security that stresses military threats is used, this operates differently because the People's Liberation Army is a party organ that exists to protect the CCP, not the state. This is a concept of security not found in the Western security literature, and reveals the importance of both ideology and political system in shaping the concept of security. There are likely parallels here to other revolutionary states, but also with some postcolonial semi-authoritarian states (see for instance Munoriyarwa, 2022: 3).
Second, as can be seen throughout the period traced here, China's security concepts do not clearly distinguish between internal and external threat. In Western theorising and practice national security has conventionally referred to external threats. 20 Although there are similarities with Western shifts in the post-9/11 era, this blurring between internal and external dates back much further in China, seen in imperial legacies as well as Maoist practice. Chinese thinking was not founded on a Westphalian notion of geographical sovereignty where inside/outside is the most meaningful distinction. In the early revolutionary period, ideology became the core distinction: the threat was anyone who was against the revolution. As the republic becomes consolidated, the threat evolves to become anyone against the regime or party-state. Throughout the period studied here, 'inside/outside' is less important for understanding threat, and in Xi's more recent discourse the impossibility of separating foreign and domestic threats and challenges is a central feature. Through most of the century covered here internal threats are considered more significant, and while the terminology changes political dissidents remain a central threat throughout. This has a clear impact on policy, shaping resource allocation: since 2011 China's domestic security budget exceeds the budget for external defence (Guo, 2012: 445). The People's Armed Police, which has duties centred on internal security, has also been restructured and incorporated under the Central Military Commission, strengthening party control. Consequently, analysing security in China through theory based on Western territoriality (see also Li, 2022) gives only a partial picture of what is happening. This shows the importance of history and culture in shaping the concept of security.

Conclusion
Despite growing acceptance that theorising about security is Eurocentric, the discipline has struggled to move beyond critiques of Eurocentrism to examine the concept of security 'elsewhere'. Writing an alternative history of security using China as a starting point shows a hybrid concept that is 'almost the same, but not quite'. Eurocentric theorising has obscured difference, and as a result security studies has underestimated the impact of history, culture, ideology and political system on the concept of security. It's hard to generalise from a single case study, but the differences present in China's concept of security suggest the universal model that underpins most current theorising is at best partial. Therefore, it has implications for theorising security, since we need to recognise the contingent nature of the concept of security. It also has implications for understanding policy: understanding security in China gives insight into how and why Chinese leaders might act in particular ways (Liu, 2012: 74). The existing literature on Chinese security politics is based on a Eurocentric concept of security and so largely emphasises external threats to the state. In contrast, this conceptual history shows that China's concept of security blurs inside/outside, producing an expansive and vague notion of threat/s. Over time it has institutionalised a permanent securitization of everything that shapes and permeates everyday life. This in turn enables an expansive and pre-emptive security policy that supports the CCP and protects the party-state. Total national security requires a strong state and centralised power, relying on a sense of chaos just beyond and in the hypothetical future, reinforcing the power of the state and legitimating the current political order.
Where does this leave us? Rather than suggest that China is uniquely different, I posit that security everywhere is differently different. Eurocentrism, hybridity, and mimicry have served to obscure fundamental differences in how security is understood and how it operates in different places. Accounting for history, culture, and ideology reveals the limits of universal theorising. Western states share similarities in terms of their approaches to security, but these similarities do not necessarily represent something universal in the concept of security: it may simply reflect similar history, culture, and political systems. There may also be more differences between Western states than we usually recognise. This has implications for the future of security studies. There is much space here for empirically grounded research that reflects on the particular, rather than the universal conceptualisation and manifestation of security. This might focus on comparative trends, like comparing security concepts in revolutionary states, or within sub-regions with connected history and culture. It could also consider how non-Western states shape global or local security practice, drawing out the complex interconnections between states in the Global South or between the Global South and the North. In the case of China one particularly under-researched area is how Chinese theory and practice has influenced approaches to security elsewhere. For example, Maoism has shaped revolutionary movements across the world, from the Middle East to South America. Meanwhile, we are only just beginning to understand the impacts of the spread of Chinese security technologies abroad (Munoriyarwa, 2022). Finally, the changes implemented under Xi Jinping have reshaped Chinese domestic politics, but in April 2022 Xi proposed a new Global Security Initiative (quanqiu anquan changyi) that may indicate interest in expanding Chinese thinking on security abroad. The Western concept of security has been universalised: the task is now to provincialize it, by taking 'other' places seriously on their own terms.