Reconceptualizing regional order: a critical/scientific realist (CR/SR) intervention

ABSTRACT This article begins from the observation that while the problem of international order is an issue of crucial importance to International Relations (IR), the very concept of “order” is still mainly treated in terms of patterns and regularities that sustain rule-governed interactions among states, thus failing to bring IR in closer contact with the more general discussion on “social order” in the social sciences. Written from the “emergentist systemic” perspective, this article attempts to develop a conceptual framework for the systematic analysis of regional orders centering on characteristics common to all societal spheres. The argument is developed in three steps. First, I focus on the system’s emergent properties and the process of self-organization to account for the dynamics of the emergence, the maintenance, and the submergence of order. Second, I discuss levels of social process and, drawing on the work of Brante, I introduce the “component – structure – context” procedure to account for the vertical differentiation of order. Third, drawing on Archer’s and Bhaskar’s theories, I attempt to construct “culture-inclusive” as well as “history-inclusive” account of order. I conclude with some suggestions as to how the proposed framework might contribute to the ongoing debate on order transition in East Asia.


Introduction
Critical Realism (CR) is a movement in the philosophy of social science starting from Roy Bhaskar's writings. Bhaskar originally called his general philosophy of science "transcendental realism" and his philosophy of social science "critical naturalism." 1 According to Collier, the term "critical realism" arose by elision of these two phrases. 2 Scientific Realism (SR), on the other hand, is a comprehensive scientific outlook that goes well beyond Bhaskar's transcendental realism, with some of its proponents believing that SR can provide an adequate account of social scientific practice. Among those committed to the role SR can play in the social sciences, proponents of Mario Bunge's systemism 3 and proponents of complexity theory 4 must be mentioned.
Although the philosophy of science makes no direct contribution to the accumulation of knowledge in the field of International Relations (IR), it is difficult to imagine the research having taken the form it has, if it had not been informed first by positivism, and later by social constructionism. 5 In addition, in recent years, a group of IR scholars have proposed academic agenda supported by realist philosophical ideas, making significant contributions to the field. 6 Following the "realist philosophy → realist-informed IR theory → actual research project (here, revisiting the concept of 'regional order')" sequence, this article attempts to clarify how realist approach to IR studies might work while applied to a specific research project. An account of regional order pursued in this paper seeks to weave into a single narrative the Bhaskar-inspired critical realist account of social reality, Bunge's systemic approach, and a range of insights from complexity theory. To borrow Wan's term, the theoretical approach to regional order proposed in this paper can be most accurately described as an "emergentist systemism." 7

Social system and its elements
Dictionary of Critical Realism defines social system as a "collection of social entities that form a whole, the behavior of which depends on the relations between the entities more than on the nature of the entities themselves." 8 According to Bunge, every material (=concrete) system may be modeled as comprising the following elements: • Composition = collection of all parts of a system 9 • Environment = collection of items, other than parts of a system, that act on or are acted upon by some or all components of this system 3 See, for example, Bunge, Finding Philosophy; Pickel, The Problem of Order; and Wan, Reframing the Social. 4 See, for example, Reed and Harvey, "The New Science," 356-79; DeLanda, A New Philosophy; and Byrne and Callaghan, Complexity Theory. 5 Positivism, scientific realism, and social constructionism are philosophies of science that build on different ontologies and epistemologies. Both scientific realism and social constructionism provide ways of thinking about the practice of science that runs counter to the positivist approaches, and, as such, are frequently covered under the umbrella term "post-positivism." For an account of the differences between those three approaches with regard to the social sciences, see Cruickshank, "Positioning Positivism," 71-82. Also, it must be noted that Wendtian social constructivism is an approach in IR which links positivist-informed approaches (e.g. neorealism, neoliberal institutionalism) with subjective and ideational ontology of social constructionism. In Wendts's own words: "Epistemologically I have sided with positivists . . . on ontology, which is to my mind the more important issue . . . I will side with post-positivists" (Wendt, Social Theory, 90). 6 See, for example, Joseph, Hegemony; Wight, Agents, Structures; Kurki, Causation; Patomaki, The Political Economy; and Wight and Joseph,Scientific Realism. 7 Wan, Reframing the Social. 8 Hartwig,Dictionary,451. 9 Throughout this paper, the terms "parts," "constituents," and "components" (of a system) are used interchangeably.
• Structure = collection of relations among the components of a system (=endostructure) or among these and items in its environment (=exostructure) • Mechanism 10 = collection of processes in a given system that make it behave the way it does. 11 Thus, in Bunge's CESM model, system is a concrete thing, whereas its structure is a set of its internal and external relations, and its mechanism is a process in the system.
Alternative ontologies offer different typologies of systems. Since this paper argues in favor of emergentist materialist ontology, a typology of systems offered by Mario Bunge will serve as a starting point for further reasoning. Bunge classifies systems into three broad types: • Material (=concrete) systems (further subdivided into natural (e.g. a nervous system), social (e.g. a family), and technical systems (e.g. a TV network)) • Conceptual systems (defined as one composed of concepts (e.g. a scientific theory)) • Semiotic systems 12 (defined as one composed of signs (e.g. a language)). 13 In Bunge's classification, East Asian Regional System (EARS) falls clearly under the "social systems" category. However, there is one important caveat to entera warning against Bunge's strict materialism. Bunge defines system as "a complex object whose parts or components are held together by bonds of some kind." 14 In the case of a conceptual system these bonds are LOGICAL, in the case of a concrete system they are MATERIAL. Further, Bunge maintains that, unlike material systems, "the conceptual and semiotic systems have compositions, environments and structures, but no mechanisms. The reason is that changeability (or energy) is the defining property of matter." 15 Proponents of CR, however, call into question Bunge's proposition, arguing that conceptual and semiotic systems not only have mechanisms but can also be mechanisms (i.e. processes in concrete systems). Wight's analysis of political participation is useful for my purpose. Doubtless, the mechanism of political participation is crucial to our understanding of democratic system (=a concrete system). But can it be fully 10 Examples of social mechanisms include the following: cooperation, competition, exploitation, exclusion (Kaidesoja,"Bhaskar and Bunge,; securitization (Guzzini, "Securitization," 329), brokerage (Tilly, "Mechanisms," 26); recognition (Kessler and Herborth,"Recognition," 155); mimicking, emulation, localization (of ideas) (Acharya, "Ideas," 195). 11 Bunge,Emergence and Convergence,35. 12 An in-depth discussion of the role of semiosis in the emergence, the maintenance, and the submergence of social order is beyond the scope of this paper. Suffice to say that this study situates semiotic analysis within the morphogenesis of culture (see Section 6) and pays particular attention to the concepts of "discourse" and "social practices." In their excellent study of cultural political economy (CPE), Sum and Jessop distinguish three levels of semiotic analysis: (1) semantics, (2) social practices, and (3) discourse. In their model, discourse refers to practices of "meaning-making" at the linguistic level. Social practices, on the other hand, can be defined as actors' meaningful actions that have both discursive and practical (=material) aspects (Sum and Jessop,Towards a Cultural Political Economy,. With regard to the East Asian Regional System (EARS), semiotic analysis can be applied to a range of issues concerning the diffusion of ideas across the region. For example, the latest discourse on the "revival of the Sinocentric regional order" and how it has been adopted, adjusted and altered in dialogue with a range of political discourses across East Asia can be fruitfully investigated through the lens of semiotic analysis. 13 Bunge, Emergence and Convergence, 33-4; and Bunge, "Clarifying Some Misunderstanding," 372. 14 Bunge,"How Does it Work?" 188. 15 Ibid., explained without referring to its conceptual space (e.g. the ideas possessed by agents) and its semiotic space (e.g. language as a means of communication)? 16 From the above it is clear that conceptual systems (expressed through various semiotic systems) are an integral (though not material) part of the social world and, as such, play an important role in social explanation. Concepts and ideas possessed by agents as well as icon and indexes incorporated by these concepts are part of the causal complexes that generate changes in a given social system and can thus constitute a system's mechanisms.
Turning now to the issue of the (alleged) absence of mechanisms in conceptual systems. Archer has challenged this part of Bunge's definition, arguing that certain logical relations between a set of ideas (e.g. relations of contradiction and complementarity) do constitute mechanisms in conceptual systems. For example, when two opposed discourses (=a relation of contradiction) both legitimate (=maintain) and challenge (=transform) a given social form, such a logical relation of contradiction certainly has causal capacities to generate change and can thus operate as a mechanism in conceptual systems. 17 Overall, the important point to keep in mind here is that an account of regional system/order pursued in this paper is based on a double premise. First, conceptual and/ or semiotic systems have mechanisms. Second, conceptual and/or semiotic systems can be part of the causal complexes that generate changes in a given social system and can therefore be conceived of as a system's mechanisms.

The problem of collective agency
In social theory, "agency" tends to be associated with social (=human) action, while "structure" with patterned relations and with constraints upon action. One of the most important developments in recent social theory has been the move towards agencystructure integration. Social theorists such as Giddens, Archer, and Bourdieu developed theories that attempt to bring together both agency and structure into one integrated scheme. 18 For critical realists (e.g. Archer's morphogenetic approach), agency and structure interact and these social interactions are cyclical phenomena. Preexisting structures emerge from a prior cycle and govern subsequent social interactions. Agents interact with these preexisting structures, so that the structures undergo change: they are reproduced or transformed. Such conception of an agentstructure relationship is often termed "analytical dualism." 19 The term "agent" is generally understood to mean a human actor with the capacity to act, while "agency" denotes the exercise or manifestation of this capacity. Like Hindess, I argue that there are two main types of actors: individual human actors 16 Wight, "Theorizing the Mechanisms," 296. 17 Archer,"Introduction,[8][9][10] For an exhaustive discussion on different concepts of agent-structure relationship, see Porpora,"Four Concepts,[195][196][197][198][199][200][201][202][203][204][205][206][207][208][209][210][211]. An in-depth analysis of the agent-structure problematique in IR is offered by Wight, Agents, Structures. For a critique of the application of the agency-structure dyad in IR studies, see Tang,"International System, In Archer's own words: "My 'analytical dualism' forms part of the explanatory program, it is a very useful contrivance for practical theorists attempting to explain something in particular, i.e. break it up into analytical phases to explain the problem in hand, but it is nevereven temporallyontological or philosophical dualism" (Archer, "Introduction," 10). and social (=collective) actors. 20 The latter includes central and local government departments, NGOs, political parties, trade unions, private firms, etc. As noted by Sibeon, social actors are characterized by mechanisms of resource allocation and decisionmaking, as well as formal networks of positions and informal patterns of norms and communication, thus being capable of engaging in decisions and actions. 21 The main point of contention with regard to collective agency is whether it can be reduced to agency of individuals involved (put differently, whether collective agents possess causal powers on their own). Consistent with Petit's model of collective agency, this paper argues in favor of granting agency status to social actors. 22 As is the case with every higher order social system (e.g. a region), it is the decisions of various collective agents that shape the social, economic, and political reality of the system, thus constituting one of the focal points of the system/order analysis. In addition, with regard to international system, the presumed agential status of the state is an issue of great importance to this study.
In the context of the postwar behavioral revolution in the US academia, the task of IR theory was not to theorize the state agency, but rather the patterns of behavior of state units in the interstate system. 23 Within constructivist scholarship, in the early 2000s, Wendt argued in favor of a non-reductive, anthropomorphic theory of the state, attributing to the state those properties possessed by persons. 24 A non-reductive, non-anthropomorphic alternative found in CR works is best articulated by Jessop's "strategicrelational approach" (SRA). As Jessop explains, the SRA derives its self-designation from its re-specification of structure and agency in relational terms. It highlights the importance of the strategic context of action and the transformative power of action. In these terms, structure consists in differential constraints and opportunities that vary by agent; agency in turn depends on strategic capacities that vary by structure as well as according to the actors involved. 25 Thus, in Jessop's model, state power results from both collective actors' strategies towards the state and from the "strategic selectivity" that is present in the relations that constitute the state.
Overall, this section lends support to the argument that the state is not "a person." Instead, it argues in favor of the state being envisaged as a collection of strategicrelational social entities, which, at a given time and space, participate in the emergence, the maintenance, and the submergence of the regional order. 20 Hindess, "Actors," 115. 21 Sibeon, Rethinking Social Theory, 120. 22 Petit's model of granting agency status to collective actors proceeds in three steps. First, Petit posits that a social entity will count as properly agential when it exhibits, in a wide range of contexts, a purposive-representational pattern. In a second step, Petit shows how this framework could be applied to "fully transparent" social entities, that is, groups where all its members have full and equal awareness of the collective goals. The last step generalizes what has been said about fully transparent groups to "less transparent" ones. The main line of the argument is that once the agential status of fully transparent groups has been granted, it is natural to extend the claim to cases where transparency is reduced (Petit, "The Reality," 68-91). 23 See, for example, Waltz, Man, the State, and War; and Singer, "The Level-of-Analysis Problem," 77-92. 24 Wendt, "The State as Person," 289-316. 25 Jessop, The State, 54.

Social system and complexity theory
Complexity theory assumes that human life exists within a context of systems that are complexthat is, systems characterized by nonlinearity, emergent properties, hierarchy, self-organization, and openness. 26 The aforementioned fundamental features of complex systems are dealt with in detail in the following subsections.

Emergentism
Central to the theory of emergence is the thought that as systems acquire increasingly higher degrees of organizational complexity they begin to exhibit novel properties that in some sense transcend the properties of their constituent parts and behave in ways that cannot be predicted from the properties found at lower levels. 27 Thompson points out that the exploration of the concept of emergence is a part of a broader "reductionism vs. emergentism" discussion and that both positions have ontological and epistemological components: • Ontological reductionism = the relations between the parts of the system are all reducible to the intrinsic properties of the system's most basic parts • Epistemological reductionism = the best understanding of a system is to be found at the level of structures and laws of its constituents • Ontological emergentism = a whole is more than the sum of its parts and their intrinsic properties • Epistemological emergentism = the best understanding of a system is to be found at the level of structures and laws of the whole system. 28 Proponents of ontological emergence consider emergent properties to be objective features of the world, their emergent status being independent of human existence and knowledge. 29 For an emergentist social ontology to be successful, it must thus offer conceptually clear and consistent definitions of the concepts of "emergent property" and "emergence." While a variety of definitions have been suggested, in this study a definition offered by Bunge will be deployed: A property of a complex object is said to be emergent if neither of the constituents or precursors of the object possesses it . . . Emergence [in turn] is a process in which new entities with emergent properties arise. 30 Since every "complex object" is either a system or a system component, it follows logically that emergent properties of "complex objects" are always properties of some system. 31 26 For a concise introduction to the "complexity turn" in the social sciences, see Urry, "The Complexity Turn," 1-14. On the application of different strands of complex systems thinking to international politics, see Gunitsky,"Complexity, On the development of the concept of emergence and its various uses in the philosophy of science, see Goldstein, "Emergence," 49-72. 28 Thompson,"Emergence and Complex Systems,. Throughout this article, the focus is on ontological approaches to emergence. 29 Humphreys, "Emergence," 191. 30 Bunge,Emergence and Convergence, Ibid., 71.
In the case of complex systems, the ideas usually associated with the phenomenon of emergence include the following statements.
• New entities with emergent properties arise in systems whose constituents have nonlinear interactions.
Linear interactions are additive and predictable. They generate "aggregative" properties that are analytically reducible to the properties of their constituents (e.g. mass). Nonlinear interactions, on the other hand, are nonadditive. They produce "emergent" properties that depend upon the mode of organization of the parts, thus being "something more" than a collection of these parts. As noted by Wimsatt, a wide variety of properties are emergent, whereas aggregative properties are rather rare. Emergent properties thus play a central role in understanding how (social) systems work. 32 • New entities with emergent properties bring into the world new causal powers 33 in their own right.
Higher level entities and their emergent properties are causally effective in their own right. 34 Examples of emergent properties of social systems include relational structures, institutions, organizations, etc. For social structures to be explained as emergent properties, we need to identify social entities that possess these emergent properties, the mechanisms by which these emergent properties were produced, and the constituents of entities that interacted in these mechanisms. Certainly, some further clarification of the above claim is needed.
First, I must distinguish between social structures, institutions, and organizations. This paper will use definitions proposed by Geoffrey Hodgson who defines structures (in a manner similar to Bunge) as "a set of social relations." The term "institutions" refers to "systems of established and embedded social rules that structure social interactions," while the term "organizations" is used to denote "special institutions that 32 Wimsatt,"Reductionism," 448,460. 33 There are many ways to think about causality. For example, Brady distinguishes between four distinct approaches to causal inference: (1) the Neo-Humean (=Hempelian) regularity theory, (2) the counterfactual theory, (3) the manipulation theory, and (4) the mechanisms/capacities theory associated with realist tradition (Brady,"Causation,. Consistent with Hempels's deductive-nomological model, positivists understand causality as the constant conjunction of events, whereby empirical event A is said to cause empirical event B. As noted by Guzzini, positivist approaches to IR tend to reduce mechanisms to a sequence of variables, leaving them purely descriptive and thus depriving them of their causal efficacy (Guzzini,"Securitization," 332). Realists, on the other hand, look beyond empirical outcomes such as "if A, then B" and instead seek to identify the underlying generative mechanisms that lead from the cause to the effect. 34 The issue here is whether social structures possess causal powers that are ontologically independent of the causal powers of people (=agents) that reproduce and transform these particular structures. Bhaskar's notion of "causal powers of generative structures" has been heavily criticized by many realists. For example, Sibeon talks about "the illicit attribution of agency to entities that are not actors or agents" (Sibeon, Rethinking Social Theory, 4). To my understanding, however, it does not mean that social structures fall completely outside the concept of causality.
Wright's analysis of working class power is useful for my purpose. Wright distinguishes between "structural power" (=power derived from workers LOCATION within the economic system) and "associational power" (=power that results from the formation of collective organizations of workers) (Wright, "Working-Class Power," 962). Then, it is correct to argue that social entities (=collective organizations of workers) do possess emergent causal powers, while social structures (=workers location within the economic system) involved in this specific type of relations both enable and constrain social actions taken by those social entities. To sum up, I concur with Sibeon's view that structures have no intentional causal powers (=structures cannot exhibit agency), but structures have causal capacities in the sense of conditioning agency and social action (Sibeon, Rethinking Social Theory, 148).
involve (1) criteria to establish their boundaries and to distinguish their members from nonmembers, (2) principles of sovereignty concerning who is in charge, and (3) chains of command delineating responsibilities within the organization." 35 From the aforementioned description, it is clear that organizations are a subset of institutions, and institutions are a subset of social structures. Second, I must clarify what is meant by "social entity." The term "social entity" refers to a group of people that remain in a structured set of relationships with each other. 36 For example, a multilateral development bank (MDB) (understood as a social system) typically includes such social entities as board of governors, board of directors, senior management, and members of advisory panels. It may also include human artefacts, such as bank headquarters buildings.
Lastly, I would like to touch upon the mechanismic explanation issue. Put simply, the mechanism-based explanation associated with SR/CR is a form of causal inference that attempts to establish how an observed social phenomenon is brought about, and the way causes are linked to effects via causal mechanisms. 37 • New entities with emergent properties exert causal effects on lower level entities from which they emerged (this idea is called "downward causation"). 38 From the vantage point of systems theory, new entities with emergent properties can be viewed: (1) as complex systems possessing causal powers in their own right and, at the same time, (2) as being composed of lower level entities with their own causal powers. With regard to the relations between the causal powers of the emergent entities and those of their components, the concept of "downward causation" is frequently raised. For example, Campbell and Bickhard believe that "anyone who maintains that there are emergent properties is committed to the possibility of downward causation." 39 But what does it exactly mean "to cause downwardly?" In the context of complex systems theory, downward causation corresponds to the influence the system's topological organization has on the behaviour of its components. 40 Applied to the social sphere, top-down causation corresponds to the influence the configuration (=topology) of the system's emerging properties has on its constituent parts. To put it in a slightly different way, the configuration of the system's emerging properties limits certain possible behaviours of the system's components, while simultaneously opening up new possibilities by virtue of the properties possessed by the system as a whole. 35 Hodgson, "Institutions," 96. 36 Elder-Vass, "Top-Down Causation," 86. 37 Throughout this paper, the terms "causal mechanisms" and "generative mechanisms" are used interchangeably. 38 The term "downward causation" comes from the American psychologist Donald T. Campbell ("Downward Causation,. In the philosophy of social science, the concepts of "downward causation" and "top-down causation" are used interchangeably (see, for example, Elder-Vass, "Top-Down Causation," 82). 39 Campbell and Bickhard, "Physicalism," 42. It must be noted, however, that proponents of SR/CR disagree over the usefulness of the term. For example, Bunge refutes it, Lawson views it as problematic, whereas the term is implicit in Archer's morphogenetic approach, and Elder-Vass deploys it extensively (Bunge, "Clarifying Some Misunderstandings," 373-4; Lawson, "Emergence and Morphogenesis," 79-84; Archer, Realist Social Theory, 14-5; and Elder-Vass, The Causal Power, 58-62). Throughout this paper, I retain the term "downward causation" to capture the global-to-local influence of a system over its components. 40 Thompson, "Emergence and Complex Systems," 68.
• The concept of emergence has synchronic and diachronic connotations. 41 The appearance of emergent properties accompanied by the process of downward causation is frequently referred to as "synchronic emergence"the emergent properties of a whole "cooccur" with the properties of its parts. In other words, the concept of synchronic emergence refers to the part-whole relation that prevails in a particular social system in a certain instant of time. "Diachronic emergence," on the other hand, refers to the historical sequence of systems, when the "old" system gives rise to the "new" system. Put differently, through the process of diachronic emergence new systems with emergent properties are historically formed.

Hierarchy
Although all social systems are ultimately composed of individuals and their artifacts, they also form hierarchies in which the system is "nested" within higher level social systems and is itself made up of subsystems, which, in turn, are composed of lower level subsystems.

Self-organization
While nonlinearity implies that "all is chaotic" within a complex system, the idea of selforganization suggests that interactions between the system constituents have the capacity to generate a sort of order and that the system has a tendency to organize in increasingly complex forms.
An extreme version of self-organization is called "autopoiesis." 42 In the philosophy of social science, the idea of "autopoietic systems" (conceived of as networks of communications) is strongly associated with Nicklas Luhmann's systems theory. 43 As Elder-Vass correctly summed up, Luhmann's account of "autopoietic system" comprises the following two claims: (1) the influence the components exert on the system can be ignored (for the purpose of explaining the reproduction of the whole) and (2) although the system interacts with the environment, it is capable of controlling its own reproduction. 44 Such a conception of a system neglects the role of systems constituents (i.e. the role of emergent processes) and, at the same time, suggests systems of communications with rigid boundaries and minimal interactions with systems environment.
PROPOSITION ① At this stage of discussion, I would like to propose that the term "social order" refers to the configuration of the system's emergent properties endowed with the quality of downward causation, by virtue of which the system impacts (=governs, influences) the behavior of its components.
Emergentist account of social systems, on the other hand, places the idea of self-organization within the domain of morphostatis (=system reproduction), allowing the system (1) a set of structures that "accommodate" external causal factors and (2) a degree of stability between the system's constituents and the whole achieved by virtue of downward causation. In other words, the emergentist account of social systems does not refute entirely the concept of selforganization, but interprets it in the following two ways: (1) as a set of mediating processes between the system and its environment and (2) as a system maintenance over time.

Openness
System openness implies permeable borders that can be traversed by other systems. 45 Thus, while discussing the idea of system openness, the issue of system boundaries comes inevitably to the fore. In the literature, the term "boundary" tends to be used to refer to those components of a system that are linked directly to the system environment. In the case of social systems, the issue of boundary setting is a complex one and does not refer exclusively to the spatial/geographical boundaries. 46 For example, Mingers classifies boundaries into three broad types: (1) boundaries that delimit the scope of the problem (=define what range of actors is to be included), (2) boundaries that delimit the time horizon, and (3) boundaries that limit what may be changed (e.g. a set of cultural norms that can shape the extent of potential changes at the social level of focus). 47 It follows from the above that system boundaries can be analyzed both as a process of boundary setting between the system and its environment and as an observer-dependent construct.
It is worth noting that the distinction between the "process (in a system)" and the "observer-dependent construct" parallels Bhaskar's division into "transitive" and "intransitive" domains of knowledge. In the transitive domain, we always deal with our own human (=observer's) construction and models. The intransitive domain constitutes the object of study itself, whether these are physical processes or social phenomena. 48 Regarding the reality of the objects of study (e.g. system boundaries), for CR the causal efficacy is the proper criterion for their existence. In other words, if we can identify possible mechanisms by which system boundaries were produced (=causal efficacy), we can take boundaries to exist in the intransitive domain.
I would like now to turn to Tilly's model of the formation, transformation, and suppression of social boundaries. According to Tilly, an in-depth explanation of social entities involved in boundary reification would comprise two types of causal mechanisms: (1) mechanisms that cause boundary change (e.g. encounter, imposition, borrowing, etc.) and PROPOSITION ② Social order epitomizes self-organization processes within a given social system. Following the spatiotemporal line of reasoning, social order can thus be disaggregated into (1) spatial component of self-organization (=a set of exostructures that "accommodate" external causal factors) and (2) temporal component of self-organization (=a configuration of the system's emergent properties endowed with the quality of downward causation. These properties "stabilize" the parts-whole relation of the system contributing to its durability). 45 Cudworth and Hobden, "Analysing Change," 633. 46 For example, Bunge describes sales representatives and PR managers of a business firm as constituting its boundary (Bunge,Finding Philosophy,. 47 Mingers, Systems Thinking, 143-6. 48 Bhaskar, The Possibility of Naturalism, 9-14. (2) mechanisms that constitute boundary change and produce its direct effects (e.g. inscription-erasure, site transfer, etc.). 49 Further, it is important to recognize that open systems are composed of multiple interacting mechanisms, and therefore, in some instances, it is the specific conjunction of mechanisms that explains a given boundary change. Given that one of the goals of a mechanismic approach is to discover typical mechanism recurring in specific types of social systems, Tilly's model provides a useful point of departure for analyzing the issue of the region's boundaries.
3.4.1. Delineation between social and natural systems Another important issue related to the idea of system openness is that of the delineation between social and natural systems. Throughout history, humanity and nature have been coevolving in a dynamic fashion and this coevolution now takes place at a rapid pace. That is precisely why the concept of "socio-ecological system" (SES) has recently been gaining prominence. 50 In the literature, the term is generally understood to mean complex systems where social and biophysical entities are interacting at multiple temporal and spatial scales and where individual/collective agents intentionally invest time and effort to develop forms of infrastructure that affect the way the SES functions over time.
In this paper, I adopt (with minor modifications) the framework proposed by Anderies et al. that identifies the relevant components of an SES and how they are linked. Figure 1 gives a schematic overview of the SES. 51 Anderies et al. distinguishes four "entities" that are normally involved in SESs. Two of these entities (rectangles) are composed of humans. Those two social entities are respectively the resource users (=the local population) and the public infrastructure providers. The public infrastructure (the oval) combines two forms of capital: physical capital (=a variety of engineered works) and institutional capital (=the rules used by the infrastructure providers to govern the system). The resource (the hexagon) is a biophysical system or a form of natural capital (e.g. water resources). Further, interactions exist among resource users regarding the harvesting rate from the resource (Arrow 1), and among resource users and the public infrastructure providers (Arrow 2), the linkages among the public infrastructure providers and the investments made in the infrastructure (Arrow 3), and among human-made and ecological entities (Arrows 4 and 5).
The above framework offers a useful methodological tool to study different types of socio-ecological subsystems in their capacity as potential sources of the system's emerging properties. With regard to the EARS, the SES model can be applied to analyze the following salient issues: (1) energy security, (2) nontraditional security threats that stem from environmental degradation (including food, water, and health security), (3) patterns of cooperation between the MDBs (e.g. the ADB-AIIB cofinancing projects), and (4) patterns of development of the natural economic areas (e.g. Greater Mekong Subregion).

System reproduction, transformation, and dismantling
The existence of a social system at any given point in time depends on morphogenetic and morphostatic causal factors. In Archer's "morphogenetic cycles," agents and structures interact and these social interactions are cyclical phenomena. Preexisting structures emerge from a prior cycle and govern subsequent social interactions. Agents interact with these preexisting structures, so that the structures undergo change: they are reproduced (i.e. morphostasis occurs) or transformed (i.e. morphogenesis occurs). In other words, morphogenetic factors bring each social system into existence (=the preexisting structures are transformed), whereas morphostatic factors sustain its existence (=the structures are reproduced). 52 With regard to the social system dismantling, I would like to begin with the idea of "properties submergence." Bunge defines the concept as "the loss of higher level properties." 53 Since properties are possessed by social entities, properties submergence is just one aspect of the system dismantling. Doubtless, every instant of social transformation (=morphogenesis) accounts for the disappearance of a set of social forms. It is crucial, however, to clarify further the above claim by inquiring into which properties of social entities cease to exist and which survive in some vestigial form.
In their timely examination of "morphonecrosis" (=system dismantling), Al-Amoudi and Latsis delineate the conditions for properties submergence in terms of structure-PROPOSITION ③ Region can be conceived of as a complex social system, that is, a system characterized by nonlinearity, emergent properties, hierarchy, self-organization, and openness.
PROPOSITION ③α Some subsystems of a given regional system can be conceived of as socio-ecological systems, that is, systems where social, human-made, and ecological entities interact at multiple temporal and spatial scales. 52 See, Archer, Realist Social Theory, Chapter 3. 53 Bunge, Emergence and Convergence, 31.
agency-culture (SAC) triad. The first condition is agential, that is, properties submergence depends on "the interaction of socially situated actors holding differing concerns and vested interests." 54 The second condition is structural and involves the dismantling of social institutions upon which actors' vested interests depended. The third condition is cultural and entails "reinscribing the social form in a different discourse." 55 With regard to the third condition, Archer made a similar point arguing that culture cannot be confined to ideas that are currently endorsed by social groups because these are always a portion of the ideas available for endorsement. Usually, socio-cultural conflict leads to the activation of some of that non-salient portion, specifically those ideas which challenge whatever is and bids to be hegemonic. Thus, unfashionable ideas can be revived. 56 As should be clear from the above explanation, Al-Amoudi and Latsis's model focuses on what might be labelled "endogenous conditions for properties submergence." It seems to me, however, that exogenous conditions are at least equally important. For example, system dismantling has a lot to do with the process of exostructures weakening that is often designed and carried out by certain "external agency." 57 Tying together endogenous and exogenous conditions for properties submergence, Table 1 outlines a tentative framework for social system dismantling.

Levels
It is necessary here to clarify exactly what is meant by "level." In this paper, the term "level" is used to refer to (1) the division of ontology into domains, (2) the stratification of the real world into emergent levels, (3) levels of social process, and (4) "level-ofanalysis" approach in IR studies. An in-depth discussion of (1) and (2) is beyond the scope of this paper. Suffice to say that this study endorses Bhaskar's stratified ontology 58 as well as the hierarchy of levels of reality (as characterized by corresponding academic subjects: physics → chemistry → biology → physiology → psychology → sociology/  58 Bhaskar argues that in order to make sense of the scientific enterprise we need to distinguish between the domains of the empirical (=experiences), the actual (=events), and the real (=structures and mechanisms). Structures and mechanisms then are real and distinct from the patterns of events they generate, just as events are distinct from the experiences in which they are apprehended (Bhaskar, "A Realist Theory," 2-3). To put it in a slightly different way, these three domains are related in that causation originates in the real world and is only detected through empirical events perceived through experiences. It must be emphasized, however, that although ontologically experiences, events, and structures/mechanisms constitute different domains, empirically they are encountered conjointly. economics/politics). 59 Let us then proceed to examine the meaning (3) and (4) of the term "level."

Levels of social process
Levels of social process refer to a vision of society as stratified reality. To use Onuf's apt observation: levels are not just a taxonomic convenience for scholars, or a methodological expedient. They are a potent metaphor . . . for marking, and thus making, wholes. In our culture, as in our field [IR], we would have difficulty getting along without the language of levels. 60 But what exactly do social scientists mean by "levels?" According to a definition provided by Bunge: A level is a section of reality characterized by a set of interlocked properties and laws, some of which are thought to be peculiar to the given domain and to have emerged in time from other (lower or higher) levels existing previously. 61 Social levels, then, are distinguished by virtue of its emergent properties and each constitutes a complex social system. Furthermore, the understanding of any level depends greatly on research into adjacent levels, both underlying and overlying ones. It must be emphasized, however, that the issue here is not how many social levels should be identified but the relationships between the levels we are interested in. I would now like to turn to Brante's model of inter-level relations. For Brante, each level constitutes the system that is distinguished by virtue of its emergent properties and by two types of mechanisms: (1) specific for each levels and (2) binding levels together. To explain social events, Brante introduces the "component-structure-context" triad. A given social event [e] is a function [f] of (1) the structure [S] that generated it at the analyzed level, (2) the components for the structure [C1] that belong to the underlying levels, and (3) the context [C2] to the structure that pertains to the overlying levels. Thus, a complete explanation of a social event [e] would look like this: e = f (C1, S, C2). 62 To take just one new topic of research that might be facilitated by the proposed procedure, consider the Asian Infrastructure Investment Bank (AIIB). To analyze potential emergent properties 63 possessed by a social entity called "AIIB" at the regional level, we need to account for the lower level components from which these properties might emerge (e.g. China's industrial overcapacity, China's "new normal" development strategy, the Western Regions Development Strategy, etc.) as well as for the context to these properties (e.g. the Renminbi (RMB) internationalization, the establishment of the New Development Bank, the Asian Development Bank's (ADB) major financial 59 For a detailed account, see Ellis, "Top-Down-Causation," 124-40. 60 Onuf, "Levels," 53. 61 Bunge, "Levels," 405. 62 Brante,"Consequences, For example, a set of social relations (=structures) that has emerged from the AIIB-ADB and the AIIB-WB cofinancing projects. Please note that the AIIB is treated here as a component of the EARS, thus emergent properties it has generated might (in a specific time-space context) account for the transformation of the East Asian regional order. At the same time, the AIIB is a social system in its own right, composed of different social entities, as explained in section 3.1.
restructuring, the World Bank's (WB) safeguard policies reform, the US efforts to prevent its allies from becoming the AIIB founding members, etc.) To sum up, the "component-structure-context" procedure can be of great help in solving the problem of the "vertical differentiation" of social order, that is, a clear-cut differentiation of the emergent properties at different levels of social process. 64

"Level-of-analysis" approach in IR studies
Levels-of-analysis approach has been present in IR studies at least since the early 1960s. Both Waltz's "the individual-the state-the international" triad and Singer's "international system-national system" distinction are rightly considered important theoretical developments in the discipline. Further, Hollis and Smith's reformulation of Singer's scheme in terms of "actors and their system of interaction" has since become a standard model in IR scholarship. 65 Recently, however, serious attempts have been made to reformulate the vertical, hierarchical image of the Waltzian/Singerian levels and (re)integrate the issue into a broader systemic approach. For example, Wight offers an alternative conceptualization of levels in IR arguing that (1) there is no need for a distinct individual level beneath that of the (Waltzian) state since human/collective actors feature in every level, and (2) each level (author distinguishes between international, nation-state, and domestic levels) requires further disaggregation into four components (namely, culture, structure, agents, and practices). 66 Thus, in Wight's model human/collective actors can act at different levels, although the properties they possess vary from level to level due to distinctive social practices generated by various structural-cultural contexts.

Culture-inclusive account of social order
Archer's theory of culture is based on the distinction between relations that exist between ideas and the ideational influences that exist between human agents. In Archer's account, the relations between ideas are characterized by the degree of logical consistency (i.e. ideas contradict/complement each other). Ideational influences between human agents (e.g. manipulation, legitimization, etc.), on the other hand, take the form of causal relations. On the basis of the above distinction, Archer differentiates between the Cultural System (=the world of ideas) and Sociocultural Interaction (which consists of the influences people have on each other with regard PROPOSITION ④ To account for the vertical differentiation of social order, the "component-structure-context" procedure has been introduced. A given social event can thus be explained as a function of (1) the agent-structure relationship that generated it at the analyzed level, (2) the components for the structure that belong to the underlying levels, and (3) the context to the structure that pertains to the overlying levels. 64 Although the term "differentiation" has been widely used in sociology since the late nineteenth century, the idea of "vertical/horizontal differentiation of order" put forward in Section 9 of this study has been greatly inspired by Professor Donelly's article on different types of structural differentiation in international politics (Donnelly,"Rethinking Political Structures,. 65 Hollis and Smith,Explaining and Understanding,[99][100][101][196][197][198][199][200][201][202]. 66 Wight, "Morphogenesis," 225-8. The term "practices" denotes here the mediating point linking action to structure. to culture). 67 As emphasized by Archer, by maintaining the distinction between the Cultural System and Sociocultural Interaction, "it becomes possible to theorize about variations in cultural integration and their relationship to variations in social integration." 68 Archer's morphogenetic approach forms the basis for developing her theory of cultural change. The morphogenesis of culture has three phases: (1) cultural conditioning, (2) sociocultural interaction, and (3) cultural elaboration (either morphogenesis or morphostasis occur). In the first phase of the cycle, Archer distinguishes between two possible cultural conditions, namely "constraining contradictions" and "concomitant complementarities." In the case of constraining contradictions, there is a necessary connection between two sets of ideas, that is, one is dependent on the other and cannot operate without it, even though they conflict with each other. Under the condition of concomitant complementarities, on the other hand, two sets of ideas are linked to each other and an ongoing clarification and confirmation of the ideas take place. As ideas have been systematized, however, it has become increasingly difficult to add new ideas. Thus, at the certain moment, new ideas are often repressed/rejected in order to protect the existing Cultural System. 69 In the second phase of the cycle, Archer's theoretical focus is on how agents interact and influence each other and thus contribute to cultural morphogenesis/morphostatis. Both constraining contradictions and concomitant complementarities prompt the formation of three types of cultural agents: the elite, the masses, and the marginalized. In a long-term perspective, sociocultural interactions between the elite and the marginalized set the process of cultural transformation in motion. 70 In the third phase of the cycle, Archer discusses the cultural morphogenesis (or morphostasis) generated by Sociocultural Interaction. To this end, Archer identifies four types of cultural conditioning each characterized by different situational logic and different developments with regard to the Cultural System and Sociocultural Interaction: (1) Under the condition of constraining contradictions, the prevailing situational logic is that of the correction of ideas. This results in the syncretisation of the Cultural System and unification at the level of Sociocultural Interaction. (2) The condition of competitive contradictions involves a situational logic characterized by the elimination of ideas (i.e. agents take up one set of ideas and simultaneously attempt to eliminate the other). Competitive contradictions lead to pluralism in the Cultural System and conflicts at the level of Sociocultural Interaction. (3) Under the condition of concomitant complementarities, the prevailing situational logic is that of the protection of ideas. This results in the systemization of the Cultural System and morphostasis in Sociocultural Interaction. (4) The condition of the contingent complementarities involves the situational logic of opportunity (i.e. agents (including the marginalized) are free to make a choice). This leads to divisions and/or specialization in the Cultural System and sectionalism at the level of Sociocultural Interaction. 71 The main point of my argument in this section is that Archer's theory of cultural change offers a useful methodological tool to study the emergence and submergence of ideational structures prevailing in a given social system. To take just one topic of research that might be facilitated by Archer's theoretical model, consider the diffusion of the "Practical Learning" discourse across the Early Modern East Asian Regional System (EMEARS). 72 The term "Practical Learning" is used here to describe a Confucian heterodox discourse guided by the spirit of 經世濟民. 73 The examination of the interactions between Neo-Confucian orthodoxy (understood as the state ideology) and the "Practical Learning" discourse in China, Korea, and Japan reveals different types of cultural conditioning characterized by different situational logic.
In the case of the Manchu regime, in the early Qing the situational logic of the "Practical Learning" discourse was that of "the elimination of ideas." At the sociocultural level, this led to suppression of the ideas represented by "practical-type" scholars of both Cheng-Zhu and Lu-Wang Schools.
In the case of the Late Choson Dynasty, on the other hand, the situational logic of "the correction of ideas" under the condition of the constraining contradictions describes the state ideology -"Practical Learning" interactions most accurately. Thus, at the sociocultural level, "practical-type" scholars had no choice but to accommodate themselves to the official discourse.
In the case of the Tokugawa regime, the situational logic was that of "opportunity" under the condition of contingent complementarities. This resulted in a plethora of doctrines and theories available at the cultural level, but led to divisions, new formations, and relatively limited role of the state ideology at the sociocultural level.
Furthermore, with regard to the cultural condition for the EMEARS dismantling, a challenge to the state-endorsed Neo-Confucian ideas from the philosophical discourse of modernity 74 can be fruitfully analyzed within the morphogenetic framework. While the submergence of Neo-Confucian orthodoxy across East Asia was sudden and agonizing, clusters of ideas related to the "Practical Learning" discourse (e.g. theory of governance, economic culture, etc.) continue to "coexists" with the hegemonic discourse of modernity, thus constituting "a portion of the ideas available for endorsement" within the Cultural System of contemporary East Asia.

Constitutive and derivative structures
The distinction between Cultural System and Sociocultural Interaction allows me to argue that higher level social systems (e.g. region) possess "constitutive structures" that pertain to the Cultural Subsystem and "derivative structures" (emerging from agents social practices) that belong to the Social Relations Subsystem. 75 Constitutive structures 72 For a detailed account, see Tanaka, "Continuity and Transformation," 183-217. By "early modern" I mean the Manchu regime in China (1644China ( -1912, Late Choson Dynasty in Korea (c. 1600-1860), and the Tokugawa regime in Japan (1603-1868). 73 Can be rendered as "administer the state and relieve the suffering of the people." 74 The term "philosophical discourse of modernity" (PDM) is used by Bhaskar to denote an ideology of the capitalist mode of production. It is further subdivided into five phases, each marked by a revolutionary upheaval or transformation (Bhaskar,"Reflections,. Although in terms of its origin the PMD is an Eurocentric phenomenon, in parallel with the globalization of capital, it has evolved into a universal global ideology giving rise to multiple modernities and varieties of capitalism at the regional level. 75 "Sociocultural Interaction" in Archer's terminology. account for the theories of human behavior, theories of order and change in society, moral philosophy, etc. Derivative structures, on the other hand, "translate" the content of constitutive structures to a specific time-space context. 76 The above distinction, however, is hedged with several qualifications. First, the purpose here is not to advocate (à la Parson) the primacy of the cultural domain over other subsystems. Second, although in this section the focus is on the constitutive and derivative structures in their capacity as the potential emergent properties of the regional system, such an approach does not contradict the commonly used division into "ideational" and "material" structures. 77 Third, it must be noted that at a given point in time only some ideational structures (most likely those endorsed by the elite) qualify as the system's constitutive structures.
Overall, the idea behind this section has been to construct a "culture-inclusive" account of social order that might be fruitfully applied to the examination of the world of ideas within a given social system at a given period of time. 78 Thus, proposition ⑤ goes as follows:

History-inclusive account of social order
At the core of the mechanismic approach in the social sciences lays an idea that proper scientific explanations should detail the "cogs and wheels" of the causal processes through which social phenomena to be explained were brought about. 79 Further, for any mechanisms to merit consideration as a source of social transformation, it must necessarily comprise structured human relations (=context dependence), human actions (=activity dependence), and human ideas (=concept dependence), that is it must incorporate SAC. 80 To explain a given macro-social phenomenon (e.g. economic transformation, regional integration, democratization, etc.) in a historical perspective, we need to identify the processes through which it was generated within the system. In other words, we need to identify a cluster of causal (=generative) mechanisms that explain a given macro-social phenomenon and describe how these mechanisms manifest themselves in concrete PROPOSITION ⑤ To construct a "culture-inclusive" account of social order, the constitutive-derivative structures dyad has been introduced. Whereas constitutive structures pertain to the Cultural Subsystem, derivative structures (emerging from agents' social practices) belong to the Social Relations Subsystem. Constitutive structures account for the theories of human behavior, theories of order and change in society, moral philosophy, etc. Derivative structures, on the other hand, "translate" the content of constitutive structures to a specific time-space context. 76 By way of illustration, the chaogong (=tribute) relations can be understood as a set of derivative structures through which human agents "realized" the content of the Neo-Confucian constitutive structures in a specific time-space context of the EMEARS. 77 Reus-Smit, "The Idea of History," 130. 78 A similar distinction into "deep structure" 深層結構 and "surface structure" 淺層結構 has been proposed by the Taiwanese psychologist Prof. Hwang Kwang-Kuo (Hwang,"The Deep Structure,. Hwang analyzes Confucian ethical system in terms of the deep structure of "benevolence-righteousness-propriety" 仁・義・禮. The relationships among such Chinese concepts as renqing 人情, guanxi 關係, and mianzi 面子are regarded as the surface structure, which is derived from the deep structure. Interestingly, in his recent article on "culture-inclusive" theories in Chinese indigenous psychology, Hwang explicitly uses Bhaskar's and Archer's theories to account for persistence of Confucian ideas in Chinese societies (Hwang,"Cultural System,. 79 On the diverse origin of the Mechanism-based explanation in the social science, see Ylikoski,"Social Mechanism," 415. On the role of the mechanismic explanation in historical sociology, see Gorski,"Social Mechanisms,. 80 Archer, "Social Morphogenesis," 4. contextsas a chain of social events. As noted by Mayntz, an overwhelming majority of social macro-phenomena cannot be explained by one particular mechanism. 81 Instead, a cluster of mechanisms jointly generating the outcome is involved. Also, it must be emphasized that not all of the component mechanisms are social sensu stricto. For example, McAdam et al. offer a typology of generative mechanisms that distinguishes between environmental, 82 cognitive, 83 and relational types. Environmental macro-level mechanisms denote "externally generated influences on conditions affecting social life" (e.g. resource depletion), whereas cognitive mechanisms operate at a micro-level through alterations of individual and collective perceptions (e.g. agents' identity shift). Relational mechanisms, on the other hand, transform or reproduce the connections among collective actors (e.g. brokerage of a coalition). 84 Overall, although in a historical perspective relational mechanisms remain of crucial importance for our understanding of macrosocial phenomena, the environmental and cognitive mechanisms need not be overlooked.
I would now like to turn to diachronic emergence. As explained earlier, through the process of diachronic emergence new systems are historically formed. My main argument here is that an explanation of the dismantling of an old system (as well as the emergence of a new system) requires an in-depth analysis of not only structural and agential conditions but also an account of "nodal points" that mark systems geohistorical transitions. 85 Analytically, the idea of "nodal points" draws on Bhaskar's conceptualization of being as characterized by the "absence-negation" dyad and thus as involving CHANGE. 86 As Hartwig aptly observed, in Bhaskar's model, any transition embraces both a "transition state" (=a spatiotemporal moment of indeterminate negation) and, within that, a "nodal point" (=the point at which it is no longer possible to say whether A is still A and not B). Thus, the concept of nodal points exploits the bivalence of "node" as "tying together yet distinguishing." 87 In A Note on History, Bhaskar urges the historian to be "concerned with the nodal points and the motive agencies in the past and present (and the likely and possible futures) of human history." To this end, he offers a typology of nodal points that might be useful in mapping out the system geo-historical transitions (see Table 2). 88 To sum up, the main aim of this section has been to utilize the concepts of macrolevel social mechanisms and nodal points to construct a "history-inclusive" account of regional order that might be fruitfully applied to the study of the historical social systems. Thus, proposition ⑥ goes as follows: 81 Mayntz,"Mechanisms," 254. 82 To put it in Bunge's terminology, environmental mechanisms correspond to processes by which exostructures are produced, whereas cognitive mechanisms correspond to psychological processes that underlie agents' specific behaviors. 83 Throughout this paper, the terms "cognitive," "dispositional," and "psychological" (mechanisms) are used interchangeably. 84 McAdam et al.,Dynamics, See, for example, Patomaki's account of the causal complex that made the First World War possible (Patomaki, The Political, 232-49). 86 There is no space here to discuss in detail the stages in critical realism development. Suffice to say that this study relies heavily on Bhaskar's philosophy of "basic critical realism," whereas the concept of nodal points introduces some themes from the "dialectical critical realism" phase (Bhaskar, "Critical Realism," 9-24). 87 Hartwig, ed., Dictionary, 482. 88 Bhaskar,Scientific Realism,217. 8. Revisiting the concept of regional order Systems Theory was originally used in the natural science, but its application quickly spread to the social sciences as well. Talcott Parsons's structural functionalism was once a dominant paradigm in Anglo-American sociology. David Easton, on the other hand, was the first to apply systems theory to political science. 89 In the 1950s, particularly in the United States, Parsons's and Easton's methodological conceptualizations of systems theory were applied to sociopolitical science of IR. The first major work that presented the systemic approach to IR was Morton Kaplan's System and Process in International Politics (1957).
Parsons's theory of social order 90 was an attempt to strike the middle ground between Lock's utopian theory of social contract (with its emphasis on the rationality of man) and Hobbes's theory of social order guaranteed solely by Leviathan (=a state, a sovereign). 91 Parsons argued that social order is possible on the basis of shared values acquired through the process of internalization. He laid particular emphasis on the concept of "normative ordr," arguing that Normative order . . . is always relative to a given system of norms or normative elements, whether ends, rules or other norms. Order in this sense means that process takes place in conformity with the paths laid down in the normative system . . . a social order . . . is one which cannot have stability without the effective functioning of certain normative elements. 92 The normative element of order resides, according to Parsons, in the cultural system, but it "governs, controls, organizes" the other systems further down in the hierarchy. 93 Although Securing the satisfaction of the enabling condition for a process PROPOSITION ⑥ To construct a "history-inclusive" account of social order the concept of nodal points has been introduced. The role of nodal points is to mark the system geo-historical transitions, thus contributing to a historically grounded explanation of the emergence, transformation and submergence of the successive social orders. 89 Easton, The Political System. 90 Important theories of order include the following: (1) Parson's structural-functionalist model, (2) Aron's "descriptivenormative" account of order, (3) Goffman's symbolic interactionism, and (4) Garfinkel's ethnomethodological approach. Parons's and Aron's models offer a "bird's eye view" on the problem of order thus leaving its mark on IR theorizing. With regard to Aron's account of order, two examples of contemporary liberal IR theory that draws heavily on Aron's theorizing include liberal democratic peace and neoliberal institutionalism. Also, as noted by Rengger, Bull's writing on the world order closely resembles Aron's analytical approach (Rengger, "International Relations," 49). On Parson's influence on IR theory, see Albert and Buzan, "International Relations," 117-35; and Goddard and Nexon, For a detailed analysis of the "utilitarian dilemma," see Parsons, The Structure of Social Action, 67-8. 92 Ibid., 91-92. 93 The other three systems articulated by Parsons are (1) social system, (2) personality system, and (3) biological organism (see, Parsons and Bates,Family,[31][32][33]. the control and determination of the lower systems through the normative factors is never absolute (the conditional factors (=motivation) cannot be ruled out), for Parsonian social order to exist the value element has to be the predominant, controlling one. From the 1970s on, Parsons's "normative problem of order" has been the focal point of controversy. As Cohen aptly summed up, Parson's overemphasis on normative factors in his account of social order stems from his negligence of the concrete ways in which order is generated in the course of the agents' social activities. 94 Furthermore, proponents of complexity thinking argued that order and chaos are conjointly present in open systems and although order exists as a result of repetitive, internal, and external interactions, it is always transitorysystems are seen as being "on the verge of chaos." 95 Anthony Giddens, on the other hand, directed strong criticism at Parsons's treatment of human actors (=agency), arguing that "human agents seem to elude the grasp of his scheme: the stage is set, the scripts written, the roles established but the performers are curiously absent from the scene." 96 Further, Giddens made an attempt to reframe the Parsonian formulation of order along the spatio-temporal line of reasoning: The fundamental question of social theory as I see itthe problem of order conceived of in a way quite alien to Parson's formulation . . .is to explicate how the limitations of individual "presence" are transcended by the "stretching" of social relations across time and space. 97 Against the backdrop of the above outlined developments in social order theorizing, building on the content of the Sections 2-7, I would like to propose the following reformulation of the concept of "regional order": • Region can be conceived of as a complex social system, that is, a system characterized by nonlinearity, emergent properties, hierarchy, self-organization, and openness. Some subsystems of a given regional system can be conceived of as SESs, that is, systems where social, human-made, and ecological entities interact at multiple temporal and spatial scales. • The term "regional order" refers to the configuration of the region's emergent properties endowed with the quality of downward causation, by virtue of which the system impacts (=governs, influences) the behaviour of its components. • Regional order epitomizes self-organization processes within regional system.
Following the spatiotemporal line of reasoning, regional order can thus be disaggregated into: (1) spatial component of self-organization (=a set of exostructures that "accommodate" external causal factors) and (2) temporal component of selforganization (=a configuration of the region's emergent properties endowed with the quality of downward causation. These properties 'stabilize' the parts-whole relation of the system contributing to its durability). • To account for the vertical differentiation of regional order, the "component-structure-context" procedure has been introduced. A given social event can thus be 94 Cohen, "Structuration Theory," 38. 95 Baker, "Chaos," 133-5. 96 Giddens, Central Problems, 253. 97 Giddens, "Comments," 35.
explained as a function of (1) the agent-structure relationship that generated it at the regional level, (2) the components for the structure that belong to the underlying levels, and (3) the context to this structure that pertains to the overlying levels. • To construct a "culture-inclusive" account of regional order, the constitutivederivative structures dyad has been introduced. Whereas constitutive structures pertain to the region's Cultural Subsystem, derivative structures (emerging from agents' social practices) belong to its Social Relations Subsystem. Constitutive structures account for the essential theories of human behavior, theories of order and change in society, etc. Derivative structures, on the other hand, "translate" the content of constitutive structures to a specific time-space context. • To construct a "history-inclusive" account of regional order, the concept of nodal points has been introduced. The role of nodal points is to mark the system geohistorical transitions, thus contributing to a historically grounded explanation of the emergence, transformation, and submergence of the successive regional orders.
9. Theories of order and the ongoing debate on order transition in East Asia: points of encounter This article, following the ''realist philosophy → realist-informed IR theory → actual research project" sequence, attempted to clarify how realist approach to IR studies might work while applied to a specific research project (here, revisiting the concept of "regional order"). It must be noted at the onset that the overarching goal of the above theoretical enterprise has not been to refute the existing approaches to regional order, but rather to establish the foundation for a more productive engagement between realist-informed IR theorizing and the core theoretical perspectives. A look at the East Asian regional order through the spectacle of CR/SR has five main implications: (1) It brings the leading IR perspectives in closer contact with the more general discussion on "social order" in the social sciences, thus helping clarify their positions on the emergence of order, the idea of self-organization, mechanisms as a source of social transformation, agent-structure interactions, and the role of culture. Further: (2) With regard to the spatial vertical differentiation of order, the "componentstructure-context" procedure offers a useful methodological tool to analyze the EARS exostructures in the context of the American-led liberal world order. (3) With regard to the spatial horizontal differentiation of order, the mechanismbased explanation facilitates the study of the formation, transformation, and suppression of social boundaries in the open regional systems. (4) With regard to the temporal vertical differentiation of order, an account of geohistorical nodal points provides an explanatory framework for the diffusion of the discourse of modernity across East Asia, with a particular focus on the conditions of multiple modernities and varieties of capitalism that emerged through the process. (5) With regard to the temporal horizontal differentiation of order, Archer's model of cultural change opens up a promising avenue for further investigation into the spread of the Confucian ideas across the EMEARS. It might also facilitate the examination of the "vestigial properties" of the historical systems in their capacity as emergent properties of the contemporary EARS.