Cultures of Anarchy as Figurations: Reflections on Wendt, Elias and the English School

The Roots of a Relational Approach

Relationism is not a school, yet. Its first roots may be found in the works of Heraclitus who wrote: ‘You cannot step twice into the same river; for other waters and yet others go ever flowing on’. American pragmatists like Dewey and Bentley are also recognised as pioneers in transactional thought. The way we use relations is synonymous with the way these thinkers used transactions. In sociology, the seminal text of Mustapha Emirbayer (1997), ‘Manifesto for a Relational Sociology’, has begun to open the door to new debates.[1] For the purpose of this paper, I shall concentrate my attention on the relational thinking developed by Norbert Elias. Several reasons lead me to this decision. First, Elias produced a prominent and diverse body of work that forms a coherent whole. The fact that he wrote over the course of 40 years allows us to see the evolution of his thought. Second, Elias did not belong to any school, nor did he want to create one. Thus, one can use his works as a toolbox to offer radically different approaches to IR or to propose a re-conceptualisation of some basic terms. Thirdly, Elias’s personal life was shaped by the traumatic world events of the twentieth century; he was a privileged and informed witness to many events that would have an important impact on the development of his thoughts. Elias used many examples and illustrations from world politics in his theoretical reflections (Linklater 2004: 41). His entire œuvre was dedicated to studying relations of power and violence, themes at the core of the discipline of International Relations. Fourth, a number of IR scholars have been influenced by Elias. Patrick Thaddeus Jackson (2006) is one of these thinkers. At the same time, one of the major thinkers of the last three decades, Andrew Linklater, has started to integrate Elias’s thought into his research. Both Jackson and Linklater are essential players in the development of relational thinking in IR as they appear to be bridge builders: Jackson with the constructivists and Linklater with the English School and with critical theorists. Indeed, as Linklater (2007b: 4) notes:

Finally, there is a group of ‘Eliasians’ who devoted most of their scholarly life to developing Elias’s thought. Robert van Krieken, for instance, wrote a crucial text on violence (1989) but also on social theory (2002); Stephen Mennell (1990; 2007), who works on globalisation and de-civilising processes as well as the American civilising process, also relies heavily on Eliasian models. Both are part of a growing international ‘figurational Sociology’ network.

Many authors recognise that ‘down the road’ it is all about processes and relations. In other words, co-determinist thinkers are often partly ‘relational’ as evidenced in Bhaskar’s work (1998). Wendt also appears to be partly relational. For example, he writes ‘Agents and structures are themselves processes, in other words, on-going “accomplishments of practice”’ (Wendt, 1999: 313). But they remain trapped in their focus on, on the one hand, an analysis of the social structures; and on the other hand, an analysis of the agents. What relational thinking can bring is a systematic analysis of the relations and processes going on between the two. In other words, we want to give priority to the ‘arrows’ instead of the isolated boxes, in order to avoid reification and offer a more dynamic understanding of social reality. The goal of the next section is to clarify what is meant by relational thinking.

Principles of Relational Thinking

Rejection of Causality and Conceptualisation of Duration

Relational thinking rejects causality because it implies an ontological distinction between two separate entities:

More importantly, relational thinking proposes a particular conception of duration.[6] It opposes any form of discourse about a ‘beginning’: ‘There is no zero-point of the social relatedness of the individual, no “beginning” or sharp break when he steps into society as if from outside as a being untouched by the network and then begins to link up with other human beings’ (Elias 2001: 27). This idea clearly contrasts with Archer’s morphogenetic schema beginning with a ‘time 0’ (Archer 1995) or with Wendt’s narrative of the first encounter between Alter and Ego (Wendt 1999). It also dismisses Wendt’s reading of Meadian individuality per se and social terms of individuality.[7] Time can be cut; duration cannot. This has two important implications. First, the ‘past’ is not seen as conditioning and determining actors’ practices and discourses, nor does it ‘cause’ social reality. Second, it allows a more dynamic understanding of reality if we consider how reality unfolds in durational processes. Thus, we avoid static thinking and reification. Actors may repeat more or less the same transactions, but this does not transform these transactional experiences into independent social structures. Privileging duration rather than time, and especially a long duration, allows us to better conceptualise change. Archer, Wendt and other scholars working in similar traditions think that most of the time we are dealing with homeostatic structures; agents are just reproducing social structures. This is because they look at a short period of time. Over a long duration, it becomes obvious that figurations are changing, transforming and constantly developing all the time.

Social Habitus and Civilising Process

One of the most important social processes studied by Elias was social habitus (see Fletcher 1997: 11).[8] The social habitus as a ‘second nature’ is made up of different layers (depending on the many figurations to which people are attached to). But ‘among them, a particular layer usually has special prominence’ (Elias 2001: 183). For instance, in many societies of the past century the national habitus has been of paramount importance. Habiti are produced, diffused, internalised and transformed through numerous transactions. No habitus is ever fixed. Some are more deeply internalised than others and, therefore, are more dominant; but one habitus may also enter into open conflict with others (for instance a more deeply internalised exclusionary/ethnic national habitus may clash with a newer open/civic national habitus, see Nieguth and Lacassagne 2009: 8-13).

Likewise, the civilising process (the other central concept in Elias’ work) is subject to change and contradictions. This process is concerned with the imposition of constraints by the actors on themselves. This can range from table manners to the development of self-restraint and self-control in the use of violence. The civilising process is also linked to the increasing differentiation of functions due to growing economic competition. People come to be bound together in ways that involve and encourage those forms of differentiation and they must adapt their behaviour and develop more self-control (Elias 2000: 365-69). In more complex societies, the external constraints become self-constraints. The civilising process is not unilinear, there can be de-civilising processes (Elias 1996), major outbursts of violence, or a return to a state in which the external constraints take precedence over self-restraint.

As a partial conclusion, relational thinking defends an ontology based on social interdependencies. There are no isolated individuals, no social things, and no social structures. There are figurations of social transactional actors. Actors are bound by many different figurations: family, clans, tribes, states, nations and humanity. Here, interdependencies should not be understood in a traditional fashion – as interdependence between two distinct entities – but rather as ‘existential interdependencies’. These interdependencies are ontological. Humans are placed back in the heart of the discourse, but not in a voluntaristic fashion - humans are not isolated actors, but rather transactional ones. Moreover, contra voluntarism, humans are not free:

Epistemologically, relational thinking rejects causality and positivism. There are no social laws but only processes that may be structured in specific ways by the sum of the transactions and may have an overall direction (but never one that is unilinear or teleological). However, no one controls the outcomes of these processes, or their direction: ‘The intertwining of the plans and actions of many people and many groups, in other words, results in social processes that are not planned or intended by any of those who help to bring them about’ (Elias 1987: 107).

Some Implications of Relational Thinking for the Study of IR

Firstly, relational thinking can help us continue with the fruitful third debate – the so-called agency-structure debate - that started in the 1980s in IR with the introduction of social theories. IR thinkers, in an unprecedented move, opened their eyes to what was going on in other disciplines. These relational perspectives could be particularly useful for critical theorists working on the idea of a cosmopolitan global political community (see for instance Linklater 2005, 2010a, 2011). The powerful idea of a de-civilising process provides an alternative to many ‘established’ discourses, because it suggests that we can never take for granted any state of affairs, including peace among nations and within nations.[9] As Elias has noted: ‘the armor of civilised conduct would crumble very rapidly if, through a change in society, the degree of insecurity that existed earlier were to break in upon us again, and if danger became as incalculable as once it was. Corresponding fears would soon burst the limits set to them today’ (cited in Dunning and Mennell 1998: 341).

Secondly, relational thinking can be helpful in overcoming issues associated with state-centrism, because it allows us to see states as ever changing social figurations. Therefore, one does not need to distinguish between different levels of analysis and decide which level should be privileged.[10] This does not reduce the importance of states as actors. In fact, not only did Elias (2000) devote much of his work to the study of state formation and always emphasise the importance of the ways in which states are and were inter-related, he was also clear that we still live in a state-centric world insofar as states are considered by most people as the last survival units (Elias 2001: 206). Yet, Elias (2001: 232) also thought that humanity was in fact today the ‘overarching survival unit’ and that humanity might well end up under the domination of a world state. Wendt (2003) makes the same argument. A relational perspective simply permits us to socialise (i.e. give priority to the social transactions), historicise and humanise the state. A large number of thinkers in historical sociology in IR have worked in the same spirit (see for instance: Hobden and Hobson 2002; Rosenberg 2006; Teschke 2011). States are no longer reified entities or homeostatic structures; they are considered as on-going social and historical processes. States did not always exist; and may not exist in the future.

Thirdly, relational thinking offers many insights into the concepts of violence, aggressiveness and emotions such as shame, empathy and so forth as well as their transformations. In other words, IR theorising can reintegrate issues of personality structures (emotions, fears and affects in particular) with social processes. For example, the civilised habitus is more important within certain states because of the monopoly of the legitimate use of violence. An Eliasian analysis of ‘civil wars’ might suggest that they are due to a failing social control of the state apparatus. A relational perspective can also explain why in some cases, wars are waged and not in others. There are places and people for whom one may have less empathy, and where it is easier to use force (former colonies for instance) (Elias 1978b). The way wars are waged and the different degrees and means of violence are related to variations in identification and empathy. For example, it can be argued that many Westerners were more invested in stopping genocidal acts in Yugoslavia than in Rwanda because the level of identification with people in Yugoslavia was higher (Swaan 2001). Finally, a relational perspective can help us understand why support for organised violence can drop dramatically as soon as people appreciate the atrocious conditions of conflicts:

That being said, it appears that a relational/processual perspective could also productively complement Wendt’s efforts of theorising cultures of anarchy.

Wendt’s Conceptualisation of the Culture of Anarchy

The publication of Wendt’s seminal book, Social Theory of International Politics, in 1999 – 20 years after Waltz’s masterpiece Theory of International Politics – offered the first overall constructivist critique of neo-realism (a long lasting IR paradigm) and had many effects on the theorisation of international relations. Because of Wendt’s influence in the field, it seems important to engage with his theory as a possible starting point for building bridges between relational thinking and IR. Moreover, most of Wendt’s theory relies on the concept of cultures of anarchy. For the purpose of the discussion, we have summarised briefly in the following table his tri-partite classification of cultures and our conceptualisation of the links between these cultures, violence and the civilising process.

From a theoretical viewpoint, the notion of culture as developed by Wendt juxtaposes three logics: dualism, duality and relationism. The logic of dualism implies a strict ontological distinction between the agents and the structures. Wendt (1999: 165) speaks of ‘interaction’ or ‘co-determination’. Wendt (1999: 166) also recognises a process of the mutual constitution of agents and structures – duality. Finally, he makes it clear that the cultures of anarchy as social structures are always in process (Wendt 1992: 395). Structural change is explained by four variables: interdependence, common fate, homogeneity, and self-restraint (Wendt 1999: 336-63). Let us contrast Wendt’s definition of these four variables with Eliasian relational thinking.

Interdependence

In IR literature, the term ‘interdependence’ is very often used to speak about relationships of dependency between ontologically defined entities as in the case of liberal and neo-liberal theories of liberal interdependence (Keohane and Nye 1977). In an Eliasian sense, interdependence instead represents the webs of social relations in which people are intertwined. Admittedly, Eliasian terminology is somewhat misleading in this context because the prefix inter- presupposes two different entities. ‘Transactions’ and maybe ‘trans-dependence’ would better convey the Eliasian perspective. That said, Wendt and Elias agree that the We-identity that is central to the existence of any state or similar survival unit is formed through the interdependence of individuals.

Common fate

Contrary to interdependence where interaction is necessary, such a requirement does not exist for the common fate according to Wendt (1999: 349). Common fate represents an external force that imposes itself upon individuals and their groups, as in the case of global warming. But the question of conscious awareness remains. People still have to become aware of this common fate and agree on the solutions to best address the issues at stake. Wendt (ibidem) argues ‘As with interdependence, this can only cause collective identity if it is an objective condition, since subjective awareness of being “in the same boat” is constitutive of collective identity, not a cause’. This statement denies the fact that people need to develop a particular understanding of a threat to perceive it as a common threat. Wendt gives the example of Nazi Germany that he describes as a social threat to European states. But unfortunately, his example does not correspond to historical reality. Some Europeans did not perceive Nazism as an ‘objective common fate’ that caused the formation of a collective identity. Two sets of inter-subjective understandings arose concerning Nazism: Nazism as a threat to the European modern ideal, and Nazism as the solution to save a conservative, anti-modernist European ideal. One made possible the development of a European democratic inclusive collective identity. The other made possible the emergence of a racial European collective identity (weak today but still in existence). This common fate cannot be external to the perception and inter-subjective understanding of the people. This common fate does not mechanically force everyone to act collectively in the most appropriate manner. The existing social figurations and webs of interdependencies are just too complex.

Today, humanity is de facto the last survival unit. Humans have become interdependent because of the threat of nuclear power and drastic climatic changes as well as global financial turbulence and uncertainty. Nevertheless, this interdependence has not yet provoked a social integration or the development of a We-identity at the scale of the humanity. Elias notes:

Interestingly, Wight (1966: 93) made a similar statement: ‘if the community of mankind is not yet manifested, yet it is latent, half glimpsed and groping for its necessary fulfillment’.

Homogeneity

According to Wendt, there are two kinds of homogeneity: one refers to the corporate identity of the actors and the second to the type variation within the corporate identity. Nowadays, homogeneity exists in terms of corporate identity: most of the primary actors are states (not clans, confederation, kinships or empires); whereas important variations still exist in regime types of states (democracies, authoritarian regimes or totalitarian regimes). ‘The hypothesis would be that increases in objective homogeneity cause actors to re-categorise others as being like themselves’ (Wendt 1999: 354)

Elias would have surely admitted the crucial role played by homogenisation in the formation of a We-identity. In The Civilising Process, Elias shows how the new habitus is diffused and how it can become homogenised within a society. But this homogenisation can happen only through the knitting together of social relations. For example, the homogenisation of table manners occurs only once the bourgeoisie develops and becomes rich enough to penetrate the court society. Therefore, the Wendtian variable of ‘homogeneity’ must be conceptually linked to the notion of ‘interdependence’. Wendt’s explanation and hypothesis are too mechanistic to be accepted. It presupposes that a kind of ‘invisible hand’ will homogenise social groups, forgetting that human transactions are necessary for the process of homogenisation and that no one can predict if such a process will indeed unfold.[11]

Self-restraint

Wendt sees self-restraint as an absolutely necessary variable in the formation of a collective identity. In the same fashion, Elias sees self-restraint as a necessary condition in the development of the civilised habitus. In this respect, Elias and Wendt have similar positions. Here, one is faced with an Eliasian concept par excellence. Significantly, Wendt’s writing (1999: 359) explicitly refers to Elias ‘[i]t is partly for this reason that Norbert Elias argues that self-control is the essence of civilization’, and references other Eliasian writers (van Krieken (1989) and Mennell (1989)). Yet, it seems that Wendt makes certain unnecessary conceptual moves. Wendt locates his explanation at the level of states as ontologically distinct social things, whereas, in fact, human actors make state figurations. States, by themselves, are not engaged in processes, real people are. Wendt puts himself in a contradictory position by recognising that all structures are processes. This statement surmises that they cannot be ontologically independent, thus Wendt’s arguments are antithetical. A state cannot simultaneously restrain itself and internalise norms because the state is not a person, but constituted by people. Only the people within a state can restrain themselves and internalise norms. The behaviour of states depends on the humans who constitute them.[12]

Two possibilities come to mind here. In the first case, when a significant number of people have internalised some habitus – like for example the non-use of torture (see Linklater 2007a) – this effectively results in a state that is ‘self-restrained’ or ‘self-limited’. In the second case, a small number of people have strong power relationships within the state figuration and can impose a greater or lesser degree of self-control on behaviour attributed to the state. Power relations then may play a more crucial role than Wendt seems prepared to allow in his theory. The resistance to a new habitus from specific segments of a society is linked to the power ratio these segments enjoy within society. This is brilliantly illustrated by Elias and Scotson (1994) in a more local context.

Cultures of Anarchy as Figurations

The cultures of anarchy can be seen as social figurations within which a specific habitus is internalised by people. In particular, the linkage with the concept of habitus can overcome some difficulties that arise with Wendt’s degrees of internalisation. Wendt describes three degrees of internalisation – by force, by self-interest and by altruism – in the respective cultures. But it is difficult to conceptualise how a Kantian culture could be internalised at the first degree – by force, and a Hobbesian culture at the third degree – by altruism. Wendt (1999: 273 and 303) is actually aware of these paradoxes. It may be more efficient to look at these internalisation processes through the tensions existing between the different layers of social habitus. This is the point of the following discussion.

Hobbesian Culture

A Hobbesian culture corresponds to a social figuration where the civilised habitus has not yet been internalised or where it has eroded due to a de-civilising process i.e. a move from self-control to external constraints. Often, this change is due to increased insecurity. This insecurity can be the aftermath of some collective disruption such as a social, economic, political or identity crisis. It is a process which looks similar to the security dilemma: a state figuration undergoing profound crises sees its degree of insecurity increasing which contributes to less self-control and thus, to more aggressive gestures, behaviours and discourses towards people, who in return, will see their insecurity increase. Elias speaks of double bind process (1987: 99 and 108). This can create a spill-over effect but one should not see a determinate mechanism. Indeed, this process can be affected by other actors that could, due to their situation of interdependence, influence the unfolding of events. For instance, the behaviour of the Bush Administration was interpreted by some as a ‘return of the state of war’ (Battistella 2008). But other actors can also intervene in the world figuration. For example, France, Germany and Russia played an active role to prevent the return of a Hobbesian anarchy. In fact, one can see that different anarchies co-exist in this figuration. It is certainly a factor of complexity; but also a potential factor of regulation of international relations.

Lockean Anarchy

The Lockean anarchy corresponds to a social figuration in which the national habitus occupies a predominant place. The civilised habitus is expressed only within the framework of the national/state-society. In other words, within the state figuration there is a strong self-regulation of aggression due to the monopoly of violence exercised by the state. While there are some forms of regulation in these inter-state relations, the balance between self-control and external constraints favours the latter, though in the case of an international society, the element of self-restraint and the we/I balance is rather different from that found in the Hobbesian case.

This Lockean anarchy can emerge in two ways. It can succeed a Hobbesian anarchy, or it can follow a Kantian anarchy. The first case can be illustrated by the development of state figurations and the establishment of the ‘Westphalian system’ in Europe. The development of self-restraint within European societies decreased people’s insecurity; who, then, decided to expand this idea. Moreover, a process of homogenisation was at play. In the second case, a major crisis can provoke a loosening of self-control, and cause a lack of confidence (in the national self and towards others). Such a case can be found in the degradation of the relations between France and the USA between the years 2003 to 2007. From friendship, rivalry developed (relations between Paris and Washington became acrimonious; but it was not enmity – a Hobbesian culture – as France was not labelled as part of the ‘axis of evil’). One could also easily imagine the European Union – a Kantian culture – disintegrating under the effects of the current social, political and economic crisis and, thus, the European states returning to a Lockean logic.

Kantian Anarchy

The Kantian anarchy corresponds to a social figuration in which the national habitus regressed to an inferior rank compared with the civilised habitus of the people. Indeed, states share a Kantian anarchy when they internalise the external constraints imposed upon them by the other states. They are capable of self-control in their use of violence, which is no longer an acceptable means to manage their relationship with other actors in the same figuration (Wendt 1999: 299). A national habitus still exists, but is less important as the We/I balance between state and international society, or better, humanity, changes. Both this collective identity and civilised habitus allow for the development of empathy and emotional identification, a psychological tool necessary for the second phenomenon stressed by Wendt: mutual aid. A Kantian anarchy can thus emerge from a Lockean culture. Within this last transition, states recognise each other as equals, learn to live together, and develop common norms, rules and values. Once these norms have been internalised and legitimised, i.e. when a Lockean culture is internalised at the third degree (by altruism), homogenisation processes begin, which allow for the diffusion of a civilised habitus, leading to the emergence of a Kantian culture.

But a Kantian culture can also emerge directly from a Hobbesian anarchy if the Hobbesian anarchy engendered such a trauma that the people feel an urgent need to implement a strict system for the regulation of violence. If one observes the case of Europe, it seems that the traumatic experience of the Second World War and the Holocaust led Europeans to move from a Hobbesian culture to a Kantian culture, i.e. a post-national (and maybe post-state) society, in a remarkably short period of time. The horrific experiences of this exceptional war forced Europeans to become conscious of their shared identity as human beings. It is because the whole système concentrationnaire (Rousset 1998) relied on the very negation of humanity that the Europeans were able to become conscious of their humanity once the war ended. The nature of the Nazi regime and its occupation of most European countries generated a shared experience of victimisation and (to varying degrees) of culpability. The rapid transition to a Kantian culture is even more understandable if we consider the time period leading up to the Second World War. In particular, the First World War was a huge trauma and was supposed to be the last war. However, the return of a Lockean anarchy in the 1920s and 1930s did not prevent an even more traumatic experience. Europeans had thus to adopt a more radical approach to their relationships which led to the development of a post-national community - the European Union - a regional Kantian culture. This has required a profound change in the personality structure of individuals. In other words, the national political communities need to be extended to integrate ‘men’ as ‘citizens’ to use Linklater’s terminology (1990).

Of the importance of history

Most English School thinkers privilege historical-sociological investigation. For instance, Martin Wight (1966: 96) wrote: ‘International society, then, on this view, can be properly described only in historical and sociological depth.’ Chapter 4 of Wight’s seminal System of States (1977) entitled ‘The Origins of our States-System: Geographical Limits’ offers an interesting discussion when compared with Elias’s own account of history and conception of time and duration. The chapter’s introduction explores the origins of the ‘Western states-system’ and concludes that several dates may be particularly noteworthy: the events in Italy in 1494 with its preceding events like the Congress of Mantua in 1457-60 described as the ‘first pan-European gathering to be frustrated by the national egoism of the powers’ (Wight 1977: 111); the Peace of Lodi and the Most Holy League of Venice in 1454 which ‘founded the Italian concert and the first system of collective security’ (ibidem). Or perhaps the date to be retained is 1522-23 during which the concept of a common Christendom is vanishing and the concept of raison d’état is gaining legitimacy. Of course Wight (1977: 113) recalls that the conventional starting-point is the Peace of Westphalia in 1648. But after this introduction, Wight (1977: 114) makes an interesting proposition: that a more appropriate historic bench mark would be 1492 and the ‘discovery of the Americas’.[14] This discussion is pertinent because it shows what Elias’s writings can offer at the theoretical level to such an historical perspective. Elias is adamant in reminding others not to look for origins or beginnings. The Western states-system is not a period with a starting date and an ending date, it is an historical process. Westphalia is the outcome of older processes and figurations of the relations of power; just as 1494 was an outcome of preceding processes. This idea was of course understood by Wight (1977: 152) when he wrote: ‘At Westphalia the states system does not come into existence: it comes of age. [emphasis added]’. The European states-system is always in transformation, ever moving, ever changing. Sometimes, particular events (1492, 1522, 1648, etc.) accelerated the pace of change and transformation. Elias, in his essay on time (1992), offers an informative discussion on the notion of time and duration. He privileges duration over time; as the école des Annales privileges longue durée over courte durée; processes over events.

What Elias’s writings offer the English School is a sociology of history or an epistemological reflection on the use of history. His theoretical insights help ground the historical work of the English school thinkers. In any case, Elias shares with these scholars a recognition that a historical approach is needed. But often Elias introduces sociological elements, while the thinkers of the English School tend to introduce legal perspectives. To summarise, the latter have more a legal-historical approach; while Elias favours a socio-historical approach. A strength of the English School’s approach is that it encourages Eliasians to take seriously law; a promise that has yet to be explored. A fruitful discussion has also started on the links to be made between the civilising process and the international society, a key concept of the English School (see table above) (see Linklater 2010b).

Overlapping cultures/figurations of anarchy

If Wendt clearly explained that the world we live in today is Lockean (Grotian), it does not carry any normative weight for him. In contrast, Wight sounds more normative as he insists on the Grotian notion of international society in the end, as we can see from the last section of International Theory: The Three Traditions (1994). However, a strength of Wight’s perspective is that he follows Grotius which affords us an opportunity to view the fact that there are different concentric circles within the system. In other words, some circles can be Hobbesian, others Kantian, others Grotian. Wendt’s viewpoint offers us a less nuanced understanding of the international political landscape in which the culture of anarchy characterises the whole system so that several cultures cannot coexist at the same time. This represents a very strong limitation and does not take into consideration the historical development of the European states-system and its diffusion to the world, how the various peripheries have been hierarchised and are still hierarchised in many respects despite decolonisation. Wight (1977: 128) importantly identified that: ‘Grotius presents the dual or concentric conception of international society.’

The term ‘dual’ is problematical because it pictures two circles located independently of one another. It is better to conceptualise them as overlapping. There is an overarching human figuration, intertwined with smaller figurations of people bound together by a historically-shaped space; they are ‘civilizational constellations’ (Delanty 2003:14), or need to be thought of as a ‘metageography’ (Jackson 2002: 458). But all humans are bound together by their humanness. Elias understood that humanity had become the last survival unit:

In the same fashion, the various civilisational figurations and the intermingling of civilising processes that have long influenced each other, exist only in relation to one another. Wight insists on the historical opposition between the Europeans and the ‘Turk’: ‘The doctrine of universality of human society found its greatest obstacle, in the sixteenth century, in the peculiar relationship between Christendom and the Turk.’ (Wight 1977: 120). This is a key-element in understanding the reluctant behaviour of Europeans to accept Turkey within the EU. But there is, on the longue durée, a deeply internalised habitus of constructing a European identity in relation to, in opposition to, the ‘Turk’. This opposition also allows seeing the intertwinement of different social processes as this process of Otherisation is connected to the process of secularisation, a fundamental process of modernity along with processes of centralisation, taxation, state-formation, individualisation, urbanisation, differentiation, etc. The historical depiction of Wight is about two religious groupings: Christendom vs. the ‘Turk’ (a word that was synonymous with Muslim and did not refer to a specific nation). Today, we face a supposedly secularised Europe vs. a secularised Turkey (legally Turkey is certainly one of the most secular states in the world (Çinar 2005; Burdy and Marcou 2008). Yet, most discourses of the opponents to Turkey’s EU membership are constructed in religious terms (although often disguised under cultural and civilisational terminologies). Some of these opponents were the very same people who wanted to include in the preamble of the defunct European Constitution a reference to Christianity (Casanova 2006). Therefore, to fully understand the opposition discourses on Turkey’s membership, one has to look at it from an historical perspective, but also in terms of personality structures (especially the persistence of an old religious habitus), and power relations (Europe is less dominant than she used to be with peripheral states that are less and less peripheral). This example illustrates how the historical perspective of the English School can be productively combined with other important insights from the process and figurational sociology of Elias to become very powerful tools to develop a deep understanding of contemporary cultures of anarchy and relationships between states.[15]

Europeans states today form a Kantian culture of anarchy that Turkey wants to join. Europe also shares a Kantian culture with the United States, Canada and Israel; she shares a Lockean culture with most states on the planet. But these figurations are very malleable. For instance, Libya very quickly became an enemy (Hobbesian) instead of a rival (Lockean). After a few weeks of military operations and the assassination of Khadafy, Libya seems to be again a rival. Several figurations co-exist and change constantly. Therefore, as I have outlined above, no one single culture of anarchy exists within the international system due to the fact that so many various hierarchies are present in the system and the dynamics of the international system are more fluid than usually assumed.