The evolving concept of liminality: symbolic interactionalism and post-structuralism

Victor Turner, still the leading name within symbolic anthropology, evolved van Gennep's perspective on the social experience of being in between states.Footnote ² For Turner, liminality was a condition of being betwixt and between socially established categories, and not simply the condition of being in the midst of two stages in a ritual. Liminality could also be the condition of being suspended or even trapped between two different sets of role expectations, a condition often leading to impassivity, or even to a social impasse. Turner inspired a whole string of anthropologists who have put the concept of liminality as a social category to good use, most famously perhaps Mary Douglas. In her book Purity and Danger, Douglas focused on the problem of categorisation. Douglas charted in detail how one animal, in one social setting, was seen to be betwixt and between categories, and so effectively beyond categories (or hors catégorie to use a language where this condition has been captured in idiomatic form). She then went on to broaden the perspective by drawing in a plethora of other examples, and landed on the conclusion that where pure categories do not apply, feelings of insecurity and danger ensue. Douglas does not put the category of liminality at the centre of her analysis, preferring to focus on the conceptual pair of purity and danger, but her theorising clearly hails from the liminal tradition. The perspective has changed a bit, too; instead of focusing on temporality, as did van Gennep and Turner, Douglas is interested in how the boundary is a place inscribed by a particular mood, namely fear.

Where there is fear, there should be IR scholars, for fear gives rise to policies, many of which are geared towards heightening security. All three articles in this Forum draw attention to the importance of liminality as a concept that focuses on process, and so has the capacity to challenge the dominant structural perspective on IR. They also highlight the familiarity of the concept of liminality to other concepts that have recently had considerable purchase in social sciences. Mälksoo mentions ‘hybridity, interstitiality, creolisation, marginalisation, and carnivalisation’. Rumelili lists ‘hybridity, margins, borderlines, interstitiality, transversality’. Most of these terms are, as Mälksoo points out, associated with post-structuralism. This is hardly surprising, given that the ‘post’ in post-structuralism is there because of a perceived need by scholars in the 1960s and 1970s to move beyond the hunt for a latent structure underlying society at large, and rather concentrate on the manifest structures, their emergence, and the relationship between them.Footnote ³ They are post among other things because they are not primarily interested in structure as ordering, but structure as productive of margins and marginalised agents. Such a project by necessity draws attention to discursive margins or borderlines or boundaries and the processes and states of being one may find there. Note the similarity between the binary opposition structural/post-structural and the binary opposition that constitutes the title of Turner's book: ‘structure and anti-structure’.

Already in 1967, Foucault hatched the concept of heterotopia for utopian (non-)spaces where social hierarchies were suspended.Footnote ⁴ This concept has been put to good use in studies of typically liminal activities such as holiday-making or the living out of a so-called gap year (a year or more between the end of high school and further education or work, typically involving the very liminal activity of travelling). In 1980, Gilles Deleuze attempted to theorise the age as one of returning nomadism, and as seen from the point of view of the sedentary, the nomad leads a liminal existence.Footnote ⁵ Liminality found its way into political science when Anne Norton nouned it in 1988, naming those who are in a liminal phase liminars.Footnote ⁶ In this Forum, Stoicescu retains the use of this concept for entities in a liminal state. I picked up the idea and applied it to European politics in the mid-1990s.Footnote ⁷ This may have been an onomastical first, but I was late to the ball. Post-structural scholarship within IR had already evolved the same kind of ideas under other rubrics, perhaps most prominently by unpacking the concepts of borders and borderlines.Footnote ⁸ Below, Maria Mälksoo suggests a fourth post-structural link, which is to link liminality up with Giorgio Agamben's concept of state of exception, with the key idea being that contemporary rationality of government is predicated on the idea that we live in unusual times – and that times simply go on being unusual. In such times, general social categories and narratives for how to do things are suspended in favour of what is supposed to be a temporally bounded state of exception. The possibility stressed by Agamben is that temporality has been moved from being interstitial to being permanent, in which case liminality becomes indefinite. It is as if van Gennep's initiators, the people who were to inculcate order onto the naked boys who were about to be initiated, had gone on to subject those boys to different kinds of trials into their manhood and old age.

The meeting between van Gennep's concept of liminality with post-structuralism is, as already hinted at, a reconciliation of sorts, for van Gennep was the unwanted element in Durkheim's school, and we may trace post-structuralism back to a break with Lévi-Strauss, whose theorising quite explicitly took place in a break with Marcel Mauss's work, and Mauss was the nephew and major successor of Durkheim. Indeed, already Douglas's work owed a lot to Mauss and his prolonged struggle with the role of categorisation in social life.Footnote ⁹ I draw attention to all this here for two reasons. First, although liminality may parade as new in an English-speaking non-anthropological setting, the term as well as the broader conceptual family of which it is part is at the heart of continental social theorising. Secondly, and relatedly, if a specifically anthropological concept of structure could be imported into IR, as did Waltz with the structural-functionalist concept of structure as conceived by Nadel and others, it must also be highly apposite, not to mention overdue, to do the same for a processual and anti-structural concept like liminality.Footnote ¹⁰

Post-structuralists and symbolic interactionists do not try to negate structure, but to pair it up with process, usually by using the concept of practice. The spirit of this undertaking is nicely captured by Stoicescu when she notes that ‘reading the world through the lenses of structures is neither good nor bad, but simply incomplete’. This Forum does not set out to shoot down other perspectives, but to complement structural perspectives by putting a concept at centre stage that sensitises us to the glitches left by structures, the positioning of certain actors within these glitches, and the strategies that they forge to deal with liminal life in sites between categories forged by others. All three authors offer readings of liminality along these lines. To Rumelili, liminality is exactly a sensitising concept that may steer our analytical focus towards studying marginalised agents and their strategies. To Mälksoo, it serves as an antidote to essentialism, privileging becoming over being. To Stoicescu, it helps us ‘understand what are the different, often contradictory, parts that make up’ an actor's identity, and also to identify the ‘tools, materials and strategies’ at that agent's disposal.

Liminality as a relational concept

It should be noted that symbolic interactionist anthropology and post-structuralism are not the only perspectives that have been drawn towards boundaries and the liminal in recent years. One prominent example concerns Andrew Abbott's work on what he refers to as things of boundaries.Footnote ¹¹ Struggling to produce relational sociology geared towards the production of knowledge, Abbott fastens on the material markers of boundaries as artifacts that are being claimed and contested by multiple parties, and so act as focal points for boundaries between social groups. His is a relevant perspective to liminality because rather than ask about the importance of the temporality of the liminal phase, as do van Gennep (within ritual) and Turner (within social life at large), or the indeterminacy of liminality, as do Turner (from the perspective of social action) and Douglas (from the perspective of social classification), Abbott stresses the importance of studying contestations over material things in order to get at the logic of the liminal.

If we look for a least common multiplier that may embrace all these perspectives, we can do better than talk in general terms about the importance of the social for global politics. More specifically, symbolic interactionism, post-structuralism and Abbottian sociology may all be put to work within what is called the relational turn in IR. In order not to essentialise social phenomena, one takes as one's starting point the relation between two actors, rather than the agency of the actors involved.

This begs for the question of how far one may take the programme of anti-essentialism. It will not do simply to declare that one should shy away from laying some kind of ground or foundation for inquiry. One has to start from somewhere. Already the use of spatial metaphors – ground, foundation – goes to show that certain parameters have to be laid down and taken for granted. This is due to the fact that without categories to order social reality, humans may not live in, let alone study, social life. Maria Mälksoo is right to stress the anti-essentialist potential of the concept of liminality in her contribution, but we need to be clear about the limits to this project. In an incisive work, Kripke demonstrates how the very act of naming people is an act of violence exactly because it essentialises being. We have countless examples of people who try to defy this logic by remaining anonymous, using aliases, etc. The rock musician Prince took this logic to successively new levels when he first insisted on being known as ‘The artist formerly known as Prince’ and then by dint of a glyph, it simply did not work. The categorisation system of naming turned out to be so integral to social life that going without a name spelt (to use an appropriate verb) disaster in terms of, again, name recognition. The upshot is, I think, that the best we can do is to adopt what Gayatry Chakravorty Spivak calls a strategic essentialist stance, that is, to adopt as our own a standpoint from which to view the social that we occupy for a limited time and for strategic purposes, and do not claim a privileged epistemological status for.

As betrayed by liminality's genealogy, liminality is a function of categories. It would by definition be impossible for a social world to exist without shared categories. If we grant that no scheme of categorisation can be all-inclusive – a post-structural credo that it has been hard for opponents to shoot down – then it follows that liminality will exist within any social order. We are left, then, with a Foucauldian insight: any social order has its marginalised groups. What liminality as a category can do for us, is to focus our gaze on these groups. It cannot preclude us from categorising these groups as marginal, or make us come up with a wholly anti-essentialist way of living or of doing research. Such an undertaking remains beyond our grasp. Liminality's focus on temporality and becoming privileges process over structure, but it is not a gateway into unfettered anti-essentialism. Mälksoo's discussion of wartime as a liminal period furthermore underlines the ethical issues involved. Mälksoo is right to stress how war is a state of becomingness and hence of liminality. It is, perhaps, the liminal social activity par excellence. But look where that may lead us. If we believe that war is the father of all things, it may easily lead to the celebration of war. Alternatively, war may be treated as a pudenda origo; a dirty origin. Either way, we are led in the direction of celebrating liminality in general, and war in particular, as something positive.

War has temporality. The similarities between the concepts of living on borderlines, being in heterotopia, and experiencing a state of exception and liminality seem to lie in the degree in which all these states of being are conceived to be temporal, but may, for better (heterotopia) or worse (state of exception), turn into permanent states. Here we have a suggestion of permanent marginality and marginalisation that has followed the one of liminality like a shadow from the very start, but which has its own history, going back at least a century, to Georg Simmel's celebrated essay on the stranger, who ‘is an element of the group itself … whose membership within the group involves both being outside it and confronting it’.Footnote ¹²

All these concepts play on the impossibility of permanent categorisation, understood as a permanent dividing up of the world into mutually exclusive bundles, and I think they point, at least implicitly, to the inevitability of heterodoxy, hybridisation, and pluralism. Small wonder, then, that these ideas have had an immediate appeal to IR scholars who are working under and trying to understand the transformative effects of globalisation.

I note a hybridisation of the concept of liminality itself here. Van Gennep was interested in tightly scripted rituals playing themselves out in traditional societies. He was, after all, an ethnologist. The way the concept of liminality is used in this Forum, it speaks to post-modern conditions. By definition, those conditions are neither tightly scripted nor traditional. The explanation to this seeming paradox seems to lie in the object of reference for the state of liminality. To van Gennep, the society from which the boys were separated and the society they returned to was the same, stable society. The situation of liminality pertained only to that little group of people who were taken out of it and returned to it. Even that phenomenon was a stable one, for boys came of age physically all the time, which meant that the ritual which would make them come of age socially had to be performed at regular intervals. If we look at the cases analysed here, there are no stable societies from which to be taken away and be returned to. With everything in flux, van Gennep's scheme of a pre-liminal, luminal, and post-liminal phase collapses, and the possibility of perpetual liminality opens up before us.

The implications of this for IR go beyond the field of identity, which was where the concept was applied in the first place. Throughout human history, certain polities have been on the move.Footnote ¹³ Some 16,000 years ago or less, the phenomenon of the sedentary polity emerged. During the first millennium BC, a major drama for all the three principal sedentary societies – the Chinese, the Persian, the Byzantine – was how to keep the nomads at bay.Footnote ¹⁴ The nomadic polity had its last moment of glory – and what a moment! – as Djenghis Khan and his Golden Kin conquered the known world. And that seemed to be the end of the nomad. The picture that we are presented with in this Forum, however, is one where, yet again, nothing stands still. The nomad, arguably the liminal element par excellence in Chinese and Persian empire formation as well as in European state formation, seems to be back in another form. If so, then Westphalian IR, which is nothing but rooted to the spot, needs all the conceptual help it can get. Read in this light, this Forum is an invitation to letting go.

The contributions

Mälksoo's article, which opens the discussion, begins with what is arguably the most tightly organised discussion of liminality to date. Hers is a programmatic piece, where she argues the case for breaking with a structural understanding of global politics, and for why liminality is an apt concept with which to do so. One notes that Mälksoo writes from a background of studying Estonia and Poland, two cases of state entities being seen as and seeing themselves as liminal to two other cultural entities: Russia and Europe.

Rumelili's case, here as elsewhere, is Turkey as seen from Europe. Turkey has been liminal in the sense of being what Stoicescu calls a ‘little self'; an entity seen as copying the self but not quite managing. As Rumelili puts it, the ascendance of civilisationist discourse in the eighteenth century and their confluence with the more particularistic discourses on empire have positioned the Ottoman Empire, and also Turkey, as liminal. In recent years, however, Turkish foreign policy has responded to this by trying to hold Europe, in institutional terms the EU, to their own standards. The liminal is becoming the umpire of proper behaviour.

Stoicescu, who has previously written about Ukraine, Turkey, and Romania, fastens on the latter case. She argues that liminality may take two forms; appearing as a little self or being at the subaltern end of incompatibility of categories. Here she is on safe historical ground, for, as stressed above, the concept began its history in van Gennep's hands, as a way of analysing little selves, and was then evolved by Turner, who fastened on general social impasses. Stoicescu's argument is that, over the last twenty years, Romania has gone from trying to insist on being not small but big (in casu, not South-Eastern Europe but Central Europe) to casting itself far more successfully as somewhat incompatible and hence interesting to Europe.

It is no coincidence that Europe's Eastern seam is the region under scrutiny here. This is an area where there has been intense contestation of identities since the end of the Cold War, and where identity politics has been paired directly to the institutional question of membership in the European Union. That has made it possible for scholars to look not only at process, but also at outcome (membership and degrees of non-membership). As seen from Europe, and also from Russia, the area has traditionally been seen, and has experienced itself to be, on the borders between, or liminal to, East and West. Not least, for the last decades, the area has been one of the key growth areas when it comes to producing IR scholars. The CEEISA, which is the Central and East European sister organisation to the British International Studies Association, which publishes the Review, has been the discipline's fastest-growing organisation in recent years. For obvious reasons, the costs for scholars of studying areas with which they are linguistically and socially familiar from childhood are rather more limited than of studying other cases. It is, therefore, somewhat over-determined that Europe's Eastern rim is the geographical area era of study here.

Liminality is, however, a concept rooted in the social as such, pointing to a phenomenon that occurs everywhere. The cases traditionally studied by means of it were, for example, African. There is no theoretical reason why the contributors to this Forum should not go on to see the concept put to use in the study of global politics everywhere.