A problem with levels: how to engage a diverse IPE^*

Abstract

Though welcome, Cohen's call for exchange across diverse perspectives in international political economy (IPE) evades the question: why have we remained unaware of or insensitive to the diversity that already exists? We follow John Hobson's claim that racism, imperialism and Eurocentrism disallow a western-dominated social science from engaging with diverse viewpoints. We argue further that a disciplinary bias towards a unit-level or atomistic understanding of social science precludes and disallows epistemological encounters in which actual diversity might be harnessed. We support this claim in two steps. First, we draw on Ghassan Hage's analysis of exigophobia, or the fear that social explanation inadvertently justifies horrendous actions and humanises their perpetrators. Exigophobia activates what we call the condemnation imperative: an eagerness to condemn an individual or group act, of fierce violence, for example, before one has tried to understand or explain it. Second, building on Nicholas Onuf's work on levels, we show that the disciplinary bias towards explanations which 'see' from the level of individual actors treats Europe or the west, in Hobson's terms, as 'self-constituting and exceptional'. When one neglects the structuring features of the whole, and assumes western 'pioneering agency', it is easy to treat non-western inferiority (irrationality, backwards culture, and so on) as an explanation of the relative successes and failures of a flattened planet of autonomous units. Though we endorse forms of social explanation that start from the whole as opposed to the parts, we favour the view that there are only simultaneous and continuous processes whose seeming mystical flow our descriptions cannot but freeze. We suggest that there are no levels, simply parts and wholes in process.

Keywords:
Eurocentrism; Condemnation Imperative; Exigophobia; Atomism; Social Explanation; Levels; Parts and Wholes; Process

Introduction

We welcome Cohen's initiative to produce a global IPE conversation, and we applaud what the editors refer to as 'the diversity, pluralism, and openness he encourages'. Similar concerns have motivated our work (see Inayatullah and Blaney 2004). We worry, however, that Cohen's geo-epistemological mapping of various IPEs may come to little if this account turns our attention away from the deeper reasons for the absence of such a global conversation. If the obstacles to a global conversation remain, the intended results will not ensue even if tools are precise, energy is abundant, and intentions are noble. Accordingly, we seek to locate the impediments to such a conversation. As we see it, the query 'how do we start a meaningful conversation that engages the diversity of IPE?' requires a prior question: 'How have we remained unaware of or insensitive to the diversity that already exists?'

John Hobson (2012, 2013a, b) produces a powerful answer to the latter question. He argues that racism, imperialism and Eurocentrism disallow a western-dominated social science from engaging with diverse viewpoints. We find Hobson's arguments and the evidence he marshals compelling. Yet good arguments and weighty evidence may not be convincing unless we begin to understand the 'how': how exactly do those who profess to pursue value-free science nevertheless produce it in a manner that is racist, imperialist, and Eurocentric? We argue that they do it via a disciplinary bias towards a unit-level or atomistic understanding of social science. It is this habit that precludes and disallows epistemological encounters in which actual diversity might be harnessed. Our task is to substantiate this answer.

Our presentation contrasts accounts that favour unit-level characteristics with those that foreground the culture, history, and context of the actions in question. We refer to the latter as social explanation.¹ This dichotomous contrast between atomistic, unit-level accounts and social explanation allows us to make sharp differentiations that may motivate and compel readers. However, in the conclusion, we move to a more nuanced and supple position that, we suspect, will strike some of our readers as rather speculative even as others may welcome its belated arrival.

What we are calling social explanation cultivates the intrinsic virtues of social theory, but it also, we argue, triggers a deep anxiety: that an explanation foregrounding culture, history and the context of actionjustifies atrocious conduct. In his reflections on the social science of suicide bombing, Ghassan Hage (2003) calls this discomfort exigophobia: the fear that social explanation inadvertently justifies horrendous actions, and humanises perpetrators. Exigophobia activates what we call the condemnation imperative: an eagerness to condemn an individual or group act, of fierce violence for example, before one has tried to understand or explain it. Specifically, the condemnation imperative is triggered when explanations of social phenomena seem to excuse actions that are ethically reprehensible, and is especially tricky for those whose profession it is to explain such acts. The logic of contextual explanation, the presence of exigophobia, and the condemnation imperative work together to produce the individualism imperative: a tendency to favour explanations tied to the characteristics of individualised units, and to devalue explanations that rely on social, historical or institutional context.

We can say all this in simpler language: theorists and lay people alike worry that understanding and explaining horrendous acts also mean defending, rationalising, affirming, or justifying them. To push away this fear, they condemn such acts before they explore their making, defaulting to explanations that place the blame and responsibility upon individuals (whether persons or larger units) rather than upon those that point to the larger context. In this way, exigophobia, the condemnation imperative, and the individualism imperative work together to produce what Hobson calls the racism, imperialism and Eurocentrism of western social science, and therefore of much of IR/IPE.

One more remark before we begin our investigation. As we expose the usual habits of dominant forms of western social science, we also affirm a standard through which we develop our critique. We refer to this ideal as the ethics of social explanation. Such an ethic abandons strategies that either reject or assimilate the other, and instead require regarding the other as both universaland particular. When the other's humanity is restored in this way, the other's similarity and difference appear first and foremost as resources for self-understanding and dialogue. Overcoming the individualism imperative may help us move in directions envisaged by Cohen's call.

Hage: social explanation, exigophobia, and the individualist imperative

Hage explains the emergence of Palestinian 'suicide bombers' in his journal article '"Comes a time we are all enthusiasm": understanding Palestinian suicide bombers in times of exigophobia' (2003). Simultaneously, he deciphers why his explanation of suicide bombing falls foul of his students and fellow academics. He discovers, much to his annoyance, that commentary on Palestinian suicide bombers must begin with a moral condemnation. Prior to saying anything about their motives, tactics and circumstances, one must first reject 'suicide bombers' as beyond the pale, as absolute otherness outside the realm of explanation. Hage confesses that this 'condemnation imperative' places him in a performative contradiction: before he can explain or assess the actions of Palestinian suicide bombers, he must first judge their actions to be reprehensible and barbaric (2003: 67). He must give the science before the science, as Marx might say. Hage accepts this predicament in order to cast light on the situation. He humanises 'suicide bombers' by proposing a hypothetical sociological/anthropological project which he ties to the term 'social determinism',² but which we will refer to as social explanation.

Along the way, he produces three intriguing insights. He discovers: (1) a fear of social explanation, or what he calls exigophobia; (2) the partisan politics/ethics of social scientific explanation; and (3) a methodological move we are calling the individualism imperative. The last requires further explanation, a task we take up in the next section. In our reading, these three strands of Hage's discovery are reducible to this question: why does dominant, methodologically individualist, western social science fear the ethics of social explanation? Producing an answer to this question requires that we move carefully through the last two sections of Hage's article.

Hage relates an encounter with a student that exemplifies his and our predicament. Having conveyed to a seminar the steps necessary to devise an adequate sociology and anthropology of Palestinian 'suicide bombers', he notes the discomfort of his students. One of them says, 'You've made it as if suicide bombers are ordinary human beings' (2003: 83-4). Hage contemplates this comment, and wonders if this conversion of the seemingly extraordinary to the commonplace isn't 'what is always at stake in social explanations' (2003: 84).³ Explanation makes seemingly anomalous events the mundane result of a recognisable process. If x, then y; in situations where one party monopolises the means of violence, and subjects a second party to an unviable loss of social dignity, it should not surprise us if the second party organises her society in such a way that some living bodies become weapons whose detonation serves to make the present meaningful, and the future hopeful (see Abufarha 2009). Such explanations can make suicide bombers seem understandable as 'ordinary human beings'.

More is at stake, however, than a familiarity with and an understanding of 'suicide bombers'. The more dangerous implication is that any society put in such a situation will respond in a similar fashion. It is this inference we may fear as we shift locating the cause of an event from the individuated unit to that actor's social and historical context. The worry produced by a contextual social explanation is that they and we are made of similar stuff, and that, when placed in similar situations, wewould respond in similar fashion. We resist this movement, argues Hage, because we fear seeing ourselves in them. Our fear of their difference (theyperform suicide bombings, we do not) is secretly a fear of our similarity (we might do the same when put in their situation).⁴ The truthfulness or actuality of this claim may not be germane. What is important is the anxiety itself that can be fully activated by the mere possibility.

This insight is followed by a second, bolder, claim: the ordinariness of Palestinian 'suicide bombers' is also a claim about the ethics of social explanation which, as Hage asserts, connects us to the mundane humanity of perpetrators:

War emphasises the otherness of the other, and divides the world between friends and enemies and good and evil. This war logic is negated in a social explanation that draws on an ethics of social determinism. By proposing that the other is fundamentally like us, social determinism suggests that given a similar history and background we might find ourselves in the other's place. When we explain an act as the product of a particular history and particular social circumstances, we give its perpetrators some of their humanity back (2003: 86-7, emphasis added).

If so, retrieving perpetrators' humanity is what is at stake in social explanation. The other's humanity is retrieved whether social explanation makes the other's difference familiar, as in this case, or makes the other's seeming sameness appear as a partial difference. Again, these 'insights' need not be verified to produce anxiety. It is the potential threat of discovering a perpetrator's normality that can produce both the fear and condemnation of social explanation.

If what we have argued so far seems plausible, it brings us to a complex position in relation to the other: the other is not us; nor is the other not-us. Rather, the other is both us and not-us at once. Assimilating the other's difference erases the particularity of the other; rejecting others' similarity to us overlooks their universality. Both moves deny the other's humanity. Social explanation can humanise others by embracing both similarity and difference. This process, Hage points out, is not an impartial act: 'In taking the side of social explanation, one is clearly not inhabiting a politically neutral position' (2003: 88). Such partiality, he quickly asserts, does not preclude our ability to condemn the very actions we are trying to explain. Having retrieved the humanity of perpetrators, we can still oppose their actions, so long as we are willing to admit that by so doing we also condemn our own potential actions, should we find ourselves in a similar situation. As we condemn their actions, we condemn our own - as forgotten past, denied present, or potential future.

Pulling these elements together, we might say that social explanation highlights particular actions in particular situations via generalisation, and that doing so requires regarding the other as both universal and particular. Simultaneously, social explanation reveals an ethics that resists both the assimilation and rejection of the other. This is the standard by which we assess western social science, and through which we hope it can live up to its own ideals.

Despite this logic, most social theorists hesitate to embrace such an ethics. We resist the humanising move, the acceptance of the vilified as 'ordinary people'. Why? Perhaps we fear a loss of judgment, an inability to take a stand, and the dissolution into a relativist quagmire. The tension between the logic of social explanation and our difficulty in accepting its consequences may be described as a hesitation, or a stuttering - a seemingly involuntary disruption of speech and thought, or a deep-seated phobic reaction (exigophobia, in Hage's terms) to the humanising role of explanation. Our desire to override such hesitation expresses itself as the 'condemnation imperative'. As Eagleton (2010: 8-11) notes: 'No western politician today could afford to suggest in public that there are rational motivations behind the dreadful things that terrorists do. "Rational" might too easily be translated as "commendable".' And, asHage (2003: 66-7) argues, the failure to condemn in scholarly discussions makes you 'a morally suspicious person', associated with, if not culpable for, the malevolence itself. Both Hage and Eagleton suggest that pursuing social explanation - explanation that humanises the other, that refuses to locate the other as absolute difference - associates the scholar with nefarious purposes. This would seem to foreclose social explanation altogether, replacing it with the cant of the condemnation imperative.

But this resistance to social explanation as a humanising strategy does not simply foreclose social enquiry; it channels it in certain directions. The condemnation of individuals or groups as evil presumes and invokes the idea of autonomous and responsible individuals whose motives and actions are separable from ours. WilliamConnolly (1991: 1-2, 99) emphasises that the language of evil is 'bound up with the issue of responsibility ... some agent must be responsible for it'. If others are responsible for evil acts, then no further social enquiry is required; the motive of individuals or groups or the context of their action is mooted. And, he stresses, by intensifying 'our search for responsible agents at the level of individual or group, we 'exclude some injuries from the category of evil' (1991: 1). More precisely, as Eagleton (2010: 143) suggests, we exclude those injuries for which an account in terms of individual responsibility is difficult to sustain, such as where 'wickedness is institutional ... the result of vested interests and anonymous processes, not of the malign acts of individuals'. These 'forms of wickedness are built into our social systems, within which individuals are deeply conditioned by their circumstances' and in which we may be implicated (2010: 144-5). Connolly and Eagleton are pointing to the kind of social explanation advocated byHage (2003: 86-7) that attends to 'the social conditions of action and the historical conditions of formation of the acting self'. They suggest that explanation which humanises others produces a resistance that prevails not only in Hage's seminars and at scholarly conventions, but also in public debate and in social theorising more generally. They also suggest that resistance is intimately interwoven with a focus on the individual or group as a separable unit, and a unique locus of agency.

The hints provided by Hage, Connolly and Eagleton imply that we can link this stutter or involuntary disruption in our political/ethical sensibilities to theindividualism imperative: the tendency to ignore all levels of analysis in favour of the individual level. Here, actions emerge not from context or structure, but from the autonomous will (or traits) of the individual unit. The individual unit is the locus of the action, and this will or these traits must be 'blamed' for an insufficient reading of what is good, right, and proper in the world. The western social scientist is thereby caught between the humanising ethics of social explanation and the ready opportunities to blame the individual or group for poor decisions. Hence the hesitation, or stuttering.

To foreshadow our point of arrival, we see two responses to this stutter. One can discount the role of individual will or traits, and move towards a full ethics of explanation. Or one can emphasise the individual unit by disavowing the ethics of explanation, or disavowing it in the last instance.⁵ In the latter case, one moves to a kind of essentialism, or atomism. This is the individualism imperative. This imperative, as we shall see, goes hand in hand in IR and IPE with a refusal and rejection of the global 'level'. This refusal also facilitates the Eurocentric myths that keep self and other separated, that produce the stutter, and resist the humanising move of social explanation. We trace the work of the individualism imperative in the following section, showing how it has come to define conventional international relations as a positive, supposedly value-free science.

Levels, the individualism imperative, and Eurocentric condemnation

We think with metaphors; indeed, we cannot think without them (Lakoff and Johnson 1980). For example, in thinking about the planet, we use visual metaphors that organise space. We imagine its three-dimensional oneness via satellite pictures of the earth as a blue globe, and we see differences within and across that oneness via maps that show how the various continents are divided into states. IR arranges these images via the metaphor of 'levels'. But IR's general bias is towards the unit or actor level. This emphasis on unit-level explanations leads international theory to neglect the very processes that produce relations of global inequality and domination. Our wager in this section is that a focus on levels helps us locate the deeper root of how IR/IPE becomes racist, imperialist, and Eurocentric, thereby undermining its own desire to engage other, more diverse viewpoints.

In his classic paper on levels of analysis in IR, J David Singer begins by admitting that levels create a problem of choice:

In any area of scholarly inquiry, there are always several ways in which the phenomena under study may be sorted and arranged for purposes of systemic analysis. Whether in the physical or social sciences, the observer may chooseto focus upon the parts or upon the whole, upon the components or upon the system (1961: 77, emphasis added).

According to Singer, the word 'choose' suggests opportunities for a pluralism of scholarly perspectives: 'For a staggering variety of reasons, a scholar may be more interested in one level than another (in parts as opposed to the whole) at any given time, and will undoubtedly shift his orientation according to his research needs' (1961: 90). But, as Onuf (1995: 35) notes, Singer boils the choices down to 'two levels - the behavioral and the systematic,' and scholars' research problems dictate whether they highlight parts or whole and therefore the level they select.⁶

Robert Cox shifts the language somewhat by emphasising the role of habit or convention in disciplinary practice:

Academic conventions divide the seamless web of the real social world into separate spheres, each with its own theorising; this is a necessary and a practical way of gaining understanding. Contemplation of undivided totality may lead to profound abstractions and mystical revelations, but practical knowledge ... is always partial or fragmentary in origin. Whether the parts remain as limited, separated objects of knowledge, or become the basis of constructing a structured and dynamic view of larger wholes is a major question of purpose and method. Either way, the starting point is some initial subdivision of reality, usually dictated byconvention (1986: 204, emphasis added).

Cox seems to suggest that one's orientation towards parts and whole is not so much chosen but acquired through disciplinary socialisation. To further invert Singer's claims, we might say that levels schemes choose the scholar, and - having been selected (or initiated into disciplinary conventions) - scholars organise their work according to a pre-decided understanding of levels.⁷ And, Cox stresses, our 'choice' of analytic stance is not innocent. We can read him to say that a discipline organised around 'parts' tends to support a 'problem-solving' mentality that takes existing structures of world order for granted. Only when scholars are disciplined to focus on the whole can they adopt a critical stance towards world orders (1986: 207-10).

We follow Cox's lead by exploring the political and ethical implications of being selected by a level. We read along with Nicholas Onuf (1995: 43), who has done the most thorough and sophisticated work on levels: 'Levels schemes are all members of a family of pictures, or framed spaces, within which we see the contents of the field of study we call international relations.' Pushing against Singer's notion of the choice of levels following from our selected problem, Onuf seems to indicate that the levels scheme precedes that selection. As he puts it, 'the "level" is just one in a family of spatial metaphors for ways of seeing. They tell us how we see, and not what we see' (1995: 41).

Onuf implies that levels schemes show us how to see the contents of the world. Put more starkly than Onuf might like, we can say the level selects the scholar, and then conditions how that scholar can see, and what counts as enquiry.

But, in apparent contrast with Cox, the implications of levels schemes seem to have little political importance for Onuf. Indeed, he suggests that levels are free of ethical implication:

As a family of pictures, levels schemes are horizontally oriented. Top and bottom lines frame the picture. Levels are lines parallel with top and bottom, each functioning as top and bottom for pictures encapsulated in the larger picture. Up and down are locational instructions; only the observer's angle of vision and focus change. While "up" generally connotes good and "down" bad ...,levels schemes have no such implications (1995: 44, emphasis added).

We want to dispute this last claim, namely that levels schemes carry no connotation of good or bad.⁸ Some of the challenge we can muster with Onuf's own words. For Onuf, levels shape not only how we see but also how we make our disciplines, and how we make the world. As he puts it:

Levels are not just a taxonomic convenience for scholars, or a methodological convenience. They are a potent metaphor, an ancient convention, for marking, and thus making, wholes. In our culture, as in our field, we would have difficulty getting along without the language of levels. In other cultures people make wholes of their own ... and mark their significance with conventions we may not even recognise (1995: 53).

For example, Onuf reports that Aristotle locates the 'complete good' in the self-sufficient community that stands above its many parts. Similarly, Christian cosmology associates the complete or final good with 'with God and the heavenly city' that stand above the 'social arrangements' which order the human beings who reside below that social order. The cosmos was visualised as a 'great chain of being ... [o]rdering all things from least to greatest, lowest to highest' (1995: 48-9). As individuals, we look up to locate the good - towards God, and the social order sanctified by God. Here, up is good.

But, by the 17^th and 18^th centuries, 'an alternative visualisation gained favour': 'After Kant, metaphorically, the great chain tipped over - levels became stages' (1995: 49). In place of God's order and God's goodness (descending from on high), we come to see nature as governed by its own laws. So too with humans: God's laws are replaced by those humans make for themselves. And a notion of sovereignty follows, as Onuf (1995) explains: 'If nature's design fosters horizontal representation in space, so also does the emergence of autonomous sovereignties' ... ; the 'flat territorialities of states' come to replace 'overarching claims to hierarchical authority'. A modern cosmology challenges the notion that we locate the good above. The locus of the good now shifts downward to a lower rung on the great chain - to individual members of the community, and the independent states that now govern this horizontal space. Down is good. Social order, which is up, appears not as a realisation of the good of individuals, but as a potential constraint on the autonomy and responsibility of individual or sovereign actions.

Thus the 'flat territorialities' of states suggest an IR of 'like units'. The claim of sovereignties suggests an IR of independent actors that appear, in principle, as formally equal. Where the units are privileged as the locus of the good and of creative action, explanatory emphasis is placed on the actions of the units as opposed to the determining effects of the structuring of the whole. Singer captured this opposition in his own discussion of parts and wholes: looking at the world from the level of the individual unit emphasises the concrete, and moves away from abstraction. Looking from the system or the whole, 'we tend to move ... away from notions implying much national autonomy and independence of choice and towards a more deterministic orientation' (1961: 80-1). We catch a hint of favouritism: an emphasis on the whole smacks of the kinds of historical arguments made by dependency theories that so violated the commonsense notion of the world as looked at from below - from the point of view of the behaviour (and freedom) of individual actors.⁹

The flat map of formally 'like units' does not preclude the assumption of graded selves or states, however. The flat map of sovereignties coexists with inequalities of position within a capitalist division of labour emphasised by theories of dependency and underdevelopment. But although those inequalities are taken to indicate 'higher' and 'lower', where higher suggests more civilised, more developed, and so on, the individual units retain the status granted by the flat map. They remain sovereign units, ungoverned or undetermined by any social arrangements that might be located at a higher level. Looking from above, however, we see how colonial rule and imperial domination produce an unequal grading of western and non-western peoples as natural. Our modern scholarly socialisation now disfavours the view from above, from the point of view of the patterned inequality. Seeing from above has been stigmatised in conventional IR because it suggests the primacy of the patterned inequality of the whole, violating the imperative to condemn or assign responsibility to the individual or group. Where seeing from below, from the level of the units, is idealised, patterned inequality is treated as secondary - as an effect, not a cause.

Thus, even if Onuf is correct that the up and down of levels do not connote good or evil in the sense suggested by a Christian cosmos, levels schemes are not politically and ethically neutral. Rather, a levels scheme that privileges looking from below dovetails with Eurocentrism in IR. Hobson (2012: 224-5) explains that seeing from the level of individual actors, in this case Europe or the west, as 'self-constituting and exceptional', we 'thereby deny the dialogical notions of an "other-generated Europe" and the poly-civilization "logic of confluence" that a non-Eurocentric approach would focus upon'. When one neglects the structuring features of the whole, neglects colonial rule and imperial domination, and assumes western 'pioneering agency', it is easy to treat non-western inferiority (irrationality, backward culture, and so on) as an explanation for the relative successes and failures of a flattened planet of autonomous units.

Though explicitly universal and value-free in its aspirations, much of contemporary IPE proudly displays the individualism imperative that reproduces and reinforces the political content of this racist Eurocentrism. The claimed scientific achievements of dominant strains of IPE, Benjamin Cohen (2008: 143) claims, have come via a focus on individuated units - 'actor behaviour' - and how those actors, usually states, manage the interdependence generated by their interactions. Cohen's story of IPE begins with the post-war recognition of growing independence among 'national' economies, including newly independent countries (2008: 21-3). International institutions (GATT and the IMF, in his account) appear as mechanisms for managing disputes among independent states. In an advance over the work of economists, Cohen announces a new era of enquiry in which '[t]he world economy could be depoliticised no longer'. Yet, as he acknowledges (2008: 42-3), the tremendous gains in knowledge production involved imitation of the 'technical sophistication and intellectual elegance' of economists - a kind of 'reductionism' (in his terms) that 'comes at a price in descriptive and practical credibility'. The explicit contrast is with earlier Marxist theories of imperialism, dependency theory, and a contemporary 'British School' that harks 'back to the tradition of classical political economy, [where a] broad comprehension of "society" - the social context of IPE - is valued more' (2008: 18-9, 44). Cohen seems to suggest that we can simply add some of the 'social' to an inadequate science of IPE focused on unit-level behaviour.

However, as we have seen, 'levels' are not neutral, or simply abandoned or expanded at will, and they foreclose the kind of conversational diversity that Cohen (2014) now seeks. To return to Hobson, the continued embrace of Eurocentric myths by international theory in the 21^stcentury need not be taken only as the resurgence of latent racism/Eurocentrism on the part of scholars or a discipline. It is not only western ignorance or discrimination that leads to the marking of the non-west as legally recognised but relatively failed states or quasi-sovereigns. The west also turns to data. The non-west's failure to develop is 'evidence' that reveals and confirms an underlying hierarchy of capacities and potentialities. Civilizational hierarchy or graduated sovereignty merely names this underlying differential in the capacity of the autonomous units. Given an ontological individualism that rules out thinking of the whole as the context within which parts and their relations emerge, this naturalised hierarchy can only be re-enforced by the continuing 'evidence' of the relative failure of the non-west to develop or live up to civilised standards of governance.

Thus the political and ethical possibilities of modern life rest upon a tension between wholes and parts within the theory of progress. On the one hand, political and economic development is seen as isomorphic with modern western civilization. Here, modernity is more than just a model; it foretells the future of all cultural spaces. On the other, political and economic development is also linked to the principle of sovereignty; a principle that separates political and economic spaces, and putatively allows each state to find its own version of meaningful development. IR theory proper concentrates exclusively on one side of this tension by focusing on the spatial demarcation of the world into separable nations or peoples or states of civilizations who follow (or not) or are capable of following (or not) the path pioneered by modern western states and the international institutions they have created. This atomistic vision posits that realising the promises of modernity (in other words, political and economic development) depends upon the character of the developing (or not) units themselves. In this way, dominant strains of international theory suppress from our thinking the larger context of historical structures/interactions that follow from seeing down from the whole - suppress a fuller comprehension of the structuring of the whole. If the western theorist 'chooses' the individual level, he seems preordained to do so. It serves his belief that he and his state do not exist in deterministic and hierarchical order; that he and his state have earned their status. Conversation ends here, precisely where it needs to begin.

Closing remarks

As we have tried to show, the dread of contextual/structural explanation, the condemnation imperative, and the individualism imperative exhibit a fear of similarity - an anxiety that we may be just like them. We cannot allow ourselves to imagine how the other sees the world lest we admit that her vision is already within us - that ordinary human beings, like us, might use our bodies as weapons, or might find our sagging prospects in a culture of competition attributed to our genes, or our deficient culture. Such sameness threatens to undermine our own space/time/cultural distinctions, and weakens claims to ourdevelopment and our advanced culture. Similarity challenges our specificity, exclusivity and uniqueness, and, in Hobson's terms, a sense of our 'pioneering agency'. As long as we don't allow ourselves to see the other as an other within, we will be tempted to explain their actions as resulting from their (projected) status as 'evil', 'backward,' 'underdeveloped', 'yet to be civilized' - in a nutshell, 'the still lacking'. And, to extend Hobson, these hesitations, stutters, erasures - all produced by the individualism imperative - are the most entrenched instantiations of Eurocentrism, Orientalism and racism.

If we cannot allow the other to be the same, we also cannot seem to bear the other's difference. Difference is always a latent critique, and we are apprehensive about what it may purport. Learning from the other inverts the usual hierarchy wherein the other must always learn from us. It is, nevertheless, no small miracle that this relationship remains reversible; that we can still learn from the other's difference. Fear of difference, then, is resistance to the need to know beyond oneself to know oneself; particularly learning that none of us can become whole on our own, that our feeling of being unique or self-sufficient depends on repressing the similarities and overlaps that make us the same. In this way, fear of similarity and fear of difference entwine.

Where we adopt a position of social explanation, sameness and difference have more to offer than destabilising our position in a hierarchy. When the other's humanity is restored, her similarity and difference appear also as resources. To recognise and accept our intimacy with difference is also to embrace the simultaneous teacher/student status of both self and other. To recognise and accept our intimacy with sameness is also to desire entering the world of the other for the sake of learning. But we hesitate, we stutter, we resist the turn to social explanation. Our trepidation exhibits itself as a fear of systematic explanation, a fear of regarding parts as always in relation to wholes.

In IR/IPE, we often translate holism as an emphasis on the global or systemic level, as we saw in Singer and Cohen. But, in this closing section, we wish to translate the global less as a level and more as a system in which whole and parts are simultaneous moments in process. Looking from above, from the system as a whole, does not exclude the parts; indeed they constitute the whole. But the parts - in this case as individuated units - do not exist apart from 'the social conditions of action and the historical conditions of formation of the acting self' (Hage 2003: 86-7). Still, the usual preference is to see from the other direction - from the parts upward. Conventional IR/IPE scholars have not produced much of a symmetrical countermove. Though unit-level analysis may require attention to context, so that explanations incorporate 'constraints,' units are treated as if they can exist apart from the whole. Here we have parts without any account of the whole, or as Schwartz (2007, 132) would have it, 'incomplete knowledge' that is filled by a 'belief in the virgin birth' of the units.

Overcoming the individualism imperative requires us to make two opposed moves: to invert the usual bias, and then to negate that very inversion. Inversion of the bias towards unit-level means emphasising the whole, the global level. But we also repudiate the dichotomy that separates parts from wholes, and unit-level from global level. This repudiation requires us to move past the language of the 'mutual constitution' of 'wholes and parts, and of different 'levels', and towards the claim that such language reifies what might be described as simultaneous and continuous processes. Taken to its logical conclusion, this means there are no parts, no wholes, and no levels. There are only simultaneous and continuous processes whose seeming mystical flow our descriptions cannot but freeze.¹⁰

Regardless of our speculations in this closing section, two conclusions emerge from our analysis. We find laudable Cohen's desire to seek out and invite diverse others to the IPE conversational summit. However, Cohen and his colleagues have yet to confront why that diversity has not already been embraced. We have suggested some of what is required in this confrontation. Absent such an effort, the good intentions of Cohen and his colleagues risk becoming empty gestures.

References

• Abufarha, Nasser. 2009. The making of the human bomb. Durham: Duke University.
• Cohen, Benjamin J. 2008. International political economy: an intellectual history. Princeton: Princeton University.
• Cohen, Benjamin J. 2014. Advanced introduction to international political economy. Northampton, MA: Elgar.
• Connolly, William E. 1991. Identity/difference: democratic negotiations of political paradox. Ithaca: Cornell University.
• Cox, Robert. 1986. 'Social forces, states, and world orders: beyond international relations theory'. In Robert O Keohane (ed), Neorealism and its criticsNew York: Columbia University, pp. 204-54.
• Eagleton, Terry. 2010. On evilNew Haven: Yale University.
• Hage, Ghassan. 2003. '"Comes a time we are all enthusiasm": understanding Palestinian suicide bombers in times of exigophobia'. Public Culture 15(1): 65-89.
• Hobson, John M. 2012. The Eurocentric conception of world politics: Western international theory, 1760-2010. Cambridge: Cambridge University.
• Hobson, John M. 2013a. 'Part 1 - Revealing the Eurocentric foundations of IPE: A critical historiography of the discipline from the classical to the modern era'. Review of International Political Economy 20 (5): 1024-1054.
• Hobson, John M. 2013b. 'Part 2 - Reconstructing the non-Eurocentric foundations of IPE: From Eurocentric "open economy politics" to inter-civilizational political economy'. Review of International Political Economy20 (5): 1055-1081.
• Inayatullah, Naeem. 2011. Autobiographical international relations: I, IR. New York: Routledge.
• Inayatullah, Naeem and David L Blaney . 2004. International relations and the problem of difference.. New York: Routledge
• Jackson, Patrick T. 2011. The conduct of inquiry in international relations. London, Routledge.
• Lakoff, George and Mark Johnson . 1980. Metaphors we live by. Chicago: University of Chicago.
• Onuf, Nicholas G. 1995. 'Levels'. European Journal of International Relations 1(1): 35-58.
• Schwartz, Herman. 2007. 'Dependency or institutions? Economic geography, causal mechanisms, and logic in the understanding of development'. Studies in Comparative International Development 42: 115-35.
• Singer, David J. 1961. 'The levels of analysis problem in international relations'. World Politics 14(1): 77-92.

• 1
  .We recognise that unit-level strategies are the epitome of the so-called social sciences in the United States, but we mean to disrupt that characterisation.
• 2
  .Nevertheless, Hage (2002: 76) eventually enunciates the very condemnation that he initially and productively resists. There is a paper to be written on how and why he succumbs to the 'condemnation imperative'.
• 3
  .Presumably, Hage would agree that the opposite, turning the commonplace into the extraordinary, is also part of social explanation.
• 4
  .A deeper fear might be that, if placed in the same situation, we will lack the 'courage' to do as they do. We owe this point to Paulo Chamon.
• 5
  .One can also do both and neither. We turn to this more nuanced position in the conclusion.
• 6
  .Singer (1961: 92) does argue that the development of 'systematic' generalisations about IR depends on resolving the problem of parts and wholes, though he does not settle on any particular resolution.
• 7
  .How a disciplinary schema or a social context 'selects' the scholar results perhaps from the scholar's location and movement within the global political economy, what we otherwise call her/his biography. See Inayatullah (2011: 1-12).
• 8
  .In recent correspondence, Onuf has agreed that levels scheme are not value-neutral.
• 9
  .Patrick Jackson (2011) challenges the presumption of neopositivists in IR that science requires the testing of proposition based on the observable behavior of individual units. Other, context-dependent, forms of explanation have equal claim to scientific status.
• 10
  .Robert Cox (1986: 204), in a passage we have referred to, speaks of the 'seamless web' of the social world, about how 'contemplating ... undivided totality may lead to profound abstractions and mystical revelations', and about the need to construct a 'structured and dynamic view of larger wholes'.
• 11
  The authors are grateful to the following for their careful reading and thorough criticisms: Kaela Bamberger, Paulo Chamon, Austin Doullaird, Duncan Fuller, Xavier Guillaume, Inanna Hamati-Ataya, John Hobson, David Johnson, Emma Kast, Anna Leander, Nick Onuf, Alina Sajed, and two anonymous referees. Versions of this paper were presented at the Centre for Feminist Research and the Graduate Program in Political Science at York University; the 40th anniversary ofMillennium at the London School of Economics; and the Department of Political Science and International Relations at the University of Geneva. We wish to thank seminar participants for their comments

Publication Dates

• Publication in this collection
  Sep-Dec 2015

History

• Received
  28 Jan 2015
• Accepted
  08 July 2015