Conflict Networks:

Collapsing the Global into the Local[1]

 

Nick Srnicek

 

Recent decades have seen a dual and simultaneous shift in conflict trends. With the end of the Cold War and superpower support, conflicts have become increasingly intrastate and increasingly localized, dependent for their sustenance upon local assistance and national resources. Yet this localization of conflict has coincided with the increasingly international aspect of conflicts, with humanitarian intervention and UN peacekeeping becoming ever more prevalent. The aim of this paper is to provide a framework for understanding these shifting relations between the global and the local. This is accomplished through an analysis of actor-network theory and its rejoinders to reductionist understandings of conflict. Rather the reducing the eruption of violence down to greed, grievance, or ancient hatred, actor-network theory aims to examine conflict networks and their specific composition of local, material, and global actors. Three aspects of these networks are highlighted in particular: the personal networks of local individuals, the material actors, and the conflict network as a system. With these clarified the final section turns to an analysis of some of the primary modalities through which global actors relate and embed themselves within local networks.

 

Since the end of the Cold War, it has been widely acknowledged that conflict is increasingly complex and increasingly intrastate with the majority of conflicts now involving non-state actors such as local tribes, militias, criminal organizations, and insurgents. There has been, in other words, an increasing localization of conflict within particular states. Simultaneously, with the rise of a system of liberal global governance, new international norms concerning humanitarian intervention, and the rising entanglement of development projects with human security concepts, conflict has taken on an increasingly global aspect (See Duffield, 2001). The largest current wars – Iraq, Afghanistan, and the ongoing war in the Democratic Republic of Congo – all simultaneously incorporate the highest levels of the global system with a radical diversity of local situations and actors. Yet despite the voluminous literature on these conflicts, and the complications arising in their local dynamics, it has been exceedingly rare for commentators to discuss how the global conflict interacts-with, is embedded-within, and passes-through the local networks that make up the terrain of actual conflicts. Moreover, this is a deficiency common to the analysis of all macro-level conflicts – whether between the West and fundamentalist terrorism, liberalism versus illiberal states, capitalism versus the multitude, or power versus resistance. All such global conflicts exist only as embodied within their local instantiations, yet the crucial role of these mediators has gone largely unacknowledged.

In the cases where the global-local link is substantively examined, three analytically distinct conceptions of the global are often invoked either explicitly or implicitly: (1) the global as container, (2) the global as the highest position in a hierarchy, and (3) the global as a level of detail (See Keohane, 1986; Putnam, 1988; Singer, 1961). The first conception of the global visually imagines it as being the larger container within which regional and local dynamics occur. The global, in such a perspective, is what provides the basic framework for the dynamics occurring inside of it. We see this most explicitly in analyses of social structure, as a limiting construct within which other processes occur. In International Relations, it has been popularized by Kenneth Waltz’s neorealism, with the international system determining the limits of domestic action through the mechanisms of socialization and competition (Keohane, 1986). Similarly, analyses which see economic globalisation as a constraint on state action also tend to subscribe to this sort of ‘container’ approach.

The second conception of the global visualizes it as being situated at the top of a hierarchy, with the regional and local placed below it. Contrary to the first conception, the other regions are not necessarily embedded within the global. Rather, what makes this visual metaphor unique is that the global is seen as operating at a largely independent level, rather than a foundational level. Each level has its own unique dynamics, which may or may not have any effect on the others. The classic reference for this position is Robert Putnam’s (1988) work on ‘two-level games’, where the domestic and international levels each constitute their own separate dynamics with interaction between them occurring at regulated points.

The third common conception of the global visualizes it as a level of detail. Like a microscope, one can zoom out to the global macro features of the phenomenon under investigation, or one can zoom in to the local details involved. Depending on whether one is interested in generalized features or a singular case study, one chooses to examine a phenomenon at either a global or local level. This clearly occurs in the compromises between case studies and large-N studies, with the latter statistically analyzing a vast array of cases while explicitly neglecting context and detail. The argument is that “if a given relationship holds across a variety of contexts, then context cannot be so important” (Hopkin, 2002, p. 255). Yet this image of the global is also explicit in David Singer’s (1961, p. 80) classic paper on the levels of analysis when he notes the “dearth of detail” that a focus on the international system requires.

The problem with all three of these conceptions of the global is that they presuppose multiple levels of reality and are intelligible only in such a framework. Yet, each level of reality produces an analytically insurmountable gap between them, or it requires willfully ignoring the connections that lead out to other levels. Moreover, whenever we go out into the field looking for these multiple levels of reality, all we see is the single, same world. One goes to look for neoliberalism, and finds economists and macroeconomic models working at the World Bank. One goes to look for financial globalisation, and finds traders and computer systems in New York and London. One goes to look for global governance, and finds diplomats arguing at Security Council meetings. Everywhere we look, we run into more and more local networks, and never some independent realm labeled ‘the global’.

What is required to overcome these difficulties and incorporate these insights is a re-thinking of the ‘global’. Conflict provides the ideal field for elaborating on the global as ‘new wars’ have increasingly involved both the local and the global (Duffield, 2001). Conflict is the immediate unity of the local and the global. Situations of collective violence also supply us with the most radical political phenomenon – one which is often taken to be chaotic and unintelligible, and one which often effects entire societies. This essay will therefore set out to understand conflict immanently – from its instantiation in local networks of human and nonhuman actors. It will proceed by briefly outlining actor-network theory and examining the basic claims of such a perspective. It will then turn to recent scholarly work which has made clearer the locality and complexity of conflict. Finally, it will analyze this work for an understanding of the global/local connections operative within conflict.

 

1 - Actor-Network Theory

 

Actor-network theory (ANT) is a unique approach to analyzing our world. While typically labeled as ‘sociology’, as Graham Harman (2009) has recently made clear, ANT also offers a thoroughly philosophical vision. In this section I will argue for three of the most important aspects of this philosophy: the denial of any a priori distinction between nature and culture; the specific definition of actors; and, the rejection of social abstractions as explanations of phenomena.[2] I then examine the notion of the ‘global’ that emerges from these ideas and follow through on its implications for social science.

 

1.1 – Actor-Network Theory

 

For ANT, sociology is a matter of studying associations, or more specifically, of associations as they instigate and translate a force (Latour, 2005, pp. 1-25). The social is not an independent realm of some ontological substance, but rather a way of relating between heterogeneous entities. This simple definition obscures a number of important implications, foremost of which is that the separation between the human and the nonhuman must necessarily be dissolved. Taken by ANT as the study of associations, sociology is as equally concerned with human-human interactions as it is with human-nonhuman and nonhuman-nonhuman interactions.[3] There simply is no fundamental difference. The apparently peculiar nature of this claim is belied by what Bruno Latour (1993) calls the ‘hybridization’ of our world – namely, the increasing degree to which humans and nonhumans intermingle in the constitution of our contemporary world. The supposedly subjective sphere of meanings, interpretations, and private thoughts, openly interacts with the supposedly mechanical sphere of technology, nature, and science. Our world is clearly constituted by the open-ended proliferation of these actor-hybrids, yet social science continually tries to separate the two into two irreducibly separate realms of Nature and Culture. Instead of seeing these two realms as the a priori ontological framework within which all phenomena occur, Latour and Michel Callon (1992, p. 349) argue that “‘natures’ and ‘societies’ are secreted as by-products of this circulation of quasi-objects [i.e. hybrids or actors].”[4]

     By contrast, the vast majority of social science presupposes precisely these two independent realms. Objects and materiality may be referenced in explanations, but are all too often reduced to being merely passive resistance against human intentions, or transparent conduits for human intentions. Objects and materiality, in other words, are rarely given their own active agency; even when they are included in a study, they rarely make a difference on their own. By contrast, the dissolution of the Nature/Culture division entails that objects also have their own agency (see Latour and Fuller, 2003; Bennett, 2005; Harman, 2009; Johnson, 1988). In this regard, both humans and nonhumans are equally actors which force other actors to act. Actors, at their most basic level, are anything which makes a difference within a situation. An actor need not be a rational, reflective, self-conscious human agent in order to be an actor. As Latour says, under his pseudonym Jim Johnson (1988, p. 299), “every time you want to know what a nonhuman does, simply imagine what other humans or other nonhumans would have to do were this character not present. This imaginary substitution exactly sizes up the role, or function, of this little figure.” But once the breakdown of the Nature/Culture divide has occurred, the more radical step is that the human/nonhuman binary breaks down into the sheer multiplicity of actors in the world[5] – rather than a single modality of interaction between human and nonhuman actors, there is a proliferation of interaction modalities (what Latour will call ‘translations’). If everything is an actor, then each interaction will be unique and call for its own investigation. In other words, the global divide between Nature and Culture breaks down into the networked divisions between a multiplicity of actors.[6]^,[7]

The arguably underdetermined definition of actors employed here has the distinct benefit that ‘who the actors are’ becomes an eminently empirical problem, rather than an a priori imposition. Moreover, the question of ‘who the actors are’ leads directly onto questions of legitimation, knowledge, expertise, and evidence since any answer is immediately wrapped up in epistemological processes. Actors appear in a multiplicity of ways, and the process of narrowing it down to a ‘fact’ is itself subject to description and study.[8] Thus for an actor who has taken on a stable existence as an object of knowledge, it can be examined how it came to be determined as a ‘matter of fact’ that this actor indeed exists as such.[9] Or the actor can continue existing as multiple, yet overlapping objects of knowledge (as, for instance, when multiple medical tests produce distinct but usually overlapping knowledges of a disease-actor; (Law, 2004, pp. 45-67) or in conflict situations, when various encounters, rumours, news reports and espionage produce overlapping but also conflicting knowledges of an enemy). This process of stabilizing an actor in knowledge is itself never finalized (in the same way that science is a constitutively incomplete project), and so the question of ‘who an actor is’ can always potentially be recommenced. The process of stabilizing an actor can include different forms of scientific knowledge, expert analysis, politically-inflected groups, as well as local knowledges, but also the tools and techniques used, the consensus practices that have become established through piecemeal construction, and the actor itself (as the invariant X[10] around which discourses circle). Epistemologies and knowledges, no less than physical objects, are actors in the world – ones which shape and produce differences themselves. Thus, for example, theories of conflict determine how peacebuilding is carried out, where resources are applied, and what events are significant and insignificant for understanding a particular conflict. As these theories circulate, they act in the world by compelling more and more actors to be organized along their lines (or resisted). With this in mind, the aim of analysis is to examine the relations between particular ideas, concepts, objects, and other types of actors through an empirical investigation, and consider how they function together in a single networked system.

This leads us to the issue of interactions between actors. The standard natural science perspective is to see a closed realm of causal interactions between physical entities, yet this viewpoint presumes the rigid and absolute Nature/Culture divide that was rejected earlier. On the other hand, the standard social science perspective is to reduce as many effects as possible to as few causes as possible (the explanatory parsimony principle). While this principle may produce elegant theoretical systems, as an ontological theory it fails, and as a pragmatic theory for producing effects, it also underestimates the complexity involved in any given phenomenon. This is a crucial flaw in attempts to make social science relevant for policy initiatives or activist movements, i.e. those forced to face up to the complexity of the world. Parsimony may be graceful for the theorist, and simplifying for a decision-maker, but if the Iraq and Afghanistan fiascoes have reminded us of anything, it is that simple theories are useless (or worse, harmful) in the real world. Moreover, the drive to parsimony and simplicity reduces knowledge to a series of abstractions that exist nowhere and that are in need of explanation themselves. As Jane Bennett (2005, p. 455) argues:

 

The active power of assemblages [i.e. actor-networks] is concealed under the rubric of (social) structures, (cultural) contexts, (religious) settings, (economic) climates, or (environmental) conditions – terms which denote passive backgrounds or, at most, states of affairs whose sole power is the negative one of constraint or resistance. Structures, surroundings, contexts, and environments name background settings rather than spirited actants.

 

To remedy this reduction of the active power of an actor-network, we raise Bruno Latour’s distinction between ‘intermediaries’ and ‘mediators’. Whereas the former refers to actors who cleanly propagate the causes that instigate them (e.g. explanations in the form of “an individual is a mere puppet of social forces”, or “the individual is playing a functional, structural role”), the latter refers to actors who transform the forces that pass through them. Rather than a social force acting smoothly on an individual (regardless of how many actors it must pass through), the notion of mediators highlights the role that each actor plays in contributing to the propagation of any action.

This entails a number of significant consequences. First, the entire chain of a network becomes potentially significant to understanding the effects. In actor-network theory’s terms, we must ‘trace’ the connections – a necessarily empirical and patient project. Second, ontologically speaking, reduction becomes not an a priori assumption (e.g. “the phenomenon is clearly caused by power relations, or by knowledge epistemes, or by balances of power, etc.”), but rather something which must itself be slowly and painstakingly constructed. The work of reduction in science is something that takes numerous scientists, and numerous experiments, to produce. Third, the division between the global and the local – the mystery that we started this paper with, becomes resolvable. The gap between the two becomes reconfigured in terms of a chain of mediators; the way in which they affect each other is through this network of actors which links them in a highly specific configuration. ANT’s renewed definition thus gives scientific meaning to the emergence of such distinctions, and provides concrete answers as to how the global interacts with the local.

 

1.2 – Flattening the Global

 

The problem we started with at the beginning of the paper was the problems concerning three analytically distinct conceptions of the global: (1) the global as container, (2) the global as the top position in a hierarchy, and (3) the global as a level of detail. As we noted, they are incapable of accurately explaining the relations between levels and ignore that when we go to look for the global’s operations, we only find more local areas.

What we have then is a single plane of existence, rather than differing levels of reality. What distinguishes the global from the local, as we will see, is the differing size of the actors – though the precise meaning of the term ‘size’ remains to be explained. What we typically call the ‘global’ is therefore not a matter of incommensurable levels of reality, but is rather comprised entirely of the largest actors in the world. Yet the existence of these macro-actors causes us to run into another problem. If, as our analysis has argued, the world consists of actors acting according to their own immanent dynamics and logic, it would appear impossible that something like a macro-actor would ever arise. The chaos of multiple, conflictual, divergent actors would be too much for something like an institution, a rebel group, a state, or a state system to ever emerge. They presuppose too many actors, acting in cooperation (though not necessarily harmony),[11] to appear achievable in a world of divergent actors. Yet macro-actors clearly do exist, and so the question becomes, ‘how?’ Callon and Latour (1981, p. 284) argue in an early article that macro-actors are constructed through a process of associating durable materials:

 

By associating materials of different durability, a set of practices is placed in a hierarchy in such a way that some become stable and need no longer be considered. Only thus can one ‘grow’. In order to build the Leviathan it is necessary to enroll a little more than relationships, alliances and friendships. An actor grows with the number of relations he or she can put, as we say, in black boxes. A black box contains that which no longer needs to be reconsidered, those things whose contents have become a matter of indifference.

 

Thus, for instance, a monarchy doesn’t rely on transient social relations, but rather develops on the basis of a palace, an array of status symbols, a mercenary force, inherited wealth, various legal declarations, claims to divine authority, papal support, property, etc. These must be slowly and patiently constructed (not always intentionally) and arranged so as to be taken as ‘black boxes’ – relatively stable conduits of force that can be relied upon under normal circumstances. The operation of global power,[12] in other words, only travels through highly specific conduits, and these conduits are the object of study for actor-network theory. Power presupposes that an entire network has been constructed through which it can be exerted.[13] Power and control over a network of actors involves (1) constructing multiple chains of actors, (2) maintaining and expanding these chains, and (3) the effort required to propagate a command through them.

But the notion of a black box highlights a significant distinction to be made between types of global actors, and implicitly points towards an extension of Callon and Latour’s notion of a macro-actor. On the one hand, there are the established (institutionalized, organized, materialized) actor-networks for creating a global action – the realm of black boxes that Callon and Latour examine. On the other hand, there are the global actions which operate without the need for a series of black boxes. In this regard, al-Qaeda perhaps exemplifies a macro-actor that need not rely on black boxes. Instead, al-Qaeda uses the tight interconnection of modern networks against those very networks, in order to act upon key nodes, which then create disproportionate effects. Al-Qaeda requires only a minimal construction of conduits through which it can exert itself reliably;[14] it only needs a wide range of actors to be affected. In the end, this is the minimal condition of globality: the capacity to affect large numbers of actors that are widely dispersed. The size of an actor is determined as much by the conduit of networks it can ally to itself, as it is by the range of effects it can carry out. In that regard, we can make a distinction between macro-actors that are founded upon a network of intermediaries (black boxes) and macro-actors that are founded upon a network of mediators (relatively independent actors). Contra Callon and Latour, what makes an actor ‘macro’ or global is not its construction of conduits for power, and the use of durable materials, but rather the range of the effects stemming from an action.[15] A single pedestrian standing in front of a tank in Tiananmen Square is therefore as global an actor as the CEO of Goldman Sachs. The Board of Governors of the International Monetary Fund is as global as the individuals responsible for the destruction of Iraq’s Al-Askari mosque.[16]

The ‘global’ as a realm is therefore not independent of the local, nor is it foundational, nor is it more general. An analysis of the global must focus on the interactions between macro-actors, specifically by tracing their actions through the local networks they have organized and affected. The global is an extension of the local, but precisely for this reason, an examination of global actors and events must focus on the local.

What all this analysis entails is that any given social field is constructed by actors of varying sizes, materials, reflexive theories, relations, and degrees of systemic importance. To look only at the individual level would be to miss the larger actors; to look only at the cultural level would be to miss the material level. The analysis of a situation must examine the actor-networks involved in their concrete occurrence.

 

1.3 – Tracing the Connections

 

The uniqueness of this program, and also its major difficulty, is to keep the full scale of local complexity in view while simultaneously taking on a global perspective.[17] There can be no scaling of perspective here, and no reduction to generalities. As a means to overcome this difficulty, we turn to what might be called ‘The Principle of Traceability’. This principle stems from a number of the ideas that have already been presented:

 

1)     the commitment to immanence and the rejection of any transcendence, either in the form of a social whole or a world of ideal laws;

2)     the pragmatic, epistemological perspective which sees the ontic realm consisting of networks of actors;

3)     the argument that these actors exist only through their actions which spread throughout the network – what Gabriel Tarde called an ‘imitative ray’, i.e. the contagion of actions, beliefs, desires, practices, objects, forces, etc. that pass through a network.

 

From these three conceptions of the world, it follows that to analyze a given phenomena, one must ‘trace’ the connections by following them as they lead the social scientist along their own path.[18] Instead of a leap of faith between different levels of reality (local, regional, global), one has to map out the actual conduits through which these areas are connected. One must trace the connections established between global actors and local actors to such a degree that one could, in principle, explain the subjective manifestations of global dynamics.[19] This means, for example, following an innovation as it progresses from a laboratory to a published paper, to a set of colleagues, to a venture capitalist, to a marketing team, to a distributor, to a collection of stores, to the public, and to their friends through word of mouth and advertising. Similarly, with conflict, one may want to (as best as possible) trace the line of a weapons shipment to its embroilment within a particular battle. The opening scene of the movie Lord of War is emblematic here: portraying the creation of a single bullet in a factory, and following it as it is checked for quality assurance, packaged into a wooden container, shipped abroad to a rebel group, loaded into a weapon, and finally used to kill a man in a conflict.

Beyond the tracing of lines, this ontology also leads one to proliferate the actors involved. Rather than reducing the world to a lifeless husk through which a few major forces play out their battles (e.g. the forces of modernization, class conflict, ethnic war, clash of civilizations, etc.), it must be acknowledge that social forces act through actors that have their own relative autonomy. Thus, accounts of conflict which attempt to explain it on the basis of greed, grievances, a ‘new barbarism’, underdevelopment, ethnicity or nationalism are weak accounts, accordingly.[20] They do not trace the pathways through which these causal factors are actually carried out and their efficacy created. It makes no difference to these theories, whether the resources in question are oil or whether they are cocaine or agriculture. It makes no difference to them whether the weapons used are machetes, assault rifles, explosives, or armored vehicles. An ANT analysis rejects this structuralism, and looks at how each actor contributes and constructs phenomena. For instance, grievances may form a part of a conflict system, but they must pass through – and be altered by – familial relations, education by religious leaders, the pervasiveness of nationalism within local textbooks, the networks of informal relations between disparate hierarchies, etc. An account which attempts to reduce a conflict to grievance or greed ultimately explains nothing.

      One final note on agency: if the empirical world is composed of human and nonhuman actors interacting with each other and inducing actions in each other, any particular human individual will be the manifestation of a (variably-sized)[21] local network.[22] In this way, actor-network theory gives an empirical and scientific basis to the oft-cited claim that actors are socially and culturally embedded. Rather than citing an empty notion of ‘context’ (e.g. “it depends on the context”), actor-network theory forces the researcher to discern, describe and reveal the power of the surrounding network. Note, though, that by saying that everything emerges from a concatenation of local networks, we are not excluding the global – our reformulation of the global means that it must be channeled through a series of localized networks, which means that macro-level actors can and do act to produce phenomena. But they act only through a particular series of conduits, and not through some abstract ‘social structure’ or ‘social force’.

 

2 – Conflict Assemblages

 

With the philosophical backdrop laid out, the question to be answered now is how does violent conflict appear if the global is flattened and nonhuman actors are recognized as making a difference? We are helped out in this endeavour by an emerging academic recognition of the complexity of conflict. Most notably, recent work has shifted the analysis of conflict from being an absence of order to being the construction of a different type of order (see Keen, 2008; Berdal, 2009; Duffield, 2001). Simultaneously, various ethnographical studies of conflict have shown the numerous ways in which local dynamics are crucial for understanding how a conflict emerges and evolves. This section will look to build off this work, and inflect it through an actor-network theory perspective, in order to come to the complex concept of ‘conflict assemblages’.

     This will proceed by showing, first, the local embeddedness of conflict, and how it both shapes and is shaped by the preexisting local assemblages. The second step will focus on the neglected study of nonhuman, material actors in conflict, and show the contribution to be made by reintroducing these elements as actors in their own right. The final section will build upon these actor-networks in order to show how conflict can be understood positively, as a peculiar order and not as an intractable chaos. This step is necessary for beginning to grasp the networks and microdynamics of conflict, and for relating them to global events. It refutes the standard idea of conflict as an intrinsically chaotic, and reveals how even the messiness of war can provide conceptual traction for thought.

All of this combined will form what we call ‘the theoretical ontology of conflict’ – which is to say, a systematic approach to conflict which presents theoretical entities (systems, nonhuman actors, local networks) as scientific hypotheses for explaining real dynamics of conflict. This theoretical ontology must be distinguished from a philosophical ontology, which aims to study ‘what it means to exist’ for any possible entity. A theoretical ontology does not aim at this level of generality, nor does it make claims about what it means to be – it merely posits certain conceptual entities as pragmatically useful and scientifically explanatory.[23]

Notable in this regard is what we are not speaking of: states, ideologies, cultures, ethnicities, classes, and formal institutions. Instead, we will attempt to explain conflict in alternative terms, in an attempt to overcome what Mark Duffield (2001, p. 141) has called “the lack of an adequate language for describing the social and organizational effects of the new wars.”

 

2.1 – Local Networks

 

Recent years have seen a growing turn towards focusing on the micro-level factors of conflict.[24] This has been assisted by increasing numbers of data sets available for researchers, as well as the embroilment of America in two lengthy wars, making the study of conflict a popular topic. Mapping the micro-dynamics has begun to reveal the diverse ways in which people respond to conflict – joining, abetting, migrating, resisting, etc. – that are missed by macro-level analyses focused on ethnicity, interest groups, socio-economic groups or any other type of ‘molar’ identity. For actor-network theory, the imperative to take into account which actors make a difference in a situation demands that we take into account the local dynamics of a network. The grand abstractions of social theory – social forces, class, nationalism, ethnicity, etc. – do not exist. Moreover, conflict does not operate in a vacuum, but rather filters through existing actor-networks. Thus local networks can be seen to provide a number of different functions, two of which we will consider here: (1) preexisting disputes and social relations and their effects throughout the conflict period; and (2) the personal networks of recruitment.

The first significant aspect of local networks is their existing network of personal conflicts and mutual support: these include disputes between neighbours (Kalyvas, 2006, pp. 346-351), resentment formed through gossip (Scott, 1985, pp. 22-23), hierarchies based on material distributions (Scott, 1985, pp. 18-19), and local systems of governance and adaptation (Berdal, 2009; Duffield, 2001, p. 146). Each of them transforms ‘global’ narratives of the conflict through a local prism. Thus, one of the most significant findings of recent micro-level literature is that “while people may kill under the pretext of ethnic ideologies, real motives and interests are often rooted in local relations and power structures.” (Fujii, 2008, pp. 570-571.) Lee Ann Fujii (2008, pp. 571-576) reports how even in the extreme situation of the Rwandan genocide, when possible, some Hutus would refrain from killing or even help a Tutsi based upon previous friendships. Through his ethnographic work, James Scott shows how practices like gossip and shaming are much more crucial for outlining local hierarchies than are post-hoc nationalist or emancipatory narratives. Beneath the molar identities of landowner, peasant, etc., lies local networks shaped by everyday interactions. Thus, for instance, some individuals are produced as ‘shameful’ and are resented regardless of their class position, while others can be heralded for their decent acts independently of their class position (Scott, 1985, pp. 1-27). Stephen Lubkemann (2005, p. 495) demonstrates that “understandings of wartime violence, assessments of risk, and consequent migratory reactions often had very little to do with the political programs or pretensions of either of the national parties to the war. Rather, residents of Machaze [in Mozambique] calculated risk and reacted to it primarily in terms of the logic of local social conflicts.” Further confirming the importance of inflecting conflict through its local networks, Mark Duffield (2001, p. 125) argues that “conflict and displacement […] often act to reconfirm or even strengthen social and cultural ties” as a means to survive. By and large, local networks involved in conflicts are determined more by matters of coping and survival than by any other logic (see Mueller, 2000, p. 42; Lubkemann, 2005; Kalyvas, 2006, pp. 116-117; Kilcullen, 2009, p. 67). Individual human actors therefore make a difference through the variable ways in which they distribute acts of violence and assistance. Put differently, entire groups do not act in any coherently homogeneous way.

On the other hand, Stathis Kalyvas’ work reveals a more insidious logic involved in local networks.[25] He shows the massive importance of disputes between neighbours in determining the use of violence in civil wars. In particular, the crucial ingredient for controlling a territory, by any party, is the requirement of local knowledge about potential opponents in their midst. Since the nature of insurgent wars means that such opponents can hide within the civilian population, the occupying force relies upon networks of informants. Perhaps surprisingly, as Kalyvas amply shows through a lengthy set of historical examples, there is never any shortage of denunciations. While some of these denunciations are correct, the vast majority are the results of local disputes, with neighbours blaming each other as a means to resolve personal conflicts. As Kalyvas (2006, p. 332) argues, “violence is often a reflection rather than a transgression of neighbourliness.” In addition, therefore, to the logic of survival and coping, there is also the logic of denunciation – all of which are based upon local social networks and which transform macro-conflicts into their own terms. Kalyvas shows that such variable and singular social networks explain a large part of the spatial distribution and even type of violence used in conflicts.

     Local networks also contribute by providing the basic elements from and through which an insurgency is constructed. One of the most striking aspects of insurgencies is their small-scale origins. A moment’s reflection confirms that this must be the case: insurgent groups – as with any other actor – do not arise fully formed, but must slowly grow and create alliances. These connections are formed amongst both other groups and with the nonhuman world, by creating narratives and knowledges linking, for example, material inequality to an embodied opponent. The failing water supplies, the collapsing houses, the lack of food, are all mobilized in an alliance with various insurgencies. Thus, it is no surprise that a look at any conflict group will find a small-scale origin.[26] The Shining Path in Peru began from a teacher and his students, and used the educational networks to propagate their ideas (Weinstein, 2007, p. 81). The National Resistance Army in Uganda began from twenty-seven “close colleagues” (Weinstein, 2007, p. 69). The RENAMO insurgency in Mozambique originated with a small group of people broadcasting criticisms of the government (Weinstein, 2007, p. 72). The latter also points to the use of nonhuman actors required to expand the group, a category which can also include pamphlets, textbooks, TV shows, oaths, and the internet. The spread of an insurgent group is necessarily channeled through the local conduits available to it; whether through clandestine personal networks, anonymous broadcasts, or public denunciations. The insurgent organization is both limited and constructed by the local assemblage.

 

2.2 – Material Actors

 

While local networks of social relations are clearly significant for explaining conflict dynamics, the sole emphasis on social relations, human actors and incentives (greed or grievance) risks placing conflict in an immaterial realm unbound from the material actors that also populate the world. Items like explosives, barter goods, mountains, Kalashnikovs, plains, paved roads, ore, dirt pathways, security walls, forests, cows, oil, and gold, all play crucial roles in determining the onset, dynamics, and outcome of a conflict. Each of these elements needs to be seen as an actor itself which contributes to the conflict dynamics, rather than a mere accessory to the power struggles amongst human actors. The material actors of local networks produce their own series of possibilities, constraints, determinants and openings, independently of any human action – yet their analysis has largely been relegated to an aside within the standard studies of conflict.[27] In order to begin to understand the power of material actors, we have to largely turn towards other academic disciplines. To compensate for their neglect the variable effects of nonhuman actors will be given extra attention here, though this is not to argue for a technological or natural determinism.

In understanding the significance of nonhuman actors, a crucial question to ask is, for example, what difference does it make to conflict dynamics if the resource of the local community is oil, diamonds, lumber, barley or coca? Each of these resources entails a different mode of extraction, a different mode of processing, a different mode of transporting, etc. The general category ‘resources’ operates at too abstract a level and fails to explain the dynamics of a particular conflict. For instance, the anthropologist Anna Tsing (2005, p. 34) has shown how the nature of lumber extraction has given rise to nomadic groups of quasi-legal and militant loggers within Indonesia. The nature of the resource allows these groups to mark their territory in the bark of the trees, while its transportation out requires the construction of new roads cutting through traditional pathways and creating new conduits for migrants, thieves, traffic and fugitives, and altering established patterns of existence (Tsing, 2005, p. 38). In other cases, the difficulty and specialized knowledge required to extract a resource can lead to a situation where major companies rely on small-scale collection by locals, leading to a sustainable system rather than destructive exploitation (Tsing, 2005, pp. 184-5). Economist Tim Mitchell (2009), meanwhile, has analyzed how the shift from a coal to an oil-based economy, along with the nature of oil transportation has made possible new forms of resistance, such as the ability of the Movement for the Emancipation of the Niger Delta (MEND) to cripple economies through low-cost attacks on oil pipelines. When resources are themselves taken as actors, we can begin to see that the entire socio-natural assemblage built up around them is an actor as well. This perspective-shift not only makes our explanations more complex and more accurate, but is also arguably more scientific and more useful to decision-makers and activists.

      Similar analyses hold for the other major material actors in a conflict. For instance, the particular weapons bought and employed entail a wide range of different effects on the conflict dynamics. Again, they are actors themselves within the conflict and not mere intermediaries cleanly transmitting human intentions. The use of different weapons entails a whole array of different tactics, knowledges, organizations, and cultural norms. The explosive power of improvised explosive devices (IEDs), for instance, is used in highly specific situations of asymmetrical warfare, the existence of routine paths by the victims, and the use of armored vehicles. On the other hand, the highly personal violence from machetes determines a particular mode of action and series of tactics, while assault rifles make possible an entirely different mode of action, and artillery yet another. Different weapons also potentially play a role in shaping local gender and social hierarchies, with some weapons being used to disfigure females and instill humiliation and fear, while other weapons become a proxy for masculinity.[28] In addition to their immediate uses, attaining and maintaining any particular weapon places precise constraints and compels certain organizational forms to arise: the maintenance and use of weapons requires highly specific skills which will shape how battles are fought and violence is used. An entire socio-technical assemblage of organizations, tools, economics, practices, supply chains, and information flows, must be constructed around the use and maintenance of each specific weapon. Insurgents in Iraq, for instance, have relied upon mobile freelance organizations of bomb-makers in order to create and innovate with new improvised explosive devices (Robb, 2007, pp. 135-137). The Irish Republican Army had to create alliances with the knowledge of chemistry, physics, and engineering, in order to continually create new and more deadly weapons, innovating in the use of ammonium nitrate, diesel and Semtex, for instance (Oppenheimer, 2008). This knowledge was then passed along to new generations through the establishment of education programs for new bomb-makers. The significance of these assemblages is that without them, it is easy for the government to outflank insurgents that have to recreate their knowledge base anew for each attack. With these socio-technical assemblages operational, however, the conflict takes on the dynamics of an arms race to the point where Iraqi insurgents are now more effective with IEDs than when the war first started (Robb 2007, p. 135).

     Beyond the knowledge and ingredients to use, maintain, and create weaponry, there is also the major issue of securing a supply of either ready-made weapons or the tools and components to create one’s own weapons. Gun-brokering has surged since the end of the Cold War with massive amounts of illicit and untracked arms flooding into the market with the collapse of the USSR (see Stohl and Grillot, 2009, p. 108; Keen, 2008, p. 39). As Mats Berdal (2009, p. 64) notes, “criminal actors and activities have played a critical role in sustaining the war effort of belligerents in contemporary intra-state conflict.” This increase in deterritorialized weapons has led to an increased influx of weapons into conflict situations, which has altered social structures, hierarchies and traditional patterns of warfare (Stohl and Grillot, 2009, p. 122). But the flows of weapons themselves rely upon clandestine personal networks, and established smuggling routes often created for the purposes of drug-running and other illicit trade (Glenny, 2008, p. 19). These preexisting channels for illicit objects (and sometimes subjects) further ease the flow of weapons, though the type of weapons being traded alter the ways in which they are shipped, the organizational structures that are involved, and the networks of corruption that line the pathway of the trade.[29] The simple point that heavy weaponry and vehicles are much harder to smuggle than small caches of weapons, means that the former require much more elaborate routes to attain (e.g. ‘missing’ cargo ships, re-routed shipments, corrupt government officials, and even the use of existing humanitarian aid transportation routes) (Griffiths and Bromley, 2009).

     Finally, we might note the massive importance of the human-made physical structures: dams, bridges, wells, and roads, but also buildings, tunnels, walls and checkpoints. The architectural theorist Eyal Weizman has made a massive contribution to our understanding of these aspects, examining the highly specific material ways in which the conflict between Israelis and Palestinians has played out. For instance, in the tight streets of Gaza’s urban areas, Israel recognized the necessity for bulldozing houses in the Palestinian territories in order to widen the streets for tank movement (Weizman, 2007, p. 70). In other words, the specific urban architecture has acted by limiting Israeli movement, compelling them to alter it, and provoking angry Palestinian reactions as a result. Similarly, Israel has continued the conflict over Jerusalem through political means, by enacting local laws that require new Israeli buildings to be built in the same architectural style as the historical sections, using stone clodding as the exterior. The result has been a blurring between settlement expansion and the traditional centre, making any future resolution increasingly difficult (Weizman, 2007, p. 28-33). The establishment of close-proximity charges and new military theory has opened up entirely new military tactics, as urban commandos can now move through walls, rather than through alleys and roads which have been blockaded (Weizman, 2007, pp. 208-210). In Afghanistan, the construction of paved roads has produced jobs for the project, lent legitimacy to local governments, improved trade capacities, and altered the placement of IEDs as they become easier to spot on paved rather than dirt roads (Kilcullen, 2009, pp. 87-105). In other words, the material actors play crucial roles in determining how a conflict is played out, yet too often our analyses focus solely on the human actors.

 

2.3 – Conflict Systems

 

With the various actors of a conflict assemblage briefly outlined, we can turn now to a more abstract level and determine the nature of what holds together a conflict as a conflict. The question here is of understanding conflict as a system; that is to say, as an interconnected and open collection of actors which attains a level of consistency that lets them function together.[30] In this regard, recent scholarship’s most significant advance has been to increasingly recognize that conflict does not entail the absence of order (‘chaos’), but rather the emergence of a new form of order – a systematic configuration that serves particular functional goals, and that develops novel systems of governance, whilst also making war a beneficial endeavour for some human actors. David Keen (2008, pp. 14-5), one of the academics most responsible for the shift, describes it as such:

 

Rather than listing the causes of war or famine and rather than portraying war as fundamentally irrational or as an aberration or interruption, it would be more helpful to investigate how violence is generated by particular patterns of development, by particular political economies which violence in turn modifies (but does not destroy). Indeed, part of the problem in much existing analysis is that conflict is regarded as, simply, a breakdown in a particular system, rather than as the emergence of another, alternative system of power and even protection. Yet events, however horrible and catastrophic, are actually produced, they are made to happen by a diverse and complicated set of actors who may well be achieving their objectives in the midst of what looks like failure and breakdown.

 

The key to this analysis is to recognize the functional aspects of conflict, and the ways in which it creates its own order out of these functions and self-perpetuates itself. Conflict, in other words, is not determined simply by the goal of destroying an enemy; it can have beneficial effects for a select few macro-actors. Keen (2008, p. 17) lists three primary goals of conflict which compete with and often undermine the goal of simply defeating an enemy: (1) limiting violence, particularly against oneself; (2) gaining immediately, through economic or psychological benefits; and (3) weakening opposing powers. However, we must be careful to avoid two things: first, to fall into a macro-functionalism, which would reintroduce the hierarchical global we removed in the last section; secondly, to see a conflict system as determined solely by human interests.

A macro-functionalist analysis would see conflict as oriented around ‘class interests’, or ‘criminal interests’, or ‘political interests’, and remain at a transcendent, structural level, rather than discerning the real, immanent dynamics. We must, therefore, distinguish between a macro-functionalism at the level of aggregates and social abstractions, and a micro-functionalism at the level of actor-networks.[31] The latter would look to a specific situation in order to understand the particular lines of alliances that are being mobilized in order to serve the interests of particular individuals. There can be emergent patterns in how these functional alliances are organized, but any given instance is its own unique and singular construction. For instance, in the chaos of post-Communist Bulgaria and through the contingencies of history, the head of the reformist group was introduced to a former wrester-turned-strongman by the head of the national security services. These three then used their combined control over established power networks (politics-muscle-security) in order to create alliances with illicit trade routes, stockpiles of weapons and surveillance tools, pro-democratic forces, secret police, and factories (Glenny, 2008, pp. 4-9). The newly established capitalist system was taken control of, and major lines of profit were then directed towards this alliance. As Deleuze and Guattari (1983, pp. 181) note, “the large molar machines presuppose pre-established connections that are not explained by their functioning, since the latter results from them.” It is these pre-established connections that macro-functional analyses are incapable of examining and that are crucial for understanding the specificities of an actor-network.

Such a focus on micro-functionalism also entails that we reject the emergent transcendence of a system ‘above’ the assemblage, which would determine and shape the dynamics of the network without in turn being determined and shaped by it. In this way, actor-network theory rejects any idea of the system as a totalizing instance which makes its elements mere effects of itself. Rather, we must follow Deleuze and Guattari’s (1983, p. 43) claim that:

 

… the Whole itself is a product, produced as nothing more than a part alongside other parts, which it neither unifies nor totalizes, though it has an effect on these other parts simply because it establishes aberrant paths of communication between non-communicating vessels, transverse unities between elements that retain all their differences within their own particular boundaries.

 

Mereologically, this means that the system is not a container within which the local is contained. The system establishes new connections by virtue of its creation, but it operates alongside its parts and not as a container. It brings together other local networks, linking them and drawing heterogeneous connections between previously disconnected and disparate networks. But these connections require work – they must be constructed and produced – and even then, the system remains merely another part alongside the other parts, albeit a part with a disproportionate range of influence.

    We must also dispense with the idea that conflict is determined by human interests alone. As the previous section on material actors showed, a major part of how conflicts emerge, evolve, and end is shaped by the surrounding objects, i.e. by the entire conflict assemblage and not merely its human aspects. In what way do nonhuman actors contribute to a micro-functional system? The standard idea of ‘functional’ is ‘for a purpose’, and in this sense, objects are precisely functional in their ability to determine their potential purposes. That being said, technical objects also have an internal consistency that is irreducible and conditions their use: as Gilbert Simondon has shown, any particular technical object has its own ‘mode of operation’ which entails establishing a consistency between all of its component parts. In this regard, they have an internal history within which they develop according to their own technological logic.[32] When placed in relation to a larger assemblage, they then begin to limit, constrain, shape, make im/possible, create, and produce new functions. As Antoine Bousquet (2009b, p. 2) argues, “Every technical object has its own specificity and presents its own resistance such that it simultaneously contributes to shaping the social field and reorganizing the various actors that are connected to it.” They provide more than just the background aspects of a conflict. One major contribution is that they alter the entire series of differential relations of movements between the components of the assemblages. Deleuze and Guattari (1987, p. 397) provide the contrast between the bullet and the tank as an example of the ways in which material objects achieve this:

 

It can happen that speed is abstracted as the property of a projectile, a bullet or artillery shell, which condemns the weapon itself, and the soldier, to immobility (for example, immobility in the First World War). But an equilibrium of forces is a phenomenon of resistance, whereas the counterattack implies a rush or change of speed that breaks the equilibrium: it was the tank that regrouped all of the operations in the speed vector and recreated the smooth space for movement by uprooting men and arms.

 

Thus, the very dynamics of war, and the ways in which conflict is organized and carried out (centralized and decentralized organizations), are radically altered both by the creation of new objects and new uses. Objects both constrain (e.g. the machine gun limiting territorial progress in WWI) and make possible (e.g. the tank providing new blitzkrieg maneuvers in WWII).

Finally, what constitutes the singularity of a system? For our purposes, we will take a conflict system to be the spatial index of a network plagued by violent conflict. This is a space that is sometimes bounded by state borders, though this is a contingent and not necessary relationship.[33] More often, a conflict system operates both beyond these borders and heterogeneously within state borders, with some towns plagued by violence, while others continue their peaceful coexistence.[34] Conflict systems create their own space, which can contain long extended networks, as well as gaps and holes within a local geographical space. The state and its borders are merely one more actor within the system, and not an a priori prime determinant of the dynamics involved. This definition therefore includes the contiguous networks which play a significant role in the conflict system, yet do not themselves partake of the violence – e.g. the arms trade networks, the surrounding governance networks (regional and global institutions), the refugee networks, the humanitarian networks, and the surrounding economic networks. The limits of a conflict assemblage are variable, open to new additions, and subject to empirical determination. What plays a significant role in a particular conflict? What is significant and insignificant? These answers are co-determined by both the real (i.e. nonconceptual) instance of conflict, and the level of theoretical sophistication. What ANT attempts to do is to open one’s theory beyond its habitual analysis of conflict, and to recognize the surprise that particular actors can produce.[35] A particular resource, for example, may not play a significant role in one’s theory of conflicts, yet upon coming to examine a specific conflict, it may be the case that the resource itself is a relevant actor. Actor-network theory’s refusal to a priori determine the nature or extent of the actors in a situation is what allows for it to remain open to such surprises.

     To summarize, conflict is not a battle between two macro-actors, but rather a micro-functional and often self-perpetuating and self-sufficient system that emerges from the connections between human and nonhuman actors of varying sizes within a single assemblage.

 

3 – Connecting the Global to the Local

 

With a basic analysis of conflict systems in place, we can now begin to substantiate some of our theoretical arguments in the first section with reference to this empirical system. Our task here will be to extract immanent abstractions concerning the global-local linkages operative within conflict situations. This will aim to give concrete meaning to the relations between the two, by showing the ways in which local, regional and global vectors pass through a conflict system, and refract each other. Our focus, in particular, will be on the links between micro- and macro-actors involved in contentious actions. A more extensive study would include the links between non-governmental organizations, donor countries, international institutions, and regional economies.

    Within this constraint, we will examine six analytically distinct modes through which the global and the local interact within an immanent plane: infection/contagion, alliance, leverage/cascade, and aggregation. The modes analyzed here are some of the ways in which macro-actors interact with their local assemblage during conflict. It refutes the standard ideas about the global’s relation to the local, which typically take the form of unexamined metaphors like the local ‘embodies’ the global, ‘reflects’ it, ‘overlays’ it, or ‘manifests’ it. All of these explanations of the relationship are premised upon the separation of the global from the local, whereas, as we have seen, this transcendence must be refused. Macro-actors must thread themselves into the local fabric, and operate from localizable positions. They exist immediately as a unity of both the global and the local.

 

3.1 – Infection/Contagion

 

The first two modes stems in part from David Kilcullen’s ethnographic work in Indonesia, Iraq and Afghanistan. It is responsible for producing what he calls “the accidental guerrilla”: individuals who are not ideologically driven, but rather acting to resist a foreign presence (Kilcullen, 2008, pp. 34-38). The production of such subjects takes place in four stages: it typically begins when a macro-actor (al-Qaeda is the most common in recent conflicts) establishes itself into a local network. The macro-actor then aims at spreading itself and diffusing its ideology to the surrounding areas, often using its newfound safe haven as a space for planning attacks. Thirdly, as a response to the macro-actors’ actions, other macro-actors (the United States military, typically) enter into the network and initiate a conflict. Finally, the local network rejects the interventionist macro-actor and sides with the first group. We see here a number of different ways in which the global immediately interacts with the local.

First, there is the initial mode of infection, integrating a macro-actor into the local network. This occurs through a variety of transferrable mechanisms, including agreements with the government, intermarriage with local individuals, setting up businesses, operating training camps, and becoming involved and partnered with any number of black market operations. In this way, the macro-actor ties itself into the fabric of a preexisting assemblage, becoming indistinguishable from it at many levels. The infection modality is unique to situations where a foreign rebel group enters into a new network and attempts to become a part of it, intertwining itself with the immediate surroundings. In modern times, this has typically taken the form of a foreign macro-actor installing itself in a local network in order to establish a new space for action, as al-Qaeda has done in both Sudan and Afghanistan. Historically, though, it has also included nomadic micro-actors, who either force their way into the borderlands of an empire, or who are recruited by representatives of an empire as protection for the frontiers (van der Pijl, 2007, pp. 82-89).

The more general second mode, contagion, operates through the intentional spread of certain ‘ideological frames’ (worldview-assemblages consisting of narratives, concepts, theories, and practices) which are transferred through texts, schools, personal networks, radio broadcasts, pamphlets, madrassas, etc. This aims at altering the relations between the local network and its environment, and its own internal self-relations. At its most radical level, even individual subjectivities and phenomenologies are altered by being framed from the surrounding assemblage.[36]^,[37] A major part of this process is the shifting of local phenomenological perceptions from their often limited local focus (i.e. events are understood in terms of the immediate surroundings and local habits) to a more global focus (i.e. where events become allied to meta-narratives that construct a particular ‘far enemy’ as the one responsible for grievances). This shift is attained through a multiplicity of affective and cognitive means, often unique to the situation at hand. Yet in any situation where an actor aims to grow, such modes are crucial for its success and for its strength.

On the other hand, the absence of these mechanisms for the state can help to explain cases of state failure. For instance, a major part of a state’s legitimacy derives from establishing a network through which certain ideological narratives are passed along, whether this be through education, stories, traditions, or symbols. In this perspective, a state’s lack of capacity is not simply the inexistence of governance, but rather the inability of the state-qua-macro-actor to extend its network of intermediaries into particular areas. Areas where the state is unable to govern effectively are revealing of the requirements necessary for the state-network to function in normal situations: the existence of things like transportation networks and communication linkages, the overcoming of geographical barriers, the ability to effectively transmit economic flows to particular areas, and the ability to monitor and enforce compliance with national programs.[38] The absence of one or more of these can make it increasingly difficult for the state to act on a large-scale level, and can reduce it to being an actor no larger than any other in the area. It effectively shows that the state is a mere difference in degree between other actors, and not a difference in kind.

 

3.2 – Alliance

 

The second major mode of interaction, alliance, refers to the tendency for micro-actors to cooperate with the local representatives of macro-actors in order to further their own aims. While this is given a prominent role in Kalyvas’ work, it arises in numerous other studies as well, suggesting it is a widespread mechanism (See Lubkemann, 2005; Kalyvas, 2006). As such, it is arguably one of the primary ways in which a global conflict is inflected through a local prism. The specific mode of alliance often takes the form of denunciation. Since the controlling force of a particular territory requires information about defectors, they turn to local individuals for assistance. Most of the time the information that emerges from these denunciations is the result of local disputes rather than accurate denunciations (Lubkemann, 2005, p. 498). People spread rumours or use the local authorities as means to further their own goals, and avoid any direct retaliation against an opponent. Despite the frequently false information, the mechanism has the effect of playing a massive role in determining the use and scope of violence, thereby playing a large role in the overall dynamics of the conflict.

     A similar dynamic occurs between occupying forces and local leaders, who come from a variety of different authority centers, “including tribal structures, insurgent or terrorist networks, local mosques, business and criminal networks, […] government structures, political and religious parties, and official and ‘unofficial’ security forces.” (Kilcullen, 2009, p. 158.) Within these competing leadership roles, the crucial individuals are those crossing over a number of different networks (Kilcullen, 2009, p. 158) – what Malcolm Gladwell (2000) has called ‘connectors’: individuals with an unusually large number of social connections and that play a major role in passing along information and in influencing others to adopt new practices and approaches. Simply because of their structural position in a social network, these individuals largely determine whether certain memes and practices diffuse and spread, or retreat and perish. Alliance, therefore, takes on another modality by connecting not with a wide range of everyday people, but by connecting with a few key nodes in the local network.

    Alliances are thus largely the composition of a unique assemblage consisting of, on the one hand, personal networks, social gossip, everyday disputes, and habitual systems of authority, and on the other hand, a large collection of weapons, trained and untrained fighters, and access to resources. The transferring of information in one direction, and the reciprocal transferring of violent or non-violent support in the other direction, is what constitutes the logic of this mechanism.

 

3.3 – Leverage/Cascade

 

While the other two modes have focused on the ways in which actors external to the conflict system interact with the local dynamics, it is also the case that small-scale actors are increasingly aiming outwards at large-ranging actions. This is particularly the case with the leverage modality, which we can see operating in terrorism and the rise of ‘global guerrillas’, along with their emerging awareness of the material and immaterial networks which constitute the fabric of a conflict system (Robb, 2007). These actors recognize that the global is intimately intertwined with the local, and that small-scale actions can have vast global repercussions. For instance, the power grids, power plants, electrical engineers, transmission lines and electrical towers in Iraq have been the subject of numerous attacks (Robb, 2007, pp. 52-4). Other crucial infrastructure networks of the state system have also been attacked, such as transportation routes, oil pipelines, and even multinational companies. In Iraq, insurgents have used kidnapping and assassination on key foreign companies (e.g. food providers or shipping companies) in order to leverage these small actions for large effects (Robb, 2007, pp. 54-57).

A particular variant of the leverage modality is what John Robb calls the cascade mode, which operates by disrupting a single node, which then forces excess flows into other channels, overloading them and ultimately leading to a cascade of failures. The most obvious example is of electricity, whereby the disruption of a single crucial substation can lead to a cascade of failures throughout the surrounding network. But a similar mechanism holds for disrupting traffic flows (as exemplified when a road is closed or blocked, and side streets become congested and jammed), oil flows (as disrupting a pipeline or blockading a shipping channel restrict the flow and drive up prices), and information flows (with Denial-of-Service attacks being a common tactic for taking down a website).

What these events share is a similar mode of leveraging, which is the exertion of force upon a key node in a (often nonhuman) network, which then has disproportionate ramifications throughout the entire assemblage. It belies the standard idea that small causes equal small effects, and is premised upon the science of complexity and networks.[39] This mechanism operates not only through intentional action, but also through unintended events, such as the massive blackout that occurred in North America in 2003 (Bennett, 2005). The key point is merely that a local scale can transform a micro-actor into a macro-actor merely by virtue of its relative structural placement within the assemblage, thus providing another mechanism through which micro- and macro-actors become related in conflict.

 

3.4 – Aggregation

 

Finally, we come to the fourth mode of global-local connection, aggregation. This typically takes the form of a sort of minimal-level conflict, using what James Scott (1985) has called the ‘weapons of the weak’. It operates through highly decentralized and widely dispersed means. Often without any coordination, tactics such as “foot dragging, dissimulation, desertion, false compliance, pilfering, feigned ignorance, slander, arson, sabotage, and so on” (Scott, 1985, p. xvi) can be easily used by people in their everyday lives to resist the imposition of various forms of order. While, on their own, any single instance of these actions would be useless and at best a nuisance, over a period of time and scale, these small acts accumulate into a much grander mode of resistance. As Scott (1985, p. 35-6) writes:

 

Multiplied many thousandfold, such petty acts of resistance by peasants may in the end make an utter shambles of the policies dreamed up by their would-be superiors in the capital. […] Whatever the response, we must not miss the fact that the action of the peasantry has thus changed or narrowed the policy options available to the state. It is in this fashion, and not through revolts, let alone legal political pressure, that the peasantry has classically made its political presence felt.

 

As we can tell from these instances, unlike the logic of leverage, aggregation does not seek to aim its force at a particular point, but rather operates in a dispersed and accumulative manner. The state’s ability to use its black boxes of power in order to cleanly propagate actions from the capital to the surrounding periphery has been disrupted as some of these black boxes slightly alter the orders – not enough to draw attention to a ‘malfunctioning’ box, but enough to aggregate into a force that leaves the original orders in ruins.

These four modes of immediate global/local connection do not constitute a full and final set, nor do they constitute universal modes inherent to every conflict. Rather, they highlight certain regularities in the relations between the ways in which small-scale network settings affect and are affected by large-sized macro-actors. Future research can and should aim to discern more, while following the general constraint that the global is not a separate realm from the local.

 

4 – Conclusion

 

The main result of the preceding discussion has been to show the usefulness of actor-network theory for understanding conflict dynamics, and to set forth a research program for future study. Reconfiguring the relation between the local and the global entails a number of philosophical, empirical, and methodological consequences that have only begun to be outlined here. Recent studies in conflict have in various ways contributed to this rethinking, but the aim now must be to synthesize this disparate work into a coherent framework for understanding systemic dynamics. Finally, the different modes through which global actors act through local networks has revealed some of the major processes involved in modern conflict.

Future research would tend in two major directions: an empirical and philosophical extension of the existing network. First is the local, situated and ethnographic analysis of existing networks of conflict and the national, regional and international actors involved. The particular network structures and channels of diffusion can be mapped out, with attention to the ways they are evolving. Understanding these would provide novel means to effect change in positive ways, focusing for example, on a detailed analysis of the crucial components that made the 2008 financial crisis possible, and the particular power structures that sustain resistance to reform of the financial system. The second extension would be to examine some traditional philosophical concepts in light of network thought. Notions of particularity and universality can be embedded within networks, and their effects traced and their constitution rethought. Similarly, traditional political concepts such as revolution and radical change must be reconceived in light of a non-structural account of politics. If the diffused agency of actors exists in opposition to power at every level, innovation and change become integral aspects of political space. The dynamics of change become increasingly less binary (reform versus revolution) and much more complicated. Working out the meaning of networks for political and philosophical thought is therefore one of the crucial tasks for future work.

 

Notes