The purpose of this article is to offer an account of the importance of ethnicity in the world today, particularly in the global periphery (what is conventionally termed the Third World —the areas further from the core of the global economic system). The theory proposed is that ethnicity is basically a means whereby the network social form which arises among the dispossessed can be recuperated by the global system and by state power. Its pervasiveness is a sign both of the vitality of networked social relations and their insurrectionary potential, and the attempts of states to reduce the danger of such networks.
Horizontal networks and resistance
Discussion of network social forms has suddenly become rather fashionable. Most of the discussion focuses on contemporary high-tech social movements, which rely heavily on computer networks and other communication networks such as mobile phones. Recent studies by the Rand Corporation for instance have emphasized the growing importance of "netwar" - struggles between or against social networks. Theorists sympathetic to social resistance such as Graeme Chesters make similar claims, attributing the ability of anti-capitalist protesters to mobilize effectively without leadership to a "swarm logic" based on distributed network forms of power. The technological aspect of this view is taken furthest by leftists such as Hardt and Negri, who view the network form of protest movements as an outgrowth of changes in production, of the primacy of "immaterial" labor, and the rise of a new kind of capitalism based on network organization.
Where this leftist reading goes wrong, however, is in linking the network form primarily to high-tech or advanced capitalist conditions. It is certainly the case that high-tech protest groups and countercultural movements use network forms, and that technologies allowing network construction are used in this construction. Hackers, open-source programmers, and online protest campaigns are examples of network social forms. It is also the case, however, that similar non- hierarchical horizontal networks arise in almost every situation where people try to mobilize or cooperate outside the framework of the state and of domination. Hunter-gatherers and other indigenous societies, peasant movements, and the urban poor of the chanty-towns and ghettos are among the most obvious examples.
In relation to indigenous societies, Rohrlich-Leavitt noted that "gatherer-hunters are generally non-territorial and bilocal; reject group aggression and competition; share their resources freely; value egalitarianism and personal autonomy in the context of group cooperation; and are indulgent and loving with children." Where distinct groups exist, they often relate in a networked way—the gift networks of the Trobriand Islands and the extended kinship networks of the Lakota being two examples. One characteristic of such societies is the non-exclusive nature of attachments and affinities, and hence the absence of an overarching identity. Even in the strongest kinds of segmentary lineage systems that come closest to fixed group identity, the existence of extra-familial affinities operates as a restriction on ingroup-outgroup patterns, ensuring some degree of social openness.
Larissa Lomnitz studies survival and mutual aid networks in Latin American chanty-towns, revealing that kinship and neighborhood relations form an entire informal economy, enabling a layer of excluded people to survive on the periphery of major cities by means of horizontal relations. Partha Chatterjee shows how the formation of Indian national identity leaves a trail of fragments—identities based on class, caste, ethnicity, region, religion, and so on—which provide the basis for entire areas of social life organized beyond the reach of the state, in private associations and homes. The power of the state is thus very much partial, constrained by and always at risk from the subcultures and countercultures emerging from the space beyond its reach.
Hecht and Simone provide a series of examples from African societies of horizontal social forms which operate invisibly to inflect, undermine, and sometimes overthrow states and formal institutions.
Rather than defining particular structures, the term civil society has come to indicate myriad invisible threads that weave the fabric of African societies together when nothing else appears to be holding them together... [such as] so-called 'popular neighborhoods' ... usually controlled through ethnic, religious, or sectarian affiliation. They produce informal, and often illegal, associations, alliances, strategies and practice, that provide an infrastructure for the community and a measure of functional autonomy.
The uncontrollability and unpredictability of these movements is the source of their strength. In Senegal for instance,
diverse groups are doing more than developing a critical language. They are taking things into their own hands... attempting to reinvent their surroundings... asking for or demanding... taxes to finance their society independently of a larger authority... creating public protests and the occasional riot.
Even in mass societies, everyday relations are often networked and horizontal, and thus implicitly anarchist—a point made clearly by Colin Ward, who goes as far as to portray "apolitical" kinds of social affiliation such as the local music scene in Milton Keynes as anarchist due to their structure, a network of overlapping voluntary associations existing for practical purposes rather than as part of a political principle of domination.
Examples could be multiplied, but the case is clear: in most of the world, the integration of the global system of domination requires the powerful to deal with a proliferating, unpredictable, subterranean type of social relation which cannot be reduced to mass culture and which, indeed, often appropriates mass culture for its own ends. It is as a means of dealing with this situation that the importance of ethnicity should be viewed.
Networks versus hierarchies
The importance of the network form is that it allows the construction of relations which do not rely on a hierarchic moment. In order for a hierarchy to be constructed, there needs to be an authority or totality to which all the incorporated people or elements submit—an overarching leader, cause, organization, idea, or some other spook around which organization is articulated. This is equally the case for reactive moralities, in which the self-deadening shoulds of self-abasing belief are grounded in some moment of authority. Networks, however, do not require any such moment of authority. They operate like a swarm, without leaders or guiding principles, and they can incorporate people and other beings in ways that bring them together in spite of, or even because of their differences. Deleuze and Guattari contrast the network (or "rhizome") model to the "arborescent" model, structured like a traditional image of a tree (though in fact trees do not follow this model very closely). Whereas in an arborescent model, everything stems from a central trunk, and the branches are given their status by their relation to this trunk, in a network there is no integrating element, only a series of non-reduc- tive and infinitely expansive horizontal connections. For this reason, networks are inherently dangerous to all systems of hierarchical power.
Already in the work of Kropotkin, a dividing-line is drawn between society, by which he largely means network logics, and the state, referring to hierarchic forms of integration. Kropotkin counterposes the social logic of networks and voluntary associations to the hierarchic political logic of statism, in which people are fragmented and controlled. While networks are bubbling with life, states bring with them death and decay, for the state has to destroy horizontal relations wherever it goes, to arrogate social power to itself and stand in for the community that no longer exists (one of the paradoxes being that the state needs to create the scarcity and competition which then act as the legitimation of its existence). Hence the first act of the state wherever it was established was to break down horizontal networks and pillage the societies they formed.
But while the State was condoning and organizing this pillage, could it respect the institution of the commune as the organ of local affairs? Obviously, it could not. For to admit that some citizens should constitute a federation which takes over some of the functions of the State would have been a contradiction of first principles. The State demands from its subjects a direct, personal submission without intermediaries; it demands equality in slavery; it cannot admit of a 'State within a State'. Thus as soon as the State began to be constituted in the sixteenth century, it sought to destroy all the links which existed among the citizens both in the towns and in the villages.
The state principle is a principle which destroys everything. The irony of a recent British law which defines gathering together in a public place as anti-social behavior would not have been lost on Kropotkin. It stands in a long tradition of state bans and attacks on horizontal association. For statists, people can only relate through the intermediary of the state; to remove this mediation is inherently threatening to it.
Either the State for ever, crushing individual and local life, taking over in all fields of human activity, bringing with it all its wars and domestic struggles for power, its palace revolutions which only replace one tyrant by another, and inevitably at the end of this development there is... death! Or the destruction of States, and new life starting again in thousands of centers on the principles of the lively initiative of the individual and groups and that of free agreement.
This thesis was provided with further empirical backing by Clastres, who argues that non-state societies construct mechanisms to prevent the emergence of systematically stratified relations.
State power and the problem of networks
State power requires the suppression of rhizomatic affiliations in order to intensify its own domination. In western societies this has reached the most extreme forms of social fragmentation and hierarchical reintegration known to history. One sees a situation where the majority are politically integrated in a predominantly symbolic way, often receiving nothing in return, and being managed politically as a Silent Majority through the mass media. This leads Trevor Pateman to argue that the idea of televised elections is misleading—instead one should refer to a "television election," in which the election itself is a televisual phenomenon, a construct of the mass media. This is certainly the maximal form of rule which the state has so far found.
This massified society is built historically on a gradual breaking-down of horizontal links and the construction of a massified society where people only relate via categories and formal relations. Thus, the so-called industrial revolution is actually a cumulative progress in alienation, a cumulative increase in regulation and separation to ensure the imposition of social control. Historians writing about successive periods chart this process over time; thus, Craig Calhoun can write of the destruction of artisan culture in nineteenth-cen- tury Britain in very similar terms to those used by Richard Hoggart in discussing the rise of mass culture in the 1960s and the resultant disintegration of locality-based forms of working-class life. The cumulative effect of such corrosion of horizontal affiliation is the construction of a society based on what Sartre terms "seriality" — a relation in which people are interchangeable, mediated by their social position. One reaches the point where Hakim Bey can argue that even to meet with other people outside the contexts of work and family is already a victory for revolutionary energies.
Such a situation is, however, unusual. It characterizes, at its broadest, western societies, and perhaps even then only a few of the most advanced (in the system's terms) not having fully affected rural communities or areas such as southern Europe. If mass society is the optimal means by which the system manages social relations, in much of the world it has to cope with a suboptimal situation where social networks remain lively and active.
In most of the world, modern ethnicity is a colonial invention. It apparently derives from some combination of nationalism—a phenomenon dating back three centuries at most, arising among Europeans and set- tler-colonists, and basically constructing spooks of sameness linked to the rise of industrial technology—with theories of biological superiority derived from discourses of aristocratic class privilege. Colonial administrators and their pet anthropologists and social scientists went to great lengths to categorize people into groups based on ethnicity — the basic function of the colonial census as a device of subject-construction, as well as to construct and promote discourses differentiating the various groups and associating them with some eternal essence. In some cases (such as Vietnam), colonizers actually went to the lengths of inventing an entire written script in order to construct the colonized population as an ethnicity.
This project continued after decolonization, and in this regard at least, the postcolonial state is far more a continuation of colonialism than its triumphant adversary. Nominally independent states (often under the watchful eye of imperial gunships and international financial institutions) do a much more extensive job of constructing and enforcing ethno-national categories than their colonial forebears. Thus one finds Algerians, Iranians, and Indians acting much the same way towards the Berbers, Kurds, and Nagas as the French and British once acted towards them. One also finds the subordinate states performing the function of integration into the world system on their own behalf, saving the imperial powers any need to expend military resources on their compliance. Imperial violence is then reduced to a kind of fire-fighting operation, suppressing those lines of flight which take particular peoples outside of the state system (so-called failed states) or which pit particular states against the dominant powers (so-called rogue states).
Ethnic networks versus affinity networks
Nevertheless, networks continue to proliferate throughout the global periphery, weakening state power and generating periodic insurrections such as the Argentinazo, the Berber uprising, the struggles in Papua and Bougainville, the repeated overthrows of neoliberal governments in Ecuador and Bolivia, the Zapatista uprising, and a thousand smaller-scale examples undermining the integration of the global system. To contain the revolt, therefore, the system needs to find ways to recuperate the networks which undermine it. From this dilemma arise the many forms of uneven development, in which network- based social forms are incorporated to a sufficient degree to enable exploitation or at least to head off revolt. Capitalists are only able to profitably exploit societies in which a capitalist infrastructure has not yet been constructed, by working with and through existing social relations; often, this means finding ways to incorporate networks. And it is here that ethnic and patronage networks become useful.
The distinction between ethnic/ patronage and horizontal/affinity networks is subtle, because the external organizational forms are often quite similar. The difference is that, whereas the latter involves horizontal links and structural openness, the former introduces a hierarchical element which is potentially system-integrative or leaves the network open to integration.
In the case of ethnic networks, this hierarchical aspect is an identity category, a strong discourse of Us and Them defining the network and its resources as the exclusive property of an authoritative social group. In patronage networks, this identity- basis is used in combination with a hierarchical situation—an asymmetrical control over resources —to integrate the network around relative privilege, under the control of an elite within the group who hold positions of power and use them to the advantage of the group (and to the disadvantage of outsiders). It is my contention that patronage networks based on ethnic, religious, and sectarian affiliations are the primary form of system-integration in the global periphery, and that these networks occupy such a role because of their proximity to the affinity-network form which arises among the dispossessed. Ethnicity is thus crucial as the primary recuperative device used by the powerful in the world system to contain the insurrection of the global poor.
The basis of ethnic identity in exclusionary categories — even to the point of a structural primacy of the other over the self, a dependence of the self's identity on its differentiation from an excluded other — has been widely noted by conflict analysts looking into ethnic conflict. Anti-colonial activist Franz Fanon put this phenomenon down to what he terms a "narcissism of small differences" — an elevation of superficial difference into something fundamental, used as a way to privilege oneself over those conceived as different—like in the parable where people were killed or freed depending on how they pronounced the word "shibboleth."
The difference between ethnic identities and the kinds of identities that arise in social struggles is that ethnicity typically maps the in-group as both eternal and privileged (or superior to others), whereas socially-located identities react more directly to social relations and conjunctures, without the same degree of mediation by abstract, mythical categories. Thus, Alfredo Bonanno discusses the rise of political Islam as a kind of mapping of the situated categories of oppressor and oppressed onto the ethno-religious categories of Muslim and infidel.
The Islamic distinction between friend and enemy, faithful and unfaithful ("mu'min" and "kafir"), corresponds to the modem one between oppressed and oppressor ("mustad" and "mustakbird"). And it is within the immense theoretical laboratory of militant Islam that disturbing similarities are appearing between civil war and military war, war of peoples to liberate themselves and war of States to impose their own domination. And Muslim fundamentalism finds a good hold where it equates oppressors with the unfaithful and the latter the most advanced, i.e., wealthiest, countries of the West. Poverty has always been short-sighted, and is a bad counselor... In particular there is a mental closure that comes into contrast with the tradition of civility and tolerance peculiar to the Muslim world which, is transforming Islam into a theodicy of dominion, a totalitarian regime.
Through this mapping, a group which is in fact subordinate fantasizes itself in a position of domination and sets about establishing this domination through the microregulation of everyday life and a generalized violence against outsiders. Political Islam is here not unique; Bonanno recognizes parallels with certain eastern European nationalisms, and phenomena such as Hindu communalism could be added to the list. The basic device here is a refusal to identify as an excluded, peripheral or minoritarian figure, instead hiding behind a myth in which one identifies as a member of a superior in-group, albeit a dispossessed and unfairly treated one, and attempts to establish this in-group as the new master, overthrowing the existing masters only to replace one domination with another. Its characteristics include an often extreme violence against outsiders —not only members of the dominant group on an undifferentiated basis, but also other out-groups, perceived traitors to the in-group, and people whose personal autonomy puts them outside the fixed essence attributed to the in-group. Islamists for instance have repeatedly targeted groups such as gay men, unveiled women, film-goers, barbers, members of the Ahmadi sect, indigenous Papuans, non-Muslim minorities such as the Balinese Hindus, and members of different branches of Islam, as well as targeting westerners, Christians, and Jews in an undifferentiated way. This generalized violence against out-groups is discursively a practice of domination, even if socially it sometimes correlates with a struggle against real oppressors.
Because of the over-determination of conflicts with active, anarchistic and reactive, totalitarian elements (resistance to oppressors and establishment of domination), peripheral networks can slip between emancipatory and repressive social forms rapidly and almost imperceptibly. An example is the panchayat or village commune model which operates in many isolated rural parts of the Indian subcontinent, such as the ungovernable highlands of western Pakistan. These agencies of autonomous local power are the locus of resistance movements such as the revolt against Pakistani state control in Waziristan, and often organize resistance to capitalism and the state, such as the expulsion of Coca Cola from Plachimada, India. However, they are also the structural basis for localized forms of domination, as in the case of Mukhtar Mai, publicly gang-raped as a punishment for breaching inter-clan barriers and damaging the honor of a privileged group. The form of local autonomy seems to produce emancipatory and dominatory effects depending on the balance of hierarchical and libertarian elements in the local power structures and dominant customs which come into play there.
The distinction between affinity networks and ethnic- based movements is clearer in the case of Manipur. In 2004, a mass social movement against emergency powers shook the Indian occupation. This movement was not based on ethnic categories, but rather, operated across the lines of the various social groups. One of its most notable features was the adoption of a fragmented, centerless, localized form of organization in which social groups, classes, villages, and so on, were able to organize their own autonomous activities. This proliferation of direct action overwhelmed the state machine. One report states that '[t]he entire stretch of the road, from Karong to Hiyangthang was dotted with such barricades, and attempts by the police to clear the road were frustrated due to the sheer number of agitators' With villagers in each area organizing autonomously, the state was overwhelmed by action. Parallels with effective anti-capital- ist and ecological direct action in the west are very obvious here.
In contrast, ethnic politics in Manipur takes the form of the operation of a number of hierarchical armed opposition groups. Each of these groups is attached to one or another ethnicity, and their methods take the form of persecution and exclusion of others. Each is fighting for some kind of state in the world system—greater privileges in the distribution of patronage, an independent state under the control of a specific group, or the institutionalization of one or another set of privileges (such as language criteria) establishing the supremacy of a particular ethnicity. While Meitei groups seek an independent state of Manipur, Naga and Kuki groups fight for separate homelands, and in contrast to the popular autonomy expressed by the social movements, the armed opposition groups operate in an extremely hierarchical way, imposing "moral codes" (such as traditional dress and alcohol prohibition) by means of violence and punishment. Armed opposition groups
regularly conduct publicity seeking exercises such as setting fire to drugs, breaking alcohol bottles and destroying video cassettes of Hindi and pornographic movies in a bid to project themselves as protectors of State's culture and moral values.
This complicity in statism and authoritarianism also leads to conflicts among the groups. "[T]he core ideology of all the insurgent groups moved around their respective distinct ethnic identity." This has led, for instance, to tensions between Naga and Kuki who inhabit the same hill territories, and between Meitei and Pangal (Muslim) groups, sometimes leading to armed clashes and deaths.
In the context of the land shortages caused by encroachment and the failure to invest in productive jobs, every group views the limited land and jobs as its exclusive right. So each community rewrites its history to claim an indigenous status and the exclusive right over resources in a given area. Ethnic conflicts are a direct consequence of such hardened ethnic identities and exclusive claims.
It is notable that there is a large amount of tension between the various armed groups, and between these groups and the social movements. Women's, peace, and human rights groups have organized protests against killings by armed groups and dialogues for peace between different ethnic groups. It is interesting to note in this context that the Indian government seems more inclined to negotiate with armed opposition groups and to rehabilitate their members than it is to engage with the demands of civil-societal groups across the northeast; and also that it seems to encourage ethnic tension, both by pursuing peacemeal negotiations with groups one at a time, and allegedly by setting up certain groups to maintain division.28 In other words, the state is quite happy with the existence of a war system or a system of negotiations with state-like bodies, in which it can use means such as patronage to pursue integration. It is far less happy dealing with movements of a type dissimilar from its own. The same can be said for the Russian state in Chechnya, which has concentrated on eradicating secular militias and covertly strengthening Islamist factions.
Athina Karatzogianni's work on cyberconflict similarly produces an empirically-based distinction between ethnoreligious and sociopolitical movements, with the former based on rigid identity categories and exclusions, and the latter notable for looser kinds of attachments. The latter are better able to use the more horizontal characteristics of information networks, whereas the former are likely to simply repeat the war model they draw from conventional politics. Thus for instance, in relation to the Iraq war, a clear difference can be observed between those movements pursuing ethnoreligious and sociopolitical reasons for opposing the war.29
A similar distinction can be made between the kind of messianic Judaism embraced by authors such as Walter Benjamin and Martin Buber, and the type espoused by statist Zionists as the basis for constructing an ethnically exlcusion- ary state. European racists found Jews threatening precisely because of their non-inscription in the state system and their resultant outsider status. It was from this position—as bearers of hybridity and as people "out of place" — that the most important radical developments of Jewish thought have arisen. In contrast, with the exception of a small neo-Nazi fringe, the normalization of Jewish identity through the creation of a state-based ethnicity has effectively defuzed anti-Semitism among European nationalists and statists. Rather, there is now a kind of fellow-feeling with Israel as a western-allied power contributing to world-system integration in an unstable region. This rapid turnaround from hostility to commonality can be explained in terms of the system-integrative functions of ethnicity. Contrary to appearances, what European statists hated about Jews was not anything specific to this particular group, but rather, the fact that a particular group (any particular group —one could also refer to the Roma here) could not be inscribed in the dominant system. The moment this exceptional status was eliminated by means of integration into the dominant system of representations, a discourse of antagonism was replaced by a discourse of similarity and equivalence.
One should thus take seriously the paradoxical position of contemporary racism—the view that ethnic others are unobjectionable as such, as long as they are within "their own" cultures or regions. Far from being in contradiction with historical racisms based on superiority and dominance, this principle has in fact always been central to the racist project (as for instance in South African apartheid, British colonial practices of native government, and the practices of the Canadian Department of Indian Affairs). Racism of all kinds rests on identity-fixity, which in turn requires otherness at a distance, as something radically outside. In contrast, the flows of hybridity and interrelation are threatening to such systems of identity-fixity because they overflow the system of management through representation. Of course, this recognized difference is also quite compatible with systematic inequality— confined to their own social or geographical space, distinct groups can then be differentiated in ways which produce social, economic, political and cultural inequalities between them. The entire immigration control system is one big apparatus for the systematic construction of inequalities through the differentiation of people as belonging to different identity categories.
Patronage power and system-integration
Though ethnicity can express itself in terms of racism and populism, its most characteristic form—especially as regards attempts to manage multicultural societies - is patronage. Patronage networks, usually based on ethnic or quasi-ethnic (e.g. regional or religious) networks and on various forms of nepotism, are the normal form of political control and integration in the overwhelming majority of the world, with the partial exception of certain western societies. This predominance of patronage politics is often pathologised in western-based literatures on development, governmentality, and comparative politics — as for instance in Richard Hodder-Williams' account of the extractive view of politics held by many Africans and its alleged contrasts with a developed form of citizenship.
In fact, a reverse view is more appropriate — the pervasiveness of patronage is a sign of the vitality of everyday life, the failure to develop overarching spooks to such a degree as to subordinate populations, and a resultant need to offer something concrete as a guarantee for support. They appear corrupt to those accustomed to western systems, partly because they are different, and partly because they render support for the state conditional on what it delivers. Statists expect an unconditional allegiance from citizens, expect them to choose between agendas or election candidates based on what is good for the state... No wonder they object to this kind of conditionality, where the state actually has to deliver to gain legitimacy. The idea of corruption often gains a quite different meaning within patronage systems themselves. When raised in the slogans of opposition movements, it is typically a protest against the distribution of resources exclusively to regime supporters, a call for inclusion in the patronage apparatus or for its dismantling.
The most obvious form of patronage consists of the unequal distribution of resources to the benefit of groups deemed supportive of the regime. Thus for instance, one often finds funding for humanitarian and development projects directed at sympathetic regions, government jobs distributed by ethnicity or affiliation (a majority of Saddam Hussein's cabinet were from the Tikrit area for example, and nearly all were Sunni Arabs rather than Shiites or Kurds), and contracts and perks given to companies and individuals associated with supportive groups.
Another example is the kind of situation where power is rendered conditional on the performance of rituals of generosity which effectively buy the support of particular groups. Taiwanese candidates are expected to throw lavish festivities during pre-election rallies as a means of winning rural votes. This can be likened in certain respects to the kind of potlatch events which are often a condition for the acceptance of social hierarchy in big-man tribal arrangements, and which are often considered to be a form, of redistribution and inequality limitation. When incorporated in a modern state system, however, these practices are actually a weak form of system-integration.
There is also a common form of inverse patronage in which violence-prone states distribute their violent actions along patronage lines. During the pogroms which preceded the independence of East Timor for instance, widely portrayed in the western media as wanton violence, there was actually a systematic pattern of targeting districts and villages which had returned pro-independence results in the recent referendum. A similar recent example was "Operation Sweep Out the Trash" in Zimbabwe, which targeted mainly the urban poor — a key constituency for the political opposition. One analyst claims that "a desire to punish the urban areas for their almost universal tendency since 2000 to vote for the opposition MDC" was a key motivation for the massive purge. In fact, from the purges against the (ZAPU-voting) Ndebele on the achievement of independence through to the land redistributions (notoriously biased towards regime loyalists), the Mugabe/ZANU regime has been a paradigmatic case for the use of terror as a means of inverted patronage. Similarly, the janjaweed militias in Sudan are typically allowed to plunder the regions in which they operate in return for services rendered to the state.
The important point to note here is that the various forms of patronage are the primary means whereby the world economic system is kept intact. Without these means of system-integration, the peripheral areas would tend to delink from the system, because the predominant horizontal forms of coordination provide little basis for integration into the system. It is crucial, however, to bear in mind the limits of delinking as previously conceived by world systems theorists such as Andre Gunder Frank and Samir Amin. These approaches tend to emphasize delinking peripheral states from the world system. In practice, however, delinked states are just as likely to pursue system-integrative strategies of patronage and identity-formation as those which are fully integrated into the world system. So-called anti-imperialist regimes such as Zimbabwe, Sudan, and Iran are among the most extensive users of ethnic patronage and identity-based exclusion. And the State Communist regimes such as North Korea and Maoist China developed their own peculiar brand of patronage politics, classifying people into relatively privileged and excluded groups based on the class origins of their ancestors, in effect reconstructing ethnic schemata on the basis of class.
Multiculturalism as ethnic patronage
The form which ethnic integration takes in western societies is the community leader phenomenon, also known as multiculturalism. Basically, this phenomenon operates by creating a stratum of privileged individuals within each disenfranchized or excluded group, whose purpose is to socially manage the group, to channel its frustrations into a positive attachment to an ethnic category, and to defuze these frustrations by means of the negotiation of this group's constructed identity within the system.
The history of this strategy can be traced back to the British Empire, which often used local leaders (religious figures, chiefs, kings, etc.) in this kind of way—a strategy which was absolutely crucial to the management of a wide- ranging empire given the small number of settlers and administrators. It was also used in nineteenth-century Italy, where it took the peculiar form known as trasformismo — the beheading of social movements through the parliamentary or administrative incorporation of movement leaders or figureheads. It reached something akin to its modem form in the Soviet Union and Yugoslavia, which were the world's first multicultural states. Each "nationality" was permitted its own local party structure, representative institutions, and so on—but its representatives, much like today's community leaders, were appointed from within the party-state apparatus, usually by the central leadership, as a means of integrating the various "national" areas. It reached its current form as a response to social crises in countries such as America, Britain, Canada, and Australia—as a strategy for defuzing the increasingly militant struggles of black people, migrant populations, and indigenous peoples. Though often counterposed to the monocultural models of ethnic-majority populism, it is in fact structurally similar, relying on a similar model of social integration through ethnic categories.
A similar strategy has been used to contain prison revolt. When the black consciousness movement first reached prisons, the resulting assertiveness of black prisoners was welcomed by the entire prison population, as something that altered the balance of power between prisoners and screws and that won important gains for prisoners. To undermine this solidarity, screws started playing favorites—giving benefits to black prisoners only, to create resentment from other prisoners, or rewarding other groups for being compliant. In this way, one can see the origins of the ethnic prison gangs which have since come into existence. These gangs can be seen as at least partly a result of divide-and-rule strategies which used ethnicity to undermine resistance.
When network social forms have outflanked control apparatuses, ethnicity can be used by states and other dominant groups in order to reestablish control. The effects of this become very clear in contexts where the state uses pogroms to defuze anti-state unrest. The Indonesian financial crisis of 1997 offers an especially clear example-state forces suppressed popular anti-capitalist, anti- state and anti-dictatorship protests, but encouraged and collaborated with pogroms against the Chinese population of Indonesia. These pogroms served as a way to channel social discontent in a way which was harmless for the state and capitalism. This kind of pogrom may be uncommon in Indonesia, but the channeling of frustrations onto ethnic groups deployed socially as intermediaries is very common—not only are the Chinese frequently exploited in this way throughout Southeast Asia, but colonial regimes frequently used ethnic minorities (the Tutsi, the Tamils) or migrant communities (such as South Asians in East Africa) in the same way, and one could even interpret European anti-Semitism along these lines. In addition, ethnic politics based on pogroms and constant conflict is a normal part of capitalist management in certain parts of Nigeria (eg. Kaduna), Indonesia (eg. Ambon) and India (eg. Gujarat). There are also similarities with the situation in Sydney, where a racist pogrom—tolerated, encouraged, and incited by state agents—followed two years of mass unrest against the state. The boundary between rigid ethnic identities and loose affiliations in revolt against oppressors is a slim one, and one which the socially excluded cross over on a regular basis; the emotional and psychological reactions generated by social and economic marginality and exclusion seem to be equally open to either kind of articulation. This fluidity is something the state exploits in order to prevent the kinds of revolts which really threaten its power.
A similar observation could be made regarding events in Britain and France in November 2005. In France, the absence of multiculturalist integration left open the possibility of revolts which crossed boundaries of ethnicity and religion, and which were directed primarily against the state. The result was a massive urban insurrection organized on a network basis against the poor, directed primarily at crackdown culture and the repressive apparatuses of the state and capitalism. There was also unrest in Lozells, Birmingham, at around the same time - an area which hosted a large anti-state uprising in the mid-1980s. In this case, however, the discontent-while clearly sparked by exclusion, poverty, and social alienation —was channeled in directions which were largely harmless to the state. Instead of taking the form of an uprising against the police, the revolt took the form of communal fighting between young men of Asian and African origin, on the basis of firm identifications with specific ethnic categories. This is the harvest the state has reaped for its strategy of multiculturalist integration—the use of ethnically targeted state patronage to solidify group identities, and the use of populism to channel concerns arising from social exclusion and economic precarity into ethnic categories.
Similar strategies have been used in occupied Iraq as part of the strategy to contain resistance to the occupation. The Iraqi elections were constructed in such a way as to encourage the formation of ethnic political blocs and their competition for state resources. Ethnic and sectarian militias have been incorporated into local state apparatuses and allowed to take control of local governmental machineries in return for collaboration. These militias have then been deployed against other ethnic groups-Shiite militias in al-Qaim, Kurds in Fallujah-to foment divisions and create a basis for colonial power in the internal structures of Iraqi society. American troops forge alliances with local tribes, using existing rivalries to undermine opposition militias.3 An American military leader adopts the dress, mannerisms and customs of village sheikhs in an attempt to gain influence. Ethnic militias attract recruits with payment and perks, while the British army effectively hands over southern towns to the Mahdi and Badr militias. Iraq's interior ministry, controlled by the Shiite SCIRI faction, refuses to deploy western-trained troops, instead delivering positions to its own loyalists. Similarly in Afghanistan, the occupying forces rely on local militias to maintain control. In both cases, it is only the use of local ethnic patronage networks that has stood between the occupier and instant collapse. The blatant use of such networks in these cases of sharpened conflict is a clear indication of their crucial function they play in the integration of systemic power at the periphery.
Conclusion: ethnicity as social control
So what is the trick which allows patronage politics to integrate networks? The answer lies in the kind of categories it uses. Patronage networks are similar to affinity networks in adopting a network organizational form; but they differ in that the integration of the network is based not on degrees of affinity nor on particular projects, but rather, on belonging to a rigid category. Most often, this category is ethnic, although it can also be religious, regional, class-based, and so on. It involves the endorsement of an idea—a Stirnerian spook—which holds the network together and which sets up a rigid borderline between the inside and outside of the network.
The trick performed by state strategists is to alter the balance between active and reactive attachments within a network, turning categories of affiliation into rigid categories and active, expansive association for practical, ludic, or survival purposes into reactive, exclusive, closed association for purposes of competition, domination, and patronage. The two mutually exclusive logics —of affinity and ethnicity - are usually already operative within social networks of the dispossessed; whereas one of them provides the seed of insurrection which renders these networks disruptive of state power, the other is the Trojan horse through which the state minimizes the threat which the networks pose.
Basically it is a particular, sophisticated form of state- led recuperation. Examining trasformismo in Italy, Gramsci argued that this kind of strategy was a way of creating social passivity by preventing the emergence of antagonistic forces. It was a means of passive revolution, which is to say, of Hegelianism in the bad sense — the synthesis of each antagonistic agent into the existing system, so that a radical break could never emerge. This account prefigures the later Situationist theory of recuperation, but the two phenomena are subtly distinct—whereas recuperation usually happens by means of symbolism (such as the commercial appropriation of the cultural symbols of dissident movements), trasformismo is more of an organizational phenomenon, integrating oppositional movements through the incorporation or creation of a malleable leadership. But the function is basically the same—bringing a flow which exceeds the system back into the system's remit, by means of reinscribing it in the system's categories.
In Deleuze and Guattari's work, this process is viewed as axiomatization.
The capitalist system is axiomatic in the sense of relying on quantification as a means of establishing equivalence between diverse phenomena. Thus, many different subcultures are integrated in capitalism by means of the addition of new axioms, of particular niche markets, or new kinds of commodities. Capitalism was able to digest the Russian Revolution only by continually adding new axioms to the old ones: an axiom for the working class, for the unions, and so on. But it is always prepared to add more axioms... it has a peculiar passion for such things that leaves the essential unchanged.
Ethnic politics is one example of this kind of phenomenon — the addition of axioms in order to include particular populations and particular social networks in the capitalist world system.
The crucial point about ethnicity is its establishment of identity-fixity. The kinds of affinity theorized by Bonanno, Stimer, Kropotkin, Ward, Deleuze, and the rest are based on types of affiliation which are immediately actual. In contrast, affiliations aroused by ethnic categories are mediated by a fixed representation which states that people belong to a particular social group on the basis of some essential characteristic or other. Thus networks are fixed and closed, rather than fluid and open. It should be viewed as a reactive kind of attachment in the Nietzschean sense —an attachment to identity-categories based on an emphasis on certain differences as exclusive, what Fanon terms a "narcissism of minor differences;" and an establishment of personal or group identity based on a primary exclusion, on an insistence that a particular Us-Them relation is primary in defining one's being. Against this, anarchist networks insist on the primacy of becoming or existence over being, the active construction of categories, and the irreducibility of people, groups, and relations to any imposed representational categories. As such, they are outgrowths of active desire. This difference, however, is a difference in the relation between self and group—not necessarily a difference in the structure of the group itself. To an outside observer, the two kinds of networks can look very much alike.
The crucial political point here is that the network form is necessary but not sufficient to liberation from statism and hierarchy. The lesson of ethnicity for anarchists is that social networks can be recuperated into hierarchic social forms by means of categories which operate primarily at an ideological or psychological level. In this way, a preponderance of social networks (as opposed to hierarchies or mass-society forms) can be rendered compatible with certain forms of state control and systemic integration. The question is not simply one of building networks of resistance among the excluded and oppressed. There is also a need to address and overcome the cops-in-heads, the categories and spooks that tie people into hierarchical identities.
 E.g. John Arquilla and David Ronfeldt, Swarming and the Future of Conflict (RAND Corporation, 2001).
 Graeme Chesters, Another World is Possible: Social Movements Confronting Capital and the State (London: Pluto 2006).
 Michael Hardt and Antonio Negri, Multitude (Harmondsworth: Penguin 2004), 54-8,145-8.
 ‘Cited in John Zerzan, Future Primitive http://www.insurgentdesire. org.uk/futureprim.htm
 Harold Barclay, "Anthropology and Anarchism," The Raven 18 (5:2), April-June 1992, p. 160.
 Larissa Lomnitz, Networks and Marginality: Life in a Mexican Shantytown (New York: Academic Press 1977)
 Partha Chatterjee, The Nation and its Fragments (Princeton: Princeton University Press, 1993).
 “Simon Hecht and Maliqalim Simone, Invisible Governance: The Art of African Micropolitics (New York: Autonomedia 1994), 14-15.
 Hecht and Simone, p. 104.
 Colin Ward, "Anarchy in Milton Keynes," The Raven 18 (5:2), April- June 1992,116-31.
 Peter Kropotkin, The State: Its Historic Role (1897), http:/ /www.panar- chy.org/kropotkin/1897.state.html
 Pierre Clastres, Society Against the State (New York: Zone, 1989).
 “Trevor Pateman, Television and the 1974 General Election, http://www.selected- works.co.uk/ televisionl974.html
 “Leopold Roc, "Industrial Domestication: Industry as the Origins of Modern Domination," http: / / www.eco-action.org/dt/ inddom.html
 Craig Calhoun, The Question of Class Struggle: The Social Foundations of Popular Radicalism during the Industrial Revolution (Chicago: University of Chicago Press, 1982); Richard Hoggart, The Uses of Literacy (Harmondsworth: Penguin, 1957).
 Hakim Bey, "Immediatism vs Capitalism," http://www.left-bank.org/bey/ imm_cap.htm
 See Benedict Anderson, Imagined Communities (London: Verso, 1983); Eric Hobsbawm and Terrence Ranger, The Invention of Tradition (Cambridge: Cambridge University Press, 1992). While these texts are the main rivals in academic discussions of nationalism, it is worth noting that they agree on the basic points—that nationalism first came into existence about three centuries ago, that it was a form of state-integra- tion based on reinterpreting history, that its basic contours are linked to industrialism and that its spread to the majority of the world was carried out by colonialism. Where they differ is on the reason for its emergence—while Hobsbawm and Ranger emphasize political machinations, Anderson stresses changes in perspective resulting from the new industrial landscape.
 “Anderson, Imagined Communities, 164-5.
 See Edward Said, Orientalism (Harmondsworth: Penguin 1995), especially chapter 1. "No merely asserted generality is denied the dignity of truth; no theoretical list of Oriental attributes is without application to the behavior of Orientals in the real world" (49).
 “Anderson, Imagined Communities, p. 128.
 E.g. Vivienne Jabri, Conflict Analysis Reconsidered (Manchester: Manchester University Press, 1996); Ali Khan, The Extinction of Nation States: A world without borders (The Hague: Kluwer Law International, 1996); David Campbell, "Violence, Justice and Identity in the Bosnian Conflict," in Sovereignty and Subjectivity, ed. Jenny Edkins, Nalini Persram and Veronique Pin-Fat (Boulder: Lynne Piener, 1999), 21-37.
 “Alfredo Bonanno, "For an Anti-Authoritarian Insurrectionalist International," http://www.geocities.com/kk_abacus/ioaa/insurint. html
 Pitch of protest hightens after partial lift," Imphal Free Press, 14 Aug.
 Routray, Bibhu Prasad (2005), "Manipur: Rampaging Militants, Mute State," SOUTH ASIA INTELLIGENCE REVIEW Volume 3, No. 26, January 10,2005 http://www.satp.org/satporgtp/sair/Archives/3_26.htm
 Routray, "Manipur..."
 Upadhyay, R (2004), "Manipur —in a strange whirlpool of cross-current insurgency," South Asia Analysis Group http://www.saag.org/ papersl3/paperl210.html
 Femandez, Walter, "Limits of Law and Order Approach to the North- East," EPW Commentary, Nov 16 2004
 “Richard Hodder-Williams, An Introduction to the Politics of Tropical Africa (London: Allen and Unwin, 1984).
 The possible argument that patronage or extractive politics is an outgrowth of poverty, while not entirely unfounded, ignores that these practices also extend to wealthy sections of the population and to relations between the state and corporations.
 Deborah Potts, '"Restoring Order'? Operation Murambatsvina and the Urban Crisis in Zimbabwe," Journal of Southern African Studies, Volume 32, Number 2, June 2006.
 Samir Amin, Delinking: Towards a Polycentric World (New York: Zed, 1990).
 John Darwin, Britain and Decolonization: Retreat from Empire in the Postwar World (London:MacMil- lan 1988).
 Hannah Allam and Mohammed al Dulaimy, "Marine-led Campaign Killed Friends and Foes, Iraqi Leaders Say," http:// www.commondreams.org/head- lines05/0517-01 .htm “
 Antonio Castaneda, "Iraq citizens deem U.S. soldier as sheikh," http://www.mlive. com/ newsflash/ international / index. ssf?/base/international- 25/1122826992255580.xml&story list=intemational
 Daniel McGrory, "Militias steal new recruits with better pay and perks," http://www.timesonline.co.uk/ article/0„7374-2159349,00.html “
 Juan Cole, "British To Withdraw from Maysan, Muthanna," http:// www .juancole.com/2006/05/bri tish-to-withdraw-from-maysan.htm
 “Iraq's Interior Ministry Refuses to Deploy US-Trained Police," GI Special 4D5.
 “Gilles Deleuze and Felix Guattari, Anti-Oedipus (London: Athlone 1984), 253.