In my previous post I discussed big macro-level processes of victory, defeat, and compromise. In this one, I want to continue using academic research to speak to policy issues, but shift focus to a more nitty-gritty question: who joins insurgent groups? There are many reasons for mass populations to grow discontent with a political situation: ethnic competition and status reversal, state weakness or collapse, shifts in economic distribution, imposition of direct rule, etc. But in most rebellions the actual number of fighters is small relative to the population. Even insurgent supporters may form a minority of the group in whose name the insurgents fight. Why do some discontented people join or support insurgent groups, while others, even with the exact same political beliefs or ethnic/class identity, stay on the fence?
The basic challenge of insurgent mobilization is the collective action problem – even if I want the insurgent movement to succeed, my individual contribution probably doesn’t matter much and it would be less dangerous to let others do the fighting for me. Insurgency is incredibly risky and costly, and so my best strategy is to keep my head down rather than risk the tender mercies of the Burmese or Pakistani military. Though there have been some challenges to this conceptualization (Kalyvas and Kocher argue that sometimes it can actually be riskier to stay neutral), it remains a useful framework for thinking about the problem. What are the existing scholarly arguments, and what might they tell us about broader questions of insurgency and counterinsurgency?
One major perspective focuses on social networks and institutions as the conduits of mobilization (Scott, McAdam, Gould, Petersen): members of a political party, school, or religious institution, friends and family, solidary villages, and other preexisting social relations can best communicate, cooperate, and monitor, making it easier to overcome the challenges of mobilization. Risk-acceptance may be higher in these contexts, since norms of reciprocity are more likely to kick in.
This is a particularly powerful dynamic when the networks are imbued with ideological content that provides a political frame justifying rebellion: nationalist church groups opposing Soviet occupation, Sunni veterans’ networks opposing Shia dominance, Islamist political parties opposing control by non-Muslims, leftist student groups targeting landlord dominance (see Wittenberg and Darden on enduring loyalties embedded in networks). The overlap of a clearly-framed political grievance with an embedded social infrastructure can make specific networks the locus of recruitment and leadership, rather than a broader, representative cross-section of the aggrieved population.
However, we also have significant evidence that people join insurgent groups due to selective incentives: benefits of protection, money, social services, redistribution, or local power that can only be acquired by becoming a fighter or supporter (see Weinstein, Popkin, Berman, Collier and Hoeffler). This solves the collective action dilemma by providing goods and services that people could not acquire if they just stayed on the fence. Attracting these new recruits is necessary if an armed group is to expand beyond its initial social milieu. These scholars focus less on political grievance or social structure than on the provision of specific material benefits in return for support and joining; a variety of fairly mundane reasons can pull people into rebellion.
The social ties/community approach and selective material benefits approach have been framed as competitors, but Humphreys and Weinstein find that both of these motivations are important, and they suggest that the two schools are compatible rather than competitive (see also Wood for a different take).
How might this compatibility work? The answer may lie in the sequencing and type of insurgent mobilization. The first-movers in rebellions – the organizers, risk-acceptant fighters, and core command cadres – seem to be drawn from robust networks, whether they represent “the people” or not. These are the people who turn riots and protests into insurgencies and their networks form the enduring social heartlands of insurgency; as Selznick refers to them in the communist context, they are “the steeled cadres upon whom the continuity and the basic power of the party rest.”
By contrast, selective incentives can only be provided once the insurgency has become at least loosely institutionalized: protection needs people who can protect, material benefits require channels of acquisition and distribution, and shifts in local status and power are contingent on the ability, both social and military, to reshape local relations. Only once these are in place are fence-sitters attracted to join the insurgency by selective incentives; selective incentives need to come from somewhere. Thus, all else being equal, different kinds of fighters will join insurgencies at different times and for different reasons. Methodologically, as Humphreys and Weinstein note, systematically understanding these temporal sequences and variations in insurgent role requires deep historical knowledge of waves of recruitment and processes of organization-building.
In terms of policy, this heterogeneity suggests that insurgent movements are likely to be characterized by different social blocs with diverse motivations for fighting. From the insurgent leadership’s perspective, this means managing the varying goals of “core” vs. “peripheral” fighters and supporters. The internal life of the insurgency thus becomes an endless battle to maintain control, balance disparate factions, and keep the selective benefits flowing while trying to satisfy the interests of the networks at the heart of the organization. Different organizations will face very importantly different types of balancing acts, but the specter of disastrous internal feuding hangs over almost all rebel groups – as Monty Python insightfully noted quite awhile ago (warning: language). When movements fall apart, peace may become more difficult to build.
From the perspective of the counterinsurgent, this creates both opportunities and frustrations. It opens the possibility of splitting the insurgency by pulling away social blocs and factions. However, it also creates enormous complexity in the incentives at play within the insurgent movement that make it hard to cleanly induce fragmentation. States tend to have awful information about the actual balance of loyalties and interests within insurgent groups (even after nine years of American war in Afghanistan, the exact relationship between the Haqqani network, Quetta shura, and Pakistani ISI remains a matter of informed guesswork) and so clever games of divide-and-rule are far easier proposed than implemented. Exploiting variation in the social underpinnings of an insurgency requires extremely fine-grained information and policy dexterity that are not generally the calling cards of counterinsurgents.
Despite these challenges, awareness of heterogeneity should encourage COIN policies like amnesty and differentiated targeting that can exploit internal insurgent contradictions. This seems to have been valuably recognized by US forces in Afghanistan, even if it is only a partial solution.
Therefore both insurgent leaders and their foes are constantly forced to deal with the consequences of different insurgent motivations. My next post will take up two directly related questions: first, why insurgents jump ship to the side of the state, and, second, the causes of collaboration with counterinsurgents.