19 February 2018

How do people rebel? Mechanisms of insurgent alliance formation

Why do people rebel against the state is a common question; how they rebel, much less so. In his PhD thesis, “No Good Deed Goes Undisputed: Property Relations and Insurgency in the Periphery”, Daniel Norfolk uses nine rural cases to provides a model of insurgent alliances, which are crucial to the survival of most rebel groups. Interview.

How did you choose your research topic?

There is something close to a consensus, in the literature, surrounding two dimensions of insurgency – defined as “a struggle between a non-ruling group and the ruling authorities in which the non-ruling group consciously uses political resources and violence to destroy, reformulate, or sustain the basis of legitimacy of one or more aspects of politics” (1) – and these two points of consensus are: first, collective action has its roots in pre-existing social networks; and, second, alliance formation is crucial to any insurgency, because it is rare that a single social collective has the capacity to independently challenge the state or a centralised, ruling authority. So the question that began to guide my research was, “How do pre-existing social relationships shape alliance formation in insurgency?”

But it wasn’t until I was conducting fieldwork in Kashmir that the interface between political territory and personal property struck me as fundamentally important. Since I was already studying networks of insurgency, I was attuned to look for the structuring power of underlying social relations, and I couldn’t help but notice that, in the rural hinterlands where I was conducting research, relations forged around land were very often the most powerful structuring force in a community. And the narratives used to explain dislocation in these property-oriented social structures nearly always conflated personal property with political territory in the context of a struggle between local communities and the state purporting to govern over or conquer them. Just a simple example is the frustration felt by families about the inability to arrange marriages or pass on inheritance due to the state’s encroachment on hereditary property.

I became very conscious that personal property and political territory can never really be mutually exclusive, and that even our contemporary conceptual distinction between the two realms is quite recent. Where states are weak and processes of industrialisation have not been imposed, the conflation of political territory and personal property in land is a very real, lived phenomenon, inscribed into daily life.

Can you describe your thesis?

The central argument of the thesis is about the interaction between centralising authorities and the communities residing in the territory they aim to control. It rests on a set of premises: first, weak states that cannot extend hard power into the rural periphery aim to manipulate or co-opt prevailing political-economic structures there. Second, dominant political-economic institutions in the periphery revolve around land, as property, which becomes the logical target of state manipulation. Finally, local resistance to the territorial encroachment of centralising authorities is conditioned by pre-existing property relations.

James Coleman’s “boat” is a well-known diagrammatic analogy for how causal influence for macro-factors works through disaggregated effects at the micro-level:

Norfolk_Property Relations and Insurgency_diagramme.jpg

State consolidation strategies targeting property regimes in the periphery, which I call the property strategy, do not simply cause or lead in any deterministic way to insurgency. They interact with structures at the micro-level, which disaggregate the response to this stimulus.

Micro-structures in the form of agrarian social networks oriented by property either facilitate or mitigate resistance to state-initiated property reform by inducing or constraining participation in collective action. I call these agrarian social structures property-networks.

So the broad argument that emerges is not that state-initiated property reform in the periphery causes insurgency. This is not a causal argument about why insurgency explodes onto the scene. It is that insurgency in the periphery is shaped by alliances that form as a result of the interaction between property strategies and property-networks.

What are your major findings and how are they policy relevant?

By applying the model in each episode, we can see how the pressures exerted by property strategies from the centre impact property-networks in the periphery and set in motion specific mechanisms of alliance formation. I will summarise three key mechanisms here.

1. Structural equivalence. By reconstructing local property-networks, it is possible to identify “equivalence classes” or collective actors who have similar roles due to their equivalent position in a system of property relations. Structural equivalence was observed as an interaction outcome in episodes of all three regions. It was a good predictor of simultaneous action but not necessarily alliance formation. The empirical evidence collected in this study points to a pattern according to which structurally equivalent property-network classes form what could be called “latent” insurgent alliances that are ripe for collective action, but which require one of the other mechanisms to scale-up into insurgency. In the absence of a secondary mechanism, equivalence classes are more likely to engage in spontaneous and simultaneous protest, riots, or other forms of uncoordinated uprising against the state.

For policymakers looking to prevent insurgency from scaling up, the key is to prevent external actors from developing bridging ties between equivalence classes of property-networks. Bridging between equivalence classes that have been pressured by the state endows brokers with significant influence over formidable latent insurgent alliances.

2. Network merger. Under pressure from a top-down property strategy, connections between previously disconnected networks in the periphery can increase. Land reforms often create new connections between individuals and groups. Geographically separate groups may be forced to migrate to the same location, for instance. This process can be sudden or gradual.

This provides a key insight: when property relations become highly exploitative (by increased taxation, for example), structurally constrained segments of the networks are likely to diversify modes of production, thereby establishing new connections with external networks. Connections between structurally disadvantaged property-networks and external groups often form the basis of alliance formation when equivalence classes do not exist.

From a practical perspective, networks that are isolated by a property strategy (meaning that they have no structurally equivalent neighbours and they are not allied with the state) tend to nurture connections and possibly form coalitions with out-groups, to whom they are not tied by property relations.

3. Brokerage. A broker spans a “structural hole” or a gap between two otherwise disconnected networks. The cases in this study have demonstrated that most sedentary rural property-networks have few bridging ties that span significant distances. Property binds individuals to a specific geographical area, which makes brokerage a crucial mechanism for alliance formation in the periphery.

An important finding of this study is that local property-networks often seek out supralocal brokers to push their local property claims and grievances to the centre. This dynamic was observed in at least six of the nine cases. In each case, local groups beset by state-led property strategies sought connections with national or transnational actors to broker alliances that would enable collective action.

In this regard, analysts and policymakers cannot afford to underestimate small, locally oriented groups in the periphery, particularly when it comes to grievances over property. Disorganised rebels are potentially more threatening than their organised counterparts due to the structural predisposition that increases the likelihood for brokerage and escalation. Policies of agrarian reform that overlook the potential negative ramifications for ostensibly insignificant peripheral communities may face resistance of disproportionate magnitude if local property-networks connect with sympathetic or entrepreneurial supralocal actors.

What is your major contribution to the academic field?

The major theoretical and analytical contribution of this study consists in the rendering of property relations as social networks. Property is a social relation that has for too long been incorporated into political science through the prism of class. Class is a useful concept for general theorising, but it is too nebulous to be applied to the micro-foundations of insurgency and civil war. I suspect that the reason why network analysis has not been applied to rural property relations, or has not supplanted class as a more tactile mode of social interpretation (particularly in the context of civil war), has to do with the academic trends of the moment. In political science, the study of rural livelihoods has fallen out of fashion along with the subfield of “peasant studies”, which once attracted eminent scholars. By rearticulating property relations as social networks connecting individuals vis-à-vis land, the researcher gains considerably more analytical purchase on the micro-level structures of rural society.

What are you doing now?

I’ve been managing a conflict mitigation programme in the Upper Nile region of South Sudan for the last two years. The dynamics I mention above are very evident here, in fascinating and rather unique ways. The majority of South Sudanese are agro-pastoralists, so their property relations are qualitatively different from most of the groups I analysed for the thesis. Now that I have more time on my hands, I’d like to apply the model developed in the thesis to the South Sudanese context.

Dr Norfolk defended “No Good Deed Goes Undisputed: Property Relations and Insurgency in the Periphery”, his thesis in International Relations/Political Science, on 15 January 2018. The jury was composed of Professor Anna Leander, president and internal reader, Professor Keith Krause, thesis director, and Professor Benedikt Korf, external reader from the University of Zurich.

Abstract: Scholars of insurgency agree that agency is located in both centre and periphery, in both the private and the political spheres. This study attempts to connect these sites by concentrating on the structuring power of property. Rather than asking why groups rebel against the state, I ask how they rebel. Specifically, I ask how pre-existing social relationships shape alliance formation in cases of rural insurgency. Alliance formation is crucial in any insurgency, because it is rare that a single social collective has the capacity to independently challenge the state. It is even more important in the rural context, where most insurgency foments, due to the logistical obstacles to mobilising a sparse population for collective action. Under these circumstances, the alliances forged to contest state authority have an immediate impact on the ability to do so. Yet, we know very little about the modalities linking distinct actors, structures and objectives. I argue that variation in insurgent alliance formation is heavily influenced by the social structural conduits and constraints formed by local property relations. In order to test this argument, I develop a deductive model that theorises how insurgent alliances are formed by the interaction between supralocal (political sphere) territorial expansion of centralising authorities and local (private sphere) property relations in the periphery. The model is tested in nine empirical cases of insurgency in three geographical regions (the borderlands of India/Pakistan, Afghanistan/Pakistan, and Uganda/Rwanda).

Full citation: Norfolk, Daniel. “No Good Deed Goes Undisputed: Property Relations and Insurgency in the Periphery.” PhD thesis, Graduate Institute of International and Development Studies, 2018.


(1)  Bard E. O’Neill, Insurgency and Terrorism: From Revolution to Apocalypse (Washington, DC: Potomac Books), 15.