Continuing to reflect on the death of Abdelmalek Droukdel, and on this recent thread from Yvan Guichaoua, led me to revisit the debate over whether and how much analysts should privilege “global jihadism” versus “local dynamics”* in analyzing jihadist movements. I’ve been a pretty strong proponent of the latter approach, although when the evidence warrants it I turn to global and transnational dynamics.
But actually, I think much of what’s at stake in such debates is not necessarily geographical scope but a more fundamental question: Is jihadism cohesive or contingent? Do jihadists proceed according to strategic blueprints, or do they react to events with just as much acuity and sloppiness as any other type of actor? Are jihadist organizations tightly organized and controlled, from the (supposed) masterminds at the top to the (supposed) dupes at the bottom, or is there infighting, improvisation, and indecision up and down the chain? Does jihadism look essentially the same in different parts of the world, or is its evolution deeply situational and contextual?
I incline heavily toward the “contingent” view – or maybe it would be better to just call it the “messy” view. This is also my view of how history and politics work in general. I might be more favorable toward the “globalist” school if it acknowledged more of that messiness, contingency, indiscipline, improvisation, frustration, negotiation, etc. – both within jihadists’ ranks, and between jihadists and the societies around them. Yet globalism in jihadism studies is often completely bound up with the idea of jihadism as a highly cohesive project, the idea that jihadist hierarchies function in a top-down, disciplined, and cohesive way.
The implication from analysts is often that a top leader’s statements can be taken as a blueprint for how the emirs heading affiliates, the field commanders, and the rank and file will all behave. In the cohesive view, at its most exaggerated, jihadists are depicted as the only actors with real agency – they may fight with each other, but otherwise they are the only actors who move in a color across a landscape where all other actors are merely bystanders, depicted in black-and-white, helpless to do much more than react to jihadists’ machinations. And even when jihadists fight with each other, the cohesive/globalist school tends to make top leaders the only figures who matter: Baghdadi versus Jolani, Baghdadi versus Zawahiri, Shekau versus Barnawi and Nur, Droukdel versus Belmokhtar, etc. It is not an accident that the privileged sources for the cohesive school of analysis are jihadist propaganda materials, leaked correspondence, etc. – it is easy and tempting to conclude the jihadists’ documents map directly onto reality. Yet as Mike Tyson famously said, “Everybody has a plan until they get punched in the mouth.” Jihadists react just as much as they act.
I would say that I’m not necessarily a “localist” but rather someone who sees contingency operating at every level – I don’t say that the local is cohesive while the global is contingent, but rather that the jihadist project is shot through with contingency and improvisation – as are, again, the projects of every other actor. In fact, the cohesive view would be equally problematic even if it were applied in a strictly “localist” framework. And the “contingent” view can be really compelling when applied in a global context – I think this is one reason I really like Barak Mendelsohn’s The Al Qaeda Franchise, because Mendelsohn foregrounds contingency and messiness in his global comparative analysis.
Many other authors’ global comparative analysis of jihadist groups, however, shows a tendency to flatten contextual differences (Mali, Yemen, and Afghanistan are all “failed states,” right? how different can they be?). That tendency is even more pronounced when analysts have a penchant for promoting the cohesive view of al-Qaida and the Islamic State at the global level. Then analysts just apply the same frameworks regardless of context. This is how you get the kind of paint-by-the-numbers analysis one often reads of jihadist groups, especially emerging jihadist groups in understudied countries, where the analysis is essentially interchangeable with analysis of a jihadist group in some other country. For example, I personally don’t know anything about Mozambique, but I bet that with 2-3 hours, a story or two from al-Naba’, and a few reports from the wire services, I could write you an analysis that would look a lot like some you may have read.
Yet if jihadist materials are merely one source among many that are necessary to understand a given conflict, then the task of analysis becomes much harder – and triangulating among multiple sources necessitates at least an implicit admission that, as Guichaoua says in the thread cited above, there is a Rashomon-like quality to the competing narratives, with jihadists’ accounts not necessarily more authoritative than any other actor’s.
The localist or contingent school of analysis tends to come out of area studies. I think area studies provides a kind of training that often inculcates a sense that events are contingent, that actors are flummoxed by events just as often as they master them, etc. A lot of area studies scholars are political scientists, but often not the kind of heavily quantitative political scientist who views the world in a highly schematic way (a way often untenable in its own right). The weakness of area studies, however, is frequently a lack of engagement with jihadist materials, sometimes because area studies scholars do not speak/read Arabic and sometimes because – even when they do – they either lack the interest in or the familiarity with jihadist ideologies and idioms that is necessary to follow what jihadist sources are saying. Sometimes that means that “localists” are too dismissive of the jihadist content, and this then opens the door for the propaganda-focused cohesive/globalist school to attract wide audiences, particularly among policymakers, by promising unique insights supposed gleaned from jihadist materials.
There is a third pole of the debate, meanwhile: structuralist analyses. The perennial discussion about identifying the “root causes” of conflict can be productive or tedious, depending on how sophisticated the effort is. The localist and globalist analysts of jihadism are often in rare agreement, though, in rejecting the over-determined, simplistic, and monocausal structuralist accounts that regularly surface in an attempt to explain jihadism. “Boko Haram is an outgrowth of overpopulation.” “Climate change explains Sahelian jihadism.” etc. With that said, though, the more sophisticated structuralist accounts cannot be dismissed, and often form part of the kind of localist/contingent analysis that I think works best. An analysis of Boko Haram that didn’t take account of population growth, poverty, corruption, urbanization, etc. would not be workable. And an analysis of jihadism in central Mali that didn’t take account of population growth, resource competition, the interaction of legal frameworks and land use, etc. would not be workable either. But thinking about those structural factors reinforces my sense that much of any particular jihadist group’s evolution is contingent, both on macro factors and micro factors.
I’m not saying that it’s all whim and accident. I do think jihadists sometimes make plans and carry them out effectively. But they often don’t – I return to the image (and who knows if this is what really happened) of Muhammad Yusuf “hiding in a goat pen” when he was captured by Nigerian authorities amid Boko Haram’s July 2009 uprising. Some master plan. And I don’t think that everything turns on the individual, but individual decisions and outcomes matter, especially when you start to add them up, and then you can think about all the different ways things might have gone. What if Chérif Gousmi had not been killed in 1994? What if Hassan Hattab had remained leader of the GSPC? What if Mokhtar Belmokhtar had surrendered to Algerian authorities in the mid-2000s? What if Muhammad Yusuf had survived the 2009 uprising? What if Iyad ag Ghali had been accepted as leader of the MNLA in 2011? What if Government Girls Secondary School in Chibok had not tried to hold exams in 2014? What if Belmokhtar had pledged allegiance to the Islamic State in 2015, instead of rejoining AQIM? Etc. etc. And that’s just at the level of the jihadist leadership. We could pose similar questions regarding politicians’ decisions, military officers’ decisions, etc. And think of all the small-group dynamics that are mostly out of sight, all the consequential micro-decisions that could go one or the other, that add up in their aggregate to the picture we see (or think we see). To think that all this can be explained through jihadist source materials is, for me, too much.
*Problematically, including in some of my own writing, what is meant by “local” is sometimes not properly fleshed out. Is it the national level? The level of some bounded sub-national unit? The level of a city? A neighborhood? I incline more and more toward the sub-national level, although I think one has to move between the regional, the national, and the sub-national (and yes, when necessary, the global) to try to capture the full picture.