GUEST POST: Jihadi Ideology Is Not As Important As We Think

NOTE: As with all guest posts, the opinions expressed below are those of the guest author and they do not necessarily represent the views of this blogs administrator. aims to not only provide primary sources for researchers and occasional analysis of them, but also to allow other young and upcoming students as well as established academics or policy wonks to contribute original analysis on issues related to Global Jihadism. If you would like to contribute a piece, please email your idea/post to azelin [at] jihadology [dot] net. Pieces should be no longer than 2,000 words please.

See here for previous posts in this exchange:

By Joshua Foust

On January 21, Daveed Gartenstein-Ross wrote a detailed argument for why the study of “jihadi ideology” is important. He raised some really interesting points that inspired a great deal of thinking on my part, and I’m grateful he took the time to respond. I have to confess, however, that I am left more confused than ever about why we should focus so much on ideology, and not other things instead. It remains unclear to me how ideology affects behavior, or how its study can help us understand or predict the decisions and choices of would-be terrorists.

For example, Daveed is absolutely correct that I erred in describing ideology as a sole cause of behavior. I admit I was a bit confused when he later called ideology, “a robust explanation for both terrorist radicalization and also terrorist actions,” as that seemed to imply a sole-cause argument, but I did not intend to argue the point so I withdraw it.

However, I found myself lost in trying to figure out the other aspects of studying the concept of ideology as an analytic construct. For example, even if ideology is only a partial cause or inspiration for behavior, we should still be able to describe how ideology causes or inspires behavior. But I have never, in the many papers I’ve since read on Jihadi ideology, read of a mechanism by which ideology inspires or causes behavior—only a lot of stories, and plenty of people who did something that they claimed was inspired by ideology.  This begs the question: if we don’t know how ideology influences behavior, then how can I analyze it or declare it important for studying jihadism? And while Daveed’s post has a great deal of information, I don’t see in it, or in the study he wrote, a description of this causal mechanism.

I think, too, that any discussion of ideology—especially when dealing with a celebrity like Anwar al-Aulaqi, as my original post tried to do—must deal with the question of false positives and false negatives. If ideology is even a partial cause of behavior, then there should be a lot of people who espouse a certain ideology that engage in a certain behavior, along with relatively few people who espouse the ideology who DO NOT engage in the behavior, and also relatively few people who DO NOT espouse the ideology who DO engage in the behavior. Otherwise we’re seeing our cause without our effect and our effect without our cause as often as we’re seeing our cause with our effect, which would suggest we’re on the wrong track.  We do see a lot of false positives and false negatives in Islamist-ideology explanations of extremism. The vast, vast majority of people who espouse “salafist jihadist Islamist ideology” do not engage in terrorism. So how can we assign ideology as the reason why the tiny minority of people do things?

Which is another problem with this discussion of ideology and jihadism: begging the question. Given the fact that many people espouse radical ideologies but do not become violent extremists, it would seem that at most ideology is a necessary, but not sufficient to cause the behavior in question. If that’s the case, then what makes it sufficient? Does that thing lead to the behavior even if the ideology is absent? If it does, then there’s no reason to take ideology as an influence on behavior. It’s like saying “X causes Y only when Z is present” and “Z causes Y even if X is absent.” If both those statements are true, then you don’t need X—ideology—in either. You just need Z to predict Y.

I’m not really comfortable discussing Daveed’s personal experiences with radical Islam. His current research, however, does warrant discussion. For example, the percentages on radicalization that he provides at best fail to support his argument. At worst, they actually undermine it. About 40% of those studied became religious and then became political. About 10% became political and then became religious. And there is apparently no data on the other 50%. Since this wasn’t a random sample, we can’t assume that the 40% and 10% are representative of the larger population. That means it is entirely possible that it’s somewhere between a 60-40 and a 40-60 split, in which case the presence of religious influences would basically amount to a coin toss.  The figures he provides make it entirely plausible that the presence of religion is a random factor, and therefore not so causal.

Similarly, I don’t understand why he relied on behavior to explain ideology. Daveed’s study took people who already exhibit a behavior, and then measures the extent to which they exhibit other behaviors he assumes are evidence of a specific ideology. There is no control group.  This gets at the heart of the problem with false positives and false negatives: by not including people in the study who did not engage in the behavior, there is no way to assess actual behaviors. It is a significant sampling bias. Moreover, I simply don’t see how any of his examples are necessarily indicative of ideology. He argues, “Absent the prevalent ideology… there is simply no explanation for why a relatively large number of people would decide to grow their beards out in a similar way, see dogs as unclean, stop making physical contact with members of the opposite sex, et cetera.” But that doesn’t follow at all. If it were the case, then all behavior is ideological. Most people tend to behave similarly to the people they interact with – they eat similar foods, they wear similar clothes, they have similar hair styles, many people believe in the same things as their neighbors – is everything the any number of people hold in common to be automatically declared ideological? Common practices just aren’t a persuasive argument that ideology explains behavior.

I’m glad to see he didn’t rely on people’s stated beliefs—an unfortunate mistake many other researchers on ideology and jihadism commit. However, that still doesn’t avoid the false positives and negatives, begging the question, the lack of mechanisms, or the possibility that everything he found is the same as what we would expect from chance. I’m afraid I just can’t find a reason to believe it, especially when he leaps from saying ideology is important to saying who espouses the ideology is important. Even if we accepted ideology as an explanation, there’s no reason to assume ideology is tightly tied to specific people. If ideology matters, then why is it not just as reasonable to ask what the major influential components of it are, no matter who disseminates it? Why is it not just as reasonable to ask in what situations it matters, no matter what its components are and no matter who disseminates it? Or, why not be really nit-picky, and ask IF there are any disseminators, components, or situations that matter more than any others – allowing for the possibility that perhaps all options matter equally?

Viewed this way, the case for Aulaqi’s importance just doesn’t make sense. I have yet to see an explanation for why alternate hypotheses don’t apply. For example: which proponents of Islamist rhetoric were more available and accessible to western radicals than al-Aulaqi? Forensics on confiscated computers often turn up a whole host of jihadi ideologues—an ecosystem of personalities arguing for jihad. Aulaqi’s material is common, but it’s not the only thing out there. But even that is immaterial: if his speeches, by design in the English language, are just more available and accessible, then their use is more plausibly an effect or a co-occurrence, not a cause, of those people’s radicalization, just as his presence on computers similarly fails to explain lots of other people’s non-radicalization.

In a more general sense, the rejection of alternative explanations should be worrying to anyone seeking a systematic, rigorous explanation of how people radicalize and why they choose to engage in behaviors. This is the problem I have with calling Aulaqi the worst-anything, or most-anything. We just don’t know, especially because we just don’t know how the ideas he spreads affect people. All we have are correlates—not correlations—just things happening kind of at the same time for some people. While that is certainly important, and definitely deserving of detailed, rigorous research, I remain utterly confused at the certainty with which we can declare jihadi ideologues the global threat they’re portrayed to be.