The Duck of Minerva

The Duck Quacks at Twilight

The Irrelevance of IR Theory?

November 21, 2005

Over at Abu Aardvark, Marc Lynch provides some empirical evidence for the fact that major IR journals have had very little to say about al-Qaeda over the past few years:

All told, these seven journals published 796 articles between 2002-2005. I found a total of 25 articles dealing even loosely with al-Qaeda, Islamism, or terrorism. That’s just over 3% of the articles. Now, there’s lots of important stuff out there in the world, and there’s no reason for the whole field to be following the headlines, but still… 3%?

Marc doesn’t provide much speculation in his post on why this absence exists (perhaps because he’s presently working on an article on the subject), although he notes that when political scientists have something to say about al-Qaeda or Islamism they tend to say it in policy-oriented journals rather than in “the most prestigious, theory oriented journals.” Subsequent comments — many collected and responded to here — put forth a variety of speculations, including a time-lag before major world events find their way into top-flight IR journals, the general US-centrism of much IR scholarship, and (perhaps most intriguingly, but I will let Dan speak for himself on this point if he so chooses) the notion that maybe al-Qaeda isn’t all that important a phenomenon.

But there’s another theme running through the discussion: a kind of anti-theory bias. Theory, we learn from some of the commentators, is basically useless (one participant opines that we need to toss out IR theory in order to learn something about the world), and is positively harmful inasmuch as it distracts our attention from the real world.

As a card-carrying member of the IR theorist guild, I feel honor-bound to respond.

Is theory useless? Regardless of whether we find certain ways of posing a question or attacking a problem to be at all insightful, it remains incontrovertibly the case that theory is inescapable. There ain’t no such thing as a theory-free way of looking at the world. This is the case for two reasons. First, there’s too damn much in the world to look at it all at once, so we inevitably select and prioritize and classify, and all of these are theoretical operations. Second, ever since late 19th century philology, and arguably ever since Immanuel Kant’s meticulous reconstruction of the implicit logic of Newtonian physics [in case you were wondering, which you weren’t, that’s a lot of what Kant’s up to in the Critique of Pure Reason…but I digress], we have realized that thought itself is irreducibly conceptual, and that the meanings of words derive not from the innate character of objects but from implicit categories that don’t simply reflect “the way things are.” So thinking is theoretical through and through. Like it, hate it, it doesn’t matter: theory is here to stay.

Does focusing on theory distract us from real-world events? Perhaps, but only in the sense that trying to engage in conceptual analysis or the clarification of tacit assumptions means that we aren’t spending our time simply applying those concepts or assumptions in an effort to explain empirical phenomena. Asking “What is terrorism?” isn’t itself conducting an analysis of al-Qaeda, but arguably having a clear concept of terrorism in hand might improve our understanding when we do get around to applying it. And it is certainly the case that scholars can become so enraptured and entranced by conceptual vocabulary that they forget Weber’s strict admonition that theories, concepts, ideal-types, and so forth are nothing but means to an end — and that end is concrete empirical analysis. So yes, such distraction can certainly happen. Is it inevitable? No.

The dearth of analyses of al-Qaeda that Marc identifies strikes me as less of a comment on IR theory per se, and more of a comment on IR scholars and the way that we have (perhaps implicitly) chosen to organize the production of knowledge in the discipline. There’s nothing intrinsically built in to realism or liberal institutionalism or rational-choice theory or constructivism or whatever that would prevent them from being applied to al-Qaeda. Even the state-centrism characteristic of realism and neoliberal institutionalism isn’t an insuperable barrier; realists and neoliberals would just want to look for states supporting al-Qaeda, and try to determine whether the strategy of terrorist attacks was a rational decision in an anarchic world. I’ll go further: any IR theory can be applied to any empirical case or phenomenon, without exception. Whether they generate any useful insight is a wholly separate matter, but the fact remains that they can be so applied. [For a good example of this, check out the marvelous new book Making Sense of International Relations Theory, edited by Jennifer Sterling-Folker, in which ten different theoretical approaches are applied to Kosovo. Full disclosure: I have a chapter in that book. Fuller disclosure: so does Marc.]

So if any theory can be applied to any case, what explains the dearth of analyses of al-Queda? To understand that, I think that we need to get away from “IR theory” and look at “IR theoretical practice,” by which I mean the concrete ways that knowledge is generated in the discipline of IR. “Theory-oriented” IR journals, which are clearly the most prestigious according to whatever ranking-system one uses, tend to publish refinements and minor modifications of existing bodies of knowledge. This is so for two reasons: first, an article contributing to an established body of literature can more easily find sympathetic reviewers, since by definition there are reviewers working in that same general area — the people who published the other articles in that literature; and second, such an article can take a lot for granted, and presume a basic agreement on certain conceptual and methodological questions, since such things are generally shared throughout a literature. This basic agreement means that articles contributing to an established literature can be shorter — and given that basically all of the major journals have been reducing their maximum word-length over the past few years, the net effect is to promote incrementalism in scholarship. Add to this other pressures contributing to disciplinary specialization and sub-specialization, and the outcome — relatively static and conservative (not politically conservative, obviously) bodies of knowledge — seems quite overdetermined.

So I ask the following question: what was the literature on terrorism like before 11 September 2001? My sense is that it was basically non-existent. I never remember seeing any discussions of terrorism published in places like International Organization or International Studies Quarterly or the Review of International Studies in the 1980s and 1990s. This means that after 11 September 2001, when all of a sudden lots of American IR scholars suddenly noticed terrorism and Islamism for the first time, there was no existing literature to contribute to. If you wanted to say something about terrorism, but the major literatures in the field were about central banks and trade policy and nuclear deterrence, you had two challenges: figure out a way to make points about terrorism to a scholarly audience unfamiliar with the basic contours of the phenomenon, and figure out how to link your contribution to some existing set of conceptual concerns.

Quite a challenge. I’d wager that unless you were already part of a scholarly community that had ongoing debates about the issue (for example, Middle East Studies scholars had been talking about currents within Islam for a long time, so there was a literature there that one could connect Islamist terrorism up with pretty easily), even if you wanted to say something, you’d have a hard time putting together a scholarly article on the issue — at least not one that would pass through the screening functions of the peer review and editorial processes. That leaves two avenues: policy journals, which do not ask authors to contribute to a literature when they write (and which in fact positively discourage authors from talking overmuch about “the literature,” which is a large part of why I have never been published in one of those journals and am not likely to be published there any time soon), and books — not because books have no peer-review or editorial processes, but because a book doesn’t have to be as incremental as an article. You can stretch out in a book, take sufficient space to make and justify your claims, and still have enough room left over to actually carry out a meaningful sustained analysis of a complex phenomenon. Hard to do that in a 10,000 word journal article, and even harder to do in the absence of a scholarly literature and its associated scholarly community.

So what’s the problem with IR theory? Nothing. What’s the problem with IR scholarship? Like every scholarly discipline, it’s dominated by a few literatures that combine theoretical, conceptual, and empirical considerations in specific ways, and that arrangement is sustained in part by peer-review and editorial processes, in part by prestige hierarchies among journals (and the ways that this hierarchy figures into tenure decisions), and in part by the shortening length of journal articles. Given these factors (and this is only a partial list; we could factor in graduate student training, the availability of research funds, the extreme disincentives that have to be overcome if an established scholar wants to re-tool and learn a new language or region or approach…etc.…) I don’t think it’s at all unexpected that the discipline is slow to change its central orientations. No serious scholarship on al-Qaeda? Not surprising to me.

Now, the question is: can and should this be changed? I think that it could be, if we either proliferated journals, relaxed editorial standards [I don’t mean “publish crappy work”; I mean “don’t require all three peer-reviewers to recommend publication, or don’t require two of the three to do so enthusiastically”], or expanded word lengths. But there’s a limit to how much of this is desirable, I think. Knowledge, as Karl Popper once put it, is like a cathedral: it’s a construction by humans, on which individuals work (and in which individuals dwell), but which is intersubjective in character: it outlasts every individual who was complicit in its production. Without some sense of scholarly responsibility to engage in a debate or dialogue with extant work, the whole construction can come crashing down, as the discipline fragments and knowledge splinters. After all, as (Saint) Weber might have said, the activity of producing knowledge is logically distinct from the activity of producing policy, and trying to fuse the two points raises the specter of unmitigated disaster.

Then again, I’m a pessimist about human knowledge — I think that we only know about things after they’re relatively complete, and only if we have access to the relevant records and data. That knowledge might be used as a basis for policymaking, but what can I say? That’s not my department.
Filed as:,, and

website | + posts

Patrick Thaddeus Jackson is Professor of International Studies in the School of International Service, and also Director of the AU Honors program. He was formerly Editor-in-Chief of the Journal of International Relations and Development, and is currently Series Editor of the University of Michigan Press' book series Configurations: Critical Studies of World Politics.