This is none too pithy, but tonight (this morning?) I’m just going to toss up a whole stream of ideas that have been percolating since the first BioNote discussion and took shape while I was attending the ISA conference in San Francisco. They relate to a very problematic final chapter I’m struggling over now in which I attempt to account for the role I’ve played (by researching a non-barking dog) in inadvertently getting the dog to bark.

Many of us do this of course – interface with the real world in ways that put us inside the subject matter we’re studying – but there are not good formulas in IR scholarship for reflecting about this explicitly (much less analyzing one’s own actions as part of our dataset) as we write up our results.

(This ties into the earlier “Bionote” discussion in two ways. First, bionote norms, I’ve argued, are simply an example of how IR encourages people to think of themselves as observers rather than participants in the worlds we are studying. Second, the autobiographical nature of the ISA panel about which I blogged was due to, not a deviation from, this set of norms. The panel was organized as an autobiographical set of war stories from policy work by academics precisely because as Janice Gross Stein pointed out, autobiography is all we have to go on because we don’t bother to do empirical studies of how we “bridge that divide” and with what consequences.)

But why don’t we? I think it is due to the very trend I think I was trying to articulate before – IR scholars have some stake in keeping up the pretense that we exist separate from international affairs. We’re encouraged through various disciplinary norms to occlude a serious empirical analysis of the way our role in conducting research, especially if we do elite interviews, to say nothing of blogging, writing op-eds, or consulting, affects the processes we’re studying.

To be sure this is probably truer of some projects than others. I’ve concluded it’s quite true of my work on “dogs that don’t bark” in transnational advocacy networks (namely the absence of children born of war rape on the human rights agenda) and I’m trying with some difficulty to account for this as I complete my current manuscript.

Consider this anecdote from the concluding chapter:

“In Spring of 2006, I presented my preliminary findings regarding the non-emergence of “children born of war” at University of Pittsburgh’s Research in International Politics (RIP) monthly brown-bag. In such circles, heavily dominated by empirical approaches, one does not present normative theory – that is, value-laden arguments about how the world should look, or policy-oriented sets of recommendations about particular problems. Rather, one identifies puzzles about the world and then goes about solving them by applying or modifying existing theories. Theories in this sense being lenses said to explain and predict major patterns in world affairs.

Therefore, I had organized this particular paper not as a problem-focused human rights argument about children, but rather as an empirical study on “issue non-emergence” within advocacy networks. I presented the subject of “children born of war” as a negative case and demonstrated why, from the perspective of agenda-setting theory this might be considered an interesting puzzle. The case, I argued, showed that we needed a different understanding of the obstacles to issue emergence.

My colleagues provided a variety of suggestions on the theory, the methods and the structure of the argument. But one piece of advice particularly sticks out in my mind. “You’d better stop talking to international organizations about this issue until you publish,” said one of my senior colleagues. “Otherwise, before you know it you will no longer have a puzzle to explain, because these children will be on the agenda.”

Note two things about this comment. First: the idea that more attention to this population should have been less preferable to me (or anyone) than the ability to advance my career by publishing an interesting paper. Second: the acknowledgement by my colleague that in researching the non-emergence of “children born of war,” I was in fact engaged in “issue entrepreneurship” myself that could alter the research findings.

If the previous chapters illustrate anything, it is that the process of researching human rights is in fact intimately connected with the practice of constructing human rights in and around a variety of policy arenas. In other words, far from existing outside their subject matter, human rights intellectuals are part of the human rights movement and actively (if inadvertently) shape it. But my colleagues did not advise me to explicitly account for this factor in my research or discuss the academy as a source of momentum or resistance to new human rights issues. To do so would have been to breach certain professional norms within epistemic communities of political scientists, norms that suggest that “real” research is distinctive from advocacy.

As explained briefly in chapter one, I have sought to exploit the recursive relationship between academics and practitioners methodologically. The research process for me has consisted largely of poking around the human rights regime asking questions about what is not on the agenda and asking practitioners to justify their answers. This has provided insight into the regime itself, as well as its silences and the cultures within which different practitioners move.

However such a method does constitutes a notable, if modest, agenda-setting function in its own right. Simply raising a new issue in a conversational setting ‘makes people think’ and stirs up dialogue. Such communicative action can lead to organizational innovation. It also introduces the individual researcher to the network of gatekeepers who can stop an issue from emerging, as well at to those ‘true believers’ who might push for it. The practitioners may come to see the researcher, through these discussions, as an expert on the substantive topic. The researcher might be invited by true believers to consult or share findings with the practitioner community.

The choice of whether and how to exercise this role has implications both for the research and for the organizations under study, and eventually for the population of concern. It is therefore impossible and irresponsible to pretend that the research process itself has not influenced the very communities of practice we study. Acknowledging this has required me to reflect on my own role in the human rights network, and that of like-minded colleagues and of academia as a whole, as part of the subject matter of the book.”

But how to do so systematically? In developing a literature review to ground the remarks in this chapter, I re-read with some interest Patrick Jackson and Stuart Kaufman’s Perspectives on Politics piece “Security Scholars for a Sensitive Foreign Policy.” I liked this article because it represented an example of IR scholars self-consciously engaging with the real world, and theorizing about the process of doing so. To wit, the authors document their role in SSSFP, an advocacy effort led by IR scholars to illuminate cause and effect relationships regarding the Iraq war while remaining politically agnostic, thus maintaining scholarly credibility. According to Jackson and Kaufman: “Weberian activism… is an appropriate stance for scholars who wish to engage in debate on public issues.”

However I found their application of “Weberian Activism” too limiting to inform the problem I’m facing, since the authors’ careful incursion into political activism comes at a much later stage of the policy process. In other words, their model cannot inform the work of scholars studying cause and effect relationships in agenda-setting – that is, how “public issues” get constructed in the first place, and particularly why some get framed off the public agenda altogether. Any scholar who “pokes around” a policy domain trying to analyze this part of the policy process, even if s/he limits herself to standard methodologies and avoids “open advocacy” like the plague, will influence the thoughts of policy-makers and possibly affect the very outcomes s/he is studying despite his/her best efforts.

And yet the argument Jackson and Kaufman make reifies the very divide between academic and policymakers about which we might be so usefully explicit. The entire article is an exercise in finding some way to reconcile the authors’ different “hats” as researchers of world politics and participants in world politics. They come down on the side of privileging their scholarly identities (maybe because they were writing this article for a scholarly journal?), even though doing so meant essentially abrogating the possibility of being effective in their activist exercise. At any rate, such a solution would be impossible if they were seeking to “encourage broad acknowledgement of facts and problems” in an area where no policy debate already existed, because in such case the research itself would play an agenda-setting role that could not be summarily negated through the logic of Weberian activism.

Probably, what we need in such cases is to acknowledge what might be referred to, following Stephen Krasner as “Kingdonian Activism.” OK, I coined this term, not Krasner, but he inspired me this week at ISA. On the “Bridging the Theory/Policy Divide” Panel at ISA about which I blogged earlier, Stephen Krasner argued that bridging the gap is the wrong framework. Instead we should be using Kingdon’s garbage can model to think about how academia interfaces with politics. Krasner suggests academics are just one group among many contributing to what Kingdon describes as the policy “soup.”

The question of our moral responsibility as persons with civic identities to deliberately engage as activists in such cases is an important issues, but it is not the subject of this post. What I’m interested in here is our responsibility as scholars to account for our role, however inadvertent, in influencing the policy processes we are studying.

Here is a more recent example of what I mean. The Campaign for Innocent Victims in Conflict (CIVIC), about whose work I blogged approvingly not so long ago, constitutes an excellent case study for my ongoing work on the dialectic between issue entrepreneurs and establishment advocacy organizations. Since I’m interested in why new ideas do or don’t get picked up by lead actors in advocacy networks, I’m curious to follow this campaign, trace CIVIC’s strategies over time and watch to see what works for them, what doesn’t and why.

However, insofar as policy elites or those within a degree of separation from them do read this blog, my blog post itself arguably has become part of the socio-political dataset on the construction of CIVIC’s platform. How / to what extent / must I account for this in my research? Do I analyze the post and comments to it as part of the total dataset? If so, should I interview myself or at least reflect on / be transparent about my motivations in posting, as I would interview others responsible for content I’m analyzing? Should I code myself as a issue entrepreneur or issue advocate for discussing this campaign, on blogs, conference panels or in the classroom? Or should I avoid any mention of such issues I’m studying in my personal or professional conduct until my research is complete?

It doesn’t matter, because even if I had never blogged about CIVIC, the very process of conducting interviews with people about issue construction plays a role in constructing issues. When I interviewed the CIVIC Executive Director, the conversation itself allowed her to think through the organization’s strategy in ways that may have changed her thinking. And interviews with gatekeepers about issues they’re ignoring force them to justify non-policies, which helps me understand their thought process but also places me momentarily in an issue-entrepreneur role. (I’ve noticed this particularly in my children born of war research – when I talk about my work on BBC, for example, I’m usually contributing to awareness raising about an understudied human rights problem, so my comments would arguably become part of the documentary record I’m then supposed to be tracing.)

Dan Drezner had an answer to this question in his comments last November at the “Who Are the Global Governors?” Workshop at George Washington University. His take on it is that researchers “just aren’t that important” in the global policy cycle. If true, then attempting to include our own impacts in the analysis comes off as vain and narcissistic (as many comments to an earlier post suggest).

But a lot of famous global norm entrepreneurs have been academics, and the whole epistemic communities literature documents the way in which scientists influence the policy process. Bridging the theory/policy divide may not be easy or prevalent in our discipline, but it is done, and in the absence of serious empirical studies of how it’s being done and with what effect/side-effect, it’s hard to know how to do it properly and where the methodological tradeoffs lie.

Perhaps those of us in this position should be trained to recognize it and to adopt participant-observation methods explicitly and transparently. This is standard practice in sociology, but rare in IR. For example, Peter Haas’ work on epistemic communities draws on conversations he has as a participant in global policy processes, but he does not spend time in his scholarly outputs discussing how his presence inside his subject matter influences his findings and impacts the politics he’s studying.

The problem is that people like myself (and, I think Peter) want to be reasonably positivist in our work – in other words, we’d like to observe phenomena and analyze them in a valid, objective, replicable way. But we also want to observe phenomena that we can only be observed from the inside out. Our observations of the phenomena itself may be reasonably empirical; but our observations of our own interactions will be necessarily interpretive.

I can see several possible solutions within our discipline to this seeming need to either degenerate into interpretivism or to deny our role in our subject matter in order to perpetuate an illusory dichotomy between positivist and reflectivist approaches.

1)Recognize and legitimate the complexities outlined above and incorporate ethnographic methods into standard IR methodology training. Some standard criteria would need to be developed for judging the circumstances under which a scholar should know they need to do this in order to retain credibility, since not all projects call for it.

2)Delegate the role of analyzing one’s own interface with one’s subject matter to a third party. In the case of my current manuscript then, I would write up the analysis of the human rights regime as I observed it as an academic, but I would turn over my field notes, published work on the topic, correspondence with political players, briefing notes and slides, consulting records, any evidence of an impact by me on the policy domain I’m studying, to one or more people – coders if you will – whose assignment would be to evaluate how much I had shaped the politics of an issue simply by researching it in such a way that the findings would be independent of my own subjectivity and, ideally, reliable across observers. Often the impact found would no doubt turn out to be minimal; but where significant the author could then refer to some evidence other than their own judgment.

Am I making any sense here, or are these merely the rantings of a seriously jet-lagged assistant professor nervous about including a “radical” book chapter in a manuscript to a university press, late at night, after a day of transitioning back from ISA with needy children, and too many glasses of red wine?

Reactions if you please.