Articulating a Research Question

The choice of human rights subject matter is one of the first moments in which human rights intellectuals exercise power over the global human rights agenda. Shall the topic of study be the determinants of torture? The effectiveness of torture? Whether a certain practice constitutes torture? These decisions are shaped by a variety of factors, including the availability of existing research or data on which to build, the availability of funding, the disciplinary norms in which the researcher is situated, the researcher's relative rank in the academic hierarchy (younger scholars are under more pressure to publish within the “mainstream”) and such mundane factors as whether the researcher is relatively mobile. Some of these factors work against one another. Younger scholars are likelier to be itinerant and adventurous, but they also face the greatest professional disincentives to study unorthodox topics.Footnote ¹⁰

A second important question is how to frame a particular topic. In the IR field, a very heavy emphasis rests on empirical scholarship that explains something new about the world.Footnote ¹¹ Students are seldom trained in or encouraged to think about normative theory; there are few positions available in political science for researchers who study or teach applied ethics. Students interested in philosophy or “political theory” might study ethics, but IR remains largely an explanatory discipline.Footnote ¹²

When I first began to dabble in the subject of “war babies” during my first year of doctoral work in political science in 1998, I immediately ran up against this disconnect between the ostensible purpose of my research—to expose the failure of the human rights community to respond to some very vulnerable individuals—and the disciplinary norms about the nature and raison d'être of academic research. My mentor and great friend Ronald Mitchell impressed upon me the importance of beginning not with a “should” question but with a “why” question that could lead to an explanation of some empirical puzzle about the world.

This emphasis on explanatory rather than normative research is a wider phenomenon in the discipline, and therefore no advisor does his or her student a service by allowing him or her to write too far outside of this mainstream.Footnote ¹³ I myself socialize my students in this way. What this means, however, is that human rights “scholarship” published in good IR outlets is often very far removed from the kinds of questions that activists are asking themselves and indeed from the very questions that motivate many young scholars to pursue political science before they are properly socialized, professionalized, and disciplined by our “expert” training.

Consider Beth Simmons' recent book on human rights treaties. Simmons does not question whether human rights treaties are adequate or whether there is something dysfunctional about the human rights regime, since it has contributed only to marginal improvements in human security over the course of the last seventy years.Footnote ¹⁴ Rather, she is interested in exploring what accounts for the observable variation within those margins. This is something that can be studied using previously compiled datasets and regression analysis. Similarly, while Joel Oestreich's book Power and Principle includes a bold chapter explicitly theorizing ethical questions about rights programming in international organizations, it comes at the end and includes a rather lengthy justification of this divergence from the mainstream subject matter.Footnote ¹⁵

I am not saying that purely explanatory work is invalid. Quite the contrary. As Paul Krugman argues in a quote that hung on Ron Mitchell's door during my years at University of Oregon, an important line must be drawn between serious scientific study, in which the scholar's mind can conceivably be changed by the evidence, and sheer advocacy.Footnote ¹⁶ Robert Keohane makes a similar point in his important essay “Political Science as a Vocation”: “In our particular investigations we need to seek objectivity … because otherwise people with other preferences … will have no reason to take our findings seriously.”Footnote ¹⁷ I too drill this into my students' heads at every opportunity.

But what I am saying, and what I take pains to impress upon my students, is that the “cult of explanation” also causes scholars to miss much of importance in the world. It not only inures us to “should” questions but causes us to miss certain “why” questions, of which “why not” questions are an important subset. The methods commonly used for explanatory work tend to lead scholars toward datasets that exist, to events that have happened, and to issues that are politically salient already. These practices direct our attention away from “dogs that don't bark” and raise the barriers to gathering new data.Footnote ¹⁸ They cause us to occupy ourselves with topics we think might be of most interest to policy-makers, rather than taking the marginalized as our frame of reference.

As critical theorists have long claimed, disciplinary norms such as these affect the nature and type of research we do. My early career can easily be used as a case in point. Presenting to my classmates and mentors an early draft of an article on “children born of war” that was later published in Human Rights Quarterly, I quickly recognized on the basis of the feedback that this was not a dissertation-worthy topic. While I did publish the piece at the encouragement of another mentor, it was made clear to me that this article, and the journal in which it was published, would not be of particular help to me in my early career as a political scientist. Since I was interested in a variety of other topics as well, it was easy to abandon this project for the four years it took me to finish doctoral work.

Funders also heavily impact the type of work that is done in the social scientific study of human rights, and how this work is framed. Returning to this topic after I completed my Ph.D. resulted largely from the availability of funding, namely a MacArthur Foundation Research and Writing grant that allowed me to conduct exploratory field research in Bosnia and in human rights hubs. Judging from the reviews from MacArthur as well as later panelists associated with the National Science Foundation, funders are much readier to invest in “sexy” but unexplored topics than are dissertation chairs or scholarly journals.

But funders also have a set of guidelines regarding how money may be used that impacts how research may be framed and carried out, and thus shapes the practice of human rights knowledge construction in the academy. One example is the unwillingness of funders to associate themselves too directly with sensitive subject matter. The National Science Foundation (NSF) is particularly concerned about this, as it is beholden to taxpayers and reliant on Congressional appropriations for its ongoing investment in the natural and social sciences. So although the NSF was pleased to fund my proposal entitled “Mass Rape, Forced Pregnancy and Children's Human Rights” in the summer of 2004, the Program Officer insisted that the title be changed before the project was entered into their database. It would not do, I was told, for the project to attract too much attention from certain constituencies, nor did the NSF have an interest in its projects sounding hyperbolic or “political.” Although by this time mass rape, forced pregnancy and children's human rights were simply descriptive terms of reference already defined in international discourse, to the Program Officer they sounded “unscientific.”Footnote ¹⁹ This anecdote shows how disconnected the language of science can be from real human beings and their struggles. This disconnect has consequences. But that abstraction is also absolutely necessary: it protects the scientific community and the idea of research for its own sake from the kind of politicization that can undermine the scientific enterprise.Footnote ²⁰

Obtaining Human Subjects Clearance

My research on sexual violence and child rights was contingent on interactions with other human beings who had more insight and expertise than myself into these sensitive subjects. Before a researcher affiliated with a major university may undertake any research involving other human beings, s/he must go through a heavily bureaucratic process called a Human Subjects Review. The process is mandated by the federal government under rules established by the Department of Health and Human Services and the Food and Drug Administration to govern medical testing on humans. In US institutions, it is now impossible to conduct research involving any sort of human interaction until the researcher undergoes ethics training and obtains institutional approval. One is barred professionally from using interviews even for a project in which clearance was eventually obtained, if the interviews occurred before the university signed on the dotted line.

The Institutional Review Board (IRB) process, while important, itself is daunting and can sometimes pose an impediment to research. If a researcher fails to dot an i or cross a t the processing of necessary forms can be held up for weeks, and grant money can be lost. Work in foreign countries can be even more complicated, sometimes requiring formal permission from the government involved—a difficult hurdle if the research project is designed to document human rights abuses against vulnerable populations.

Early in my career, I welcomed the human subjects process as an opportunity to educate myself, prepare professionally and receive oversight from specialists in how to manage ethical dilemmas in my work—something else for which there was no required course in my social studies graduate education. Over time, however, I clued in: the process is largely designed not to protect human subjects, but to protect universities from malpractice suits. Because human subjects clearance is based on addressing the types of dilemmas faced by medical practitioners, neither the training nor the oversight is of much relevance to scholars conducting interviews among traumatized or violently divided populations in post-conflict societies. Nor does the IRB concern itself too much with these kinds of questions.

Nothing better illustrates this than my decision to interview a certain minor child in the former Yugoslavia. This was a decision of considerable gravity for me. I had decided early on not to include interviews with children. I knew there would be problems gaining access to an appropriate sample and I was concerned about stirring up contention within their families and communities. I preferred to leave primary data-gathering on children themselves to others who were trained in social work or had the resources to maintain an ongoing presence in the children's lives. So this limited my methodology to those involved in the human rights regime around the children.

I reconsidered this general decision when I realized that this particular boy had explicitly become such an actor when he publicly launched a documentary about his life. It seemed incongruous to write about the social construction of “children born of war” as a political problem, and champion a child rights perspective, while excluding those players in this process who happened to be under 18. This young man had identified himself as a child of war, and signaled a capacity and willingness to tell his own stories. I felt it only ethical to include him in my respondent pool—not as a war child but as an activist for children born of war. Since my original application for human subjects clearance had specified only adults in the respondent pool, I had to submit a “revised protocol” in order to secure permission to interview a minor child.

I contacted my source at the IRB office before submitting the revisions to make certain I had all my t's crossed. But to my surprise, he seemed nonchalant about the revisions. He seemed to interpret his job as not to ensure that I conducted myself ethically, but rather to streamline the process for me in any way possible. He asked, “Is this the only child you'll be interviewing?” Yes, I said. “Will you be testing any hypotheses?” No. “Will you be generalizing from the interview?” Hardly. “In that case,” he said, “You don't need permission. You're not really doing scientific research, so it doesn't really come under our jurisdiction.” One child's well-being as a research subject was too insignificant for the IRB office at University of Pittsburgh or the US federal government to be concerned with.

Never mind the fact that I would very much have liked to have had my hands tied! New to research in violently divided societies, untrained in ethnographic methods, with no background in social work or child psychology, I felt rather unprepared for the work of befriending and interviewing a war-affected family. Of course, I was grateful to have the go-ahead, since I very much wanted to corroborate information I'd already gathered from other sources. But when issues did arise with this family, I had little support from my university in sorting out how to balance my professional identity with what I felt was a growing web of personal friendships.Footnote ²¹

The process of human subjects clearance should, in principle, be the scholarly community's version of what international organizations call “rights-based programming,” which as Joel Oestreich writes, involves “thinking about whether policies are morally defensible and whether they help individual people.”Footnote ²² The institution of human subjects clearance certainly constitutes the former: it is a set of rules, procedures, and ethical standards that aim to protect the dignity and rights of individuals who contribute to research studies. One might imagine, therefore, that the IRB process would invoke the latter as well, that it would be a site for an ongoing consideration of the relationship between human rights and the scholarly enterprise and how the latter could be brought to bear on improving the former. It is thus an irony that the human subjects process itself can pose such barriers to rights-based research on international human rights violations.

The Data-Gathering Process

Framing a research topic and securing permissions affects the nature of science, and many have argued that this process is affected by academic norms, bureaucratic rules and the political economy of funding. But the selection and implementation of a methodology for pursuing the study further delimits the nature of the research enterprise, the types of questions that can be asked, and the findings that will emerge, and to a large extent dictates the type of written output that will result. And as with the question formulation stage, the academy privileges certain methodologies over others.

This has implications. If the subject of research is human rights, the choice of method as well as question affects what can be said about human rights. For example, my decision to never interview children as such shaped what I could and could not say about children born of war; my decision to expand my methodology beyond interviewing to participant-observation affected not only my conclusions but also the ways in which I later interpreted my interview data. Both decisions also affected my caché with human rights practitioners, sometimes negatively and sometimes positively.

In the IR subfield specifically, there are many debates about what constitutes an appropriate methodology for what question, and scholars are under normative, strategic, and practical pressures to carefully consider and justify their methodological choices. Post-positivist scholars have often lamented the weight attached to getting it “right.”Footnote ²³ I experienced an example of what they are talking about when I described this project idea to a senior faculty member at a job interview near the beginning of the research, when I was thinking of the book primarily as normative theory. He asked me: so what is the research question, and my answer at the time—to document how the babies were faring—tallied poorly with the language and raison d'être of IR as a discipline.

Over time the project evolved in a direction that better fit the theoretical concerns of my subfield: it became less about the babies per se and more about what their absence from the human rights agenda meant for the study of international regimes. This was not only a result of various subtle career pressures, but also the path-dependency of Western academia's emphasis on specialization. My original project would have required a set of methodological tools in which I was ill-trained. The latter enabled me to engage the methods of elite interviewing with which political scientists are slightly more familiar and in which I had some experience.

Interview research involves establishing a conversational context with another human being. From this, one derives a transcript that may be plied for substantive facts or studied as a narrative artifact. A complication is that aspects of one's social-professional identity and status themselves affect the interview process, whether or not by choice, and must be accounted for in evaluating the results. As Carol Cohn has written of her ethnographic work in the national security establishment: “There was an ‘I’ who asked the questions, and inevitably, who I am shaped not only what I noticed and was able to hear, but also what people would say to me and in front of me.”Footnote ²⁴

My various identities—American, political scientist, professor, woman, mother—were interpreted differently in different research contexts and always affected the data I was able to gather. In many cases this worked to my advantage. As a woman (and a woman interested in children), for example, I may have seemed less intimidating to male policy-makers; being a woman with some expertise in gender-based-violence issues simultaneously helped me when I interviewed women's organizations.

But these very same attributes may have induced some trepidation by diplomats associated with the Holy See who I interviewed about their role in blocking reproductive health legislation in multilateral settings. (In those cases, I stressed my status as a mother and interest in family issues and children's rights.) Elsewhere, I was repeatedly reminded of my outsider status—as an American in Bosnia-Herzegovina and as a non-practitioner when I entered the professional spaces of the UN system. This status enabled me to “play naïve” in such a way as to gain information that might not have been articulated as such to insiders, but it also limited what people would tell me openly. And to the extent that I was naïve—that is, never as fluent in the subtle norms and jargon of each community of practice through which I drifted as were my informants—there are certainly many things I missed or misread. To compensate for this, I asked those informants with whom I was able to maintain a correspondence to review drafts of my work as it emerged, a practice I continue on my new project.Footnote ²⁵

In any case, my personal attributes surely colored what was shared with me and how, as well as how I interpreted it. Interviewing is never objective and interview data never entirely accurate. This became clear to me in part because I had been hoping that the opposite would be the case—I went looking for information on war babies not from the children or families themselves but from gatekeepers in civil society surrounding them. As it became more and more obvious to me that second-hand stories of war babies could not be relied upon to provide credible details about specific cases, my agenda shifted to identifying broad patterns and more importantly, narratives about what rights violations against war babies meant to those with whom I spoke. Thus, the choice and limitations of my methodology ended up reconstituting my research questions.

Besides these kinds of normative, tactical, and social influences, the research process may also be affected by more pedestrian matters. Grant funding influences academics, research processes, and scientific outputs, by governing the types of expenses that may be written into a budget. For example, there are limits to what may be considered an appropriate payment to a research subject for the time it takes to contribute to a study and the cap is invariably lower than what populations in conflict zones feel their time and stories are worth. Similarly, dependent care expenses are not typically considered reimbursable research expenses by funders, so young parents conducting fieldwork in the academic profession must choose between covering the costs of their minors to accompany them in the field, or leaving them at home.Footnote ²⁶ In my case, I opted for many short trips to the field rather than long periods away from my children.Footnote ²⁷

Beyond questions about the validity of a methodological approach, a different set of concerns drives ethical critiques of research methods in IR. Among qualitative researchers, feminist IR scholars and those cynical about the overly-positivist leanings of political science more generally, a strong norm toward participatory research has emerged.Footnote ²⁸ A central tenet of this approach is that research participants themselves should be provided some degree of control over the research process. As Julie Mertus wrote in her endorsement of participatory research for future studies of children born of war, “by having the research process conducted … with significant input from local actors, the potential for participating individuals to be empowered in defining their role is greater.”Footnote ²⁹ On balance, humanitarian practitioners agree with Mertus; our focus groups found that any consideration of a global study carried out by international agencies was conditioned on the assumption that beneficiary populations should be empowered to assist in designing, conducting, interpreting, and disseminating the research.Footnote ³⁰

I found, however, that such a research strategy is complicated by work on topics like children born of war. Children are the population of concern, but it may be inappropriate to require their close involvement in research design. Nor can “the local community”—generally meaning adult gatekeepers—be uncritically empowered when the goal is to tease out the experience of marginalized groups. My research subjects generally were adults, but my research topic was children. I could hardly have allowed these adults to design and interpret my results while remaining true to the goals of the study. In his concluding chapter to Born of War, a separate volume of essays I edited on the human rights of war babies, Michael Goodhart articulated this dilemma clearly: “Participatory research … presumes a model in which clear, objective distinctions exist between members of the oppressing and oppressed groups … in cases involving children born of war, where oppression of children and mothers stems from oppression of the wider group to which the mothers belong, this felicitous alignment of values and interests breaks down.”Footnote ³¹

In addition, I found that research participants often do not wish to be burdened with analysis of data or questions about research design they often feel are beyond their competence. Though I offered all informants an opportunity to review drafts or hold veto power over how their words were to be used, very few availed themselves of these opportunities. In the end, I was satisfied that my research informants from the organizations that articulated the clearest child-rights approach seemed comfortable with my methods. As a psychotherapist from one woman's organization told me, “I am happy that your research methodology is thoughtful and careful and you're not harassing these victims for their stories.”Footnote ³² My constant second-guessing of my methods, however, was strong evidence to me of where disciplinary norms about research in conflict zones are aligned, and I wonder how these may impact researchers' willingness to tackle complex subjects for which these norms were not designed.

Besides participatory research, a current norm among researchers of violently divided societies is to “give something back” to those being studied.Footnote ³³ And for good reason: a participant in one of my focus groups said, “People often say to me … that they feel extremely exploited by the research community. Yes, they came out, they saw us, and they never talked to us again. Yes, they came out, they put their names on whatever they wrote, and we don't know anything about it.” So researchers who take seriously the humanity of their subjects will avoid “parachuting in” on conflict zones and then disappearing, and will rather aim for a more meaningful and politically active engagement with those individuals whose time they share through the research process.Footnote ³⁴

In this too, I discovered that ethically appropriate fieldwork was not quite as cut-and-dried as book chapters in methodology volumes might suggest. In some cases, it was easy to negotiate ways that I and my research could give back. The most clear-cut case was Medica Zenica, who asked simply for me to send copies of my writing when it was done, to read draft grant applications, and to share contacts I might have with donors as needed. But sometimes it was complicated. The film director who produced the documentary whose politics I analyzed offered me an interview on the condition that I would write a review of his new film for publicity purposes. Although I was not told it had to be a positive review, the implication was obvious, as was my presence as a human agent within my own subject matter.

I chose to accept. I wanted the interview, and I felt confident I could write an honest review.Footnote ³⁵ However, the incident started me thinking about how political the process of “giving back” can be. There is nothing uncomplicated in providing endorsements for one party or another's construction of a social problem, particularly when a debate is afoot within a community. My rule of thumb on this project was that I would never do or say anything in exchange for information that I did not genuinely believe served the cause of promoting children's human rights as I understood them.

Finally, there is a fine line between “giving back” to the community one studies and becoming part of that community in a way that causes a rethinking of one's methodology. When I was invited by UNICEF twice to consult on the development of a fact-finding study, and invited to advise the War and Children Identity Project (WCIP) in their efforts, I faced choices between continuing to observe and collect narratives versus actively participating in and shaping those narratives.

I chose to reconstitute my project as action research. It seemed unethical to me to study a human rights non-issue without contributing to well-intentioned efforts to turn it into an issue. I decided my insights into the complexity and pitfalls could be of help to issue entrepreneurs at UNICEF and WCIP. But this hardly represented a sacrifice of my research project to normative concerns. Rather, I realized that there was no better way to study the barriers to issue construction than to participate in efforts at issue construction and document the resistance encountered.

Publishing in the Academy

When they heard I was writing a book, my respondents in Bosnia-Herzegovina usually assumed that I had already secured a publisher and had been awarded a large advance to conduct research. As academics quickly learn, scholarly writing is not done for money and is rarely considered for publication until it is complete. And successfully publishing an academic book on human rights often requires altering the message to fit the perceived market, the conceptual and normative orientations of the reviewers, and the editor's vision of how the book should “fit” with others in the house's series.Footnote ³⁶

The process of publishing an academic book manuscript begins with a short proposal to a publishing house in the discipline. If an editor is interested, the manuscript goes out to a number of reviewers selected for their expertise in the area, or often on the basis of their willingness to read the manuscript.Footnote ³⁷ The author does not know the reviewers' identities; and the reviewers attempt to provide feedback on the manuscript solely on their assessment of its suitability for the press and its scholarly merit. The type of factors reviewers consider in determining this “fit” can play a strong role in shaping the social construction of scientific “knowledge.”Footnote ³⁸

Feminist scholars, for example, have long emphasized the way in which gendered assumptions about “acceptable” types of knowledge and “appropriate” research methods have mitigated against studying the gender dimensions of world politics, or, more generally, those on the “bottom rungs.” In a landmark study on social norms, Martha Finnemore and Kathryn Sikkink cite the paucity of gender scholarship for 50 years in the flagship journal of the IR discipline, International Organization, as an example of “norms” at work: “We suggest that there may have been a well-internalized norm (with a taken-for-granted quality) that research on gender and women did not constitute an appropriate topic for IR scholarship.”Footnote ³⁹

But feminist scholars themselves now serve as gatekeepers for certain topics in the discipline. If a manuscript is perceived by an editor as a “gender” manuscript, it will invariably be sent near-exclusively to IR feminists for review because of the assumption that feminists are the “experts” when it comes to gender work, even though feminists have long encouraged mainstream IR scholars to also develop gender expertise. This gives establishment feminist scholars a tremendous amount of influence over the way that younger scholars, feminist or non-feminist, can write or think about topics perceived as being relevant to “gender studies” in IR. Studies that do not fit the mold are often rejected.Footnote ⁴⁰

Though I developed this book as a child rights project rather than a work of gender theory, when I began sending out parts of the project for review, I discovered that editors would assume that a study concerning wartime rape is a subject best reviewed by feminist scholars. And feminist gate-keepers were not uniformly enthusiastic about the work, because it did not take as its starting point the perspective of women. So gate-keeping practices within the academy, as well as within the human rights network more broadly, pose obstacles to reframing human rights in new or unexpected ways.

There is also a more general prohibitionary norm in academia of identifying oneself too clearly as a real-world person rather than as a professional writing objectively about some subject matter. One bumps into these norms when writing up professional biographies to accompany scholarly books or articles. A detailed genealogy of my dissertation work, drafted as a preface to the book version, was cut down by the editor, who insisted in particular that I not admit the book had previously been a dissertation. The editor of a respected IR journal insisted I excise a reference to my role as a mother from my professional bio before an article of mine appeared in the journal. And although I argued with him over this, I found myself making similar demands on the contributors to my edited volume as we prepared the manuscript for submission.Footnote ⁴¹

How much do these practices cause IR scholars to disassociate ourselves from the worlds we are studying? How much does this color an accurate interpretation by our readers of our interrelationship with the subjects of which we write? How much does it contribute to a false distinction between the sheltered world of Western academia and the world “out there” we writers, teachers and thinkers help to construct through our painting of it? How much do they convince us to “study” human rights violations without taking responsibility for building a world in which they are lessened?

When I was in the early stages of field research, one of the key questions asked by friends I developed in Bosnia-Herzegovina was who would publish my book. Another was will it be translated into Bosnian? I was at a very naïve and idealistic stage of my professional development, and my initial reaction was to think I'll insist on a publisher who will translate it. As the tenure clock ticked, however, I found myself worrying much more about whether the press that published the book would be considered “good enough” by my tenure committee.

I had learned through trial and error as a junior scholar that the hierarchy among academic publishing houses weighs heavily on one's tenure prospects and market value in the discipline. A number of colleagues have said to me that my one professional “misstep” was not criticism of the feminist canon (nor blogging while untenured, nor showing up at an International Studies Association panel dressed as a Cylon), but rather publishing my first book with a commercial press. In 1999, Larry Goodson, Bradford Dillman, and Anil Hira conducted a survey of political scientists' understandings of the press rankings in the discipline. They found that “major university presses enjoy the highest level of prestige and greatest familiarity.”Footnote ⁴²

Some commercial presses bridge the gap between scholars and practitioners adroitly, but this is not an attribute that is rewarded in the academy. Such publishers seek policy-relevant work that is academically informed but practical, and pitched in a jargon-free style so as to be understandable by those working in policy rather than in the classroom. My decision to publish the volume of essays Born of War with Kumarian instead of a university press lay partly out of a sense of duty to get the research in the hands of practitioners.

But one's career advancement in political science hinges on publishing with university presses, as I was gently but firmly reminded by some (not all) senior faculty at University of Pittsburgh in the aftermath of signing the Kumarian contract. And publishing a book in a top university press (particularly if you are a junior scholar) generally means making it relevant to other social scientists, not to policy makers.Footnote ⁴³ The editor at Kumarian liked the edited book idea and immediately began thinking of how to sell it to the humanitarian sector. By contrast, editors I spoke with at university presses were concerned with which theoretical debates my manuscript was speaking to and in which kinds of political science classes it might be assigned. Publishing in the academy requires making one's work relevant to professional political scientists and their students, not to the policy community, not to everyday people who might be interested in understanding an issue better, and certainly not to those being researched. In short, the pressure to talk and think and write about human rights in particular ways poses subtle pressures on academics. This contributes, I think, to scholars' tendency to focus on certain topics—data-rich, well-theorized, suitably abstract—and not on others.Footnote ⁴⁴

Hanging Out on the Theory-Policy “Divide”

Much is made of the theory-policy gap, and I think it is fair to say that the academy institutionalizes barriers to policy-relevant research in multiple ways. It privileges value-neutrality, whereas policymakers want prescription.Footnote ⁴⁵ It incentivizes careful, lengthy, esoteric analysis, while policymakers need concise, jargon-free, actionable proposals.Footnote ⁴⁶ Academics work up to several years on each paper; policymakers need short briefs produced on much shorter timelines and they care less about scientific rigor than about the nature of the specific issue at hand.Footnote ⁴⁷ As Christopher Hill has said, “The more scholars train for policy relevance, even if only to justify our existence to society at large, the more difficult it becomes to maintain intellectual integrity.”Footnote ⁴⁸ This has led some IR scholars to argue that “what is needed is a conscious effort to alter the prevailing norms of the academic IR discipline.”Footnote ⁴⁹

I'm not so sure. In Spring 2008, I gave a job talk at the University of Massachusetts-Amherst. Because my audience was a department of political scientists, I gave the scholarly version of my talk, based on the theoretical papers I had recently published in the journals International Studies Quarterly and International Organization. As I had done at my own university, I spoke of advocacy networks, issue emergence, negative cases, hypothesis-testing and generalizability. I made cursory mention of the topic of children born of war, and framed the issue as an interesting puzzle rather than a dreadful human rights problem. After an array of comments from the audience about my methods and conclusions, one senior member of the audience raised his hand and said, “The subjects you are raising are very disturbing. You talk about these terrible things so matter-of-factly, in this language of science. I think you have some great armor built up around you to help you do this. But I want to ask you what you actually think about it. What should be done?”

At first, I was taken aback by the question. Later I realized it was one of the richest I had received in years of giving that presentation at universities and conferences. I answered, “I don't mean to seem morally agnostic about the problems of these children. But I've tailored this talk to a scientific audience. Perhaps you'd like to see the other version of my talk, the briefing I give to humanitarian workers, to UNICEF, and to conferences on children and armed conflict.”

I meant don't worry. I'm not merely cold and rational about these things—only when I perceive that it suits the context. We wear different masks in different settings. But what I myself realized as a result of answering his question is that while it's sometimes hard to speak the languages of science and policy simultaneously, it's neither impossible nor, as my colleague at Pitt had feared, undesirable; and certainly it's possible for the same researcher to alternate between pitching the same argument in different ways for different audiences.

And the entrenched academic norms do in fact leave even junior scholars some room to maneuver. I have encountered practitioners willing to listen to me as an academic, and academics hungry for me to articulate a clear policy stance. Joseph Lepgold has written, “There is a wider range of IR groups than the theorist-practitioner dichotomy implies, and the infrastructure that supports their activities has created a framework for useful exchanges … more such interactions take place already than is commonly acknowledged.”Footnote ⁵⁰ Indeed, I have found on this project that the gap between theory and policy, between academics and practitioners, is as much one of habit as of necessity. And habits can, and should be broken when unhelpful—or, at least, reconsidered and renegotiated.

It's all a question of learning the right skills. Unfortunately, these skills are seldom taught in political science graduate programs. I was lucky to land my first teaching job in a professional school of international affairs, where my students would go on to work for the intelligence community or State Department, rather than to become researchers. Teaching in a professional school helped me develop the skills needed to translate insights from social science into meaningful prescriptions for policy-makers. These are indeed different audiences speaking different languages, but the barrier is not insurmountable—as Alexander George puts it, it is “more of a chasm than a gap.”Footnote ⁵¹ Indeed, as some colleagues and I discussed at a recent ISA roundtable, University of Massachusetts-Amherst's IR program is striving to incorporate bridge building between the theory and policy worlds actively into our doctoral curriculum.

But experiments of this type make clear that bridging this chasm is not without its consequences for the way academics think about research, interpret the results, and make sense of our responsibilities as political creatures. What skill sets do we need to cultivate, and to pass to our students, if we are to do so successfully while avoiding traps? I discuss here three ways in which I sought, as a researcher, to interface with the practitioners I was studying, and then I conclude with how these interactions affected the research itself.

Disseminating Outputs to Policymakers

Although over time I redefined my project as an academic book rather than a policy treatise, I resolved to disseminate versions of my findings to the practitioner communities with some capacity to solve the problems I was documenting. As I moved through the spaces of the humanitarian network, popping into UN and NGO offices, gift shops, and waiting rooms for interviews, I took note of the sorts of reading materials that lay on desks, shelves, or coffee tables. Few academic books by Cambridge University Press appeared in these spaces. But I saw lots of paperbacks published by “low-tier” presses such as Kumarian, many practitioner-oriented journals like Disasters, and many, many sleek, colorful research “reports” with images, graphs, charts, call-outs, and sidebars. I began to envision a version of my research findings that was not about building IR theory at all, but about protecting children born of war, that would speak of the gap in programming and advocacy not as a critic or a theorist, but as a helpful observer with some concrete suggestions.

Who would publish such a piece? Not conventional IR journals.Footnote ⁵² Even if they did, who would read it? Other scholars perhaps, their students if I were lucky, but not likely these busy professionals in policy and planning departments of humanitarian agencies. To create a slick version of the argument, I realized I would need funds to self-publish. Though it hadn't occurred to me to put such a line item in the original budget, the NSF permits researchers to request up to 20 percent of the amount of their grant as a “supplement” to carry research further, and one of the permissible expenditures is “other direct expenditures related to publication, documentation and dissemination.”Footnote ⁵³ I was able to secure matching funds from the University of Pittsburgh's Graduate School of Public and International Affairs (GSPIA), which was only too happy to put its logo on a sleek policy report authored by one of its junior scholars, as well as both cash and in-kind assistance from the Ford Institute of Human Security, a research institute at GSPIA.

Self-publishing on a sensitive, politically salient topic necessitates a number of important and intensely political decisions by the researcher. For example, what policy recommendations could my team conceivably put forth? The focus groups on which we were reporting in the document essentially showed a lack of consensus among practitioners as to how to proceed, as well as a lack of serious prior consideration of the problem. In such a venue, what could researchers without field experience credibly suggest? Plus, by taking explicit policy positions, would we not cross the line from research to advocacy, complicating our role as objective experts?

In the end, we limited ourselves to describing the state of knowledge and practice in the humanitarian sector, and arguing simply for some policy attention to the children, without concrete proposals.Footnote ⁵⁴ We provided basic information on children born of war, using illustrative case studies rather than broad generalizations, and to the extent possible, allowed the practitioners' comments to speak for themselves. We concluded that serious fact-finding was the appropriate next step to generate a set of policy options from which humanitarian planners could choose.Footnote ⁵⁵

Ethical dilemmas also arose. Since reports of this nature rely on imagery as much as analysis for their impact, they constitute an exercise both in scholarship and marketing. The cover photo represented the most difficult problem. University of Pittsburgh's graphic designer proposed a variety of evocative images. Most included a crying child in a war-torn land, a tearful female refugee, or a mass of women and children in a camp. I disliked all of them. I wanted something haunting, but non-descript; something uplifting. I wanted a picture of a child playing, not weeping. Or perhaps a child watching other children play. I wanted anonymity, a blurry image: no actual child's identity should be used for this report, I insisted.

Back and forth I went with the Marketing Office. In the end we settled on a photo of a child looking out a door, most of his face hidden. It could have been a child anywhere; these were in fact orphans in Sri Lanka and not children born of war at all. It was the closest I could come to evoking the kind of thoughtfulness I hoped the report would engender without the absurd, voyeuristic, pitying sensationalism I saw in much media coverage of war children. I wanted the report to make humanitarians go, hmm, without cringing: here is a population who is hiding in plain sight. I had learned from the focus groups what a delicate task that could be. It was the limit of my role as a researcher, but that I owed it to my research subjects to do that much.Footnote ⁵⁶

Another strategy I used for engaging with the human rights and human security community as a scholar during the life of this project was through political weblogs. I was inspired to blog toward the end of this project, in the fall of 2007, largely due to encouragement from other political bloggers in the profession.Footnote ⁵⁷ Of course, until very recently, blogging had been rather a suspect activity for political scientists, especially those not yet tenured. A somewhat alarming article entitled “Bloggers Need Not Apply” was published pseudononymously in the Chronicle of Higher Education in 2005, outlining the potential risks to a scholar's professional reputation of spouting off political views in cyberspace.Footnote ⁵⁸ These alleged risks, coupled with the growing trend toward maintaining weblogs, particularly among junior scholars, has led to a number of commentaries on the topic.Footnote ⁵⁹ Supposed “risks” of blogging for junior scholars include offending senior scholars, appearing too partisan, or being perceived to invest too much time in a non-scholarly activity to the detriment of research, teaching, and service.

Such factors were mitigated in my case, however, by my location in a school of public and international affairs, rather than a political science department, in which it was generally considered not only acceptable but desirable for me to communicate my research in a policy-relevant way. My faculty generally saw blogs as a legitimate way to do this, even hip and cutting-edge. (Or, at least, so they said openly, and I believed them.) In fact, I had been invited to contribute to Security Sweep, a blog established by the Ridgway Center for Security Studies at GSPIA. So I felt less pressure than some junior faculty, and at any rate, confined my activities (at first) to a group blog (Duck of Minerva) with minimal time or content commitments.Footnote ⁶⁰

The bigger dilemma for me was not some perceived trade-off between teaching, writing publishable articles, doing committee work, and blogging. Rather, I worried about the political impact that my blog posts might have and the way that I was affecting my subject matter by writing about my work still in progress. In a sense, it was just another way to disseminate my understanding of children born of war to a wider audience—not just the human rights community per se, but human rights-minded liberal policy elites in Washington who periodically read the ruminations of the Duck of Minerva bloggers. But as before, I reconciled my dual role by allowing the grim facts to speak for themselves. I did not openly advocate for global attention to children born of war in these venues, but rather simply suggested that the lack of it was a very interesting puzzle.Footnote ⁶¹ The same is true on my new project following other low-profile campaigns; I am likelier to write analytically about the campaigns or the issues themselves than to make the case for any particular policy solution.

Finally, I disseminated my findings directly to policy practitioners when I presented briefings to the human rights and humanitarian community. On several occasions I was invited to brief members of this community: at the UNICEF Innocenti Center in Florence, Italy; at UNICEF's Child Protection Division in New York; at the Experts' Meeting in Cologne, Germany; and at the United Nations Office of the Special Representative to the Secretary-General for Children and Armed Conflict. One can think of such moments as data-points in a timeline of gradual issue emergence. The solicitation of such briefings by practitioners suggests an emerging acknowledgement of a problem and interest in solving it. It also provides opportunities to collect data on the current spaces for articulating new human rights, and on the boundaries of those spaces, the points at, and ways in which, policymakers listen, and the reasons for which they push back.

But such briefings are undeniably also moments of political engagement.

The role of the responsible scholar, surely, is to avail oneself of every possible opportunity to communicate one's findings to those with the power to change the world with them.Footnote ⁶² Yet to maintain the objectivity on which the authority of experts is based, the researcher must be seen as objective, not political. This poses no conundrum, in fact, to the extent that the researcher's project is about the problem. The objective research process applies when studying the problem; a position is taken when presenting the findings.

But what about when the research project is actually about the reaction of the policymakers, as in this case? Though I was usually brought in to brief practitioners about the children born of war, this was always a limited portion of my project. My underlying interest was the reaction of the international community. And if the point of the research project is to document the failure of the international community to react, then efforts to provoke such a reaction to some extent would appear to muddy the distinction between researcher and researched.

To the extent that I solved this problem at all, I did so at first by limiting my incursions into the policy arena. I provided the information asked—usually about children born of war themselves, not the absence of a policy response; I let these facts speak for themselves, and then observed the reactions of the audience. In situations where I was invited to participate as well as present, I erred on the side of observation. Of course this was harder in focus groups or conversational settings where it must have been obvious what I was fishing for. In those cases, I learned much by genuinely listening and trying to understand why practitioners might not see the child rights issues as clearly as I felt I did. In the end, that is what a researcher can do: keep an open mind.

Handling the Media

A second mechanism for disseminating one's research findings beyond academia can be, if your topic is timely and sexy enough, the media. Seeking (or acquiescing to) media attention can only be described as a fraught and precarious enterprise. When writing policy reports, blogging or doing briefings, the researcher's audience is the non-scholarly community, but the researcher always maintains control over the message.

This is not true at all when you agree to allow the media to “cover” your scholarly pursuits. National papers are on the lookout for research on new, interesting, or galvanizing issues that resonate with readers. Local papers want to report on what local intellectuals are studying and how they are connecting to the world beyond the municipality. Universities want to publicize the research activities of their faculty. All these players seem to assume that a faculty member has a stake in “getting the news out” and “taking the credit” for research. None of these players has a stake in nuance or in timing the release of stories in such a way as to calm, rather than enhance, the jitters of practitioners working to come to a consensus about a new policy problem. And they often underestimate the way the wrong sort of news coverage can kill the careful deliberation that is often percolating behind the scenes on difficult or sensitive issues.

For these reasons, I avoided media attention for my work as long as possible. When I released my stakeholder report to the humanitarian community, for example, my hope had been that this would be disseminated widely amongst themselves. But I had hoped the story of the report would be muted in the media, because I suspected it would be taken less seriously by humanitarian organizations if the media blew it out of proportion or framed it carelessly. When I returned from the briefing at UNICEF at which I distributed the report and presented the findings, I found that a casual remark to my Dean about the urgency of printing copies in time had turned into a major university news story about “Pitt Professor Going to the United Nations.”

After a few experiences like this, with reporters calling me up, dropping by my office for sound-bytes on the “war babies” work, and sometimes taking my remarks out of context, I developed a set of guidelines for myself in interfacing with the media. If the source was print media, I demanded one of two options: 1) the right to edit the draft story, which was rarely allowed; or 2) to answer questions in writing instead of orally, so as to choose the wording carefully. If I were to do a radio interview, I insisted on having the questions in advance—though, I learned, on a live show this is no guarantee that one won't be thrown curve balls. Though many of my colleagues told me it's much better to go on live than pre-taped “because you can speak truth to power and they can't edit you out,” I discovered that pre-taped shows gave me more flexibility to shape the content. If I knew that only the best sound-bytes would be used, I could simply refuse to answer certain questions or provide awkward, incoherent answers to those where I changed my mind half-way through.

Such are the strategies by which a scholar can maintain some control over the uses to which his/her work is put in the popular consciousness. In each case, though print and radio journalists inevitably wanted me to speak to unanswerables such as “how many of these babies are there” or to prognosticate about whether “more orphanages for these children would help,” my story remained simply the lack of programmatic attention to the children. I did not see it as my role to make specific policy recommendations as much as to draw attention to vulnerabilities, expose programmatic gaps, and articulate complexity.

Consulting

Of course, nothing blurs the line between research and advocacy more greatly than becoming a consultant on a policy project in the policy domain whose work one is studying. So it is surprising, given the phobia academics are said to feel toward integrating too obviously with the world of practice, that so many take consulting positions. In choosing to involve myself as a consultant on projects that I was following, I functioned not only as an observer but also as a direct player in agenda-setting efforts regarding children born of war. The extent to which I was able to contribute in a rights-oriented way to the development of this discourse was very much a product of the banal, everyday dilemmas associated with developing and overseeing a research project and drafting written results.

My foray into consulting with UNICEF began as an accident. As a result of the “poking around” by myself and earlier researchers, the Sarajevo UNICEF field office became interested in the possibility that UNICEF was missing a category of vulnerable children. But the Program Officer needed to delegate the task of data-gathering to someone, and an academic “expert” from whom the organization could distance itself if the project turned out to be a failure was a logical choice.Footnote ⁶³ It was a win-win situation for me because I would have access to any data gathered by UNICEF. Moreover, as I explicitly outlined in the memorandum of understanding I negotiated, I could use my insights on the project to better understand the relationship between UNICEF and the issue construction process.Footnote ⁶⁴ The following year, when UNICEF's Innocenti Centre picked up on the Bosnia draft study and began hatching ideas for a follow-up study in New York, I was again asked to write a background document that would set the stage for an initial meeting in New York to launch such a study. Now committed to an action research approach, I again agreed.

It seems like a simple and an obvious thing to be willing to do as an outside expert: assist policymakers in developing a framework for beginning to address a problem. But I found it difficult to maintain a specific identity as an independent, objective researcher—an “outside expert”—once accepted as a player within a policy domain.Footnote ⁶⁵ Whereas I had previously been observing UNICEF as it functioned in post-war Bosnia and beyond, suddenly I was inside the game in a way I had been able to pretend I wasn't when I was “simply doing interviews.”

This shift to a more participatory methodology caused me some mental discomfort, as the lines blurred between myself as a researcher and as a practitioner. My field notes became much more reflective and the process of interpreting data more complex. I had to take stock of how my dual role was affecting the subject I was researching, how to make sense of that methodologically and even how to reconcile conflicting roles. I minimized this tension as best I could by maintaining a distinction in my mind, and with UNICEF, between my own independent book project and my work as their agent. The memorandum of understanding between myself and UNICEF distinguishes my long-term, theory-oriented, human-rights-regime-focused book project from UNICEF's short-term, practice-focused, war-baby-focused study and demarcates the areas of overlap between them. It sets out ways of sharing data between the two projects and clarified which data would be cited as UNICEF's rather than my own.Footnote ⁶⁶

Yet this distinction was more an artifact of our mutual understanding (and the norms of my profession) than a reality, and I found it was subject to reinterpretation by UNICEF officials to suit the needs of the moment.Footnote ⁶⁷ My former translator had originally been hired to assist with my independent interviews. When I recruited her to get involved with the UNICEF fact-finding study as well, I naively assumed she would compartmentalize the two projects as I did. But an exit interview later made it clear that she had never really disassociated the UNICEF project from my broader book project, and saw both of them as inter-related. In fact, I got the sense that she saw the distinction as nonsense, and from a perspective outside Western academia, perhaps she was right.

There was also an interesting politics in consulting, a politics regarding the social construction of knowledge. For one thing, standard scholarly norms governing research methods, ownership over data, and vetting of results were out the window.Footnote ⁶⁸ One of the key points of tension between myself as a supposed “independent expert” and the organization for whom I was producing written outputs, was the purpose of research and expectations about research design. In discussions over the scope of the Bosnia study, for example, UNICEF officials stressed that the kinds of conclusions we could reach on the basis of the interviews had to be related to the types of programmatic services UNICEF was willing to offer.Footnote ⁶⁹

By contrast, I naively viewed the project as drawing on the knowledge of civil society actors who had some understanding of the families' actual needs, in order to determine a policy direction. I asked whether it wouldn't be more appropriate to determine vulnerabilities first, then decide what programs to develop. But the response was that UNICEF had only certain capabilities, and it couldn't publish a report defining vulnerabilities in a way that it had no capability to meet. Since a different program officer had taken over by the time the draft was ready, I wrote up an exhaustive list of proposals from the survey nonetheless, partly to test what UNICEF as an organization might be willing to leave in the document, and how much of “UNICEF's” perspective was really one individual's perspective. My report included such suggestions from civil society actors as “direct payments to families.” Sure enough, the incoming program officer insisted, as his predecessor would have, that these types of recommendations be stricken from the report.

In the end, I aimed to provide as objective and independent a perspective as possible, while deferring to the organization when it pushed back. I was willing to omit things from written outputs that I felt needed to be said, but I drew the line at writing things I genuinely disagreed with.Footnote ⁷⁰ I therefore played some role in constructing UNICEF's understanding of these issues, but also “went along with the gatekeepers” when it came to certain frames and arguments. In this sense, it's not entirely clear to me whether I played a genuine role as an issue entrepreneur through my consulting work, or was essentially co-opted by human rights gatekeepers.Footnote ⁷¹ Or, for that matter, how much of my uncompromising commitments as an outsider actually impeded UNICEF's ability to find an advocacy frame that suited its organizational ethos. I may never know.

Epistemological Trade-offs

It is, in fact, nearly impossible for a researcher to assess these kinds of impacts on him or herself because of his/her own subjectivity. Nonetheless we can at least reflect upon these matters. Thus, we come full circle to the trade-off which my colleague highlighted in the anecdote that I began with. Robert Keohane argued at the International Studies Association Conference in 2008 that IR scholars have a responsibility to ask questions and do research “that might conceivably matter in the real world.”Footnote ⁷² To exercise that responsibility we must engage with the real world. But by doing so, we become part of the processes we wish to study as objective outsiders. By embedding oneself in one's subject matter—by disseminating outputs to policymakers, writing for venues practitioners will read, speaking to the media or consulting—to what extent does the IR theorist abdicate any claims to objectivity, replicability, and methodological rigor?

These are important questions. But I came to acknowledge that, depending on the particular project, researchers are never so separate from the world we are studying as some of us would like to think. The responsibility lies in acknowledging this interrelationship and how it affects the research, rather than pretending it away, and interfacing with the policy sector in a way that is morally consistent with the raison d'être of the project: in my case, in a way that itself is rights-oriented.

In the end, I chose to participate in efforts at issue construction as well as to observe, in order to document the barriers encountered. The access to the human rights community this engendered assisted, rather than compromised, my project methodologically. But it did also alter it. People are likelier to talk openly once they get to know you as a person. A conversation over drinks after a briefing is much different than a formal, professional research interview. But “informants”/colleagues are also likelier to anticipate your questions, to be cognizant of what you're fishing for and how you may interpret their statements. They learn to frame their message to fit what they understand of your normative and causal orientation to the subject matter.Footnote ⁷³ And the research becomes a two-way street, almost a collaboration among colleagues about meanings, strategies, and goals.

I found that the research became richer as it became less, in all these ways, strictly “scientific.” I had to abandon my pretense that my project could be replicated by others, or that my interview data could be rigorously coded according to the scientific standards of inter-rater reliability that I had applied to some of my other work. The project became more interpretive, as the insights I gathered were based not on an analysis of text transcripts of multiple standard interviews, but rather based on an ongoing series of interactions—sometimes correspondence, sometimes snippets of conversation over lunch, sometimes snatches of speech overheard in a women's bathroom in the United Nations, sometimes off-the-record remarks.

In this sense my own thoughts, impressions, and roles in these conversations became very much part of the dataset. The practical issues faced when marketing research findings to the humanitarian sector taught me about the nature of issue construction in that sector, about framing, about ethics, about social networks, and about what sorts of language to adopt and what to avoid. These insights reflected and helped me understand much of what I'd been told in my interviews. The process of dealing with the media as an “expert” supplemented my content analysis and historical research on the role of the media in drawing attention to issues. These experiences increased my understanding of what both journalists had told me and what human rights advocates had told me about journalists, even as my own remarks and commentary then became part of the text-data-set I was studying. Consulting provided me practical knowledge of the institutional norms within UNICEF and the humanitarian sector broadly. Such context-based knowledge gained through everyday practice is very different from what one can derive from an interview. It was valuable, even though it required me to theorize my own influence on the communities I was studying.

So I concluded that interfacing with the policy community in order to disseminate research findings itself constitutes a form of participant observation that can supplement and enrich earlier research findings derived from multiple other methods. And participant-observation is a valuable methodology for scholars studying transnational spaces or other policy arenas. But it requires IR scholars to take seriously our own presence in the world of IR. This is not something we've always done well: many IR scholars have alluded to this dynamic, but few have studied this interface empirically, and rare is the IR work that reflects on the author's role in the policy world under study. I argue that such reflection is necessary if researchers are to aim at “the use of evidence to adjudicate between truth claims.”Footnote ⁷⁴ The inevitable evidence of researcher influence and researcher bias must be incorporated into analysis, not denied, in order to maintain the independence and objectivity privileged by the mainstream academy. And researchers have a responsibility both to be biased in favor of making the world a better place, and to be transparent about their normative goals and these goals' impact on their empirical findings. I conclude by fleshing out what this means for scholars studying the politics of human rights.