Scholarly article on topic 'Reflexivity in Research on Civil Society: Constructivist Perspectives'

Reflexivity in Research on Civil Society: Constructivist Perspectives Academic research paper on "Political Science"

0
0
Share paper
Academic journal
International Studies Review
OECD Field of science
Keywords
{""}

Academic research paper on topic "Reflexivity in Research on Civil Society: Constructivist Perspectives"

International Studies Review (2008) 10, 708-721

Reflexivity in Research on Civil Society: Constructivist Perspectives

Cecelia Lynch University of California, Irvine

This article explores the ethical relationship between researcher and research subject. In order to address these issues, it examines what reflexivity entails in constructivist research on civil society actors, then discusses briefly how it can differ from highlighting the ethical dimensions of research within other paradigms like realism, liberalism, and feminism. The article also analyzes the types of ethical issues confronted by constructivists, and drawing from the practices of anthropologists, political scientists, and the author's own experiences interviewing religious humanitarian activists, assesses the tasks at hand for constructivists who are serious about understanding the ethical dimensions of their work.

Constructivist perspectives in international politics are supposed to offer the advantage of reflexivity (Kratochwil and Ruggie 1986; Keohane 1988). Substantive constructivist research on globalization, human rights, security, and other issues has also opened up questions of agency in relation to civil society to a much greater degree than during the heyday of neorealism, neoliberalism, and structural Marxism. When we combine issues of reflexivity with studies of agency, however, the ethical dimensions of scholarship become slippery, even while they remain absolutely critical to understand and expose.

Constructivist scholarship on civil society actors runs into two major types of ethical issues: (1) those concerning the ideological frame or worldview of the researcher, and how this worldview can shape research questions, methods, and even results, and (2) those concerning the ethics of conducting research on marginalized peoples, or on populations having significantly different identities, experiences, and resources than those of the researcher. In addition, constructiv-ists also experience ethical tensions encountered by comparativists and others in political science whose access to information is controlled, restricted, or even denied by powerful actors (who can include government officials, donors, or others) or who are viewed with suspicion by their research subjects. How well do constructivists navigate these terrains, and if they fall short, what can be done to remedy the situation?

In this article, I make a plea for greater reflexivity on the part of scholars and civil society actors alike regarding the ethical implications of our work. Constructivists have, to be sure, made strides in incorporating reflexivity, but progress is inconsistent. At stake, I argue, is the very quality of the research product, as well as our understanding of the ethical relationship between researcher and research subject. In order to address these issues, in the first section I examine what reflexivity entails in constructivist research on civil society actors, then discuss briefly how it can differ from highlighting the ethical dimensions of research

© 2008 International Studies Association

within other paradigms like realism, liberalism, and feminism. In the second part of the article, I analyze the types of ethical issues (articulated above) confronted by constructivists, and drawing from the practices of anthropologists, political scientists, and my own experiences interviewing religious humanitarian activists, assess the tasks at hand for constructivists who are serious about understanding the ethical dimensions of our work.

Much constructivist work highlights agency. The heavy emphasis by neorealism and Marxism on structural determinants of action and behavior in the 1970s and 1980s left little room for people to act intentionally and even less for those actions to ''matter'' in world politics. As a first step, the ontology of social construction was articulated by Kratochwil (1989), Onuf (1989), Wendt (1987, 1999), and Ruggie (1998), and Kratochwil and Ruggie called explicitly for greater reflexivity in their famous 1986 article, ''International Organization: A State of the Art on an Art of the State.'' Kratochwil and Ruggie pointed out that epistemologies that attempted to demonstrate the existence or strength of international regimes through counting rule violations or establishing covering laws missed the point of a concept founded on an ontology of social construction. This ontology instead required analyzing regime institutions, actors, and norms as constitutive of each other rather than established through monocausal relationships bounded by scope conditions. The type of reflexivity advocated by Kratochwil and Ruggie, however, emphasized epistemology rather than ethics, with an attendant call for greater awareness and flexibility regarding methodological choices.

Constructivist work on ontology and epistemology mirrored and was influenced by that of sympathizers who forged their own innovative conversations between positivist and nonpositivist methodologies (especially Alker 1996). The next wave of constructivist work focused more specifically on agents at all levels of analysis: international organizations (Finnemore 1996), epistemic communities (Haas 1992), transnational networks (Keck and Sikkink 1998), global civil society (Wapner 1996), and social movements (Lynch 1999a,b). Constructivist work on agency also mirrored in many ways long-standing feminist research (for example, Enloe 1990, 1993; Tickner1992, 2001; Peterson and Runyan 1999), in that it highlighted marginalized and ignored voices, and to some degree rehabilitated voices that had been castigated as naive or nefarious. These voices included those of women, peace movements, human rights activists, and environmentalists, among others. Drawing out the impact of these types of agents, however, also came with a cost. Emmanual Adler, for example, responded to the criticism that constructivists only dealt with ''the good guys'' by arguing that its precepts were equally capable of addressing nefarious actors in international relations (Adler 1997). While Adler's point was well taken, constructivists had not yet made strong inroads in examining the negative aspects of agency or assessing the ethical implications of their work. In particular, what were some of the presuppositions made by constructivists about the value of nonstate agency in international politics?

Exploring this and related questions requires a philosophical understanding of reflexivity. A reflexive ethical stance by constructivists requires, in turn, an awareness of the existence of the hermeneutic circle, and an acknowledgement of our place as scholars within it (Klotz and Lynch 2007). Dvora Yanow (2006) explicates two understandings of the hermeneutic circle, both of which have implications for constructivist ethics, so I will quote her explanations in detail.

The first understanding addresses the methodology of interpretation:

The term has been understood as meaning both a process of reasoning and interpreting, and the community of ''readers'' (interpreters) engaged in that process and sharing the interpretation of the text under study. As a description

of the process of meaning making, it departs from a linear model (such as the steps of the scientific method), instead depicting a circular, iterative sense making in which initial interpretation starts at whatever point is available or accessible, with whatever one's understanding is at that point in time. One makes a provisional interpretation of the text (or other focus of analysis), with the reflexive awareness that one's interpretation is likely to be incomplete and even possibly erroneous. One then engages the material in further study, at which point one revises one's initial, provisional interpretation ... and so on and so on. (Yanow 2006:15-17)

As Yanow points out, a second, related but slightly different take on interpretation focuses on the context in which it takes place:

[A] text can only be understood within its ''con-text'' whether this is the author's intent and personal background, the history of the times, other associated or contrasting texts, or something else. The hermeneutic circularity resides in reading back and forth, iteratively, between text and context. ... Further layers of understanding are added as each new insight revises prior interpretations in an ever-circular process of making meaning. Interpretations are, therefore, always provisional, as one cannot know for certain that a new way of seeing does not lie around the corner. (Yanow 2006:16)

One way of thinking about the process of interpretation and the ''intertextual-ity'' between sources and researcher is to relate it to standard notions of scientific research, which encourages triangulation of data and treating conclusions as always providing 'directions for subsequent research.' But another way, which is more pertinent to constructivist (and feminist) ontologies rather than those related to other paradigms in international relations (IR), zeroes in on the position, views, and experiences of the researcher herself. This entails

[c]ombining the hermeneutic focus on texts as vehicles for conveying meaning with the phenomenological consciousness that researchers, too, act from an experientially informed standpoint. . Research designs, formulations of questions, choices of observational sites and persons interviewed, analytic frames, and writing all construct perceptions of the subject of study, rather than objectively reflecting it. (Yanow 2006:16, emphasis in the original)

To address the question about the presuppositions of those who conduct research on civil society actors, we need to think about our positionality as scholars in a variety of sometimes overlapping contexts in world politics. For example, all but the youngest IR scholars experienced the transition from the Cold War to the post-Cold War era as both political subjects living the transition, and analysts, or ''experts,'' trying to make sense of it. Two further outcomes of this transition also influenced researchers' views of civil society actors: first, the increasingly visible role played by social movements and nongovernmental organizations (NGOs) all over the world, from democracy movements in Eastern Europe to the new roles and prominence of transnational NGOs in conflicts and peace settlements; second, an overwhelmingly liberal global ideological and economic context in which Western markets and financial institutions shaped the democracy-building and postconflict development agendas promoted by many civil society actors.

Different types of research questions result from these contexts, depending on the researchers' worldviews and assumptions and hence, which context they emphasize. The explosive growth of NGOs in the post-Cold War period at first made civil society initiatives to assist the marginalized sectors in both the First and Third Worlds appear innovative and valuable, especially vis-a-vis

dysfunctional state governments. Soon, however, scholars began to raise questions about the efficacy, funding prerogatives, and hence accountability of NGOs. I will address later the worldviews and ideologies that are shaped by these contexts and that in turn influence such research questions. For the moment, however, I emphasize the idea that focusing on the ethical dimensions of our research reminds us that civil society actors, as well as the kinds of questions we pose about them are implicated together in ethical analysis.

These issues have much in common with the ethical concerns of feminist research (at least its interpretive variants) because they draw on feminist insights and because both constructivist and feminist insights draw on similar concepts and themes in social theory (Priigl 1999; Klotz and Lynch 2007; Ackerly and True 2008). They also have commonalities with classical realism and political liberalism, though perhaps less so than some of their advocates might think. And the focus on the positionality of the researcher is also a theme that emanates from variants of Marxism (such as a Gramscian stance) that allow for inten-tionality on the part of scholar-activists.

The major difference between the ethical concerns of realists and liberals, on one hand, and constructivists on the other, is that constructivist interpretations point to the variety of ethical possibility and implicate the researcher (to a greater or lesser degree) in its realization. Both realism and liberalism have a teleology that determines their ethical bent, as well as the way they view and assess other ideological frameworks. Realism privileges the good of the state in its focus on-state ''interests,'' while liberalism privileges the good of the individual, in the sense of her progress and welfare (economic, political, and moral). The ''neo'' variants of realism and liberalism are similar to each other in that they privilege the good of the state, but they disagree on how to achieve it. The ethics of neo-liberalism focus on ways to provide incentives for states to participate in interstate cooperation. This is because of its teleology of pragmatic cosmopolitanism, in which the state does not disappear but rather becomes a vehicle for enlightened self-interest. Neorealism, in contrast to neoliberalism, emphasizes the virtue of prudence in decisions about whether or not states should cooperate with others and adheres to the belief that cooperation is not an end in itself, but only the means to achieving or maintaining state power (Keohane 1986).

Inherent in these teleologies is also an epistemological assumption that research can and should be conducted ''objectively,'' an assumption which separates, rather than implicates, the researcher in the analysis. In fact, ''good'' research, according to the underlying positivist epistemology of realism and liberalism, requires such a separation, whereas faulty work stems from researchers' biases, which infiltrate research designs in problems such as ''selecting on the dependent variable,'' and circular causality. While many types of feminist research are similar to constructivism in their ontological presuppositions and methodological choices, the main point of feminist research—understanding and explaining the gendered nature of the question at hand—is pre-constituted. This gives priority to ethics of equality and justice for women, which may or may not be present in different types of constructivist work (Priigl 1999).

Constructivists, however, have different views about how significant the scholar's positionality is for research questions as well as findings. As Klotz and Lynch point out, an important question for constructivists is how far interpretation goes, and as a result, whether the researcher must include an analysis of his or her positionality as part of high-quality research (Klotz and Lynch 2007:58-63). I argue that, whether acknowledged or not, the questions asked, the methods followed, and generally the way the questions and methods shape the research findings, all reflect the scholar's (initial, at least) ideological presuppositions. This intentionality and ethical stance cannot be separated from the research procedures or results. This is true of all research, but constructivists, because we

acknowledge the influence of positionality, bear a special responsibility to highlight the ethical implications of our analytical assumptions and eventual results.

Consequently, the first thing we should do in examining the relationship between civil society actors and the ethics of global governance is to shine a light on our own assumptions about these actors and about global governance, because all assumptions embody ethical ideas and judgments. However, we also need to recognize that the researcher's intentionality and ethical stance may evolve or change substantially over time, especially if the scholar acknowledges her worldview and assumptions, and is genuinely open to new ideas and learning from her research subjects. Such learning does not result in removing the researcher from the hermeneutic circle, permitting her somehow to achieve ''objectivity.'' Instead, it demonstrates the movement that can occur within the research process that results in potentially more ''trustworthy'' (Schwartz-Shea 2006) research findings, which in turn reinforces a more reflexive stance vis-a-vis the researcher's ethical assumptions. I argue that this type of reflexivity should be a consistent goal of all constructivist research.

In the following section, I relate the issues of ideology, research context, and positionality vis-a-vis research subjects to the study of civil society actors in world politics, teasing out the complex ethical concerns and tensions that arise in this relationship.

Scholars and Ethics: The Questions We Ask and the Sources and Methods We

Use to Answer Them

Constructivism in IR is based on the assumption that global politics is socially constructed. Klotz and Lynch (2007: Chapter 1) argue that this assumption of social construction requires attention to context, intersubjectivity, and power. But beyond these three aspects of social construction, many paths to research on civil society actors are possible, depending on the ideological proclivities, training, and degree of adherence to academic conventions of the researcher. In particular, constructivist analysts of civil society actors tend to take one of two ethical stances: one that celebrates civil society actors as providing an alternative ethics to those of raison d'etat, or one that critiques civil society actors as implicated in and enabling the ethics of contemporary forms of neoliberalism.

The first framework often focuses on the growth of civil society actors and their influence in specific issue-areas. This path tends to be based on liberal understandings of democracy and accountability, as well as a critical stance towards the state. Researchers coming from this type of worldview highlight new types of epistemic communities, social movements, NGOs, and transnational advocacy networks, and examine their influence vis-a-vis ''traditional'' (read realist) conceptualizations of state interests (Haas 1992; Keck and Sikkink 1998; Klotz 1995; Lynch 1999a,b; Finnemore 1996; Wapner 1996; Priigl 1999). This type of research generates findings that demonstrate the activities, resources, and effects of civil society actors on state and Inter-Governmental Organization (IGO) policies and international treaties, the norms that underpin governance processes and their legitimation, or the identities which influence the construction of state interests (Katzenstein 1996). The liberal framework tends to emphasize the role of civil society actors in increasing the power and influence of individual and social group agendas, actions, and hence ethics vis-a-vis the state, and asks questions about whether these agents influence particular norms, interests, identities, and/or policies, how their influence can be traced, under what conditions it is manifested, and what effects their influence has on particular issue-areas such as the environment, human rights, and peace. Highlighting the agency of these actors, according to these liberal assumptions, often results in diminishing the state's ''room for maneuver'' according to traditional notions of

sovereignty. Thus, civil society groups are important because they check the powers of governments (both liberal and illiberal), expand democratic debate on contentious issues, and/or advocate new norms and practices.

Criticisms and questions about these actors which emanate from this framework tend to focus on whether civil society actors are sufficiently democratic because they lack accountability (Donini 1998; Wenar 2006), or whether they are efficient or even necessary providers of social goods. Civil society actors, unlike liberal governments, are not answerable to the general public, but only to their own constituencies. Moreover, funding concerns often drive their goals and priorities. Focusing on the limited constituencies of civil society actors vis-avis elected officials underscores the importance of representation in democratic theory and ethics, but it also assumes that representative government is in fact more accountable to the general public and less driven by money concerns than, for example, NGOs. These are at best questionable assumptions. Moving to an economically based ethical variant of democratic theory privileges questions about the efficacy of civil society actors in providing social and political goods. While constructivists tend not to privilege efficiency over other possible values, contemporary trends among NGOs may heighten their interest in related issues. For example, there are movements among development NGOs to consolidate their work in order to decrease inter-agency competition for funds and increase economies of scale. NGO business consultants are promoting such consolidations (MIDI 2005), which mimic the merger mania of for-profit corporations. If such mergers become a significant trend, future numbers of civil society actors may well decrease in some societies rather than continue their exponential growth of the 1990s, and market-based considerations can reinforce ethical questions about the relationship between liberal democracy, transparency, and efficiency.

The possibility of a growing trend of NGO mergers begs additional questions emanating from the assumptions of the second major ethical framework for analyzing civil society actors. This framework focuses on the growth of these actors as part and parcel of the hegemony of economic and democratic liberal discourses (Sending and Neumann 2006). The work of Mustapha Kamal Pasha and David Blaney pinpointed concerns about the ''elusive paradise'' of global civil society in the late 1990s (Pasha and Blaney, 1998). Others subsequently argued that, especially since the end of the Cold War and the advent of the most recent phases of globalization, states (industrialized as well as less-developed) have encouraged civil society organizations to take over much of the provision of humanitarian relief and social welfare, resulting in policies such as the Bush administration's ''faith-based initiatives,'' the Clinton administration's funding of humanitarian NGOs through the United States Agency for International Development (US AID), and the encouragement of NGO work by Third World countries' governments (Lynch 1999b, 2003). Consequently, the exponential growth in numbers of civil society actors can be traced to government policies. While traditional notions of state sovereignty might still suffer from this growth, the motivating force for such a change, according to the critical framework, resides in the constitutive power of neoliberal mechanisms of governance rather than the altruistic norms promoted by humanitarian, peace, and rights-oriented NGOs.

The types of research questions that emanate from a critical framework tend to privilege the cooptation of civil society actors into such neoliberal governance mechanisms. This work follows on the body of research conducted by critical constructivists such as Weldes et al. (1999), who focus on the way in which discursive representations of identities, threats, and power constitute states, elites, and transnational actors (Doty 1996; Grovogui 1996; Latham 1997). These analyses stem from an ethic that criticizes both hegemonic state power and global

market practices for actively constructing the marginalization of disadvantaged social groups by creating and reinforcing unequal power relationships.

The assumptions driving critical constructivist studies of civil society actors can also produce research questions that emphasize these actors' complicity in producing outcomes that perpetuate, rather than challenge, the status quo. Such research questions flip the focus on civil society actors as agents of change, which underpin liberal assumptions, to a focus on mechanisms (if not structures) of constraint which shape their actions and intentions. Some scholars have pointed to the risk of obliterating agency altogether (for example, Milliken 1999; Klotz and Lynch 2007), begging the question of whether greater reflexivity on the part of scholars can mediate between the liberal tendency to accord power to civil society actors separate from the state and market discourses within which they act, and the critical tendency to view civil society actions as largely shaped by these state and market logics, discourses, and practices.

Some scholars have begun to make substantive and conceptual inroads that attempt to accord intentional agency to civil society actors while (1) tracing the liberal logics and mechanisms that shape their goals and practices, and (2) evaluating the meaning, including the ethical consequences, of their actions (True 2001; Grovogui 2006). This type of work, however, leads to the necessity of examining the ethics of conducting research on social groups and movements that may be marginalized politically, economically, or socially.

The Ethics of Constructivist Research on Marginalized Actors and Social Groups

Constructivists have often sought to draw out the influence and voices of marginalized civil society actors, both from the assumption that such actors matter in world politics and from the assumption that the power relations involved in their marginalization need to be studied—and changed. In contrast, realist ethical imperatives often treat civil society actors as unimportant, naive or ignorant. If unimportant, they simply are not worthy of analysis; if naive or ignorant, they can promote courses of action that are dangerous from the point of view of furthering relative state advantages (Lynch 1999a,b). And liberal ethics assume that civil society actors all share the same value hierarchy, consequently they have trouble accommodating cultural difference.

But, while many constructivists (though not all) foreground the study of civil society actors as significant in world politics, they are just beginning to address the phenomenological, or experiential, aspects of the researcher himself that implicate him in the interpretive context. Scholars in related fields such as anthropology have made greater strides in this domain (as have feminist scholars, shown by Ackerly and True 2008), demonstrating how the values and expectations (''hypotheses'') the researcher brings to the work, as well as the research process itself, shape the investigation and its conclusions. Twenty years ago, James Clifford and George Marcus published an edited book that proclaimed the crumbling of ''an ideology claiming transparency of representation'' (that the scholar could transparently, and hence ''accurately,'' represent the social and cultural life of subjects through ethnographic methods), instead focusing on ''the historical predicament of ethnography, the fact that it is always caught up in the invention, not the representation, of cultures'' (Clifford, in Clifford and Marcus, 1986:2).

Studying identities of ''the other'' (Said 1978; Todorov 1984; Goff and Dunn 2004) and crossing cultural, religious, class, gender, and political boundaries is tricky when we purport to highlight ignored actors or give voice to the marginalized. But more often than not, given our interconnected political, social, and economic spheres, we study social groups whose identities overlap as well as differ from our own. While in IR as well as related fields, the rise of the

''indigenous ethnographer'' (Clifford 1986:9)—or political scientist, sociologist, lawyer, etc.—is, happily, increasingly common, ''being a native'' unfortunately does not ensure accuracy in representation or analysis.

One level of anthropological reflexivity regarding marginalized civil society actors is found in Erica Bornstein's analysis of the religious NGO World Vision's Child Sponsorship program in Zimbabwe. Bornstein's ethnographic analysis of the dual effects of child sponsorship explores the rationales and practices of the NGO itself as well as the children, families, and community affected. Her research design and procedures are comprehensive and sensitive to the context and goals of all of the various actors involved. She finds through her study that

Such humanitarian practices have two sides. First, they have the truly transformative potential of relationships formed by sponsorship: of material improvements in lives, made possible by gifts of education and opportunity, and relationships with the potential to transcend distance, class, and culture. And second, they have the potential to create localized experiences of lack that stand in the face of benevolent attempts to bridge distance and that may inadvertently be enhanced by humanitarianism itself. (Bornstein 2001:614)

The ''localized experiences of lack'' include the knowledge on the part of other family and community members that the sponsored child receives money and gifts not available to them, without any apparent logic or criteria of merit. Sponsorship can therefore, undercut the very family values promoted by the evangelical Christian NGO, because the money and gifts are channeled not through the child's parents but through the local adjunct of the organization. Thus parental decision making and authority, as well as the dignity of the child's family, are diminished, and perceptions of wealth, education, and well-being are conjured from afar.

In this article, Bornstein does not offer introspection into her own ideological or religious assumptions. However, she employs the common anthropological disciplinary practice of keeping notes that include reactions to the ''artifacts'' of her study, as in the following passage:

On one occasion, I was invited to the sponsor relations room and offered an opportunity to read mail from sponsors to their children. ... As I opened the envelopes, I could feel the exoticism of something from far away. I could imagine how excited a child in rural Zimbabwe might be to receive a paint set, a toy car, a coloring book, or even a photo of a family in Canada, Germany, or the United States. (Bornstein 2001:602)

Bornstein acknowledges that ''the context did not allow for rigorous content analysis of letters and packages,'' but she still discusses the affective response that can be provoked by these artifacts of the child sponsorship program. Inviting the reader to partake (or not) of the emotive aspects of such materials is one way to include types of ''low data'' (Weldes 2006), which might otherwise be ignored, in the analysis. Bornstein's article provides a sensitive and thorough analysis of the child sponsorship program from a wide variety of participants' perspectives. Yet, it might be helpful to have more information about Bornstein's reflections on her own assumptions and identities before we can judge whether our affect corresponds with hers. More information could also encourage our reflections as readers on her potential biases as well as our own.

Saba Mahmood's (2005) study of the women's mosque movement in Egypt illustrates the pull and push of overlapping yet distinct identities in ethnographic research. She also adds an additional layer of reflexivity to her analysis by including an assessment of her own situatedness and assumptions. Mahmood, also an anthropologist, takes the reader through her own intellectual formation as a

well-educated Pakistani woman; an intellectual, ''progressive feminist,'' and activist on the left who believed in ''some form of critical Marxism.'' Each of these proclivities shaped Mahmood's initially negative reaction to the rise of Islamist movements in Pakistan and the Middle East. Over time, however, the explanations that she and her cohort had derived for the Islamist phenomenon—that it was superficial, due to ignorance, or created and maintained by Western aid flows—appeared, she says, ''inadequate.'' Instead, she began to see the variety of political stances represented by Islamist movements and political parties, and how these stances often responded to

popular demands for democratization of the political arena, for an end to singleparty rule, and for a more critical stance toward US hegemony in the region. Moreover, Islamic welfare organizations around the Muslim world have increasingly stepped in to fill the vacuum left by postcolonial states as these states, under neoliberal economic pressures, have withdrawn from providing social services to their citizens. (Mahmood 2005:.x-xi)

Mahmood explains her own and other progressives' tardiness in casting off their initial explanations for the rise of Islamism as a product of

our profound dis-ease with the appearance of religion outside of the private space of individualized belief. . This fear is accompanied by a deep self-assurance about the truth of the progressive-secular imaginary, one that assumes that the life forms it offers are the best way out for these unenlightened souls, mired as they are in the spectral hopes that gods and prophets hold out to them. (Mahmood 2005: xi)

In other words, Mahmood's own experiences, shaped and reinforced by class and educational institutions, also influenced the way she analyzed political and cultural trends in her own and others' societies, and prevented her (initially) from seeing significant factors that would aid her analysis. The rest of Mahmood's study of three women's mosque groups in Cairo is a finely-honed and detailed analysis of the social, spiritual, political, and economic issues confronted by these women and their reactions to them, but what makes it even more memorable and deeply insightful is the frequent elaboration by Mahmood of her own assumptions and reactions to the movement participants' judgments and actions when they differ from her own. Interlocutor and subjects share some formations and identities (gender, non-Western backgrounds in Muslim-majority societies), but differ profoundly in others (religious adherence, localized experiences). Through this exposure of the relationship between herself as interlocutor, her reactions and questions to the women regarding aspects of their statements and actions that make little sense to her, and her reporting of their reactions to her questions, we see a form of ethical reflexivity in action. The degree of transparency through which this reflexivity takes place—laying bare her own assumptions and details of her background and multiple identities—can never be complete, but the sensitivity and intelligence of the effort makes Mah-mood's analysis and findings regarding the women's mosque movement all the more trustworthy, in Schwartz-Shea's terms.

These insights show that constructivists must be aware of the ethical responsibilities entailed in their research. It is critical in ''giving voice'' to actors previously ignored or marginalized to reflect before, during, and after the research process on one's own assumptions, background, and knowledge base, and to assess not only what the process teaches about the civil society actors, but also what it teaches the researcher about the validity and trustworthiness of his own presuppositions.

Ethical Considerations and Constraints on Research: Access, Power,

and Positionality

The November 2007 issue of International Studies Perspectives contains a symposium on the issue of academic freedom. The authors in the symposium debate whether and to what degree peace research, scholarship on the Middle East, and work critical of the so-called War on Terror are not only vilified in the United States today, but also actively constrained by a loose coalition of US officials who make scholarly travel and communication difficult or impossible, and Campus Watch movement participants who propagate negative and faulty information about these scholars and their work (Brand 2007; Evangelista 2007; and Falk 2007). As Laurie Brand points out

Over the years we have seen our colleagues living under authoritarian regimes . dismissed from teaching positions, prevented from conducting their research, arrested, imprisoned, and sometimes tortured because of their actions. However, largely as a result of responses to the September 11, 2001 attacks, in the last several years threats to academic freedom have dramatically increased both in number and intensity in the United States as well. (Brand 2007:384)

In this section, I address the relationship between official and unofficial constraints and constructivist insights and ethics.

One prominent constructivist insight is that dominant narratives about the international system, the role of states, and the role of researchers help to shape our research questions and concerns. More pragmatically, political factors at home and abroad result in issues of access that attempt to prevent us from asking some types of questions (or, sometimes, pressure us into asking others). This happened during the Cold War in the United States, when the IR establishment denigrated ''peace studies'' and promoted a view of international politics as a zero-sum game (Oren 2003; Evangelista 2007), and in the former Soviet Union, where access to government documents was usually impossible. Today, research on civil society organizations may be constrained by undemocratic governments who wish to prevent scholarly access to local populations, or by large donors, who decide on funding priorities that then shape research agendas for years to come.

These constraints are not specific to constructivists, and scholars working from a variety of disciplines and theoretical paradigms cope with such constraints by (1) acknowledging them in the research methods section of their analyses, (2) making every effort to pursue their original questions despite obstacles (for example, Brand), or even (3) recasting the research project to try to please the agenda of funders. I do not argue that these strategies are always effective (or ineffective), but simply wish to point out that both (1) and (2) comprise a significant degree of transparency and persistence.

However, an additional layer of reflexivity regarding (1) our own assumptions and intentions, and (2) how much of our experience and background we reveal in the research process, becomes necessary to establish rapport with our research subjects, especially in contexts that give rise to suspicion of our agendas. The positionality of the researcher affects her reception by civil society actors, whether they are relatively powerful or marginalized. Melanie Cammett (2006) discusses the challenges of doing field research in ''deeply divided societies, where religious, ethnic, regional, or other types of cleavages are highly politicized and may even constitute the ostensible basis of political violence.'' Cam-mett's work, on the relationship between faith-based social service providers and sectarian identities in Lebanon, is based on a research question that can easily become implicated in highly contested and often violent political settings. She

explains that, ''In deeply divided societies, informants may view members of other groups, such as nonco-religionists or nonco-partisans, as well as foreign researchers, in extreme terms: either as opponents or potential converts.'' Moreover,

Researchers from some foreign countries, then, may face the added challenge of presenting themselves as independent of their governments. In most Middle Eastern countries, for example, it is critical for US-based researchers to stress their nonaffiliation with governmental or quasi-governmental agencies because of widespread antipathy towards US policy in the region. (Cammett, 2006:16)

Because of these reactions to the scholar's background and position, Cammett advises several strategies to convince informants of the researcher's neutrality, including ''framing the project in neutral language'' and using technical terminology rather than obviously political language to describe the research goals. ''For example, presenting my research in terms of how people gain access to healthcare, schooling, or social assistance rather than how they become supporters of a sectarian organization emphasizes a more neutral yet nonetheless central aspect of my project.'' Despite these efforts, Cammett cautions that efforts to portray oneself and one's research as neutral are not always persuasive, especially when subjects have strongly negative views and experiences of their country of origin, such as the United States.

My own experience in conducting research in divided societies leads me to agree with Cammett that finding nonthreatening means of expression is necessary to put interviewees at ease (and satisfy Institutional Review Board (IRB) requirements not to cause them discomfort). At the same time, however, it raises additional questions for constructivists regarding whether neutrality is possible or even desirable. In my interviews of representatives of religious humanitarian organizations in Central and East Africa and the Middle East, for example, I find that attempting an apolitical stance sometimes causes interviewees to treat the research questions less seriously and more perfunctorily, and the interview as a whole goes nowhere. Conversely, when I give reasons for the questions I pose, including relating aspects of my own background and experience, or acknowledging the political or cultural leanings which lead me to ask my questions, I find that interviewees are ready to engage and react, as long as I also begin from a basis of respect and let them know in so many words that I am there to learn from them to what degree my own assumptions can be supported or need revision.

Nevertheless, because of my training as a ''social scientist,'' whether and how much of my own biases and experiences to reveal remains a constant worry. Do I tell the evangelical Christian NGO representative in Kenya that I am uncomfortable with proselytizing? Do I let the pro-Israel activist know that I am sympathetic to the plight of Palestinians? Do I acknowledge to the Iraqi member of the Mahdi movement that I am not persuaded by his assertion that the movement has espoused nonviolence? Or do I acknowledge to the Kurdish human rights activist that I disagree with the US war in Iraq?

The answers to these questions are not set in stone. Nevertheless, I argue in favor of considering research on civil society actors as a conversation among mutually interested parties. Interviewees do not agree to a meeting unless they find something of interest in the research project. In the resulting conversation, the researcher's primary duties include (1) beginning from a position of respect towards the research subject, (2) maintaining an openness, not only to cultural difference, but also to evidence or sentiments contrary to one's proclivities and expectations, and (3) constantly reassessing these proclivities and expectations in light of the research experience.

Finally, it is important for constructivists who study civil society actors to recognize that, not only are they themselves responsible for understanding their research results as ''provisional, as one cannot know for certain that a new way of seeing does not lie around the corner'' (Yanow 2006), and not only are their own views and experiences constantly up for reinterpretation according to reflexive constructivist ethics, but the constitutive nature of their interaction with their subjects affects these actors' own perceptions, experiences, and future actions. Mahmood's mosque movement women, Bornstein's NGO and family actors, and Cammett's faith-based representatives all take new information away from their encounters with these scholars.

My own research exchanges often result not only in increased information for me, but also new knowledge for my interview subjects. My interviews begin from a prepared list of questions but I also inform the interviewee that the process will be open-ended, and I always welcome questions posed to me as well. When I asked two Masai community members whether they were part of a Masai NGO I had found on the internet, they knew nothing about the organization, which was based in another part of Kenya and founded by someone who had traveled to the United States. I gave them a printout of the information, which they wanted to take to their elders for discussion. Whether their experience resulted in communication with that NGO, the creation of yet another new NGO, or simply a revised attitude towards NGOs in general, I do not know, but it is clear that such interactions always constitute two-way channels of knowledge construction. Many interview subjects wish to maintain contact in order to learn more about the networks, religious ethics, or even my analyses of NGOs operating in their own regions as well as different parts of the world, continuing this process of mutual knowledge construction.

Conclusions

Constructivist ethics on civil society actors implicate the researcher as part of the constitutive process of knowledge-building. Their point of departure is the existence of the hermeneutic circle and phenomenological experience, which compels the researcher to view his assumptions, questions, procedures, and conclusions as part of a broader and constantly-evolving interpretive field. Consequently, it is necessary for constructivists to learn to incorporate reflexivity into their research.

Anthropologists, among others, have shown that reflexivity requires a sensitivity to cultural difference. It also requires introspection into the reasons for the researcher's own assumptions and the questions and procedures that result from these assumptions. Scholars should always make their methods and constraints clear to the civil society actors they study as well as their eventual audience. But they should also reflect on the effects of revealing relevant aspects of their own background, experience, and working assumptions to their interlocutors and readers. While there are not currently strict guidelines for what information to provide and how to impart it, it is clear that the research process, following con-structivist insights into the nature of interpretation, influences both scholars and the civil society actors they study. The constitutive, ongoing nature of this process compels constructivists to continue to probe the relationship between ethics and knowledge-construction and to identify and debate further, based on our research experiences, the characteristics of reflexivity necessary to our work.

References

Ackerly, Brooke, and Jacqui, True. (2008) Reflexivity in Practice: Power and Ethics in Feminist Research on International Relations. International Studies Review 10(4):693-707.

AdLer, EmmanueL. (1997) Seizing the Middle Ground: Constructivism in World Politics. European Journal of International Relations 3:319-363.

ALker, HaywarD. (1996) Rediscoveries and Reformulations. Cambridge, UK: Cambridge University Press.

Bornstein, ERica. (2001) Child Sponsorship, Evangelism, and Belonging in the Work of World Vision Zimbabwe. American Ethnologist, 28:595-622.

Brand, Laurie A. (2007) Middle East Studies and Academic Freedom: Challenges at Home and Abroad. International Studies Perspectives, 4:384-395.

CAmmett, Melani. (Fall 2006) Political Ethnography in Deeply Divided Societies. Quantitative Methods. Newsletter of the APSA Organized Section On Qualitative Methods, 4:2.

Clifford, James. (1986) Introduction: Partial Truths. In James Clifford and George E. Marcus, eds., Writing Culture: The Poetics and Politics Of Ethnography. Berkeley: University of California Press.

Donini, ANtonio. (1998) Asserting Humanitarianism in Peace-Maintenance. Global Governance 4:8196.

Doty, Roxanne Lynn. (1996) Imperial Encounters: The Politics of Representation in North-South Relations. Minneapolis: University of Minnesota Press.

EnLoe, Cynthia. (1990) Bananas, Beaches, and Bases: Making Feminist Sense of International Politics. Berkeley: University of California Press.

Enloe, Cynthia. (1993) The Morning After: Sexual Politics at the End of the Cold War. Berkeley: University of California Press.

EvangeLista, Matthew. (2007) The Dangerous Study of Peace and the Risk to Academic Freedom. International Studies Perspectives 4:376-383.

FaLk, Richard. (2007) Academic Freedom Under Siege. International Studies Review, 4:369-375.

Finnemore, Martha. (1996) National Interests in International Society. Ithaca, NY: Cornell University Press.

Goff, Patricia, and Kevin Dunn, eds. (2004) Identity and Global Politics: Empirical and Theoretical Elaborations. New York: Palgrave MacMillan.

Grovogui, SibA N'ZatiouLA. (1996) Sovereigns, Quasi-Sovereigns, and Africans: Race and Self-Determination in International Law. Minneapolis: University of Minnesota Press.

Grovogui, Siba N'ZatiouLa. (2006) Beyond Eurocentrism and Anarchy: Memories of International Order and Institutions. New York: Palgrave MacMillan.

Haas, Peter M, ed. (1992) Knowledge, Power, and International Power Coordination. International Organization. Special issue, 46:1.

Katzenstein, Peter J, ed. (1996) The Culture of National Security: Norms and Identity in World Politics. New York: Columbia University Press.

Keck, Margaret E., and KAthryn Sikkink. (1998) Activists Beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press.

KEohane, Robert O. ed. (1986) Neorealism and Its Critics. New York: Columbia University Press.

Keohane, Robert O. (1988) International Institutions: Two Approaches. International Studies Quarterly 4:379-391.

Klotz, Audie. (1995) Norms in International Relations: The Struggle Against Apartheid. Ithaca, NY: Cornell University Press.

KLotz, Audie, and Cecelia Lynch. (2007) Strategies for Research in Constructivist International Relations. Armonk, NY: M. E. Sharpe.

KratochwiL, FRiEDRich. (1989) Rules, Norms, and Decisions. Cambridge: Cambridge University Press.

KratochwiL, FRiEDRich, and John Gerard Ruggie. (1986) International Organization: A State of the Art on an Art of the State. International Organization 4:753-775.

Latham, Robert. (1997) The Liberal Moment: Modernity, Security, and the Making of Postwar International Order. New York: Columbia University Press.

Lynch, CeceLia. (1999a) Beyond Appeasement: Interpreting Interwar Peace Movements in World Politics. Ithaca, NY: Cornell University Press.

Lynch, Cecelia. (1999b) The Promise and Problems of Internationalism. Global Governance 5:83-101.

Lynch, CeceLia. (2003) Agendas of Contemporary Social Movements. In International Law and the Quest for Security, edited by Richard Falk and David Krieger. Santa Barbara, CA: Nuclear Age Peace Foundation.

MahmooD, Saba. (2005) Politics of Piety: The Islamic Revival and the Feminist Subject. Princeton, NJ: Princeton University Press.

MIDI (2005) M & A in the NonProfit World. MIDI Newsletter. Columbia Business School.

Milliken, Jennifer. (1999) The Study of Discourse in International Relations: A Critique of Research and Methods. European Journal of International Relations 2(4):225-254.

Mustapha Kamal Pasha, and David L. Bianey. (1998) Bianey Elusive Paradise: The Promise and Peril of Global Civil Society, Alternatives, 23(4):417-450.

Onuf, Nicholas. (1989) World of Our Making: Rules and Rule in Social Theory and International Relations. Columbia: University of South Carolina Press.

Oren, Ido. (2003) Our Enemies & Us: America's Rivalries and the Making of Political Science. Ithaca, NY: Cornell University Press.

Peterson, V. Spike, and Anne Sisson Runyan. (1999) Global Gender Issues, 2nd edition. Boulder, CO: Westview Press.

PrUgl, Elisabeth. (1999) The Global Construction of Gender: Home-Based Work in the Political Economy of the 20th Century. New York: Columbia University Press.

Ruggie, John Gerard. (1998) Constructing the World Polity: Essays on International Institutionalization. London: Routledge.

Said, Edward. (1978) Orientalism. New York: Vintage.

Schwartz-Shea, Peregrine. (2006) Judging Quality: Evaluative Criteria and Epistemic Communities. In Interpretation and Method, edited by Dvora Yanow and Peregrine Schwartz-Shea. Armonk, NY: M. E. Sharpe, 89-114.

Sending, Ole Jacob, and Iver B. Neumann. (2006) Governance to Governmentality: Analyzing NGOs, States, and Power. International Studies Quarterly 50:651-672.

Tickner, J. Ann. (1992) Gender in International Relations: Feminist Perspectives on Achieving Global Security. New York: Columbia University Press.

Tickner, J. Ann. (2001) Gendering World Politics: Issues and Approaches in the Post-Cold War Era. New York: Columbia University Press.

Todorov, Tzvetan. (1984) The Conquest of America: The Question of the Other. New York: Harper & Row.

True, Jacqui. (2001) Gender, Globalization, and Post-Socialism: The Czech Republic After Communism. New York: Columbia University Press.

Wapner, Paul. (1996) Environmental Activism and World Civil Politics. Albany, NY: State University of New York Press.

Weldes, Jutta. (2006) High Politics and Low Data. In Interpretation and Method, edited by Dvora Yanow and Peregrine Schwartz-Shea. Armonk, NY: M.E. Sharpe.

Weldes, Jutta, Mark Laffey, Hugh Gusterson, and Raymond Duvall, eds. (1999) Cultures of Insecurity: States Communities, and the Production of Danger. Minneapolis: University of Minnesota Press.

Wenar, Leif. (2006) Accountability in International Development Aid. Ethics and International Affairs, 20(1):1-24.

Wendt, Alexander. (1987) The Agent-Structure Problem in International Relations Theory. International Organization, 3:335-370.

Wendt, Alexander. (1999) Social Theory of International Relations. Cambridge: Cambridge UP.

Yanow, Dvora. (2006) Thinking Interpretively Philosophical Presuppositions and the Human Sciences. In Interpretation and Method, edited by Dvora Yanow and Peregrine Schwartz-Shea, Armonk, NY: M. E. Sharpe.