Opening the Space Between Innocent and Oppressive Ways of Knowing: Challenges and Opportunities in Doing Research with Diverse Communities

© Currents: New Scholarship in the Human Services, 2008 ________________________________________________________ Opening the Space Between Innocent and Oppressive Ways of Knowing: Challenges and Opportunities in Doing Research with Diverse Communities Purnima Sundar and Sarah Todd Abstract The growing ethno-racial diversity reflected in Canadian society has prompted increased academic interest, particularly in the field of social work, in understanding how people from different ethno-racial groups experience and perceive the world. In this paper, we talk about the challenges of creating such knowledge, or engaging in “crosscultural research”. We focus this discussion on three main dimensions of the research process: the goals and values underlying the research; the nature of knowledge negotiated in the research relationship; and the way that power is structured in the researcher/participant relationship. We begin by describing each of these with reference to “traditional” ways of doing cross-cultural research, and articulate how such approaches work to sustain the colonialist project. We then discuss the growing trend towards using “Participatory Action Research” (PAR) as an alternative approach to conducting research in the social sciences that is respectful, liberating, and geared towards social change. We suggest, however, that the idealization that PAR can somehow create an innocent or non-oppressive space for research is an illusion. Instead, we draw on the work of post-structural educator Elizabeth Ellsworth (1997) to enrich existing work on PAR approaches to cross-cultural research. Introduction Even before the first colonizers arrived, Canada was a multicultural nation reflecting a range of Aboriginal communities with different cultural experiences and values. As European settlers, and by the early 1600s, African peoples, made Canada their home, the nation became even more heterogeneous (Thomas-Bernard & Moriah, 2007). This trend of diversification continues today. By the year 2001, 18.4% of the Canadian population was born outside of the country, the largest proportion in over 70 years (Statistics Canada, 2002). In addition, the increased number of newcomers from non-European source countries like Asia, Africa, South and Central America, and the Caribbean has had the effect of transforming Canada’s racial composition quite dramatically (Foster, 1998). Sundar and Todd 2 The changing ethnic, racial, and cultural makeup of the country has presented Canadian society with many positive opportunities, including the emergence of a strong “multicultural” identity created by official policies at the federal level (Jansen, 2005). Such a shift, however, has introduced challenges to mainstream social service agencies that are striving to create culturally appropriate and relevant services to meet the needs of all community members, including those from diverse ethno-racial and cultural groups (Herberg, 1993). In the field of social work, then, there has been growing interest in understanding how people with diverse ethnic/racial/cultural backgrounds experience and perceive the world so that knowledge can be used to provide theoretical guidelines for the provision of services across groups (Maiter, Trocmé, & Shakir,1999). In this paper, we talk about the challenges of this relationship with knowledge, through a discussion of “cross-cultural research”. We focus this interrogation on three main dimensions of the research process: the goals and values underlying the research; the nature of knowledge negotiated through the research process; and the way that power is structured in the researcher/participant relationship. We begin by describing each of these with reference to “traditional” ways of doing cross-cultural research, and articulate how such approaches work to sustain the colonialist project. We then discuss the growing trend towards using “Participatory Action Research” (PAR) as an alternative approach to conducting research in the social sciences that is respectful, liberating, and geared towards social change. We suggest, however, that the idea that PAR can somehow create an innocent or non-oppressive space for research is an illusion, and propose that the contributions of post-structural educator Elizabeth Ellsworth (1997) could enhance PAR approaches to cross-cultural research. Knowledge Creation and “Traditional” Cross-cultural Social Work Research Cross-cultural research is understood as a body of work that compares personal, social, and group processes between different ethnic/racial/cultural groups. While such research has a long history in anthropology and psychology, these studies have only recently started to surface in the field of social work (Rubin & Babbie, 2001). Prior to the 1960s, social work academics and practitioners assumed a commonality of needs across clients, and viewed attending to ethnoracial and cultural differences as only minimally important in the provision of services (Al-Krenawi & Graham, 2003). Social work interventions, then, were designed in accordance with the perceived needs of mainstream service-users. Although the role of one’s ethnoracial and/or cultural background in shaping their experiences with © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 3 social work services had been acknowledged for some time, a focused attempt to study ethnicity, race, and culture only began in the 1960s (Tsang & George, 1998). Over the last several decades, events like the Civil Rights Movement in the United States and the introduction of an official Multicultural Policy in Canada have underscored the importance of looking at how people’s ethno-racial and cultural roots affect the way they move through and experience the world (Tsang & George, 1998). These aspects of our selves, along with our class, gender, and sexual orientation shape our social location in ways that influence both how we are understood and how we make sense of others. Correspondingly, research studies that focused on “cross-cultural social work” began to emerge in the 1970s and 1980s in order to 1) develop theories explaining the experiences and needs of people with diverse backgrounds, and to 2) shape social work practice with minority individuals and families. Much of this early literature appears to have been rooted theoretically in a modernist paradigm that assumes that as researchers, we can communicate a particular “truth” by observing the world and describing the basic properties of the social universe (Ritzer, 1992; Rubin & Babbie, 2001). From this perspective, the study of ethnicity, race, and culture, requires that we take for granted that ethnicity, race, and culture are essential characteristics shared by all members of a particular group (Dean, 2001), and that these are stable constructs that are understandable through the application of “objective” techniques. Given that all social workers were assumed to have white, anglosaxon ethno-racial and cultural backgrounds, this knowledge could be used to assist them in dealing appropriately with members of diverse (i.e., non-white, non-anglo-saxon) groups. The goal of such studies, then, has been to (as neutrally and objectively as possible) measure a series of pre-determined outcomes that were understood to be important in shaping service provision (see, for example, Burger, 1972; Fishman, 1979; Waring & Kosberg, 1978; Brownlee, 1978). The researcher developed a set of tools (usually quantitative and standardized), and devised a particular method aimed at proving or disproving a hypothesis in the least biased and most unobtrusive way possible. The researcher, then, worked to gain knowledge from a “subject” for the benefit of the academic and professional community, with secondary benefits for the service users. Great emphasis was placed on “scientific” values such as rationality, neutrality and objectivity, since (consistent with the modernist paradigm) these were considered the means by which accurate knowledge could be apprehended (Tarnas, 1991). The researcher was assumed to be “culture-less” and “race-less” (or at least able to control the extent to which culture and race impinged on the research activities), which helped in safeguarding her/his © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 4 impartiality. Such research tended to produce a particular kind of knowledge, one which was typically based on an understanding of culture as fixed and unchanging, and which distinguished between those with culture (read “people of colour”) and those who were culture-less (read “white”) (Dean, 2001; Park, 2005). Culture was understood as a construct that could be reduced to a measurable “thing” which could then be used as the frame or lens through which the behaviours and experiences of “others” was comprehensible (Fanon, 1967). This knowledge was used to create models of service provision to guide white people’s work with people from “other” backgrounds (e.g., Adams, 1980; Sartorius et al., 1980). In much of this work, the focus on ethnicity and race also erased other aspects of identity such as class, gender, sexuality and ability that created diversity and commonality among clients and workers. White, middle class, heterosexual, Anglo-Saxon norms and values were viewed as the standard, and those from diverse backgrounds who deviated substantially from this standard were seen as needing to be taught how to behave like members of the mainstream (Park, 2005). Philosophers Edward Said (1978) (who wrote on the colonization of the East by the West) and Frantz Fanon (1967) (who discussed first-hand experiences with colonization by the French) argue that controlling the content, structure, and distribution of knowledge safeguards the power of those in control, and keeps those who are marginalized oppressed. Knowledge and conceptions about the “truth” reflect the interests of those who are engaged in both its production and transmission. In other words, those who control society’s knowledge create and reinforce the rules by which people are expected to live. This knowledge never challenges taken-for-granted assumptions about reality, but instead names and describes what is important in society (i.e., qualities that those in power tend to posses), which in turn legitimates and sustains the inequalities that arise from the uneven distribution of wealth and resources. The particular arrangement of power that characterized the researcher/participant relationship in “traditional” cross-cultural research reflects this imbalance. As the “expert”, the researcher was positioned as “knower”, and more powerful in relation to the “subject”. Ethnicity, race, and culture were used as indicators of different positions in a social hierarchy that understood the white majority as the reference point against which all “others” were judged. The salience of race and culture obscured all other aspects of one’s social location. Within this context, the role of the mainstream researcher was to understand the problems of the “culturally different” (Sue & Sue, 1999) in order inform practitioners (also assumed to be members of the majority) about how to resolve and/or control these issues. Again, the social worker (constructed as a © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 5 benevolent “knower”) was seen as being able to save the “cultured” individual who does not “know” as well, and cure her/him from those problems (which were likely related in some way to her/his ethnoracial/cultural background) (Park, 2005). Traditional ways of doing cross-cultural research in social work have been criticized for producing several problematic outcomes. The goals and values underlying such studies, the way knowledge is negotiated, and the unequal power arrangement that is reflected in the researcher/participant relationship all contributed to creating a certain way of thinking about the experiences of people from diverse ethnoracial and cultural backgrounds. First, models of service provision based on such research tend to otherize diverse clients. A dichotomy is created in which those who do not reflect the mainstream are constructed as “other”, and seen as inherently pathological and inferior to the white, Euro-Western norm (Anzaldúa, 1985; Bannerji, 1995; Fanon, 1967; Herberg, 1993; Tsang & George, 1998; Park, 2005). Indeed, textbooks with titles like “Counseling the Culturally Different: Theory and Practice” (Sue & Sue, 1999) and “Community Organizing in a Diverse Society” (Rivera & Erlich, 1998) reinforce this notion of a white, “culture-less” norm from which those who are “culturally different” deviate. This “traditional” approach to research was founded upon the specter of the white Anglo-Saxon, typically middle-class social worker who works with clients who differ from her/himself ethnically, racially, culturally and/or in terms of class. The idea that the social worker her/himself may in fact be a racialized or cultural “other” was generally not considered; in such cases, s/he was required to simply leave any ethnic/racial/cultural baggage at the door. Second, this approach to research has been used to develop practice models that rely on the consistency of certain attributes across members of a group. The result is a homogenizing of all of these “others” into one group that shares the collective identity of being different from the mainstream solely in terms of ethnicity, race or culture. In other words, all ethno-racial and culturally diverse groups are lumped together without acknowledging the differences that might exist both within and across diverse groups (Maiter et al., 1999). Despite “cautionary notes” suggesting that practitioners should always take into account individual differences, such research has provided a foundation for the development of standard guidelines that can be used by social workers who encounter those who are “culturally different” (e.g., African Americans, Native Americans, Latinos, Asians, and even mixed race people etc.) in their practice (e.g., Dhooper & Moore, 2001). Finally, this type of research has produced interventions that are based on an essential version of people from diverse groups, one that views them only in terms of their ethno-racial and cultural © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 6 characteristics and ignores other critical elements of their identities that might also play an important role in shaping their circumstances (Dean, 2001; Fanon, 1967). Identity dimensions such as class and gender, for example, can contribute to personal challenges that are largely determined at a structural level (Mullaly, 2007). “Traditional” forms of cross-cultural social work research have been instrumental in moving the field towards 1) acknowledging the importance of race, ethnicity, and culture in shaping people’s lives and experiences, and 2) attempting to address how these might impact service provision. Reflections on the challenges of doing this type of work, however, have typically focused on issues related to the process of carrying out the research (e.g., how best to access potential “subjects”, how to create designs and tools that are relevant to different cultural groups) (Letiecq & Bailey, 2004). More recently, however, there has been growing attention to the limits of this type of research for social workers seeking to produce projects that consistent with one of the field’s key objectives: to effect social change (Rubin & Babbie, 2001). In fact, we offer that whether intentional or not, it is clear that “traditional” approaches to cross-cultural research work to sustain the colonialist project. There is, however, an approach that has grown to represent a more viable alternative: participatory action research. Contemporary Practices of Research with Diverse Communities: The Participatory Action Research Approach Today, cross-cultural research in social work continues to work towards providing practitioners with knowledge, skills, and guidelines for becoming more “culturally competent” in their practice (Este, 1999; Lum, 1999; Williams, 2006; Yan & Wong, 2005). Such research, however, has grown to include an increased focus on the impact of racism and oppression on diverse groups, and efforts to explore how people can be supported in responding to these challenges (e.g., see Al-Krenawi& Graham, 2003; Este, 2007; Fong & Gibbs, 1995). While examples of research reflecting modernist objectives continue today (e.g., Chang et al., 2006; Ying, 2005; Arnsberger, 2005; Sung, 2004; Ben-Ari & Lavee, 2004), the range of “acceptable” methodologies has been expanded beyond hypothesis-driven, quantitative studies to include more qualitative, inductive research (see, for example, Lidchi, 2006). Given the limitations of traditional approaches described in the previous section, a progressive alternative has emerged as influential in shaping the direction of cross-cultural social work research. The “Participatory Action Research” (PAR) approach is an applied, collaborative methodology that works to ensure that those who are affected by the process and findings of the © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 7 research are involved at each stage of the investigation. With roots in critical theory, PAR pursues explanations that incorporate social, political, and historical conditions into the description of human systems, and adopts a dialectical view of society (Cresswell, 1998). Its purpose is not only to create knowledge, but to use it to transform individuals and social structures that have historically been dominated, alienated, or rendered powerless as a result of racism, sexism, classism and so on. Thus, PAR is a political project with the goal of reshaping power relations in society in a more equitable, socially just way. According to Nelson et al., (1998), PAR is the result of the fusion between participatory research and action research. Participatory research first emerged in the 1960s from the work of educator Paulo Freire and his colleagues in Brazil. People with limited or no access to power (in this case, poor peasants who were being oppressed economically and socially by Latin American elites) were supported in participating fully in a critical analysis of their situation in order to organize and act to ameliorate their circumstances (e.g., Hope & Timmel, 1987). Its history is deeply rooted in practices of adult education. However, over the past number of decades Freire’s participatory approach to knowledge creation has spread throughout various social sciences including social work. Action research was first introduced in the 1940s by German-born American psychologist Kurt Lewin, who believed that the greatest way to learn about social systems was to engage in concerted efforts to change them. Action research has also held appeal for those in a wide variety of disciplines who are interested in facilitating social change. The blending of these two approaches has resulted in PAR, which is “a research approach that consists of the maximum participation of stakeholders, those whose lives are affected by the problem under study, in the systematic collection and analysis of information for the purpose of taking action and making change” (Nelson et al., 1998: 885). Willms (1997) refers to PAR as “based on (a) liberating understanding of the nature of inquiry…(where) individuals and groups (research) their personal beings, social-cultural settings and experiences” (p. 7-8), all in an attempt to achieve social and political justice. The overall goal of PAR is two-fold: 1) to encourage the full participation of those most affected by the process and findings of the research in all aspects of the investigation, and 2) to stimulate social and political change (Fals-Borda & Rahman, 1991). As Susan Smith (1997) notes, this involves education, the development of consciousness and mobilization for action. Typically, such studies emphasize the process of the research (rather than simply its outcomes), and rely on multiple methods (both quantitative and qualitative) to address the problem under study. In addition to © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 8 effecting social and political change, a secondary product of the research is mutual learning (Hope & Timmel, 1987). Rather than the researcher gaining information from the “subject”, both individuals work in dynamic ways to discover a shared reality together. When working with diverse groups, then, the benefits are expected to extend to both the researcher(s) and group members equally (Maguire, 1987). While efforts to preserve the researcher’s neutrality and objectivity were paramount in “traditional” cross-cultural studies, within a PAR framework, values that emphasize empowerment, supportive relationships, social change, and ongoing learning are critical (Nelson et al., 1998). In using a PAR approach in research with diverse groups, the hope is that as “co-researchers”, participants (who can often experience marginalization and little access to power as a result of their ethnicity, race, and culture) would feel an increased sense of empowerment (Fals-Borda & Rahman, 1991). Through collaboration and the building of partnerships, the PAR processes anticipate the development of egalitarian and authentic relationships between those involved in the project. In addition, those adopting this approach trust that the research process will effect social change by generating useful knowledge and bridging the gap between this knowledge and action (Nelson et al., 1998). Finally, by emphasizing risk-taking and the merit of “failing forward”1, the hope is that PAR will result in ongoing learning for both researchers and participants (co-researchers). Here, the previously “culture-free” researcher is encouraged to embrace her/his culture and understand its contribution to the process of the research. The PAR approach facilitates a particular relationship with knowledge. Consistent with Freire’s (1970) notion of “conscientization” (as the process of gaining critical awareness, learning about one’s strengths, and tapping into personal power) the hope is that the research process yields knowledge that is critical of existing systems and helpful in effecting social change. While this knowledge has both intellectual and personal benefits for the researcher, those working from this perspective aim to ensure that there are greater benefits (i.e., liberation, change) for the participants (co-researchers). When working with diverse groups, then, the relationship that people have with knowledge is reworked from being a source of oppression, into an instrument of liberation (Hope & Timmel, 1987). While power is arranged unevenly in the researcher/participant 1 The notion of “failing forward” is introduced by Kathryn Church, who drawing on her work with psychiatric consumer/survivors, discusses how using a PAR process allows participants the freedom to take important risks and make mistakes; these provide great opportunities for learning (Church, 1997). © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 9 relationship in “traditional” cross-cultural studies, within the PAR framework the goal is to create a research setting in which not only are both the participant and researcher equal, but both work together to share knowledge and effect change (Patton, 2002). While the researcher certainly has particular valuable skills that s/he brings to the research, the contributions of participants to the research process are considered to be equally important. In studies with diverse groups, sustaining this equality is seen as key in disrupting the “expertsubject” dynamic. In particular, the superiority and power of the “culture-less”, mainstream researcher over the “cultured” subject is challenged, and replaced with a relationship that focuses on inclusion, participation, and reflexivity (Reid et al., 2006). The Limits of Participatory Action Research Clearly the PAR approach has generated a significant shift in the field of social work research. PAR has challenged the purposes for which social science research is employed, and has challenged researchers to work with a community to collect information relevant to the group itself, in order to create progressive change in their lives and across society (Fine & Torre, 2006). In addition, this approach has encouraged us to not only identify and make visible the values that guide our research, but has helped us to celebrate the way this shapes the questions we want to ask and the ways we seek to answer them. Finally, it has done a great deal to challenge the expertise and authority of the researcher, and resituate participants as experts, not only in their experience but also in the research of their experience (MacGuire, 1987). There is a sense that PAR is perhaps the most progressive approach to research, and is therefore somewhat insulated from critique. We suggest, however, that PAR has not managed to resolve several of the key tensions involved in the practice of cross-cultural research. So, while we respect the important achievements of the PAR approach and its research practices, we describe three concerns that remain in attempts to produce cross-cultural research, even when PAR methods are employed. The first tension we identify is related to the goals and values underlying the research. For the goals and values of PAR to be fully realized, the community and its membership should be somewhat established, and the relationship between researchers and “coresearchers”/participants’ needs to be cultivated to ensure trust and genuine involvement. This can be a lengthy and involved process that may sometimes take several months or even years. In contemporary Western society where the vast majority of research is funded and managed through the state, universities and non-profit agencies, those using PAR often find themselves reworking PAR to fit the context in © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 10 which the research is happening. What remains unclear is how much adaptation is possible before the essence of PAR is lost. The tensions and problematics that emerge from these compromises, and the doubts we might have about the extent to which our research is truly egalitarian and participatory remain only in private accounts, removed from public discussion and documentation (Cooke & Kothari, 2001). Despite many good intentions, then, the goals and values of PAR (which are presented as inviolable) are clearly vulnerable to larger systemic influences that make it difficult to preserve the spirit of this approach. Although we might be well-intentioned researchers with a clear commitment to contributing to liberatory rather than colonialist research efforts, we are still human beings with sometimes contradictory goals and behaviours. When we make claims to virtue and do not allow acknowledgement of the complex nature of our intentions and actions, we do not allow ourselves the freedom (and in fact, the responsibility) to identify and process these sometimes competing values and goals. The second difficulty in viewing PAR as an “innocent” solution to the problems posed by traditional approaches to cross-cultural social work research relates to the nature of the knowledge negotiated in the research relationship such studies. While many of the theorists whose work forms the basis of PAR were cautious about the dangers of this approach, the model is sometimes employed in such a passionate way as to assume its mere use ensures resistance to oppression. A true PAR process is implicitly understood to result in the creation of knowledge that is an accurate reflection of the thoughts and experiences of a community’s members. Regardless of the specific topic, the goal of such an approach to cross-cultural social work research is to unsettle the dominant (white) narrative by foregrounding the stories of racially/culturally/ethnically diverse people, in their own words. The effect is assumed to be a “real” or “true” depiction of their lived realities, which can in turn help to shape service provision. The problem with this idea is that PAR relies on the belief that the stories and experiences and activities of the marginalized are “recognizable truths”. Joan Scott (1992) provides the following critique of such an approach to knowledge: When experience is taken as the origin of knowledge, the vision of the individual subject (the person who had the experience or the historian who recounts it) becomes the bedrock of evidence upon which explanation is built. Questions about the constructed nature or experience, about how subjects are constituted as different in the first place, about how one’s vision is structure… are left aside (Scott, 1992: 25). © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 11 Depictions of members of marginalized communities are structured through dominant narratives, reflected in and woven through the webs of the power relations that shape our society. Their stories and active involvement, then, not only need to be understood as entangled in dominant narratives, but also must be viewed as having been shaped by the local power relations that are easily obscured when focusing on the macro level relationship of oppressor and oppressed. In other words, even those who speak and act in ways that are not, at first glance, supported by macro power relations, are not outside of these relations and may, in fact, enact exclusionary positions within local interactions. The final tension we identify here centers on the belief that PAR is inherently advanced and thus able to transcend (or at least manipulate) the power relations that shape participation and exist between researchers and participants. The very assumption of inclusivity that underpins PAR presents two problems. First, it often obscures the internal dynamics of communities in which some members are more articulate, authoritative, confident and/or accepted than others. As a result, their views are often the ones that are captured through participatory processes, while those that are considered unwelcome or are actively or passively silenced in the community remain absent. This can have the effect of reifying the agendas of more powerful actors in the community (Kothari, 2003). The second problem is that non-participation as a healthy, selfaffirming choice becomes impossible. The more inclusive our processes, the easier it is for us to see non-participation as apathy or uninformed, rather than as a legitimate and valuable practice of resistance (Kothari, 2003). Inclusive processes, in and of themselves regulate knowledge and behaviour; PAR does not get outside of power relations, but reproduces them in new, and arguably equally dangerous ways. In addition, local power relations, such as those that emerge between the researcher (who retains vestiges of the “expert” identity) and the participant (whose experiences serve as the object of interest) can have just as significant an impact on shaping people’s lives as macro relations, and are often obscured by focusing on those structures that shape a community from the outside. As Cooke and Kothari (2003) warn “‘local knowledge’ far from determining planning processes and outcomes, is often structured by them” (8). PAR relies on the notion of equality between researcher and “coresearcher”/participant, and views any problems that emerge around power inequalities as resolvable with the application of the correct skill or technique. Kothari (2003) refers to this as “tinkering” with participatory practices. A serious interrogation of this dynamic, however, acknowledges that as academics, researchers serve as a © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 12 “conduit” between people with power and those without. Therefore, they play a critical role in the PAR process, one that places power squarely in their hands in relation to research participants. Specifically, researchers are still the ones who are responding to a certain research agenda (determined by their field), and taking the lead on deciding which research questions to ask and how they should be asked, choosing the framework for analysis and writing up the results. Even with the best of intentions PAR projects do not happen in a vacuum; they are within a context that, in itself, determines the process. These dynamics are not removed just because we use a more collaborative style of working with co-researchers/participants. Despite the emphasis on knowledge as being a tool for liberation and social change, we suggest that it is possible that PAR does not lead to freedom, but rather provides the democratic window dressing to institutions (be they state, non-profit, transnational) that do not tend to respond effectively to the interests and needs of their most vulnerable members (i.e. racialized bodies and the poor). By looking at the goals and values of PAR projects, the type of knowledge produced, and the continued power differences that characterize the researcher/participant relationship, there appears to be a gap between our intentions and what actually happens in the research process. For those of us who are uncomfortable with such a disjuncture, we suggest that it might be useful to look to the field of pedagogy, and draw on ideas introduced by post-structural educator, Elizabeth Ellsworth. Enriching PAR: Embracing Uncertainty and Discovering Possibilities for Justice in Research In her 1997 book, Teaching Positions: Difference, Pedagogy and the Power of Address Elizabeth Ellsworth draws upon her early work in film studies in an attempt to unravel some of the tensions and possibilities of pedagogy. We propose that the several of the concepts she finds useful for thinking through pedagogy are helpful for understanding PAR in relation to cross-cultural research. As someone who is interested in adult education and social change, Ellsworth’s work is consistent with the ideas of PAR theorists, but her poststructural focus drawing upon film studies provides a new perspective to these ideas. She suggests that we embrace spaces of disconnect, and rather than seeing them as gaps that must be reconciled, view them as opportunities for alternative ways of being to emerge. In line with this, we suggest an opening of the space between innocent and oppressive ways of knowing to help us to acknowledge our limitations and work towards responding to them, rather than abandoning them entirely or imagining them as resolved. In her work, Ellsworth (1997) explores how modes of address, or who the film thinks you are, is always slightly off its mark, “the © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 13 viewer is never only or fully who the film thinks s/he is” (26). Here, in addition to observing how modes of address in film and pedagogy misfire, she reminds us that there is no essential “who” that can be correctly aimed at with a mode of address. Ellsworth (1997) draws on the helpful example of a feminist film scholar, Judith Mayne, who enjoys Arnold Schwarzenegger films. Those who made and promoted the film never imagined her as their audience, and she doesn’t imagine herself as those who the film is aimed at, but there is a space between the film’s intent and the viewer’s response that is unpredictable and unknowable. Our selves, desires and interests are often contradictory and in excess of whom we, and others, imagine ourselves to be. Ellsworth (1997) argues against the practice within the social sciences that strives to secure a more “accurate” mode of address, and instead suggests that all modes of address miss their mark. More importantly, she maintains that it is in the space of this disconnect that possibilities for agency emerge. In referring to pedagogy she suggests that it is a much messier and more inconclusive affair than the vast majority of our educational theories and practices make it out to be….what saves pedagogy from being completely closed, permanently othering, lifeless, passion killing, and perverse in the sense of already knowing what is best for us (Phillips, 1993, p. 108) is that the pedagogical relation itself is unpredictable, incorrigible, uncontrollable, unmanageable, disobedient (Ellsworth, 1997: 8-9). In her reflections on pedagogy, Ellsworth (1997) embraces uncertainty. Borrowing from her work, we suggest that the productive possibilities of PAR grow out of the following paradoxes that Ellsworth (1997) imagines in pedagogy: there is no “certainty about what consequences our actions” as researchers will have; that it is impossible to “designate what actions or knowledge” is needed; that PAR when it “works” is unrepeatable and cannot be copied; and that PAR is a “performance that is suspended (as in interrupted, never completed) in the space between self and other” (Ellsworth, 1997: 17). In the following paragraphs we explore what these uncertainties mean in relation to the specific tensions we explored earlier in this paper. Rethinking the Goals and Values Guiding Our Research Following Ellsworth’s (1997) approach, we argue that as social work researchers looking at issues like ethnicity, race, etc., we may be wise to shift our focus away from searching for that “certain methodology” that works to control the cross-cultural encounter that happens in research. Instead, we suggest (as Ellsworth does) that we would benefit from addressing research participants and audiences “in a way that doesn’t require them to assume a fixed, singular, unified position within power and social relations” (Ellsworth, 1997: 9). This shifts the © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 14 goals and values of PAR. Rather than having to account for our work as a practice of liberation that can be known in advance, this revised project works towards social justice, but sees the process of getting there as far more uncertain. The uncertainty an inevitable failure in achieving justice in all moments can be understood as the very moments in which the agency of participants can emerge (as resistant) and, in the unpredictability of the process, justice as an ideal which is unknowable in advance, emerges. Respecting Participants as Unknowable Traditional cross-cultural research imagined persons with non-white, non-Anglo heritage as “other”, non-knowing objects who could be examined, categorized, known and controlled. Conversely, PAR envisions racialized persons as active, engaged people who are oppressed. In the previous section, we discussed how this new, libratory mode of address often misses its mark, and can be obscuring, reifying, and unstable across context or temporal periods. Essentially, we suggest that there is likely a disconnect between how PAR imagines participants and how participants imagine themselves. As a process of creating these subjective realities, PAR is always an unfinished project, with both participants and researchers failing at being the type of subjects imagined by this approach. To address this, we suggest that it is within this space of ambiguity that radical subjectivity is possible; it is here that both participants and researchers are able to resist the objectifying tendency of all research processes, including PAR. There isn’t another, better strategy to offer to close the gap; even if there was, it should be avoided for to close this gap would be to lose the possibility of resistance. PAR as a Relationship Ellsworth’s (1997) analysis is also helpful in making visible the ways in which PAR is not an object, but exists instead as a relationship. It is the ongoing interaction of a number of aspects of the research project’s “form, style and narrative” (Ellsworth, 1997: 39). Competing modes of address, including those of academic institutions, funders, and more personal and historical stereotypes, as well as each actor’s fears, hopes and desires shape the relationship through which PAR is constituted. By rethinking PAR as a relationship between participants, researchers, discourses and institutions, we are able to attend to those situations in which the relationship is impossible. We can begin to question the liberal discourse of participatory methods that suggests that tensions can be resolved through rational dialogue or tinkering with skills and techniques. As Ellsworth (1997) suggests, engaging in a dialogue across difference is troubled by “cognitive © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 15 uncertainty, forbidden thoughts, unreliable and unstable perceptions” (42). The path to unsettle structural relations is, in itself, fraught. The encounter with the “other” cannot be cleaned up, but instead remains messy. Thus the impossibility of the project is not something we have to resolve, but is only something we can attend to in our relationships with one another. When PAR is considered to be a relationship, it is no longer an object that can be reshaped to fit any context, but is a dialogue that must always be engaged. Conclusions By rethinking the goals and values that shape our research, challenging the extent to which we might ever produce “true” or “real” knowledge, and by reconstituting PAR as a relationship instead of a set of practices designed to ensure liberation, we suggest that Ellsworth’s (1997) analysis provides us a pause from which we can re-approach PAR. What Ellsworth (1997) ultimately argues for is a genuine respect that the world and our fellow human beings are always more complex than our processes and interventions can possibly imagine. While PAR may resolve the tensions in one mode of address, it opens up a series of others that when closed would just provide new openings and disconnects. With this assurance in mind, we are able to reconsider the possibilities of PAR in cross-cultural research and acknowledge that it does not and (in fact cannot) transcend the unequal power relations that over-determine encounters with “others”. Instead of searching for the innocent path for such work, we can begin to consider what can emerge if we stay with PAR, just “sit with it”. By acknowledging these tensions and possibilities, we open the space between “innocent” and oppressive ways of coming to understand “the other”; we can become less invested in the surety of the outcomes of such research endeavors, and more able to critically reflect on our practices. In the end this further strengthens the hopes that PAR theorists have for working towards ways of knowing that encourage “an ever-deepening understanding of the many complexities of reality” (Smith, 1997, p. 176). References Adams, J. (1980). Service arrangements preferred by minority elderly: A cross-cultural survey. Journal of Gerontological Social Work, 3(2), 39-57. Al-Krenawi, A. and Graham, J. (Eds.) (2003). Multicultural social working in Canada: Working with diverse ethno-racial communities. Don Mills, ON: Oxford University Press. Anzalduà, G. (1985). Speaking in tongues: A letter to Third World women writers. In G. Anzalduà and C. Moraga (Eds.), This © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 16 bridge called my back: Writings by radical women of colours (2nd Ed.), pp. 165-174. New York: Kitchen Table: Women of Colour Press. Arnsberger, P. (2005). Best practices in care management for Asian American elders: The case of Alzheimer’s disease. Care Management Journals, 6(4), 171-177. Ben-Ari, A. and Lavee, Y. (2004). Cultural orientation, ethnic affiliation, and negative daily occurrences: A multidimensional cross-cultural analysis. American Journal of Orthopsychiatry, 74(2), 102-111. Bannerji, H. (1995). The sound barrier: Translating ourselves in language and experience. In H. Bannerji (Ed.), Thinking through: Essays on feminism, Marxism, and anti-racism, pp. 159-186. Toronto: Women’s Press. Brownlee, A. (1978). The family and health care: Explorations in cross-cultural settings. Social Work in Health Care, 4(2), 179198. Burger, H. (1972). ‘Ethno-lematics’: Evoking ‘shy’ Spanish-American pupils by cross-cultural mediation. Adolescence, 7(25), 61. Chang, I., Pettit, R., and Katsurada, E. (2006). Where and when to spank: A comparison between US and Japanese college students. Journal of Family Violence, 21(4), 281-286. Church, K. (1997). Because of where we’ve been: The business behind the business of psychiatric survivor economic development. Community Development Corporation, Toronto, ON, Canada. Cooke, B. and Kothari, U. (2001). Participation: The new tyranny? New York, NY: Palgrave. Creswell, J. (1998). Qualitative inquiry and research design. Thousand Oaks, CA: Sage Publications. Dean, R. (2001). The myth of cross-cultural competence. Families in Society, 82(6), 623-630. Dhooper, S. & Moore, S. (2001). Social work practice with culturally diverse People. Thousand Oaks, CA: Sage Publications. Ellsworth, D. (1997). Teaching positions: Difference, pedagogy and the power of address. New York, NY: Teachers College Press. Este, D. (2007) “Cultural competency and social work practice in Canada: A retrospective examination” Canadian Social Work Review, 24(1), 93-104. Este, D. (1999). Cultural competency and social work: an overview. In G. Yong-Lie and D. Este (Eds.), Professional social services in a multicultural world (pp. 27-45). Toronto, ON: Canadian Scholars’ Press. Fals Borda, O & Rahman, M (Eds.) (1991). Action and knowledge: Breaking the monopoly with participatory action research. New York, NY: The Apex Press. Fanon, F. (1967). Black skin, white masks. New York, NY: Grove © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 17 Press. Fine, M. and Torre, M. (2006). Intimate details: Participatory action research in prison. Action Research, 4(3), 253-269. Fishman, R. (1979). A note on culture as a variable in providing human services in social service agencies. Human Organization, 38(2), 189-192. Fong, L. & Gibbs, J. (1995). Facilitating services to multicultural communities in a dominant culture setting: An organizational perspective. Administration in Social Work, 19(2), 1-24. Foster, L. (1998). Turnstile immigration: Multiculturalism, social order, and social justice in Canada. Toronto, ON: Thompson Educational Publishing, Inc. Freire, P. (1970). Pedagogy of the oppressed. New York, NY: Continuum. Herberg, D. (1993). Frameworks for cultural and racial diversity: Teaching and learning for practitioners. Toronto, ON: Canadian Scholars’ Press. Hope, A. and Timmel, S. (1984). Training for transformation: A handbook for community workers, Book 1 (pp. 16-27). ITDG Publishing. Jansen, C. (2005). Canadian multiculturalism. In C. James (Ed.), Possibilities and limitations: Multicultural policies and programs in Canada. Halfax, NS: Fernwood Publishing (pp. 21-33). Kothari, U. (2003). Power, Kknowledge and social control in participatory development. In B. Cooke and U. Kothari (Eds.) Participation: The new tyranny? New York, NY: Palgrave (pp. 139-152). Letiecq, B. and Bailey, S. (2004). Evaluating from the outside: Conducting cross-cultural evaluation research on an American Indian reservation. Evaluation Review, 28(4), 342-357. Lidchi, V. (2006). Reflecting on researching the family biogram: The challenges and opportunities of working in violent contexts. Reflections: Narratives of Professional Helping, 12(1), 23-29. Lum, D. (1999). Culturally competent practice: A framework for growth and action. Pacific Grove, CA: Brooks/Cole. MacGuire, P. (1987). Doing participatory research: A feminist approach. The Centre for International Education, University of Massachusetts. Maiter, S., Trocmé, N., and Shakir, U. (1999). Fabricating the tools of resistance. Proceedings of Qualitatives ‘99. New Brunswick, Canada. Mullaly, B. (2007). The new structural social work (3rd Ed.). Don Mills, ON: Oxford University Press. © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 18 Nelson, G., Ochocka, J., Griffin, K., and Lord, J. (1998). “Nothing about me without me”: Participatory action research with selfhelp/mutual aid organizations for psychiatric consumer/survivors. American Journal of Community Psychology, 26, 881-912. Park, Y. (2005). Culture as deficit: A critical discourse analysis of the concept of culture in contemporary social work discourse. Journal of Sociology and Social Welfare, 32(3), 11-33. Patton, M. (2002). Qualitative evaluation and research methods (3rd Ed.). Newbury Park: Sage Publications. Phillips, A. (1993). On kissing, tickling, and being bored: Psychanalytic essays on the unexamined life. Cambridge, MA: Harvard University Press. Reid, C., Tom, A., and Frisby, W. (2006). Finding the ‘action’ in feminist participatory action research. Action Research, 4(3), 315-332. Ritzer, G. (1992). Sociological theory (3rd Ed.). New York: McGrawHill, Inc. Rivera, F. and Erlich, J. (1998). Community Organizing In a Diverse Society. Boston, MA: Allyn and Bacon. Rubin, A. and Babbie, E. (2001). Research methods for social work (5th Ed.). Austrailia: Wadsworth Thompson Learning. Said, E. (1978). Orientalism. New York: Vintage Books. Sartorius, N., Jablensky, A., Gulbinat, W., and Ernberg, G. (1980). Acta Psychiatrica Scandinavica, 62(285), 204-211. Scott, J. (1992). Experience. In J. Butler and J. Scott (Eds.) Feminists theorize the political. New York, NY: Routledge (pp. 22-40). Smith, S. (1997). Deepening participatory action research. In S. Smith, D. Willms, & N. Johnson (Eds.), Nurtured by knowledge: Learning to do participatory action research. New York, NY: The Apex Press, (pp. 173-264). Statistics Canada (2002). Ethnic diversity survey: Portrait of a multicultural society. www.statscan.ca. Sue, D. W., and Sue, D. (1999). Counseling the culturally different: Theory and practice, (3rd edition). New York, NY: John Wiley. Sung, (2004). Elder respect among young adults: A cross-cultural study of Americans. Journal of Aging Studies, 18(2), 215-230. Tarnas, R. (1991). The passion of the Western mind: Understanding the ideas that have shaped our world view. Toronto, ON: Random House of Canada Ltd. Thomas-Bernard, W. and Moriah, J. (2007). Cultural competency: An individual or institutional responsibility? Canadian Social Work Review 24(1), 81-92. © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008 Sundar and Todd 19 Tsang, A. and George, U. (1998). Towards an integrated framework for cross-cultural social work practice. Canadian Social Work Review, 15(1), 73-93. Waring, M. and Kosberg, J. (1978). Age, race, life conditions, use of social welfare services, and morale of the elderly. Journal of Sociology and Social Welfare, 5(6), 781-791. Williams, C. (2006). The epistemology of cultural competence. Families in Society, 87(2), 209-220. Willms, D. (1997). “You start your research on your being”. In S. Smith, D. Willms, & N. Johnson (Eds.), Nurtured by knowledge: Learning to do participatory action research. New York, NY: The Apex Press (pp. 86-110). Yan, M. and Wong, Y. (2005). Rethinking self-awareness in cultural competence: Toward a dialogic self in cross-cultural social work. Families in Society, 86(2), 181-188. Ying, Y. (2005). Variation in acculturative stressors over time: A study of Taiwanese students in the United States. International Journal of Intercultural Relations, 29(1), 59-71. Authors’ notes Purnima Sundar, PhD, Research and Knowledge Exchange Consultant, Provincial Centre of Excellence for Child and Youth Mental Health, 401 Smyth Road, Ottawa, ON, K1H 8L1. Address correspondence to lead author at psundar@cheo.on.ca. Sarah Todd, PhD, Associate Professor, School of Social Work, Carleton University, 1125 Colonel By Drive, Ottawa, ON K1S 5B6. © Currents: New Scholarship in the Human Services Volume 7, Number 1, 2008