Governing through community allegiance: a qualitative examination of peer research in community-based participatory research

Abstract

The disappointing results of many public health interventions have been attributed in part to the lack of meaningful community engagement in the planning, implementation, and evaluation of these initiatives. Community-based participatory research (CBPR) has emerged as an alternative research paradigm that directly involves community members in all aspects of the research process. Their involvement is often said to be an empowering experience that builds capacity. In this paper, we interrogate these assumptions, drawing on interview data from a qualitative study investigating the experiences of 18 peer researchers (PRs) recruited from nine CBPR studies in Toronto, Canada. These individuals brought to their respective projects experience of homelessness, living with HIV, being an immigrant or refugee, identifying as transgender, and of having a mental illness. The reflections of PRs are compared to those of other research team members collected in separate focus groups. Findings from these interviews are discussed with an attention to Foucault's concept of ‘governmentality’, and compared against popular community-based research principles developed by Israel and colleagues. While PRs spoke about participating in CBPR initiatives to share their experience and improve conditions for their communities, these emancipatory goals were often subsumed within corporatist research environments that limited participation. Overall, this study offers a much-needed theoretical engagement with this popular research approach and raises critical questions about the limits of community engagement in collaborative public health research.

Introduction

Power constantly asks questions and questions us; it constantly investigates and records; it institutionalizes the search for the truth, professionalizes it, and rewards it. (Foucault 2003a, 25)

The disappointing results of many public health interventions have been attributed in part to the lack of meaningful community engagement in the planning, implementation and evaluation of these initiatives. In an attempt to redress this situation, community-based participatory research (CBPR) has emerged as an alternative research, and evaluation paradigm for collaboratively addressing health disparities. CBPR and related approaches ‘emphasize the participation, influence and control by non-academic researchers in the process of creating knowledge and change’ (Israel et al. 1998, 184). The W.K. Kellogg Foundation's Community Health Scholar's program describes CBPR as:

… a collaborative approach to research that equitably involves all partners in the research process and recognizes the unique strengths that each brings. CBPR begins with a research topic of importance to the community with the aim of combining knowledge and action for social change to improve community health and eliminate health. (cited in Minkler and Wallerstein 2008, 6)

The benefits of community involvement in research are well documented. They include: greater representation of marginalized groups in research; data that are more representative of community needs; and increased opportunities for local capacity-building and empowerment (Israel et al. 1998; Higgins and Metzler 2001; Israel et al. 2005; Wallerstein and Duran 2006). What started as a marginal research movement has now entered into the mainstream with a growing number of researchers adopting the approach and receiving support from funders (Horowitz, Robinson, and Seifer 2009). This has resulted in a proliferation of productive community–university partnerships, demonstrated by an increase from one published article mentioning CBPR in 1987 to nearly 400 in 2010.¹ Overall, CBPR is premised on the need to challenge traditional research practices through more direct involvement with partners outside of the academe.

A growing number of CBPR projects are adopting a ‘peer research’ approach in which members of the target population are directly involved in the research process. In some cases, peer researchers (PRs) partner in all facets of a research project and are members of the core research team. In others, they are instrumental in one or more aspects (e.g. participant recruitment or data collection, or both). Peer research approaches have been used for researching diverse communities and health issues, including: adolescent sexual health (Flicker et al. 2010), HIV/AIDS (Greene et al. 2009; Logie et al. 2012), and drug use (Elliott, Watson, and Harries 2002; Coupland and Maher 2005). To date, there has been little critical discussion about the involvement of PRs in CBPR. Rather, the focus has been on the ethical dimensions of using peers and whether they are able to maintain ethical requirements when working in their own communities (Bean and Silva 2010; Constantine 2010; Simon and Mosavel 2010). Such critiques are not surprising considering that CBPR raises complex ethical issues (Minkler 2004; Munoz and Fox 2011). However, the dearth of data on peer research in practice has left unanswered questions regarding the authenticity of community participation, how power differentials are addressed, and how participation may affect community members in unanticipated ways (Roche 2008; Greene et al. 2009).

With these critical questions in mind, we conducted a study to examine CBPR projects that used a peer research approach. Our goal was to better understand how CBPR principles are being operationalized in these projects; we have reported elsewhere on the practical and ethical aspects of using a peer research approach (Flicker, Roche, and Guta 2011; Guta, Flicker, and Roche 2011; Roche, Guta, and Flicker 2011). Here, we shift our focus to offer a more theoretically informed engagement with our data. Specifically, the objective of this paper is to offer a Foucauldian-informed reading of accounts of being a PR, or of organizing and managing them. We take our lead from critical work in the participatory action research literature (Healy 2001; Fine 2006a) and international development literature (Cooke and Kothari 2001). Critical scholars in these two fields have expertly interrogated their own practices and exposed contradictions and tensions in their work. While CBPR shares some philosophical roots with both, we have not observed this same level of critical engagement. We attribute this, in part, to the greater influence of medicine and public health in the evolution of CBPR. Our aim is to contribute to greater theorizing and reflexivity in the CBPR literature.

Theoretical orientation

The growing inducement for communities and researchers to undertake CBPR may be explained through Foucault's (1980, 39) concept of ‘governmentality’ (the ‘conduct of conducts’) that accounts for the ways micro forms of power insert themselves into actions, attitudes, and discourses. Foucault (1978, 82) rejected traditional accounts of power as solely dominating and instead proposed an analytics of power equally attuned to its productive qualities. Foucault (2007, 108–109) offered the following definition of governmentality during his 1977–1978 lecture series:

First, by ‘governmentality’ I understand the ensemble formed by institutions, procedures, analysis and reflections, calculations, and tactics that allow the exercise of this very specific, albeit very complex, power that has the population as its target, political economy as its major form of knowledge, and apparatuses of security as its essential technical instrument … it is the tactics of government that allow the continual definition of what should or should not fall within the state's domain, what is public and what is private, what is and what is not within the state's competence, and so on.

Governmentality is a complex process through which rationalities based on notions of scientific truth, state sovereignty, economic calculations, and of the governed themselves, are in a perpetual state of overlap and struggle (Foucault 2010, 313). Foucault encouraged looking past obvious points of struggle and conflict towards the seemingly mundane or ordinary. Examining the government ‘of children, of souls, of communities, of families, of the sick’ makes it possible to identify and problematize taken for granted operations of power on a larger scale (Foucault 2003b, 138). Foucault (1988, 18) was further interested in the ‘technologies’ that serve to shape human knowledge and action, including the technologies of power (that serve to conduct and dominate) and technologies of the self (which individuals use to stylize themselves) to understand the effects of power and resistance. In other words, Foucault was concerned with the multi-dimensional ways that power manifests and inserts itself into quotidian actions, attitudes, and discourses.

The modern epoch is characterized by neoliberalism, a political rationality that offers new modes of relations between the ‘governors and the governed’ and extends economic analysis into a theory of human capital (Foucault 2010, 218–219). According to Foucault (2010, 226) within this logic, the subject becomes an ‘entrepreneur’ of its own productive and consumerist potential, with the effect of making it ‘eminently governable’ (270). Neoliberalism can be seen in economic policy that privileges free-market logic, a changing orientation to political power that reduced direct state intervention (i.e. the decline of the welfare state) and the emergence of new cultural ideologies associated with these changes (Centeno and Cohen 2012). Broadly, neoliberalism has re-oriented numerous aspects of life to fall within the realm of the economic to make them calculable, predictable, and governable. Issues like health and ‘public hygiene,’ previously considered outside the domain of economics, become rethought as central to improving human capital (Foucault 2010, 230). A number of scholars have used neoliberalism to explain changes to public health and health promotion practice in recent decades (Petersen and Lupton 1996; Bunton and Petersen 1997; Lupton 1999). The primary focus here has been the deployment of risk discourses and market approaches to managing health and health resources – the public's health becomes something to optimize through a growing range of interventions. These interventions must be low-cost and offer a high rate of return on investment to be considered successful within this market logic. Within neoliberal discourses, the individual's goals and aspirations become realized through the supposed ‘freedom’ to make rational choices and engage in self-improvement (Rose 2000). This logic inevitably benefits some more than others. Those individuals who are unwilling or unable to benefit then become marked as undeserving, unworthy, and even dangerous.

Often overlooked in these kinds of discussion are those forces and institutions that mediate the relationships between state policies and individuals, or that make neoliberalism operable at different levels. Foucault (2010, 296) explains that subjects are made manageable through shared economic and political interests realized through ‘civil society,’ which is itself a technology of government. For a brief period, this produced the postwar ‘welfare state’ and the promise of collective economic security. These protections have been steadily eroded in the wake of neoliberal restructuring, and have resulted in growing health inequities (Esping-Andersen 1996; Coburn 2000; Jessop 2002). Notions of political citizenship have been replaced with neoliberal notions of community, where ‘values, identities, allegiance, trust, and mutual dependence’ have become strategies of governance (Rose 2000, 1401). Miller and Rose (2008, 89) argue the ‘birth of the community’ marks a departure from a collective social body to recognizing difference that can be expertly governed:

Communities became zones to be investigated, mapped, classified, documented, interpreted … to be taken into account in numberless encounters between professionals and their clients, whose individual conduct is now to be made intelligible in terms of the beliefs and values of ‘their community’.

This process required the redistribution of the former ‘social’ into ‘communities,’ and resulted in a new sense of ‘ethical’ responsibility to one's community, and new forms of ‘identification’ to make and re-make forms of allegiance (Miller and Rose 2008, 91–93). Communities are now mobilized through participatory frameworks that make health education a shared endeavor to identify issues of importance and introduce notions of democratic decision-making between community and professionals (Gastaldo 1997, 115). Community-based organizations offer a range of health and social services and have become an intermediary between the state and individual subjects (McDonald and Marston 2002; Marwell 2004). Here, market rationalities have dramatically changed the ways service providers respond to client needs (Baines 2006, 2010; Rathgeb Smith 2010). In turn, service users have had to adopt new ways of thinking about their health and the health of others with whom they share a similar risk profile. Similar themes are promoted in CBPR, and are evident in the seminal work of Israel et al. (1998) which promotes the community as a unit of identity, various forms of capacity-building, and partnerships involving community members, organizational representatives and researchers. We consider the implications of this theoretical work as applied to CBPR, through a reading of empirical data on the process of mobilizing marginalized communities to address issues of health inequity through research. We will demonstrate that despite its emancipatory agenda, CBPR may inadvertently promote forms of governance that unintentionally advance neoliberal agendas.

Methods

The data presented here were collected as part of a multi-method exploratory qualitative study (a study of other studies) that sought to examine how popular CBPR principles were being taken up in individual projects using a peer research approach.

Sampling and recruitment

A purposive sampling strategy was used to identify projects that had been funded by the Wellesley Institute and used a peer research approach. When it provided seed funding, the Wellesley adopted the W.K. Kellogg Foundation's Community Health Scholar's Program's definition of CBPR, required applicants to demonstrate a commitment to involving community in their funding applications, and encouraged them to attend a popular series of workshops on CBPR covering conceptual, methodological, and ethical issues. Toronto is a hub of CBPR, with many community-based organizations partnering with the city's universities and hospitals, and it is an important research site to understand CBPR in Canada and beyond.

We began with a working definition of peer research as the process of involving members of the target population who are trained to participate as co-researchers. This definition functioned as an important starting point and reflected our observations as researchers engaging in and supporting community-based research at the time. In the course of this study, however, we learned that the definition of peer research and PRs roles shift according to context, the community of interest, the nature of the project, and the research team's understanding and commitment to CBPR.

We contacted project leads of previous Wellesley-funded studies to solicit their participation and aid in locating their current and former PRs. PRs were recruited for individual semi-structured interviews (n = 18) to discuss their experiences of involvement in CBPR studies. Some of the PRs did not have regular access to telephones or email for recruitment purposes. In these cases, we relied on service providers who had worked with PRs and were able to negotiate an interview time for us. PRs were provided with an honorarium and two transit tokens. The interviews lasted an average of 90 minutes.

The PRs who participated reflect a diverse group in terms of age, gender, sexual orientation, socio-economic status, and ethno-racial identity. We provide some demographic information about participants in Table 1. However, in many cases, linking individual's gender, age, history of mental/physical illness, or community affiliation to particular quotes could make participants identifiable. Consequently, we have provided limited information about participants. In this paper, they are identified only as PRs, and given gendered pseudonyms. The gender of the person speaking is represented with either an M or F next to their quote. Except in the demographic table, the four transgender participants are listed as they identified during their interview.

Table 1.

PR characteristics.

Gender
Male	5
Female	9
Transgender	4
Total	18
Community affiliation
Newcomer/refugee	4
Aboriginal	2
Member of a ‘racial/ethnic’ group (non-aboriginal)	6
Health/mental health diagnosis	4
Experience of homelessness	4
LGBT	6
Youth	3

Note:

^∗Categories are not mutually exclusive and do not add up to 18

Academic leads and community partners who adopted a peer research model in their projects were invited to attend one of two focus groups to identify and discuss methodological, ethical, and practical issues (n = 10). Most of those who participated worked as research managers or staff at non-profit agencies in Toronto that were broadly engaged in addressing the social determinants of health. The academic leads on these studies came primarily from the health sciences, but the teams they represented included representation from the social sciences. Demographic information was not collected from them. They are described as community-based researchers (CBRs) and given pseudonyms.

Data collection procedures

Interviews and focus groups were audio-recorded and transcribed verbatim for coding and analysis. For interviews with PRs, the questions explored: the participant's relationship to their community, motivations for becoming involved, their understanding of peer research, their role in the project, what they thought a peer approach contributed, and any recommendations they might have for other CBPR project teams considering this approach. For focus groups with CBRs, participants were asked similar questions, but framed in terms of their organization's experience with CBPR, why they had adopted a PR approach, issues that emerged from their perspective, and their recommendations. The majority of the individual interviews were conducted by the lead author, and a few were conducted by the third author. Interviews were conducted at the Wellesley Institute, with a few exceptions, including one interview conducted in a participant's home. The focus groups were led by the second author and conducted on-site at the Wellesley Institute.

We acknowledge the limitations of our small sample. It is likely not representative of the diversity of CBPR studies. Nevertheless, this exploratory qualitative work offers important insights into a range of issues that have not been adequately discussed in the CBPR literature. We do not claim that our findings are generalizable in the traditional sense, but borrowing from Fine (2006b), they offer provocative and theoretical generalizability beyond the limits of our sample and research context. Our goal is to encourage CBPR teams to not only celebrate the benefits of a PR approach but to also consider the possible challenges and consequences.

Analytic tools and strategies

The research team developed an initial coding framework based on the CBPR literature, team members' experiences conducting CBPR, and a reading of a subset of transcripts. Data were coded and used to develop major themes for subsequent comparison and analysis (Boyatzis 1998a, 1998b; Fereday and Muir-Cochrane 2008). An ‘audit trail’ of key decisions was kept about categories and themes (Rodgers and Cowles 1993; Koch 2006). We presented preliminary findings in presentations, workshops, and in a three-part community report series. Our initial focus was on the applied and practical elements of our data that could be taken up by research teams and used to inform their research practice to better support PRs. However, with some time having passed from our original analysis, we started to ask new questions of our data. In this paper we reexamine our data by drawing on Foucauldian notions of governmentality, and theoretical advancements that link these ideas to the emergence of community. Our current goal is to surface tensions, contradictions, and marginalized perspectives in the use of peer research in CBPR. Using Foucault's work to re-read and re-analyze data has proven useful for surfacing previously overlooked meaning in qualitative research (Payne and Nicholls 2010; Kaufmann 2011).

CBPR and the making and deployment of communities of shared marginality

Our interviews and focus group discussions raised a range of issues related to the inclusion of PRs in CBPR projects. While we were interested in hearing about PRs' roles and interactions with their team members, CBRs' narratives centered on the administrative aspects (finding the right people, getting them ready to do the work, and using them in the right ways). This was surprising, considering the emphasis in CBPR on empowering communities to realize their goals. These findings encouraged us to turn to writings about community outside of the CBPR literature to help make sense of what we were hearing. There we encountered Miller and Rose's (2008, 94) description of how communities have been made governable through a process of: (1) the reshaping of strategies for governing economic life; (2) the establishment of a new ‘ethics’ of activity that simultaneously makes and differentiates citizens; and (3) a reshaping of the relationship between expertise and politics. Their analysis advances Foucault's early writings to account for the re-organization of traditional modes of governance and the development of new programs and interventions that bring together political rationalities and technologies for regulating the self (Miller and Rose 2008, 80–82). This orientation to governance targets the most ‘abjected subjects’ at the margins of society for reform in the hopes of making them marketable (Miller and Rose 2008). We take these domains – economy, citizenship and expertise – and use them to read our data in relation to popular CBPR principles. We then focus on a participant whose narrative troubles and problematizes the dominant discourse of CBPR as empowering and transformative.

Strategies for governing economic life and the logic of peer research

We asked about participants' experiences with CBPR and their motivations for adopting a peer research approach. Israel et al. (1998) have identified a number of rationales for undertaking CBPR in the literature, including: enhanced relevance of data, maximizing skills and expertise, research quality, improved capacity of the partners, overcoming distrust of research by some communities, aims to improve health, and many others. The CBRs in our study echoed many of these benefits, and framed their decision to use PRs as an evolution from earlier CBPR efforts. Clarice, a service provider, describes undertaking a peer research approach without ‘knowing’ what to expect:

I think we were really kind of playing it by ear, we originally only envisioned having the peer research team do the interviews, and kind of knowing that we'd love to have them involved in other ways, but not really knowing, and you don't know too what the groups going to be like, how interested they're going to be in various aspects of it … we just didn't know what to expect and it definitely exceeded our expectations for sure … (CBR. FG 1. F)

This excitement to experiment with a PR approach without having considered the consequences was common among our participants. The sentiment seemed to be that if CBPR is ‘good’, then having PRs must be even better. Bruce, a researcher, explained that his agency had been experimenting with different approaches for some time with mixed results and wanted a more inclusive model. However, unlike Clarice and other participants, he described starting the process with some reservations:

[I had] mixed feelings. One of excitement because we heard about all the peer-led research models, and thought ‘What a great way to do research and involve community members,’ because [name of agency] had been struggling with how to engage community members in a meaningful way, and we thought ‘Oh, the peer-led model seems to have worked in other places, let's try it,’ so we were on the one hand excited, and on the other hand … aware of the challenges of doing on the ground, community engagement type of work, and we hear a lot of politics, and a lot of tensions, and challenges that are existing within the community that can get refracted and brought out … We wanted to find a better, more inclusive democratic model. (CBR. FG 2. M)

Here, the use of a PR approach was seen as representing a more authentic form of community engagement. However, this quote also alludes to the existence of community-level tensions and challenges that need to be explored in greater depth. What are these tensions and for whom do they represent a challenge? In the following sections, we surface some of these issues and consider their implications.

Corporatist rationalities and the entrepreneurial spirit

We were interested in why PRs decided to participate in CBPR and how they became involved in their respective projects. For many PRs, these projects offered an economic incentive in place of formal work or the promise of future work. This was particularly important for those who were economically marginalized and/or accessing social assistance (welfare, disability support, etc.). Miller and Rose (2008, 96–97) have observed the growth of techniques to govern the unemployed that oblige them ‘to engage in a constant and active search for employment and for the skills that will provide employment.’ While ‘employment’ was not the primary goal of these initiatives, they tended to reproduce particular forms of organization and relationships. Jason, who had been involved in research with the transgender community, described some of the issues that emerged for him:

Well, instantly with CBR when an academic hires a peer researcher you've got a power imbalance … right off the bat. As soon as it happens, the peer researcher doesn't feel comfortable and the academic is in a position of authority, instantly. You don't have the dynamic that you need to get the work done, and that's the problem. (PR. Interview 15. M)

This imposition of authority was often first noticed by PRs during the formal interviews that were prerequisites to joining some projects. These interviews were a means to assess their community authenticity (in some cases having other community members present to help assess this). Zahir, who is living with HIV, described his interview experience as follows:

I was interviewed by four different people in the same room, which was a bit intimidating. I must say. I felt that I was okay there, but it was intimidating having been a long time since I [had] been in a job interview. I mean, they were very friendly, but they each went around the table and asked me certain things. I guess they all had areas of expertise. I'd never been interviewed by a committee before. (PR. Interview 7. M)

In addition to recruiting authentic peers, these interviews were also a means of gauging whether individuals would be able to meet the requirements of a research project and the constraints of a workplace. Mira, a service provider with experience running peer-to-peer programs, described what she looks for when interviewing peers:

Mainly it's soft skills, having a good personality, being outgoing, someone that someone would want to talk to, because the other stuff you can teach. And they have to believe in the project; and they have to have a certain inherent respect for the people they are going to be interviewing; and they have to be able to work well with others, and show some recent history of being able to be punctual, to make some commitments, to have some things in place that kind of organize their personal life … I'll ask the question, ‘How do you get to places on time?’ If they can't answer that question, that kind of tells me a little about where they're at, it's important to understand the community you're hiring from. (CBR. FG 2. F)

A distinguishing feature of CBPR is that unlike most researcher/researched relationships, it is meant to be ‘a co-learning and empowering process that facilitates the reciprocal transfer of knowledge, skills, capacity, and power’ (Israel et al. 1998, 179). This capacity-building is meant to be undertaken at multiple levels: to build the organization's capacity to use and produce research, and also individual community members' capacity to participate in the research process. Bruce describes the sophisticated training program developed for the PRs working on his projects:

We involve community members, our target community members. From the very beginning, their involvement is shaping the research, but also extending beyond that to actually training them to conduct the research, to do the analysis, to co-write together, co-present together, and to follow up on that commitment. We've tried to establish a good training program as well … We sort of realised early on that if we are to use a peer-led approach, training would have to be essential, so we sort of made a commitment from the beginning to have a good in-house training program, and we've done, so we have ongoing trainings, reflective of the different projects, the stages they are in. (CBR. Focus Group 2. M)

These training programs were usually intended to build PRs' capacity in ways that were directly related to the project by way of computer literacy and credentialing (completing workshops and earning certificates). This form of ‘empowerment’ was not always of interest to PRs or necessarily useful to them outside of these research projects. An alternative account of ‘capacity-building’ was offered by Zahir, who questioned the motivations behind the team's goals and the benefits to him and other peers:

I feel a bit tricked into capacity building as they call it in a way, you know? I don't know. I mean, I'm really glad it happened, but I feel like capacity building is, I guess, part of their agenda, we're unaware of it in a way. (PR. Interview 7. M)

Drawing on Miller and Rose (2008, 97), we observed that behind the talk of community engagement was an ‘emphasis upon the individuals as active agents in their own economic governance’ through the capitalization of their lived experience and community affiliation. This ‘qualification’ to work brought them into the reach of ‘a whole new set of vocabularies and devices for managing individuals within the workplace in terms of the enhancement of their own skills, capacities, and entrepreneurship’ (Ibid., 97). In some cases, these projects served to divide community members by their skills, or ability to acquire skills, and then made them compete with each other for the few available positions. Many PRs, while dependent on social assistance, were not subject to this kind of direct training and supervision until they joined a CBPR project. Overall, these formalized employment approaches are at odds with the ambiguity and shifting boundaries inherent in community-based research and practice (Chiu 2008).

Bonds of allegiance – a new ‘ethics’ of activity and shared responsibility

Another central feature of CBPR is that ‘community is recognized as a unit of identity’ with a shared ‘sense of identification and emotional connection to other members, common symbol systems, shared values and norms, mutual – although not necessarily equal – influence, common interests, and commitment to meeting shared needs’ (Israel et al. 1998, 178). However, Miller and Rose's (2008, 104) work invites thinking about this as a process of differentiating between the affiliated and marginalized, and creating the conditions for the latter to be ‘re-unified ethically and spatially’ into the domain of community where they are made governable through constructed bonds of allegiance. Commonly discussed by participants was the belief in the importance of ‘lived experience,’ and that a PR model provides an authentic interaction between two individuals who share a culture and identity. The benefits are demonstrated in these two quotes: first, Alain emphasized the importance of the bond created by a shared experience of living with HIV:

I think overall the use of peer research is invaluable just because there's an instant bond there. When you talk to them on the telephone, making it the first interview process, you disclose that you're [HIV] positive and there's an instant bond. You both speak the same language'. (PR. Interview 5. M)

Michael, a PR with an experience of homelessness and substance use, added an embodied element to the importance of shared language by showing his injection marks:

They just know I'm a peer researcher. They know that we had experience with homelessness. With some of the interviewees, they're questioning, and as an ex-drug addict I'd go up and show them my track marks and go, ‘Hey, I know what you're talking about,’ and talk the language about them, and give them some advice too. (PR. Interview 1. M)

These relationships make it possible to achieve another goal of CBPR, to build on the strengths and resources within the community by harnessing the ‘skills and assets of individuals, networks of relationships characterized by trust, cooperation and mutual commitment, and mediating structures within the community such as churches and other organizations where community members come together’ (Israel et al. 1998, 178). This was seen in PRs' ability to negotiate social spaces that researchers could not otherwise gain access to, or where they would not be able to develop trust easily. In the following two quotes, PRs describe their ability to navigate community spaces: first, Michael described his skills and role as an intermediary between the community and research team:

I mean not to say that we are special, but we have special skills. To be able to go out and talk to somebody in the language of the street and understand what they're talking about, and be able to put it back into a language where everybody else could understand it. (PR. Interview 1. M)

This is echoed by Bernadette, a woman with an experience of homelessness and substance use, who describes her knowledge of street life:

That's when I felt like I was really drawing on my knowledge of the streets. Previously, by knowing, because you see so many fights break out in the bars, in the drop in centres, and you can just sort of wing it, you can eyeball it, by body language, you can tell when they're gonna flip. (PR. Interview 6. F)

PRs become intermediaries between the research team and their own community, able to access community spaces, and translate community knowledge. This is especially important in light of the fact that CBPR ‘seeks to build a broad body of knowledge related to health and well-being while also integrating that knowledge with community and social change efforts that address the concerns of the communities involved’ (Israel et al. 1998, 179). While CBPR strives to include community knowledge, it is not always easily acquired because of the complex relationship between some communities and researchers. This is especially true in communities that have been marginalized and stigmatized by research. Mandeep, who had experienced immigration and economic insecurity, described how the PR role is able to overcome community suspicion to collect better quality data:

I felt that they were more open because those people, they feel kind of suspicious. When you ask them, they don't know where that data is going. I think the fact that we were PRs, they were more comfortable. That's what I can say. And I felt that they were willing to talk. I think it actually improved the quality, the fact that they were very comfortable. So they started talking, and they were open, and they felt free with us. (PR. Interview 3. F)

Justin, a PR with an experience of homelessness, expressed the same sentiment and added that a shared experience encouraged some participants to open up in ways that could not be achieved by professionals they encounter:

I guess one thing that, that I would anticipate is that for somebody staying in a shelter or going to a drop-in, I think they'd be more inclined to be a little more open with somebody that they knew had been there. They'd probably provide a bit more detail that they otherwise might be apprehensive about doing with just a social worker, or, you know, I think for them having that knowledge that that person has been through similar things would probably make them a bit more open. (PR. Interview 4. M)

To harness this dynamic, many of the projects we heard from asked PRs to disclose their health/mental health status, community affiliation, and/or drug-using habits in order to help determine whether they were an authentic peer. The requirement to disclosure was acceptable for some PRs, as demonstrated by Alain who was very comfortable disclosing his HIV status:

I've never had a problem with disclosure. I disclosed from day one. This is me; I'm HIV positive. If that's a problem for you, that's your problem not mine. (PR. Interview 5. M)

Other PRs had a more complicated relationship with disclosure, as demonstrated in Bernadette's experience with another woman on her research team. Both were homeless, but they had different relationships to disclosure:

I know that you're supposed to get a sense of belonging in the group, but sometimes it seemed like there's conflict going on, and one of the PRs, I had introduced her to [a local politician], but she said, why did I introduce her as homeless? That I shouldn't say anything about her, because a lot of the street people are very, they're very worried that people will find out their problems, they had bad experiences, but [project coordinator] said, ‘The whole project is about homelessness, so why do you have to hide that the researchers have been homeless?’ (PR. Interview 6. F)

This talk of shared experiences and community bonds may be obscuring that the goal of many of these projects is to change community practices and behaviors (e.g. drug use patterns). For Miller and Rose (2008, 93), this form of government ‘even when it works upon pre-existing bonds of allegiance, transform them, invests them with new values, affiliates them to expertise and reconfigures relations of exclusion.’ A PR approach allows community bonds to be mobilized for the purpose of extracting community knowledge to make previously hidden practices visible, knowable, and thus governable. Foucault (1978, 59) explained that such ‘confessions’ play a central role in justice, medicine, and education, and even familial and intimate relationships. The PR provides important information about his/her community (determining what should be asked of the community) but is also able to collect the answers on behalf of the research team. Dreyfus and Rabinow (1982, 174–175) explain that:

through the expansion of methods of science the individual has become an object of knowledge, both to himself and to others, an object who tells the truth about himself in order to know himself and to be known, an object who learns to effect changes on himself.

In CBPR, confessionary practices have become a necessary precondition for both understanding and changing the community. While CBPR has a strong social justice orientation, its growing acceptance in medicine and public health may have more to do with the promises of accessing marginalized and distrustful communities and obtaining better quality data (Leung, Yen, and Minkler 2004; Jones 2007). This raises important questions about whether these, and other competing goals in CBPR, can be reconciled.

Sharing power – the reshaping of the relationship between expertise and politics

One of the most appealing claims of community-based research is that ‘all parties participate as equal members and share control over all phases of the research process, e.g. problem definition, data collection, interpretation of results, and application of the results to address community concerns’ (Israel et al. 1998, 178–179). Miller and Rose (2008, 93) explain that these new forms of economic governance and citizenship have created space for new, more subtle forms of professional governance and expertise through the use of empowerment. Despite claims that projects were community-driven and that peers played a central role, PRs described limited levels of engagement and a lack of communication. This understandably caused feelings of frustration and, in some cases, concerns of being exploited. Justin described his role as being limited to determining some aspects of the data collection instrument:

I didn't have much of a role in planning or informing the process, not really, I mean aside from just sort of reviewing what was in the questionnaire or the resource schedule. And we had some sessions where we all looked at it and they sort of changed some questions around based on the input of the PRs. (PR. Interview 4. M)

Zahir had a similar experience that caused him to question his role in the team and what the broader project was actually about:

I think I got a ‘don't worry about it.’ I feel a little bit of a disconnect between what the coordinators know and what's filtered down to me. So, I feel a little bit of, like, they're withholding knowledge somehow. I feel a little bit on the outside. Like, that I'm part of the experiment, and that doesn't sit that well with me. (PR. Interview 7. M)

These accounts do not reflect the expectation that CBPR projects are community-led, and instead reaffirms the power and authority of the professionals involved. The professionals decided what would be included and had it vetted through community insiders. In many cases, PRs had less of an impact on the research process than the research process had on them. While PRs had to be authentic and able to talk about their experiences, this had to be balanced with the professional requirements of the role. This necessitated certain comportment on the part of PRs, as discussed by Mira below:

The other thing I really to do in my trainings and my supervision is ‘how do you manage your personal feelings, that are gonna come up in the project,’ because, you know, some of the questions they may be asking people are things they may be struggling with, things they're dealing with, things that they do, you know, especially around drug use, they may be talking about risk factors they may engage in themselves, so how do you manage your self-talk and your personal triggers, and still stay a little removed from the person being interviewed but still stay connected with them, and I think that's a very difficult balance for people. (CBR. FG 2. F)

The techniques promoted here are similar to the ones social workers have used to help clients meet their ‘personal goals’ (Chambon, Irving, and Epstein 1999; Villadsen 2008), but are now being used to link personal development with community development. This form of governance relies on the expert knowledge of professionals, which merges positive knowledge and morality, to give them authority to conduct the behaviors of others (Miller and Rose 2008, 149). PRs are given strategies to manage and sublimate their feelings into something more productive and useful for the research team. Wanda, a PR with an experience of homelessness, insisted that she and the other women involved participated for the ‘right’ reasons, and emphasized how committed they were, as evidenced by their going above and beyond the project:

The interesting part, of course the honorarium was involved, but the wonderful thing was they just loved the project so much, they didn't care staying extra time to clean up and were always waiting to hear about the next phase of the project. One girl who struggled with particularly heavy and harsh addiction, even when she felt sick, she busted her ass to be there. (PR. Interview 10. F)

However, this raises the question of what would have happened if this substance using PR was not able to manage her addiction. What if she could not stay through the entire meeting, did not come at all, or missed too many meetings? At what point do her personal needs exceed the project's resources? Mira described having to discipline PRs who are not meeting her expectations:

I have fired many peer workers, and it's always awful. It's one of the reasons I generally have a contract … and so I find the policies are really important, and I always have a kind of light-to-heavy kind of discipline model, so everything is always at first about supporting the person and letting them know what the expectations are, or what the next steps of the discipline or support will be. (CBR. FG 2. F)

Here, we see a paradox in the use of PRs: on the one hand, they have to be representative of their community and speak authentically about their experience; on the other, they have to be able to function in ways that are conducive to a structured working environment. The ‘workplace’ is where neoliberal actors demonstrate their human capital and ability to internalize and perform market logic (efficiency, competitiveness, and personal responsibility) (Fenwick and Somerville 2006; Urciuoli 2008). Or, drawing on Mira's earlier quotes, this is where the PRs must demonstrate that they have the right combination of skills and self-management. Those who are unwilling or unable to adopt this logic become subject to neoliberalism's other logic of cost-cutting (‘unprofitable’ liabilities are discarded). The contradiction that is created by the two competing discourses of empowerment and efficiency is further illustrated in the case example offered below.

The case of an unruly body (empowerment and efficiency collide)

We turn our attention to Bernadette's narrative in greater depth to demonstrate what can happen at the precarious intersection of empowerment and efficiency. Bernadette's interview included more than her involvement in a CBPR project, and by her choosing covered decades of street involvement, substance use, and experiences in the shelter system. In all, she represented the disorganized and ‘unruly’ subject that social services are designed to manage (Moore 2009). Unlike the other participants in this study Bernadette did not come recommended by her research team.² This is understandable considering the experience she shared.

Bernadette described how she was ‘extremely flattered’ to have been chosen by her research team because in her case she did not have to apply for the PR position. Rather, she was chosen because as a homeless woman she was ‘qualified’ for the job. Bernadette was certainly a fit for the project, but the communication strategies she used, while appropriate in other areas of her life, brought her into conflict with the research team and the social service agency that housed the project:

I said, among the street people, the way you do it is, you say ‘fuck off’ and leave. Well, [the project coordinator] goes, ‘What do you mean you say ‘fuck?’ She said she never heard anybody in her organization say ‘fuck off when they're having an argument, never, and so there's a different way. (PR. Interview 6. F)

The message that Bernadette received is that her experience of homelessness was important, as was her ability to navigate the social spaces homeless people occupied, but that she needed to manage those ostensibly desired behaviors when dealing with project stakeholders. Unlike some of the more professionalized PRs we heard from – those who were able to maximize their training and capacity-building – Bernadette was unable to alter her behavior in ways that would allow her to be integrated into a research team. She ‘failed’ to re-stylize herself as a more productive and better functioning citizen. Bernadette experienced numerous conflicts that culminated in her being terminated from the project. Recounting the final event that led to her termination, Bernadette described leaving her home for a research-related event and promising her male partner to bring him something back:

I said ‘I'll try to bring you something,’ and that's where the line got blurred. I wasn't thinking anymore about what I was doing, I was just thinking about getting the stuff for this guy and [the coordinator] said, ‘No! You can't, we can't have the research pay for this.’ Like, they have so many pops, so many orange juices. It's just sitting there and I didn't see a lot of problems. I guess people were looking and I didn't stop, but I'd never done it that way before, so I didn't know the difference. Okay, it was just something that happened, but I didn't think that was warranted being taken off the job – they took me off of everything, and I said, ‘Well okay, next time I'll buy a case of pop, so I don't have to feel tempted to take pop [from the venue]’, and [the coordinator] said the damage has already been done, and there's nothing I can do to repair the damage. (PR. Interview 6. F)

Bernadette explained that in her experience, it was perfectly acceptable to go to a community drop-in center, take a sandwich, put another in your pocket, go back a few minutes later, and take some more. This is just one example of the survival strategies that homeless and street-involved people rely on. However, in this case, she was representing a research team and the rules were different, unbeknownst to her. Further reflecting on what happened; Bernadette considered her value vs. her challenges:

I thought [the coordinator] was valuing me, my skills, looking at the value, just not wanting the problems but maybe could have been a little more understanding that these are real issues we have to face, and there's barriers there that you can't just – boom, solve – because the whole society is set up and that's one of the things I learned when I got sober is because when I was drunk, everything's in disarray. (PR. Interview 6. F)

We find it troubling that Bernadette was dismissed from a project that purported to address the needs of women like her, for being too much like one of those women. For PRs like Bernadette, their identities came into conflict with the formal expectations of a research project, and their marginalization was only heightened by the expectation to be organized, stable, and highly functioning. While notions of authenticity and lived-experience are promoted as benefits, these come at odds with the requirements of projects with fixed budgets and tight timelines. The clashing of different administrative cultures, and in particular the role of funding, has been documented elsewhere in the Canadian context in relation to community-based research initiatives (Williams et al. 2005). We add to this by demonstrating the impact that it can have on an individual, and the ways it can serve to undermine the original goals of a project.

Concluding remarks

Our findings are consistent with those of Carlisle and Cropper (2009) who found that the role of the ‘lay researcher’ is complex and has many competing demands. The current analysis has highlighted the ways this complexity can be traced back to particular governmental strategies, relations, and discourses. The movement of CBPR from the margins to the mainstream has resulted in unexpected consequences for both community members and research teams. Despite aspiring to create change, many CBPR practitioners have taken on a governing role by organizing community members in research projects with bureaucratic hierarchies and constructed norms. Community members with a shared experience of economic and social marginalization are drawn into these initiatives by promises of ‘community engagement,’ but then find themselves confronted with the logic of human resource management, research budgets, and fixed timelines. They are required to increase their human capital through various forms of skill-building and credentialing to ‘learn’ to manage their own interests in relation to the goals of the project. This is necessary if they are to effectively perform the ‘friendly face’ of the project while engaging in the surveillance of their own communities. The joining of researchers and communities to challenge health inequities through the production of ‘evidence’ can be understood to represent the active citizenship ‘promoted by the neoliberal virtues of autonomy, responsibility and calculated prudence’ (Grundy and Smith 2007). The political rationality that has contributed to the marginalization of many communities in recent decades is also evident in the practice of CBPR. Unfortunately, this has resulted in otherwise well-intentioned CBPR projects contributing to the state's ability to penetrate the community and voluntary sector through partnerships that obtain community consensus to reinforce core market orientations (Larner and Craig 2005).

The aim of this paper was to offer a theoretically informed reading of our empirical data and challenge some increasingly taken for granted aspects of CBPR. The purpose of this critique has not been to discount the transformative possibilities of academics and communities working together, and of learning from each other (Rapport et al. 2008). Rather, it has been to challenge some of the ways community empowerment has been operationalized without taking into consideration the dual nature of empowerment. Cruikshank (1999, 2) has observed that ‘the will to empower contains the twin possibilities of domination and freedom’ and that the technologies of citizenship must be understood as operating within the field of power. Our contribution has been to explore the dangers associated with the will to empower through capacity-building that does not reflect the aspirations of the individuals' targeted for change. Requiring PRs to change their behavior to fit into a research project, and excluding those who are unwilling or unable, is a form of domination that undermines the social justice and emancipatory goals of CBPR. This analysis is intended to promote critical discussion that celebrates the potential of this approach while remaining sensitive to the possibilities of reproducing the very forms of social inequities it aspires to challenge. Shining a critical lens on CBPR makes it possible to resist its co-option and to imagine ways of engaging communities that better reflect their needs and aspirations.