Confronting espoused theories with theories-in-use: Challenges of participatory action research with marginalized communities in contributing to social change and theory building

Abstract

This article reflects on a participatory action research process in partnership with segregated Roma communities in Hungary. It will focus on the “non-positivist good theory”-building capacity of participatory action research in situations where social distance between participants is high and where action-oriented cooperation involves numerous actors, continuously extends to new stakeholders and areas, and aims to contribute to long-term and general social goals. Special attention will be paid to the effects of three phenomena: extreme poverty, extreme egalitarianism, and community hierarchy. We show that as cooperation shifts from discussions to actions and theories-in-use start dominating the process, PAR might become a complex and fuzzy process, characterized by numerous pragmatic and ethical challenges and contradictions. Thus, in a setting described above, it is a rather challenging task for PAR to create a “non-positivist good theory”: one which enables and empowers community members to make pragmatic and sustainable changes in their lives.

Keywords

Participatory action research segregated Roma communities non-positivist good theory espoused theories theories-in-use extreme poverty extreme egalitarianism community hierarchy

Introduction

Action research (AR) follows a “participatory worldview” (Reason & Bradbury-Huang, 2008) that requires any research activities and all stages of a research process to be approached as a co-creation by diverse research participants. In addition, action researchers’ commitment to meaningful participation brings power into the forefront in significant respects. First, as participation in research is enacted, the differences in the capabilities and skills possessed and brought to bear by different social groups (potential participants of AR) should be considered. If AR wants to be inclusive and empowering, its research design should be attentive to those differences that are relevant to enacting participation. Second, since power is everywhere, all communities action researchers work with and for can be characterized by the differences in power among community members. If AR wants to live up to the democratic ideals it entertains (Reason & Bradbury-Huang, 2008), researchers should consciously work with and reflect on the existing power relations within the community in question and the manner in which power is influenced by the AR process itself.

This paper reports on a participatory action research (PAR) process, in which marginalized communities and social science researchers work together for a positive change in the life of the marginalized. The three major issues discussed in this paper are all linked to the issue of power as enacted in marginalized communities. The focus will be on the three phenomena of (1) extreme poverty, (2) extreme egalitarianism, and (3) community hierarchy. It will be argued that all three components will contribute to the difficulty of bridging the existing social distance between community participants and researchers and will influence the potential of the research process with regard to coming up to the high expectations of the major principles (Reason & Bradbury-Huang, 2008) and theory-building aspirations of PAR (Friedman & Rogers, 2009).

The PAR process involving activist–researchers working in conjunction with Roma people living in segregated urban areas in Hungary which is reported here started and has been ongoing since 2011. The process has undergone numerous cycles and involved numerous co-initiated actions and cocreated results, with all of them producing significant insights into collaboration between marginalized communities and researchers. Here, however, we would like to reflect on only one of the initiatives (the so-called “patron program”) which sensitized our awareness to certain characteristics of socially marginalized communities tied to power and may hold significant lessons for PAR in general.

The paper is structured as follows: first, the situation of extremely poor, stigmatized, segregated Roma in a European and Hungarian context and the phenomena of extreme egalitarianism and community hierarchy will be introduced and linked to the social context of PAR. Next, a general overview of the PAR process and related activities will be provided, prior to discussing data collection and analysis methods. Then, the controversial nature of the PAR process will be analyzed as one that has been supported by participants and perceived to have great potential, on the one hand, and as a contradictory process of cooperation, particularly from the viewpoint of the core values of PAR, on the other hand. Finally, the social and theoretical utility of PAR is examined based on the “good AR theory” concept developed by Friedman and Rogers (2009).

Background: Social context, process and activities

Roma, extreme poverty, extreme egalitarianism, and community hierarchy

The context of the present PAR process is strongly influenced by the social situation of Roma in Europe and Hungary. The stigmatization and marginalization of Roma have a centuries-long history in Europe (Powell & Lever, 2017). Roma people are seen as Europe’s stigmatized “outsiders” (Powell, 2008; Powell & Lever, 2017), discriminated against and even “dehumanized” by the European majority, thus leading to a strong European “Romaphobia” (van Baar, 2011).

While other ethnic minority groups are experiencing a tendency toward desegregation, the trajectory of the Roma in Eastern Europe is, in contrast, toward ghettoization (Marinaro, 2003, 2015, 2017; Powell & Lever, 2017; Steger, 2007; Wacquant, 2012)—i.e., spatial segregation. Although the social and economic situation of Roma is rather diverse in the Central Eastern European region and in Hungary, a recent study has found that approximately 3% of the Hungarian population (300,000 people, the vast majority being Roma) live in segregated environments (“Gypsy settlements”) (Domokos & Herczeg, 2010) and that at least 1633 segregated areas exist based on ethnicity. Marginalized and stigmatized Roma living in segregated environments often suffer from extreme poverty, “a condition characterized by severe deprivation of basic human needs, including food, safe drinking water, sanitation facilities, health, shelter, education and information” (UN, 1995).

In addition, there is a social phenomenon of extreme egalitarianism (or collective egalitarianism) described by researchers working with segregated Roma communities in Hungary (Ladányi & Szelényi, 2004) and Slovakia (Grill, 2012). Segregated Roma communities are characterized by strong solidarity and an expectation of a radically even distribution of resources within the community—including resources and benefits flowing in from outside. Development projects that do not manage to live up to these expectations might create conflicts and result in participants’ leaving the cooperative process (Ladányi & Szelényi, 2004). Without questioning strong community integration, recent literature also suggests that, in certain settings, poor segregated Roma communities are growing increasingly fragmented and group solidarity is becoming weaker (Durst, 2002; Mariano, 2015; Messing & Molnár, 2011) and there is constant competition and self-positioning vis-à-vis others, perhaps paradoxically, “amidst the ethos of collective egalitarianism” (Grill, 2012, p. 10).

In addition to community solidarity and egalitarianism, segregated Roma communities are also described as being marked by a strong hierarchy (Durst, 2002; Grill, 2012; Mariano, 2015; Szombati, 2011): internal power relations might marginalize subgroups even within the marginalized communities. Such power relations suggest the potential presence of gatekeepers: actors with the power to undermine any cooperative projects (Smith et al., 2010). Gatekeepers might be self-declared (Seidman, 1998): not considered legitimate representatives by community members and having independent motivations with regard to any cooperation.

The present paper reflects on a PAR process (Figure 1), which is being carried out by researchers in conjunction with the inhabitants of two urban segregated neighborhoods characterized by extreme poverty, extreme egalitarianism, strong community hierarchy, and the presence of self-declared gatekeepers. Cooperation appears among (1) academics/activists or “scholar/activists” (van der Meulen, 2011, p. 375); (2) Roma representatives (members of the Roma Local Government); and (3) extremely poor Roma families living in the two segregated urban areas in a southern city of Hungary. Out of the 160,000 inhabitants in the city, the number of Roma is estimated to be around 4–5000, while approximately 400 people live in segregated areas: in ethnically almost totally homogeneous ghettos (Wacquant, 2012) (“Gypsy rows”), which are even separated from the residential areas occupied by Roma families that are financially better off.

Figure 1.

The PAR process.

Most families living in segregated neighborhoods are extremely poor, including housing conditions, (lack of) job opportunities, a low level of education, lack of security, access to information and prospects for the future, and lack of equal treatment—they are “marginalized” residents (Mullett, 2015). Communities in urban segregated Roma neighborhoods share a similar fate: the inhabitants are born into these areas or move in due to some socioeconomic pressures (e.g., extreme poverty and stigmatization). Communities in Roma neighborhoods—just like other communities (MacQueen et al., 2001)—share common perspectives and values and engage in joint action, but are also characterized by fragmentation and lack of unity, diverging self-interest among members and internal conflicts.

At first glance, extreme egalitarianism seems a basic value within the marginalized community. As a community member expressed it, “We’re all gypsies, we’ve all got the same blood in our veins. As long as you treat everyone squarely, I’m with you. But if you treat us differently, that’s no good.”

The community is also characterized by internal inequalities and a substantial community hierarchy, leading to the emergence and presence of self-declared gatekeepers. As one of them made clear to us, “You don’t get to decide who lives in this neighborhood. I’m the boss here.” Community members, however, do not consider it a legitimate position. As one of them put it, “No one’s chosen him as a boss, but he still acts like it. He oppresses people.”

Goals and activities

The cooperative process started in 2011 and was initiated by local academics/activists. The process began as qualitative research with semi-structured interviews and group discussions with local Roma leaders and experts working with Roma. The aim was to (1) gain an understanding of the situation communities are facing and (2) initiate cooperative action planning in conjunction with the members of the segregated neighborhoods.

As a result, work on the establishment of an after-school program—a project chosen in the first phase—was started together with local Roma organizations, families living in local neighborhoods, and education experts. By 2012, researchers had formed a working partnership with the local neighborhoods and all the participants realized that they have more goals in common. The Roma people identified numerous problems to resolve: discrimination and stigmatization; extreme poverty; low quality and insecure housing; lack of stable job opportunities; and inability to provide their children with prospects for a better future.

Academic and activist participants realized that their commitments are similar, even if formulated in different terms: breaking the poverty trap (arriving at a point where people, families, and generations are not locked in extreme poverty) and reducing extreme local inequalities. To contribute to these goals, cooperation has been action oriented ever since, continuously expanding and including previously nonexistent activities connected to (1) social welfare, (2) capacity building, and (3) political activism for anti-oppressive work (Mullett, 2015), including:

establishing and running two community centers and after-school programs;

enhancing the capabilities of local Roma civil society organizations (CSO) established to support marginalized families as regards their basic material needs;

empowering local Roma CSOs to be able to obtain financial resources to further their goals;

strengthening the voice of local Roma CSOs in the local press and toward the local city council;

building previously nonexistent connections (bridges) between urban middle-class and marginalized families; and

supporting community organizing and awareness-raising activities.

In addition, a “team” of approximately 20 Roma and non-Roma residents have emerged and organized itself as a “sustainable movement around marginalized participants” (Letiecq & Schmalzbauer, 2012). This group is committed to the process of cooperation on a voluntary basis to reach the agreed long-term goals.

Actors

There are two groups of actors in the PAR process in addition to extremely poor Roma families. The PAR process was initiated by local academics/activists. We have been involved in academic careers and in social activism in Hungary for many years. Some of us are initiators of and involved in numerous actions within the present PAR process and have a strong emotional and moral commitment to it. Others maintain a relative distance and participate in discussions via e-mail lists, joint reflection activities, and sometimes also in actions—they act as critical companions (Snoeren et al., 2012).

Another group of participants are recruited from two local Roma organizations. Most of them grew up in the same communities as the marginalized participants, but by now live as Roma citizens socially integrated into “mainstream” Hungarian society and still committed to supporting their marginalized peers. We refer to them as “Roma representatives.”

Members of these two groups (academic participants and Roma representatives), who are also the authors of the present article, belong to the local middle-class based on their socioeconomic status. Therefore, we refer to them (ourselves) as “middle-class participants.” Although middle-class and extremely poor participants are part of the same locality, enormous differences in material well-being, social status, and capabilities make them outsiders with regard to each other’s lifeworld in a number of ways (see also Smith et al., 2010).

Evaluation of the process

The present analysis builds on three sources of data. Semi-structured interview-based evaluations were carried out by academic participants on a regular basis to ascertain participants’ perceptions on the ongoing cooperative process.

There were events organized for structured reflection activities, including (1) regular group discussions among middle-class participants and (2) regular discussions among middle-class and poor Roma participants. Membership here is fluid: not everyone is present at every meeting. Minutes of these activities provide second-person voices as regards the process.

Having academics involved in numerous actions together with marginalized Roma families, who “do not split their work from their life” (van der Meulen, 2011, p. 370), allows the conventional “researcher vs. subjects” roles to be increasingly enriched (and replaced) by the peer perspective, thus enabling all voices to be expressed and to address “undiscussables” (Bradbury & Reason, 2003, p. 165). These experiences are recorded in the research diaries maintained by three of the authors—serving both as second- and first-person voices.

During reflection within the middle-class group, we identified (1) extreme poverty, (2) extreme egalitarianism, and (3) community hierarchy as significantly affecting the cooperative process.

After identifying this focus, we carried out qualitative content analysis (Titscher et al., 2000) on our data sources. Although the focus of the present analysis is determined by the problem identification of middle-class participants, our observations show that these issues and their consequences represent a significant concern for all involved.

Because of the presence of participants with diverging views, we always endeavor to clarify whose voices are being expressed during the analysis. If not stated otherwise, “we” refers to the middle-class participants, including academics and Roma representatives. To better express the voices of marginalized Roma participants, quotations will be used.

The effects of extreme poverty, extreme egalitarianism, and self-declared gatekeepers on the “reality” of PAR: The emergence of the patron program

With the patron program (started in 2014), local middle-class and extremely poor Roma families are interlinked. The former provides a fixed amount of financial support for a certain period of time (at least one school year) for the latter. Beyond monetary relations, families maintain face-to-face contact, form personal relationships, and establish different forms of cooperation as needed, including assistance with educational and health issues and job searches. Reasons for implementing the program are grounded in both social theories and experiences from the first years of our PAR process.

Segregated neighborhoods serve as areas of high-level internal social integration for their inhabitants, but families lack significant social ties beyond their neighborhood. Similarly, local middle-class families hardly ever have any meaningful contact with local marginalized Roma. It is striking that bridging social capital (Putnam, 1993) is missing locally.

During the first years of the PAR, we experienced how a cooperative, action-oriented process with shared goals and actions yields trust-based personal relationships and long-term commitment toward each other as well as cooperation. The difficulties of sustaining the operation and the initial results of PAR if the middle-class actors leave the process have also been experienced (see also Aziz et al., 2011). The patron program therefore aims to grow a “sustainable movement” (Letiecq & Schmalzbauer, 2012) around marginalized actors. Most importantly, the emergence and organization of the patron program are significantly influenced by our experience related to extreme poverty, extreme egalitarianism, and community hierarchy.

Extreme poverty: Emphasis on the social instead of the political and the need for extra resources

As the emphasis within the cooperative process shifted from a research orientation to joint action (see also Arieli et al., 2009; Travers et al., 2013) by 2013, the relationship between academic and poor Roma participants has changed from formal to informal and the latter have started to express their needs toward the cooperative process, primarily in terms of concerns of short-term material well-being. Most of these concerns have been related to basic needs and capabilities, including food and clothing, basic furniture and household appliances, commuting, medicines and basic medical aid, school expenditures and basic holiday opportunities for children, firewood, administrative issues, and school- and job-related support. This way a duality emerged between the consensual long-term goals of the participants, e.g., ending discrimination and poverty (see Goals and activities section), and immediate needs generated by daily hardship.

Daily hardship caused by extreme poverty seems to be unbearable from a middle-class perspective: it is like a hidden “caste system.” How could anyone think of, or treat, people as “partners” if they are starving and freezing, while people on the other side of the partnership are well-fed and live in well-heated homes? Can this be a “true” partnership and if so in what sense? Our initial perspective of long-term thinking and investments in the future (which was reinforced in the initial, discursive and more formal phase of PAR in, 2011 by the local Roma elite and poor Roma families in their choice of establishing an educational project) has started to seem rather “narrow-minded” in light of our new experience and much rather mirrored our lifeworld and mind-set than those of our partners living in their segregated neighborhoods.

Taking the “equality” requirement of PAR (Bradbury-Huang, 2010) seriously means that academic participants have to be open to the perspectives of “the other” partners, even if some elements of their perspective initially seem to be more part of the problem—e.g., “providing services not directly related to the action research” (Arieli et al., 2009, p. 282). Obviously, there may be some deviations from a previously agreed long-term “empowerment-focused” and “political” agenda. Recognizing the immediate needs of marginalized participants, however, plays a crucial role in building trust and is therefore a prerequisite of long-term, engaged cooperation. Neglecting these and pushing for our own agenda soon resulted in “pseudo-participation” (Arieli et al., 2009).

The extent of extreme poverty (the number of families affected) was also a shocking experience for us and cooperation shed light on the extreme inertia of the poverty (and stigmatization) trap, showing that reaching our goals related to empowerment in such cases will be a difficult and long-term process (Dick, 2015), which will require the mobilization of additional resources. Therefore, it seemed necessary to involve new actors in the cooperative process while also making a shift “to put the focus on the social rather than the political” (Letiecq & Schmalzbauer, 2012, p. 254), constituting, at the same time, a step toward community-driven issue selection (Minkler, 2004).

Extreme egalitarianism, community hierarchy, conflicts, and individual solutions

Extreme egalitarianism was expressed by community members as a strong and widely shared value within the marginalized community (see Roma, extreme poverty, extreme egalitarianism, and community hierarchy section), implying that the benefits of any interventions are expected to be distributed evenly within the community by community members. However, its actual form, functioning, and effect on the (PAR) cooperative process are far more complex compared to what is expressed by community members and suggested by previous research (see Roma, extreme poverty, extreme egalitarianism, and community hierarchy section).

Here, the notions of espoused theories and theories-in-use may provide some deeper insights. While

espoused theories are what actors say or think they do and are used to describe or justify behavior, … theories-in-use represent a kind of tacit knowledge which can be employed almost automatically to produce behavior that is usually, though not always, effective. Theories-in-use can only be inferred from observed behavior because actors are generally unaware of these theories or cannot articulate them. (Friedman & Rogers, 2009, p. 35)

Cooperation (common actions and observing theories-in-use) has shown that the actual appearance of extreme egalitarianism is by far not as unambiguous as expressed by community members (as an espoused theory). “Community” is a multilevel concept, not a clear-cut concept of a discrete and stable group of individuals (e.g., inhabitants of a certain segregated Roma neighborhood). Expectations for even involvement and distribution of benefits appear within families (at the individual level), among families (at the family level), among subcommunities within the segregated communities (based/organized on wider family relations or common social situation), between segregated neighborhoods, and between poor local Roma living in and outside of these neighborhoods.

This multilevel nature of the interpretation of community means that even though someone is involved in and benefits from a given cooperative process on a certain level, s/he might still feel “left out” or “less involved” than others. The temporal aspect of engagement further complicates this issue: even if someone feels included at a certain point in time, she might feel excluded at another specific point in time. The result is that during a longer process of cooperation, a certain (and always changing) group of community members feels excluded from time to time, and this feeling of exclusion leads to frequent frustration and conflicts.

Furthermore, if someone feels included, s/he might not complain about others’ feeling of being excluded: egalitarianism within the community clearly has its limits. However, egalitarianism is still recognized by all as providing a moral legitimacy for arguments for inclusion. The practical functioning of egalitarianism is further complicated by the presence of community hierarchy and self-declared gatekeepers (see Roma, extreme poverty, extreme egalitarianism, and community hierarchy section). Certain community members—parallel with sharing extreme egalitarianism as an espoused theory as a moral basis for the inclusion of every community member—declare that they should be the main beneficiaries of the cooperative process because of their distinguished social position within the community. While these actors are evaluated as “self-declared bosses” (gatekeepers) or “oppressors” by numerous community members, they often act in solidarity with other community members, typically members of certain subgroups in the community.

Extreme egalitarianism is thus a widely shared, expressed value (espoused theory) within the community, but in practice (as a theory-in-use) it reveals paradoxical dynamics. When it comes to egalitarianism, community becomes a multilevel concept for members; it clearly has its limits for those who already feel included or enjoy a higher social position within the community hierarchy. However, egalitarianism still serves as a morally legitimate basis, shared by the whole community, for the inclusion of (and sharing with) those who feel left out.

The parallel presence of egalitarianism (enhanced solidarity) and community hierarchy might relate to community members’ strong social ties (bonding capital) within the community and weak ties outside (bridging capital) the community. This “closedness” of segregated Roma communities create enhanced—and often fragmented—solidarity, egalitarianism, and expectations of sharing, on the one hand, and “constant competition and self-positioning vis-à-vis the others” (Grill, 2012, p. 10), on the other hand. The parallel and interdependent presence of enhanced solidarity and competition, egalitarianism, hierarchy, and fragmentation complicates community-level cooperation within PAR to a significant extent. This is characteristic of each and every cooperative project aimed at the level of the (segregated) community, from the initial afternoon school initiative through community bathrooms to community firewood projects. In addition to the general agreement concerning their necessity (people all want better education and leisure-time opportunities for their children, comfortable bathrooms and affordable heating), such projects are always characterized by conflicts when it comes to realization.

Discussion: The reality of PAR

We introduced how personal and community perspectives and needs, shaped by extreme poverty and stigmatization, extreme egalitarianism, and community hierarchy, contributed to a shift within PAR from the political to the social and to individual/subcommunity-level solutions, instead of “working with the community” as a whole. The emergence of the patron program is a powerful example of this shift. The program is evaluated as being of vital importance for participants, since it secures emotional and financial “support” for poor Roma families in terms of their most acute material needs by forming bridging social capital and long-term (“honest,” trusted,” “accepting,” “loving,” “connected”) personal relationships and commitment, even contributing to certain marginalized Roma families’ “having plans now” for the future, which were lacking previously. It thus contributes to a sense of control over one's own lives (Clover, 2011), and, eventually, it might reduce learned helplessness (Crondahl & Karlsson, 2015). At the same time, the patron program is a rather contradictory initiative from the perspective of certain PAR core values.

The act of “help”: Paternalism versus equality

PAR is egalitarian in nature (van der Meulen, 2011), where all participants are supposed to be equal (Bradbury-Huang, 2010). It is characterized by even dialogues and reciprocal relationships (Snoeren et al., 2012). However, differences of power in society enter into PAR processes (van der Meulen, 2011). In the present case, only middle-class participants have access to social capital enabling them to form new relationships within the patron program. The relationship between the families offering and receiving support is necessarily uneven because of huge differences in capabilities. As a patron expressed it, “Although we’re friends, I’m the one who can invite her for a coffee if we go into the city together. She can’t afford it. She depends on me. It’s necessarily uneven, and I can’t do anything about that.”

A number of further questions or concerns arise. Can a benefactor–beneficiary relationship be genuinely equal or is it fundamentally unequal (Tofteng & Husted, 2011)? How does the act of support appear in the mental models of the benefactor and the beneficiary? Can such a relationship truly be empowering (while most participants actually evaluate it as being like that)? What is to be done if we realize the contradictions of benefactor–beneficiary relations, but changing them is beyond our capabilities?

Support becomes extremely contradictory when expectations emerge in the supporter. Do patrons have (unconscious) expectations in exchange for their support? What happens if these expectations do not coincide with the goals of the family receiving the support? What should be done if families emphasize the need for support in fulfilling their subsistence needs (e.g., food), while patrons evaluate “development” or “taking a step forward”—reflected in improved school performance, for example—as success? What happens if the patron leaves the relationship because of this divergence in perspectives? Such concerns are subjects of continuous group reflection among patrons to “help illuminate the unintentionally patronizing attitudes that can lie beneath the charitable intentions” (Smith et al., 2010, p. 411).

Individual solutions, extreme egalitarianism, conflicts, and empowerment

PAR is committed to community engagement and empowerment (Dick, 2015) and working with the community (Travers et al., 2013). As Lewin (1946) argues, the group—of which individuals are members and which is the basic building block of the wider society—is the most important level of intervention to contribute to enduring changes (see also Clover, 2011). While this may work in theory and often in practice, the actual process is often characterized by the dominance of subcommunity- or family-level solutions.

The patron program is working with the community via one-to-one, family-to-family relations and individual relationships—a contradictory approach if evaluated from the perspective of the long-term goals of cooperation. It is possible to reach certain empowering goals on different levels (Mullett, 2015): from the individual to the community one. The questions clearly arising here are, for example, the following: How do levels of empowerment relate to each other? What happens if only certain families are empowered? Do family-level changes contribute to community-level ones? What happens if those who become empowered leave the community—a risk expressed by the local Roma elite during the initial semi-structured interview phase at the very beginning of PAR and observed as an actual process in Hungary (Ladányi & Szelényi, 2004)? Furthermore, although one of the main reasons for working via family-to-family relationships was to avoid/reduce potential conflicts raised by community-level solutions, having such an initiative in place does not mean that no conflicts emerge.

Participants have formulated diverging judgments in discussions about what to do in situations where initiatives cannot involve all community members. Some marginalized participants feel that “if initiatives can’t involve everyone, nothing should be done.” Others have argued that “it’s important to provide equal opportunities for everyone because everyone is poor and needs help,” but the fact that initiatives are not able to do so “doesn’t mean that nothing should be done.” Having no clear-cut solution for the problem, related decisions are always made after consultation and group reflection on a case-by-case basis. In the case of the patron program, it meant that it was started only in one of the segregated neighborhoods and only six families were initially involved out of the eighteen families living there. Launching the program immediately caused community conflicts and conflicts between some community members and middle-class actors. Families left out were angry at middle-class participants. As they expressed it, “What you’re doing here in the neighborhood is really no good.”

Hierarchy and self-declared gatekeepers: effectiveness versus emancipation

In the “ideal PAR,” power relations are exposed with the aim of changing them (Bradbury-Huang, 2010). PAR is directed toward emancipation and liberation (Aziz et al., 2011). However, participants might lend a pragmatic orientation to PAR, instead of a critical one (Johansson & Lindhult, 2008). Workability becomes more important than emancipation and reflection on issues of power.

Since it was impossible to start the patron program with all the interested families, the decision was made to initially involve families with “louder voices,” including self-declared gatekeepers. Having taken a critical approach here (e.g., involving families of the voiceless) could have meant that the program could not be introduced due to the opposition of powerful gatekeepers.

Such decisions, however, make the process even more contradictory: taking a pragmatic approach might reinforce existing power relations within the community and does not contribute to critical reflection and liberation (Johansson & Lindhult, 2008). In addition, such resource allocation is clearly unjust and contrary to the moral judgment of most participants. Why cooperate with people with more power rather than with those in greater need?

Taking a critical or pragmatic approach is an ever-present concern. The difficulty of this situation is shown by a conversation between an academic participant (A) and a woman living in a segregated neighborhood (M):

M: Everyone should get the same amount of support and be equally involved in the program!

A: I agree. But what can we do if there are only six patrons and twelve families in the neighborhood?

M: Then take those who are most in need. The poorest ones!

A: OK, but you know that everyone is poor and needs help.

M: Yes, that’s true. But you know that X [a self-declared gatekeeper] isn’t poor. And they are still included while others aren’t!

A: I agree with you. But what can we do if X says that he won’t let the program start if he’s left out?

M: Well… That’s really very difficult…

To moderate the unjust consequences of such decisions, two types of efforts have been made. First, most middle-class and marginalized participants are involved in ongoing reflection about this situation, issues of justice, power and oppression, and potential solutions. Second, we have endeavored to extend the program to every family, so by now it also reaches out to the most marginalized.

Community-level transparent decision making versus multilevel participation

Ideal PAR processes are fully transparent (Bradbury-Huang, 2010). PAR secures mutual ownership of the process and major decisions (van der Meulen, 2011) and transparent and equitable decision making (Travers et al., 2013). Contrary to our aims, we often cannot satisfy these standards. Participation is multilevel. As for academics and Roma representatives (middle-class participants), participation is close to the PAR ideal. Trusting human relations, shared goals, and commitments allow us to be involved in such a relationship. This means that there is a group of actors where interactive participation (Jacobs, 2010) takes place. There is also direct, frequent cooperation between middle class and marginalized participants ranging from a daily to a monthly basis depending on ongoing projects, circumstances, and the need for cooperation. However, within this relation, we cannot live up to our participatory ideal.

Reflection activities here happen in smaller groups and are more informal and fragmented. Beyond reasons highlighted by other researchers reflecting on similar settings where academics work with marginalized groups (Arieli et al., 2009; Clover, 2011; Letiecq & Schmalzbauer, 2012; Mullett, 2015; van der Meulen, 2011), limited participation also stems from decisions made by middle-class participants. The reason is the same one that pushes the process toward subcommunity and individual solutions: it is felt that it is impossible to live up to the expectations created by extreme poverty, egalitarianism, and hierarchy.

Marginalized families are also aware of potential dangers here. The answer on an individual level is also often to conceal the information in such situations; as a community member involved in the patron program formulated it, “I’m doing it right. I’m not telling anyone.” The result is an unequal and low level of participation—consultation and functional participation (Jacobs, 2010)—which is far from our democratic ideals and intentions.

PAR, theory building, and social change

Earlier, we attempted to show how extreme poverty, extreme egalitarianism, and a strong community hierarchy as factors that contribute to great social distance between PAR participants when middle-class researcher–activists align with extremely poor, closed Roma communities might lead to a PAR process which is characterized by ambivalences and contradictions if evaluated based on core PAR values, but which, at the same time, corresponds with the needs and values of partnering marginalized participants. The essential paradox of PAR in our case is that it is exactly this deviance from the PAR ideal which allows the process to go on for a long time with the participation of numerous (and a growing number of) committed actors and, thus, may enable the long-term “empowering” effects of PAR to be realized.

These contradictions and uncertainties give us pause to reflect on the theory-building capacity of the present PAR process according to the “non-positivist” good theory as introduced by Friedman and Rogers (2009). In this sense, a good theory is sensitive to the meaning making of participants but also goes beyond and explores unseen causal dimensions of their behavior and the environment and the interaction of the two. This way, it co-creates shared knowledge, enabling “clients” who are becoming empowered to make practical and sustainable changes in their life. Good theories are thus respectful towards espoused theories (what actors say or think they do and use to describe or justify behavior) but are also able to explore and critically reflect on theories-in-use in a way that theories-in-use, actions based on them and the behavioral world—social contexts in which actors live and act—are affected.

Different social worlds (diverse social identities) (Smith et al., 2010) meet within the present PAR process (unlike the context in which Friedman & Rogers, 2009) demonstrate their example of good theory-building). In such circumstances, setting shared goals and visions, finding common espoused theories at the initial phase of a cooperative process on “What is happening around us?” and “How do we want to change it?” might be relatively easy. Poverty, stigmatization, poor housing, and lack of opportunities and the need for change might be easily shared concerns and goals for all the actors involved.

PAR usually begins with discussions among participants, and thus espoused theories dominate its initial phase. The same is true for the present PAR process, which started with interviews, group discussions, and cooperative action planning. During this initial, primarily communicative phase, it was relatively easy to agree about pressing problems and shared goals related to the more abstract concept of empowerment (see Actors section).

Cooperation changes when participants enter at the action phase of PAR. Here, actions and theories-in-use become dominant drivers of interactions, and PAR as action-orientated cooperation forces participants to explore their theories-in-use. Action-orientated cooperation on an interpersonal level brings to light (1) potential differences between espoused theories and theories-in-use and (2) differences among theories-in-use and behavioral worlds among participants with diverging social backgrounds.

PAR thus forces participants to confront their theories-in-use compared to espoused theories and opens up spaces for critical self-reflection and for testing and refining existing social theories. As individuals, middle-class researcher–participants all share a critical approach to the oppressive nature of certain existing economic and social structures (e.g., capitalism and the consumer society) and certain institutions (e.g., the school system and the labor market). However, on an emotional level, enhanced integration into these structures, or steps toward better system integration (Lockwood, 1964), are seen as a step toward empowerment. Reflection on contradictions between espoused theories and theories-in-use might aid participants (including researchers as individuals) in better understanding and possibly changing their own inconsistencies as regards espoused and actual values, theories and actions.

Exploring theories-in-use through cooperation by action and observing them also provides opportunities for researchers to test existing theories, as introduced in Extreme egalitarianism, community hierarchy, conflicts, and individual solutions section on extreme egalitarianism. The usefulness of long-term action-orientated cooperation, especially in the case of the hard-to-reach social group of stigmatized, marginalized, and segregated Roma, has also been confirmed by recent studies that use intense and long-term observation processes combined with activism and personal engagement to better understand the situation of marginalized and segregated Roma (Grill, 2012; Lancione, 2017; Marinaro, 2015, 2017).

However, creating non-positivist good theories as suggested by Friedman and Rogers (2009) also entails critically reflecting on and changing theories-in-use in a way that is reflected in action and leads to an enhanced capacity for practical and sustainable changes. Our experiences suggest that creating a non-positivist good theory through action-orientated cooperation, where the social distance between participants is high, numerous actors are involved, and the goals of the cooperative process are rather general and long-term might not be an easy, clear-cut, and unambiguous process.

In line with Friedman and Rogers (2009), our experience suggests that among actors with relatively similar behavioral worlds—e.g., those of researcher–activists and Roma representatives in the local middle-class—putting cycles into practice that lead to good theories that in turn produce an “increase in scope for personal and collective agency” (Friedman & Rogers, 2009, p. 37) is probably a less difficult issue. Similar norms (of communication) and similar experience with (factors of) “success” support this process and lead to an enhanced capacity on the part of Roma representatives to further their own practical goals related to empowering their marginalized peers. Elements of such an enhanced capacity to establish practical goals are shown by Roma representatives’ capacity to run after-school programs and community centers, secure funding for community activities, become more embedded in the local network of CSOs/non-governmental organizations (NGO) and assert the local political interests of the extremely poor local Roma—none of which existed before the PAR process.

However, cycles of building non-positivist good theories start again on different levels as the level of cooperation extends to new areas, directions, and actors. As new participants enter the process and actors—potentially again and again characterized with high social distance—move toward one another once more and become involved in individual level cooperation and joint action in new areas, the level of understanding shifts anew from espoused theories toward theories-in-use and related differences. Potentially conflicting theories-in-use will be manifested. Placing emphasis either on the political or on the social (see Extreme poverty: Emphasis on the social instead of the political and the need for extra resources section) is one of these differences in the present case. The realization of these differences provides opportunities to better understand each other’s behavioral worlds and reflect on our own narrow-mindedness. However, it might lead to negative judgments about certain everyday actions on the part of the “other.” As individuals, middle-class researcher–participants all agree at the level of espoused theories that every individual has the right to lead a life they have reason to value—in accordance with the “capability approach” developed by Sen (1999). However, on an emotional level, we still “judge” certain forms of (e.g., “irresponsible” or “oppressive”) action if they contradict our theories-in-use. Such enormous differences in behavioral worlds, theories-in-use, and related judgments have the potential to cause frequent conflicts during the PAR process, which are not necessarily easy to resolve, since theories-in-use are deeply rooted in our lifeworld and enculturation (Friedman & Rogers, 2009).

Living one’s life as marginalized, excluded, and stigmatized represents multiple social and economic disadvantages, and in such a situation theories-in-use are rather difficult to change in the sense that they lead to an enhanced capacity to make practical and sustainable changes, even if these changes are critically reflected on. Marginalization, exclusion, and stigmatization are all significant limiting factors here ([see Bodorkós & Pataki, 2009] in a Hungarian context) that lead to the presence of a poverty, stigmatization, and segregation trap. Putting the concept of non-positivist good theory into practice in a context marked by great social distance among participants, where researchers partner with numerous marginalized and stigmatized actors to form a community characterized by specific internal norms and power relations, might pose numerous practical and moral problems. What happens if some community members (and researcher–activists) are deeply involved in the process of cooperation and this involvement opens up spaces of critical reflection on theories-in-use, enhanced agency and change, while other community members are critical or even hostile toward the cooperative process and the presence of researcher–activists in the life of an otherwise relatively closed community? How can we reflect on theories-in-use in a way that brings about an increase in scope for collective (community-level) agency (Friedman & Rogers, 2009, p. 37)? Have we developed good theory if PAR contributes to enhanced agency for some within the community, while others feel that their situation has not changed or has even worsened? These are only a few of the many questions that remain after seven years of PAR cooperation.

Footnotes

Acknowledgements

The authors are grateful to the “Elszalasztott 1000 Év” Roma Association and numerous marginalized Roma families for their committed long-term participation in the process and the research and activist support of the Environmental Social Science Research Group (ESSRG) and the Community-based Research for Sustainability Association (CRS).

We also thank Hilary Bradbury-Huang and Maria Teresa Castillo-Burguete for leading the review process for the article and the anonymous reviewers for their critical engagement with our work, which provided an invaluable learning experience to the authors.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The present PAR process is not directly supported by any research grants, however, two of such grants (PERARES: Public Engagement with Research and Research Engagement with Society FP7, 2007–2013, grant agreement number: SiS-CT-2010–244264; Hungarian Scientific Research Fund, grant number K-109425) supported some of the researcher-activist participants to be able to dedicate a significant amount of their time and energy to the process during certain periods.

Author biographies

György Málovics, PhD in Economics, is an associate professor at the Faculty of Economics and Business Administration, University of Szeged, Hungary, and member of Community-based Research for Sustainability Association (CRS) (Hungary). Since 2011 he has been working with marginalized, stigmatized and segregated urban Roma communities for empowering social change in the approach of participatory action research.

Judit Juhász is a PhD student at the Faculty of Economics and Business Administration, University of Szeged, Hungary, and member of Community-based Research for Sustainability Association (CRS) (Hungary). Her main research interests include poverty alleviation, human development and social transformation. Her approach to social research is significantly influenced by the capability approach.

Boglárka Méreiné Berki is a PhD student at the Faculty of Economics and Business Administration, University of Szeged, Hungary, and member of Community-based Research for Sustainability Association (CRS) (Hungary). Her research and activism is focused on stigmatized, segregated Roma communities in Hungary and the investigation of their social networks, especially through participatory and action-orientated approaches.

Barbara Mihók holds a PhD in forest ecology. She is a member of Community-based Research for Sustainability Association (CRS) (Hungary). She has strong interest in interdisciplinary research approaches. Her goal is to combine evidence-based academic knowledge with critical and continuous self-reflection while aiming to serve the well-being of the (local) human communities and their natural environment.

István Szentistványi is trained in pedagogy, a member of Community-based Research for Sustainability Association (CRS) (Hungary) Since 2002, he has been involved in evidence-based local community activism for ecological sustainability and environmental and social justice as a local activist and since 2010 he is an independent local representative in the city council of Szeged, Hungary, working on local sustainability and social justice issues.

György Pataki has a PhD in organisation and management studies and works as an associate professor for the Department of Decision Sciences, Corvinus Business School, Corvinus University of Budapest and as a senior research fellow for the Environmental Social Science Research Group (ESSRG Ltd.), Budapest, Hungary. He is committed to participatory research in the fields of sustainability transformation and biodiversity conservation. He loves to teach following service learning and action learning pedagogy.

Mihály Nagy is a member of the local Roma community, president of “Elszalasztott 1000 Év” Public Association and the Roma Ethnic Local Government of Szeged, Hungary. He is a committed Roma activist and representative driven by the goal of support and empowerment of local stigmatized, poor and in part segregated Roma community in Szeged, Hungary.

Janka Tóth is a PhD student at the Faculty of Economics and Business Administration, University of Szeged, Hungary, and an activist of “Elszalasztott 1000 Év” Public Association. Her main interests as a researcher-activist include poverty, stigmatization and marginalization, and the potential of social enterprises in poverty alleviation.

References

Arieli

Friedman

V. J.

Agbaria

(2009). The paradox of participation in action research. Action Research, 7(3), 263–290.

Aziz

Shams

Khan

K. S.

(2011). Participatory action research as the approach for women’s empowerment. Action Research, 9(3), 303–323.

Bradbury

Reason

(2003). Action research. An opportunity for revitalizing research purpose and practices. Qualitative Social Work, 2(2), 155–175.

Bradbury-Huang

(2010). What is good action research? Why the resurgent interest? Action Research, 8(1), 93–109.

Bodorkós

Pataki

(2009). Local communities empowered to plan? Action Research, 7(3), 313–334.

Clover

(2011). Successes and challenges of feminist arts- based participatory methodologies with homeless/street-involved women in Victoria. Action Research, 9(1), 12–26.

Crondahl

Karlsson

L. E.

(2015). Roma empowerment and social inclusion through work-integrated learning. SAGE Open, 5(1), 1–10.

Dick

(2015). Reflections on the SAGE encyclopedia of action research and what it says about action research and its methodologies. Action Research, 13(4), 431–444.

Domokos

Herczeg

(2010). Terra Incognita: Magyarországi szegény- és cigánytelepek felmérése—első eredmények. Szociológiai Szemle, 3, 82–99.

10.

Durst

(2002). “Innen az ember job hogyha meg is szabadul”. Megélhetési stratégiák egy kisfalusi cigány közösségben. Esély, 13(4), 99–121.

11.

Friedman

V. J.

Rogers

(2009). There is nothing so theoretical as good action research. Action Research, 7(1), 31–47.

12.

Grill

(2012). Going up to England’: Exploring mobilities among Roma from Eastern Slovakia. Journal of Ethnic and Migration Studies, 38, 1269–1287.

13.

Jacobs

(2010). Conflicting demands and the power of defensive routines in participatory action research. Action Research, 8(4), 367–386.

14.

Johansson

A. W.

Lindhult

(2008). Emancipation or workability?: Critical versus pragmatic scientific orientation in action research. Action Research, 6(1), 95–115.

15.

Ladányi

Szelényi

(2004). A kirekesztettség változó formái. Budapest, HU: Napvilág Kiadó.

16.

Lancione

(2017). Revitalising the uncanny: Challenging inertia in the struggle against forced evictions. Environment and Planning D: Society and Space, 35(6), 1012–1032.

17.

Letiecq

Schmalzbauer

(2012). Community-based participatory research with Mexican migrants in a new rural destination: A good fit?. Action Research, 10(3), 244–259.

18.

Lewin

(1946). Action research and minority problems. Journal of Social Issues, 2, 34–46.

19.

Lockwood

(1964). Social integration and system integration. In: Zollshau

K. G.

Hirsh

H. W.

(Eds.), Explorations in social change (pp. 244–257). Boston: Houghton Mifflin.

20.

MacQueen

K. M.

McLellan

Metzger

D. S.

Kegeles

Strauss

R. P.

Scotti

Trotter

R. T. I. I.

(2001). What is community? An evidence-based definition for participatory public health. American Journal of Public Health, 91(12), 1929–1938.

21.

Marinaro

C. I.

(2003). Integration or marginalization? The failures of social policy for the Roma in Europe. Modern Italy, 8, 203–218.

22.

Marinaro

C. I.

(2015). The rise of Italy’s neo-ghettos. Journal of Urban History, 41, 368–387.

23.

Marinaro

C. I.

(2017). The informal faces of the (neo-) ghetto: State confinement, formalization and multidimensional informalities in Italy’s Roma camps. International Sociology, 32(4), 545–562.

24.

Messing

Molnár

(2011). Bezáródó kapcsolati hálók: Szegény roma háztartások kapcsolati jellemzői. Esély, 2011(5), 47–74.

25.

Minkler

(2004). Ethical challenges for the “outside” researcher in community-based participatory research. Health Education & Behavior, 31, 684–697.

26.

Mullett

(2015). Issues of equity and empowerment in knowledge democracy: Three community based research examples. Action Research, 13(3), 248–261.

27.

Powell

(2008). Understanding the stigmatisation of Gypsies: Power and the dialectics of (dis)identification. Housing, Theory and Society, 25, 87–109.

28.

Powell

Lever

(2017). Europe’s perennial “outsiders”: A processual approach to Roma stigmatisation and ghettoization. Current Sociology, 65, 680–699.

29.

Putnam

R. D.

(1993). Making democracy work: Civic traditions in modern Italy. Princeton, NJ: Princeton University Press.

30.

Reason

Bradbury-Huang

(Eds.). (2008). The SAGE handbook of action research: Participative inquiry and practice (2nd ed.). London, UK: SAGE.

31.

Seidman

(1998). Interviewing as qualitative research. A guide for researchers in education and the social sciences. London, UK: Teachers College Press.

32.

Sen

A. K.

(1999). Development as freedom. Oxford, UK: Oxford University Press.

33.

Smith

Bratini

Chambers

D.-A.

Jensen

R. V.

Romero

(2010). Between idealism and reality: Meeting the challenges of participatory action research. Action Research, 8(4), 407–425.

34.

Snoeren

M. M. W. C.

Niessen

T. J. H.

Abma

T. A.

(2012). Engagement enacted: Essentials of initiating an action research project. Action Research, 10(2), 189–204.

35.

Steger

(2007). Making the case for environmental justice in Central and Eastern Europe. CEPL-HEAL-The Coalition for Environmental Justice, Budapest. Retrieved from http://www.env-health.org/IMG/pdf/Making_the_case_for_environmental_justice.pdf

36.

Szombati

(2011). A részvétel problémája az akciókutatásban: A mezőcsáti fejlesztőmunka tanulságai. In: Pataki Gy & Vári A (szerk.): Részvétel—akció—kutatás. Magyarországi tapasztalatok a részvételi-, akció és kooperatív kutatásokból. MTA Szociológiai Kutatóintézet, Budapest, pp. 52–83.

37.

Titscher

Meyer

Wodak

Vetter

(2000). Methods of text and discourse analysis. London, UK: Sage.

38.

Tofteng

Husted

(2011). Theatre and action research: How drama can empower action research processes in the field of unemployment. Action Research, 9(1), 27–41.

39.

Travers

Pyne

Bauer

Munro

Giambrone

Hammond

Scanlon

(2013). Community control’ in CBPR: Challenges experienced and questions raised from the Trans PULSE project. Action Research, 11(4), 403–422.

40.

UN. (1995). Report of the world summit for social development. Retrieved from http://www.un.org/documents/ga/conf166/aconf166-9.htm

41.

van Baar

(2011). Europe’s romaphobia: Problematization, securitization, nomadization. Environment and Planning D: Society and Space, 29, 203–212.

42.

van der Meulen

(2011). Action research with sex workers: Dismantling barriers and building bridges. Action Research, 9(4), 370–384.

43.

Wacquant

(2012). A Janus-faced institution of ethnoracial closure: A sociological specification of the ghetto. In: Hutchinson

Haynes

B. D

(Eds.), The ghetto. Contemporary global issues and controversies (pp. 1–32). Boulder, CO, USA: Westview Press.