Participatory action research in neoliberal academia: An uphill struggle

Abstract

Participatory action research (PAR) is described in the literature as a valuable method for enhancing the power of marginalized individuals and communities by collectively producing knowledge to transform the inequalities they experience. This deviates from most social science research, where such actors are largely the subjects of data extraction. This paper reports on our experience of using PAR to examine existing food systems and ideas regarding ‘just food system transitions’ alongside Non-Governmental Organizations in Brazil, Sierra Leone, the United Kingdom and Zambia. We describe our efforts to encourage these partners to participate in the research design, data collection and analysis in line with PAR ideals. Our experience fell short of our expectations for a PAR project. While some limitations relate to the Covid-19 pandemic, this paper focuses on the structures of contemporary neoliberal academia, which, we found, actively obstructed the realization of the optimistic claims of the PAR literature.

Introduction

Participatory action research (PAR) is often described as a method that enhances the power of individuals and communities through their inclusion as partners in research. The literature regularly provides examples of such enhanced empowerment through PAR among excluded groups such as women (Aziz et al., 2011), people with disabilities (Tanabe et al., 2018) and youth (Brydon-Miller and Maguire, 2009). The underlying theory is that the ability to generate knowledge confers power. In traditional research practice such power is aggregated to the academic, and, subsequently, other actors are not afforded the power to generate knowledge themselves. Instead, they are seen as those about whom knowledge is generated. PAR, on the other hand, ‘attempts to put forth a different form of knowledge’ (Gaventa and Cornwall, 2013: 178). It does this by including non-academic stakeholders throughout the research process and so changing ‘who participates in the knowledge production process in the first place’ (Gaventa and Cornwall, 2013: 174). As Amsden and VanWynsberghe argue, ‘PAR is an approach to research which empowers the community to define their own questions, lead the process of investigation, and create their own solutions for change’ (2005: 359).

However, as PAR has grown in profile and influence, scholars have also noted an array of challenges. These include problems regarding ethics, validity and time, as described below. However, of more concern to us here are issues related to the neoliberal nature of contemporary academia. By this we mean the underlying economic model that drives the demand for ever more output and efficiency from research activity, as will be further described below. Dawson and Sinwell note, for example, that PAR originally developed among ‘active civil society and social movement bases’ and, in being taken up by academics at a later point, found itself faced with the ‘limitations and the prescriptions of the academy’ (2012: 185). In this context time pressures can force a reversion to more traditional forms of applied research (Jacobs, 2010: 374), and PAR may either not be valued at all or may be ‘co-opted by institutions’ if it appears to attract funding (Bettencourt, 2020: 163). These concerns indicate the potential friction between PAR in theory and when practiced within the academy as it is broadly structured today.

In this paper, we are going to focus on this friction as it became evident in a PAR project implemented between April 2021 and September 2022. The project had various goals. These included exploring local understandings and perceptions of food systems and ideas regarding how to implement just food system transitions, or the restructuring of food systems to be less unequal while shifting to a low carbon economy, in specific local sites in four contexts (Brazil, Sierra Leone, the UK and Zambia). It also included building a mutual exchange and support network with participation from larger ‘advocacy’ organizations from Brazil and Europe. In this way, the hope was that the project would bring together institutions with similar concerns to share knowledge and experience, develop networks for mutual learning, and engage in future collaborative and proactive change, such as exchanging lessons learned or developing novel just transition strategies. The project brought together four academics in the UK and the Netherlands and staff of one Non-Governmental Organization (NGO) in each of the four contexts who would participate in research design, data collection and analysis. This paper focuses on the substantial challenges we encountered in implementing this project within the context of contemporary academia.

The remainder of this paper is divided into four parts. The first part reviews the literature regarding PAR as a process, the various positive claims made by its proponents and the challenges the approach faces. The second part turns to a review of the literature describing the neoliberal turn within academia, which forms the context for university-based research today. The third section discusses our project and describes a variety of different ways that the structures, processes and priorities of academia formed barriers to the success, and perhaps even the possibility, of doing truly engaged and collaborative PAR in this case. While it was sometimes difficult to disentangle the influences of the Covid-19 pandemic from the more general environment of the academy, we argue that many of the dynamics that we can identify as antithetical to PAR as an approach exist independent of the pandemic and are symptomatic instead of the contemporary neoliberal university. The final section then summarizes and concludes our argument by restating the key findings and making some suggestions for how the academy might become a more supportive host for PAR in the future.

Participatory action research

PAR has a long history spanning many decades and scholars today identify its origins in different fields and disciplines. However, whether we identify the core ideas in the work of Kurt Lewin (Adelman, 1993), Sol Tax (Stanley, 1996), Paulo Freire (Campos and Anderson, 2021), Marja-Liisa Swantz (Nyemba and Mayer, 2018) or Orlando Fals Borda (Rappaport, 2020), we can still recognize quite specific central features. As the name implies, first comes participation. PAR stands in opposition to more traditional and extractive forms of social research in that the scholars noted above moved towards the co-production of knowledge ‘with’ or ‘alongside’ communities as opposed to continuing the tradition of conducting research ‘on’ or even ‘for’ communities. The approach as a whole is rooted in ‘respect for people's knowledge and ability to participate as equal partners in the research process’ (Brydon-Miller et al., 2020: 106). As Ponzoni argues, including a wider range of people in the production of knowledge shifts power relations, ‘including the power to determine what knowledge is useful’ (2016: 558). Such approaches seek to erase the hierarchy of knowledge production and allow non-hierarchical processes of knowledge generation which may then produce non-hierarchical knowledge (Enria, 2016: 327).

The focus on participation, however, is secondary to the primary end goal of PAR, which is action to overcome the injustices experienced by oppressed groups (Brydon-Miller et al., 2020). This orientation towards action harks back to the roots of PAR, which was in activism and not academia (Bryant et al., 2019), and many PAR scholars today have one foot in both of these worlds. Gutierrez and Lipman, for example, argue that PAR is ‘action-oriented’ and ‘reject the binary of rigorous research and political involvement’ describing themselves as ‘activist researchers’ (2016: 1242–1243). This may not, of course, be a title that all PAR advocates would identify with, but PAR scholars across a range of disciplines consistently connect PAR with the goal of action contributing to social justice. PAR, as a method, therefore, ‘embodies a social justice agenda’ because its underpinning theoretical foundations ‘attend to issues of power, oppression, and injustices through embodying democracy and addressing emancipation and social transformation’ (Benjamin-Thomas et al., 2018: 2).

In such claims we see the connection that PAR proponents make from specific PAR projects to efforts to deconstruct broader structures of inequality, coloniality and oppression. Susana Caxaj provides an insightful discussion of how and why PAR can be a ‘rigorous and ethically robust’ process for interrogating and unpacking ‘assumptions about the world and the nature of knowledge that originate in settler/colonial practices’ and contour conceptions of what can and cannot count as legitimate knowledge (2015: 2). Fahlberg also makes a strong case for the value of PAR to embody a decolonial practice by including ‘the voices of marginalized groups outside the academy’ and undermining what she calls ‘epistemological hierarchies’ (2023: 99), or the valuing of some forms of knowledge over others. This is another common claim in the literature, that PAR has the potential to work against the violence or injustice inflicted on communities when their own ways of knowing and generating knowledge about the world are marginalized by processes of knowledge production and knowing legitimated by global hierarchies of power (Janes, 2016; Tardieu et al., 2023; Walker et al., 2019). Specifically, we mean here the universalizing claims to objective knowledge embodied in some social research.

This resistance to universalized knowledge claims is a key strength of the PAR approach as it means that PAR provides not less but more rigorous and trustworthy findings because they are locally and contextually specific (for discussions of rigour in PAR, see Lennie 2006; Melrose 2001). European epistemological assumptions, which frame the dominant conception of what counts as legitimate modes of knowledge creation, have long made what Tsing calls ‘promises of universality’ (2005: 8). But they are themselves very particular and historically contingent local forms of knowing. This leads to distinct blind spots when research constructed on those conceptions is applied in cultures and communities systematically excluded from the discourses which defined the rules by which these assumptions operate. As Tsing noted, ‘those who claim to be in touch with the universal are notoriously bad at seeing the limits and exclusions of their knowledge’ (2005: 8). PAR provides a potential corrective to this epistemological closure because, due to its engaged and inclusive nature, it is inherently involved in generating ‘culturally specific, local/place-based knowledge’ (Atallah et al., 2018: 491; see also Cordeiro et al., 2017: 397) and so it provides more accurate knowledge about social phenomena. There are some limits here if what you want to make are universalist claims. However, as Kong argues, it is crucial to understand what is happening within specific contexts and local settings, which then allows you to better understand the ongoing interactions between the local and the global (2018: 269).

Despite being quite eclectic methodologically, PAR projects are united in their ability to bring individuals together, to allow discussion and dialogue, and to identify problems to be solved and avenues for change. While not all PAR lives up to this ideal, many proponents describe the process of engagement, dialogue, problem identification and solution generation as ‘iterative’, and of the ideal PAR process itself as characterized by ‘cycles of action and reflection’ (Bryant et al., 2019: 1278). Calabria and Bailey have recently described this as an iterative cycle of ‘planning, action and reflection’, to ensure a balance between activism and contemplation about issues, which then ‘allows for the involvement of participants, ideally in every stage of the research cycle’ (2021: 5). This echoes earlier work describing this as ‘the action research spiral’, characterized by repeating phases of planning, acting, observing, reflecting and pre-planning (Kemmis et al., 2014: 19; see also Martí, 2000).

The iterative nature of the ideal PAR design forces us to consider the deeper requirements of the approach, and particularly the time necessary for such projects. It highlights the need for sustained commitment, and the requirement to develop ‘relationships, trust, and reciprocity’ that last over time (Smith et al., 2021: 133). As Mallan et al. describe, it requires ‘non-exploitative relationships between the research team and the research communities’ (2010: 257). However, this task can be very challenging, particularly in studies where there are broad differences between the academic researchers and participants, as is common in research spanning race, class, gender and other inequalities, or that include actors from the Global North and Global South. In such projects ‘engagement and motivation’ within a diverse research team can be ‘problematic’ and negotiations between the different research partners can be difficult (Braye and McDonnell, 2013: 270).

Unlike more traditional social research where there is less demand for close relationship building, the requirements of PAR projects echo the feminist demand for an ethic of care in research practice (Evans, 2016: 214). Feminist scholars propose that the bonds that sustain those practices are ideally rooted in a reciprocally constructed attachment between the participants (Hamington and FitzGerald, 2022: 1). Palmer et al. argue that this commitment should transcend the specificities of concrete research projects and create ‘spheres of engagement’ based on relations of trust, care and commitment that extend both into the past and the future (Palmer et al., 2020: 754). For these reasons, PAR requires a commitment of time and emotional energy unlike more distanced or ‘objective’ social science methods.

Neoliberal academia

As PAR has become more widely practiced, scholars have also identified various challenges and limitations. These include problems related to ethics (Salazar, 2022), validity (McTaggart, 1998: 2007) and time (Meredith et al., 2022: 661), but we are particularly concerned in this paper with more structural issues related to contemporary academia. As mentioned above, Dawson and Sinwell noted that PAR originally developed within civil society and only later found itself faced with the ‘limitations and the prescriptions of the academy’ (2012: 185). In this context, Jacobs describes how the time pressures faced by academics, academic institutions and funders working to increasingly short-term timelines can force a reversion to more traditional forms of applied research even when PAR had been the initial research plan (2010: 374). Bettencourt similarly articulated a concern with the currency of prestige in academic institutions, and the potential that PAR may either not be valued at all within academia, or, conversely, may be ‘co-opted by institutions’ if it appears to lead to funding opportunities (2020: 163). Indeed, Smith et al. have recently enumerated a list of 12 ‘challenges and barriers’ within academia to what they call co-produced research (2023: 169). These include, for example, problematic timescales, with PAR simply taking far too much time (2023: 170), and grant requirements, which require ‘predetermined research design and outcomes’ (2023: 169). They also include university structures, which can negatively assess co-produced research (2023: 170) and enforce overly bureaucratic administration of grants (2023: 176), and academic norms, which de-value experiential knowledge (2023: 170).

Interestingly, Wilson et al. (2022) note a number of challenges located not within academia, but with community member's perceptions of or experiences with academic researchers. These include concerns about whether, for example, academic researchers may simplify and romanticize the local community, ‘overestimate the time community members can devote to PAR,’ and whether they might ‘underestimate the competing priorities’ community members and organizers may have (2022: 15). Janes’ contribution is particularly troubling as, while noting the potential or what she calls the ‘critical hope’ of PAR to overcome divisions and inequalities, she is concerned about the tendency for academic epistemic privilege to constantly ‘re-inscribe’ itself in the research endeavour. PAR, she argues, ‘obscures the privilege of the academic researcher’ and serves, in the end, to ‘re/produce’ it (2016: 75–76). These critiques within the PAR literature, therefore, demand a more careful discussion of the context in which research happens.

In UK higher education (HE), neoliberal policies started to be deployed in the early 1990s. These policies functioned to reduce public investment in HE over time and to transfer the responsibility of financial sustainability to universities themselves. From that moment universities had to generate income by endorsing ‘market-like behaviours’ and ‘market behaviours’ competing for public and private resources (Slaughter and Leslie, 2001: 154), offering profitable services, and attracting students to generate income (Kallio et al., 2016: 688; Lynch, 2006: 5; Olssen and Peters, 2005: 326). To achieve this, universities have changed their orientation from institutions primarily responding to public interests towards becoming ‘consumer-oriented corporate networks’ (Lynch, 2006: 2). This has radically transformed universities’ internal structures and the role of academic staff.

As Dearlove noted 25 years ago, neoliberal policies applied new pressures on universities and ‘rattled established ways of organising academic work’ (Dearlove, 1998: 111). In response to these reforms, universities incorporated ‘managerialism’, a new ideology emerging in the wake of neoliberal reforms which sees management offering specialized techniques to run any kind of organization. Since in this ideology managers are the bearers of ‘advanced knowledge and know-how deemed necessary to the efficient running of organizations’ (Klikauer, 2015: 1104), managerialism argues that ‘they alone are best suited to run society’ (Wheeldon et al., 2023: 344). As a result, the balance of power in governance structures is shifted toward them. In universities, this has meant ‘stricter hierarchies, more powerful chief executives and the introduction of university boards’ that include members external to the university but involved in key decisions regarding management and leadership (Frølich and Caspersen, 2015: 381).

The managerial reorganisation of tertiary education has clear goals. The main goal is to achieve ‘greater faculty productivity’ promulgated on the ‘need for greater accountability’, ever-increasing ‘global economic competitiveness’ (Slaughter and Rhoades, 2000: 78), and the demand for high performance to access public funding (Dougherty and Natow, 2020). Productivity does not stand alone, however. Under neoliberal hegemony, a positive relation between income and expenditure is a core organizational principle. In HE this has been implemented through ‘greater stress and parsimony in resource use’ (Chandler et al., 2002: 1054), that looks always for greater efficiency in the quest for profit. Those who promoted neoliberal policies found in the public nature of HE financing the perfect excuse to demand more accountability for the efficient and productive use of those funds (Dougherty and Natow, 2020). The efficiency also impacted workers’ conditions. To ‘use’ this ‘resource’ efficiently staff are increasingly contracted in temporary and part-time positions, often outsourced and without benefits (Flynn, 2020; Ivancheva, 2015).

This profound restructuring of universities has had enormous impact on those who work within them because it has reframed the role of academic and scholarly agents. According to neoliberal theory, individuals are self-interested agents pursuing their own goals. Subsequently, organizations are expected to use ‘explicit contracts, monetary incentives and performance monitoring’ to align their members’ goals with their own (Dougherty and Natow, 2020: 459). In HE, management have implemented a range of mechanisms to measure academics’ performance (Berg et al., 2016; Kenny, 2017), based on ‘metrics and indicators’ legitimated by an ‘external standardization of the criteria for excellence’ (Kallio et al., 2016: 90), which together constitute the contemporary audit culture within academia (Shore and Wright, 1999; Sparkes, 2007). From a Foucauldian lens, Ball argues that the underlying logic is to produce a self that ‘take[s] responsibility for working hard, faster and better as part of our sense of personal worth and the worth of others’ (Ball, 2012: 19–20), that aligns them to the higher level of productivity expected from academics by universities. Since publications are one of the main indexes of productivity, ‘publish or perish’ has become a widely accepted mantra (Cederström and Hoedemaekers, 2012; Lund, 2012; Sparkes, 2007).

The toll of this system on academic staff is far from insignificant. Unceasing audits, paired with the performance culture, creates a moral system of self and mutual vigilance which consistently raises the productivity bar for academics (Ball, 2012: 19). Compliance with this system, together with the need to establish largely instrumental and ‘productive relationships with fellow academics’ (Burton and Bowman, 2022: 503) serves to ‘disembody and isolate’ scholars (Ruth, 2008: 107), which is amplified by the loss of previous forms of collegiality (Puāwai Collective, 2019: 34). Adding to this, the growth of managerial and academic support roles detaches academics from their specialized counterparts, such as administrative staff (Macfarlane, 2011), and further disempowers them within the structures of the academy.

These problems, together with the need to juggle competing responsibilities (Griffin, 2022: 2197), leave academics in a fragile state that impacts many aspects of their everyday lives. They experience an intensification and acceleration of time due to increased workloads (Gill, 2016: 49; Valovirta and Mannevuo, 2022: 1311), mounting administrative tasks (Tight, 2010), and higher expectations regarding outputs (Anderson, 2006; Vostal, 2015) that are at odds with the rhythm of research (Guzmán-Valenzuela and Di Napoli, 2015; Mountz et al., 2015). As a result of these dynamics, feelings of extreme insecurity, exhaustion, stress and anxiety related to pressure and precarious labour conditions abound among academic staff (Anderson, 2006; Berg et al., 2016; Knights and Clarke, 2014; Loveday, 2018). Usually, those feelings are related to a sense of, or a fear of, failure in many dimensions of academic work (Horton, 2020; Turner, 2020; Zielke et al., 2023). The lack of support in the face of these tensions makes universities a place where researchers do not feel cared for and valued (Gill, 2016; Gill and Donaghue, 2016). Finally, it is worthwhile noting that this scenario negatively affects the professional development of traditionally marginalized groups in the academy (and society more broadly) such as women and racialized populations (Gill, 2016: 51; O’Keefe and Courtois, 2019).

The reader will no doubt spot the tension here; that the time and care necessary for PAR research as described above clearly stands in contrast to prevailing descriptions of the environment in which academic researchers operate on a day-to-day basis and within which their research is carried out. This tension calls into question the extent to which PAR can live up to its potential within the context of contemporary academia. Examining these challenges, and identifying potential solutions, requires an examination of how PAR unfolds within the academic context we have described and how the structures of power delimit its potential for work such as PAR.

Struggling uphill: PAR in the neoliberal academy

The project we discuss here was implemented with support from a research funder (hereafter referred to as ‘the funder’) and was hosted by a UK University (hereafter referred to as ‘the university’). The project was initiated by four academics who met at a ‘Sandpit’ session related to a specific funding opportunity regarding ‘just transitions.’ The shared idea the four converged around was a project with relatively clear goals. These included the exploration of local understandings and perceptions of existing food systems and ideas regarding how to implement just food system transitions (the restructuring of food systems to be less unequal while shifting to a low carbon economy) among producers, consumers and food justice organizations, working in four local contexts. It also included building a mutual exchange and support network across the four contexts, including larger ‘advocacy’ organizations who could serve as advisors, mentors and support for those local organizations. This would bring together the local civil society, international civil society and local communities, where local civil society partners would each hold a series of three workshops focusing on local perceptions of current and ideal food systems and transitions from the former to the latter. In this way the hope was that the project would bring together institutions with similar concerns at different scales to share knowledge and experience, develop networks for mutual learning, and engage in future collaborative and proactive change, but within a relatively bounded and focused project achievable within a 12-month window. Given our focus on ‘just transitions’ across scales and the strengths of PAR as described above, the research team felt strongly that the underpinning axiological, ontological and epistemological positions of a PAR approach was best suited to the task. However, as we will describe, this turned out to be much more difficult than imagined.

A caveat before proceeding

This section could be interpreted as being quite critical of both the funder and the university. We believe it is important to stress that these are not critiques of these two institutions per se. Indeed, that would be a much less substantive point to make. Instead, we seek here to describe the challenges we experienced in this project to highlight much broader patterns within academia that echo the findings of existing studies regarding neoliberalization as described above and which, we argue, make the successful implementation of PAR projects very challenging and serve to disincentivize researchers from pursuing this kind of research. We feel that we have a responsibility to note the challenges of the context in which we work, if for no other reason than to be honest about our own research and to advocate for more suitably structured funding and research environments. We will describe these challenges in three categories: (1) limitations of time, (2) structures and lack of support and (3) undermined motivation and emotion.

Limitations of time

This project was already problematically restricted by time at the design stage. The four academics met and discussed the ideas that would become the project initially at a ‘Virtual Sandpit’ organized by the funder spanning three sessions totalling less than 7 hours (only a minority of which was spent developing this specific project) in late February and early March 2021. By the end of those sessions enough commonality of interest had been identified that a shared PAR project seemed possible. However, instead of having substantial time to develop a research project, the teams formed during the sandpit were then given only 7 days to submit their proposals (this restriction was not known beforehand). Within this time, in the middle of the teaching period, research teams had to develop research questions, design a suitable methodology, locate and confirm potential research subjects or partners, and draft and submit a research proposal. Such a timetable emerges from the implicit ethos of efficiency that demands that researchers deliver outputs within short timescales. We only really realized at a later point that this was just not enough time to develop a PAR project.

There are different reasons for this. First, PAR projects simply take more time to develop than more traditional projects. As the iterative model describes, it is expected that partners participate in the research planning as early as the development of the research questions, which has the benefit of incorporating non-academic individuals or communities into the process more fully as partners. Ideally, this would have been possible in our project, with the non-academic partners helping to define the research questions and design the process (Calabria and Bailey 2021). Unfortunately, this was simply not feasible in 7 days. As it was, the academic researchers reached out to organizations they already knew to confirm nominal commitment to participate based on our initial drafts of the project proposal. This led to our choices of partners in Brazil, Sierra Leone, the UK, and Zambia (where we collectively had connections due to earlier research projects), but it also led to substantial problems regarding the divergent capacities and resources of the partners that had implications for later stages of the research, from communications and trust building through to data collection and analysis.

The Brazilian and British partners, for example, were based in urban areas with relatively reliable connectivity to the internet, thanks to which they were generally able to participate fully in meetings, both listening to and being heard by other participants. This was not always the case for the partners in Sierra Leone and Zambia, as they were both based in rural or semi-rural areas in countries with less reliable internet access. As a result, engagement and communication with these partners was not always fluid and it was hard in online meetings to develop a shared sense of community. As noted above, nourishing trust among all participants in a PAR project is crucial for facilitating collective work (Armstrong et al., 2022; Palmer et al., 2020: 752–753). The literature also acknowledges that building such relationships is even harder in projects without previous working experience between the stakeholders (Martí, 2000: 52). This challenge was pronounced in our project because the restricted time available for project design and proposal development led to the selection of partners with widely divergent capacities and resources. That restricted time simply did not allow us to develop fully trusting and reciprocal relationships with these partners.

Uncertainty regarding the continuity of the project was another factor related to the bounded time-period of the funding which undermined the development of trusting relationships and a shared vision. This uncertainty prompted anxiety among the research team as we struggled to balance the constrained reality of the project and our partner's expectations of doing more after the end of the 12-month timeline. Even though we were clear with the partners from the beginning about the scope and duration of the project, and that decisions regarding continuing were not in our hands, questions about continuity often emerged: what would happen after the workshops were completed, would there be more research, more engagement, more activity? This shows that partners were interested in developing the research further, but, in line with Smith et al.'s articulation of relationships in PAR as reciprocal and extended over time, they were also concerned about the nature and goals of our initial short project. The spectre of neoliberal timelines was not restricted, therefore, only to those of us working within academia, but contributed also to the anxieties and uncertainties of the non-academic partners.

Another core problem, however, lies somewhat outside the funding mechanism (although interacting problematically with it) and is related to workload and assumptions regarding time and productivity within academia. The assumption is, of course, that academics have substantial uninterrupted time within the university to carry out research, and hopefully there are still today universities where some academics maintain this kind of freedom. However, due to the implementation of neoliberal policies and governance, many have seen a significant decline in the amount of time available for research that is not itself funded by grant income. In the case of this grant, limited as it was to approximately £15,000, there was no option to ‘buy out’ academic time, or to release the academic researchers from other university roles such as teaching and administration. As such, the time committed to this project largely had to fit around those other roles and, perversely enough, around other research commitments, such as securing other grants that would ‘buy out’ time.

These experiences evidence and illustrate the claims made in earlier studies by scholars such as Slaughter and Rhoades (2000) regarding the negative impacts of the drive for economic competitiveness among academic institutions and the constant demand for efficiencies and productivity as described by Chandler et al. (2002) and Dougherty and Natow (2020). While we believed initially that we had designed a project that could be implemented within the 12-month window, it turned out that this was not enough time, because of the various additional and unexpected challenges that we experienced, to develop relationships with partners within each of the four contexts, for those partners to carry out all of the data collection procedures and participate in some form of collaborative analysis, and then for the local partners, the academic researchers, and the international advocacy organizations to develop a mutual exchange network for just food system transitions. The ideal iterative process of research, action and reflection described by Calabria and Bailey (2021) was even less possible within such a limited time-period. Indeed, as the project became more complicated through delays, communication problems and administrative challenges, more and more time was committed not to the substance of research, but to just keeping it all afloat.

Structures and lack of support

That so much time ended up being spent on problem solving instead of achieving substantive research outcomes points to broader problems within the academy in relation to PAR efforts. These come down to the structures in place to govern research practices and, to a great extent, the way those structures work against projects that attempt to hold to PAR values. As already said, this is not an indictment of any one institution. However, it is important to note how much particular structural issues within the university primarily involved in this project undermined the fundamental elements of a PAR approach. One example is how these structures affected the possibilities for building trusting relationships among participants, as is pivotal for PAR projects (Mallan et al. 2010; Palmer et al. 2020; Smith et al., 2021). To some extent, of course, this was already undermined by the lack of time described above and the fact that the project was always going to be virtual, with no opportunity for face-to-face interaction on the ground. However, the various ways the university's research administration procedures further undermined these relationships was substantial.

First of all, in order to allow local civil society organizations to participate in the project, and particularly to fund the workshops within each of their local contexts, a substantial portion of the project budget was earmarked for transfer to the local partners. Understandably there would be some paperwork involved in such a process. However, we were nonetheless surprised when the actual paperwork was provided. To start, partners were asked to sign ‘sub-award’ letters, each more than six pages long and filled with legalese describing the commitments between the parties. Each partner had to confirm, for example, that it would ‘undertake not to conduct itself (whether by act or omission) in such a manner that would cause [the university] to be in breach of its obligations’ and ‘shall indemnify [the university] from and against any liabilities, losses, costs, or expenses incurred’. Statements like this had to be agreed to by civil society partners working in their second or third language, some of whom had no easy access to legal advice or recourse.

Once these forms were signed each ‘partner’ then had to complete a form that was titled the ‘supplier form’. This was an excel based form which was actually five separate sheets which included various ‘terms & conditions’, and questions about banking, worker status and procurement policies. It is, of course, logical that some kind of form will be necessary for money to be transferred from a university in the UK to a civil society organization elsewhere. Yet there seem few things that could be more disruptive and inherently antithetical to the development of non-hierarchal relationships of trust across complicated divides of culture, status and hierarchy, than to explicitly label the less privileged actor the ‘supplier’ in that relationship.

After this, the ethical approval forms also contributed to identifying the partners as something less than equal participants, as they were framed as chiefly focused on the voluntary participation of people in the research project, and the agreement of the terms about how their data would be collected, protected and used. These forms too, therefore, illustrated a lack of trust between the apparently senior actors (the university or the academic researchers) and those lower in the assumed hierarchy (the local partners). In line with Foucault, we read these practices as technologies of power that produce and reproduce structural inequalities between agents (Foucault, 2005). In this context these technologies situate the research partners as service providers in a commercial relationship with the university. Any attempts to build trusting relationships between actors in and outside an academic setting, therefore, means dealing with practices that routinely re-enact power hierarchies, which counters Ponzoni's descriptions of PAR undermining such hierarchies (2016: 558) while confirming Janes’ concerns about the re-privileging of academia as a result of PAR efforts (2016: 75–76).

But even this was not quite the end, as having successfully jumped this hurdle and become a ‘supplier’ did not mean the money could then be distributed. It only meant the partners were eligible to receive the money at some point. Once each of the workshops had been completed, the university then required that partners provide proof that they had spent the money correctly. This was a confusing and time-consuming process that put an extra burden on us, diminishing our capacity to work on other core tasks of the project, and on the local partners who were compelled to return a myriad of administrative forms. The labyrinthine nature of these processes even translated into two organizations having to wait almost six months to be reimbursed for activities they had already completed. This was symptomatic of the limited capacity within university structures to really support PAR projects, which require sensitive and alternative approaches to partners such as local communities, NGOs or social movements. These situations highlight that PAR requires sensitivity to many different issues that distinguish it from other research and which, in a very broad sense, university structures and administrative norms do not allow sensitivity to. Much of this goes back to questions about what universities are supposed to do, or what kind of work they are structured to support.

As we noted, PAR is not new, but it is still marginal. In a context in which universities are incentivized to streamline their work in favour of the most efficient activities (Jacobs, 2010; Dougherty and Natow 2020), they are also incentivized to avoid providing services or support tailored to marginal activities with little ‘profit’. Sadly, this is very firmly where PAR currently fits, as an activity that is far outside the scope of the normal research for which academic structures provide suitable support. To connect back to the concerns of Janes (2016), we see these kinds of structures as inherently problematic for any PAR projects, and as being precisely the dynamics that lead to the re-inscription of hierarchies in the creation of knowledge. If alternatives to normal, extractive, supposedly ‘objective’ science are always experienced by researchers as an uphill struggle, then they are always being pushed towards what is inherently, if unconsciously, supported and supportable within the systems that exist. Expanding this out from the level of individual university structures to those of academia more broadly, we can also see precisely how the funding mechanisms, including those discussed here, further structure the environment in which universities function. Particularly in research attempting to build relationships with disempowered or marginalized groups, with whom sustained engagement will be necessary, as described by Evans (2016: 214), the lack of mechanisms allowing ongoing support for PAR projects fundamentally hinders the potential for such research.

Undermined motivation and emotion

The experience of implementing a PAR project within the temporal constraints and bureaucratic structures of contemporary academia has been different for each of the academic team members. Overall, however, the main emotion we collectively settled on to describe this experience is one of ‘disillusionment’, as we have seen the lofty goals of PAR as described above consistently undermined by the neoliberal context. To some this may not seem like a pertinent point to make. What have our feelings got to do with our findings, after all? However, we believe quite firmly that it is very important to note that our motivation for and emotions regarding this project were greatly impacted by how the institutional settings we are entangled within do or do not align with what have been substantively defended as more ethical research practices. Because PAR requires time, attention to relationships, and practices of care (Smith et al., 2021; Mallan et al., 2010; Palmer, Burton and Walsh, 2020) that are at odds with the environment of competition and surveillance that dominates contemporary academia, it is very hard to make the two compatible. For us, the structures imposed by the current environment in HE meant that implementing PAR was always an uphill struggle.

For each of us this endeavor was a side project that, despite having the potential to expand in time into a bigger project, had a limited scope. It was always framed as a pilot project, a feasibility test. One of the reasons for this was simply that the very limited budget of the project did not allow any funds to buy out our time, as mentioned earlier. As a result, any activity on this project was piled on top of a myriad of other commitments and our individual involvement was dictated by other priorities and responsibilities. Despite our intentions to try to develop collaborative, non-hierarchical, supportive and trusting relationships, our contacts with the partners in the workshops and via email became limited, as time went on, to the practicalities of planning the project, collecting the data, or, as problems emerged, of overcoming administrative hurdles. While it may be that having sustained spaces of collective non-hierarchical work is what enables researchers to develop alternative relations of power in PAR (Kesby et al., 2007 : 24), we did not find this to be feasible within this project. To sustain the motivation, energy and commitment necessary to craft such an environment would require very different systems and structures of support within academia.

The key point to communicate here, however, is that the emotional state that the neoliberal university generates is not bounded within academia. We witnessed how it was transferred also to the non-academic research partners. As the project wore on, for example, it became clear that the local partners were also struggling to fully participate in the process, that their engagement was below what we had initially expected. This may evidence Wilson's concern with the tendency for academics to over-estimate how much time non-academic team members will have to devote to PAR projects (2022: 15), or Rosen's warning that researchers have to learn to cope with partners’ messy temporal forms of participation (2021: 7–8). However, we feel it is also evidence of the lack of time we were able to commit to develop relationships and the lack of accompaniment and mutual capacity building overall that we could engage in. The choices made at the start of the project impacted on the time and capacity each organization had to commit. In addition, none of this was helped at all by the lack of face-to-face engagement between academics and partners on the ground in each context (restricted due to the funding, but unfeasible anyway due to the Covid-19 pandemic). However, we are also very confident that our inability to commit more time to online engagement, to ensure the local partners of any future or ongoing collaboration, and to develop more trusting and non-hierarchical relationships, led to less commitment to and motivation for the project on their end.

How the outputs of this project, including this paper, were produced also reflects how the pressures of the neoliberal structure of the academy influenced our work and our decisions. While we proposed analysing the data and writing outputs in partnership with the non-academic researchers, when most of the partners were not enthusiastic about this, we did not force the issue. For us, it was easier to acquiesce. After all, co-writing would mean extra responsibilities that we were not able to take on, such as doing extra workshops for analysis and writing, circulation of additional drafts and extra proofreading. For academics, the easiest fallback position is to accept the role we are expected to play, of knowledge producer and author. In this sense, we were not able to overcome the researcher/participant divide and the majority of the partners were more comfortable with and more prepared to work within the classical division of labour, where the academics conducted the analysis and writing. The ghost of the ‘academic as expert’ haunted us and we were unable to banish it.

All of these dynamics led, in the end, to feelings of disillusionment, or that we had fallen quite short of the PAR ideal. That is not great, of course, but it seems more important methodologically to note not how we felt after, but how we felt during the project. Our motivation and emotions during the project had real implications methodologically, as they clearly impacted our energy and enthusiasm for the project, how the project was progressing and how our partners were, or were not, engaging with us and with the research. As PAR depends on relationships, and relationships are inherently two-way, the approach relies on academic researchers working within contexts that allow them to construct positive relationships with people outside the academy. This, our experience indicates, is extremely difficult within the context of contemporary neoliberal academia. This points, therefore, to various ways that the institutions that structure research practice lead precisely to the kind of outcomes already identified in the literature, including the retreat to traditional forms of research and the re-production of hierarchal inequalities around knowledge production as noted by Jacobs (2010: 374).

Conclusion

This paper makes a straightforward argument about the limits of conducting PAR research – with its emphasis on care, co-constructed knowledge and iteration – in the context of the neoliberal university. The paper proceeded in three steps. First, we reviewed the literature regarding PAR and noted that it makes very robust ethical and political claims about the benefits of the approach in recognizing marginalized actors' and communities’ knowledge and the relevance of co-created knowledge for resisting and transforming structural inequalities. Further, PAR has the added benefit of generating more accurate knowledge about local actors’ needs and problems. This approach is, however, dependent on academic researchers being able to commit to PAR processes that take time, develop non-hierarchical and trusting relationships with non-academic research partners, and abdicate the power usually retained by academic researchers regarding many of the choices about research, including the questions, the methods and the research ‘outputs’ to be developed. In the second step we outlined the contours of neoliberal academia, reviewing literature that describes a context in which efficiency in teaching, research and publications are paramount, in which an audit and surveillance culture has been constructed to incentivize research activities deemed more ‘valuable’ or even ‘profitable’ for universities, and in which managerialism ensures that the majority of researchers do not diverge from the usual modes of science.

The final step was to illustrate how our project struggled with implementing its PAR methodology. We did so by describing the challenges posed by the neoliberal academic context in which we work in three categories: (1) limitations of time, (2) structures and lack of support and (3) undermined motivation and emotion, and illustrating how each of these connects to or illustrates existing findings in the literature. In general, to avoid simply reiterating points, both the funding institution and the workload imposed on us by our universities limited the time we could commit to this PAR project, the structures and administrative dynamics were not designed to support PAR research, and the context in which we work, as a whole, undermined the motivation and emotion necessary to engage suitably with the process. While certainly not impossible, therefore, implementing a PAR project within this context was an uphill struggle in which we as academic researchers were pushing against the mechanisms of managerialism already established to incentivize ‘normal’ science which, not by design but by default, function to disincentivize truly non-hierarchical participatory approaches. So, what lessons can be drawn from this experience? How can funders and universities better support the implementation of PAR projects? Our recommendations revolve around issues of time, structures and support.

First, on the issue of time. To better support PAR projects, funders must allow much more time for the development of projects, more time for the implementation of research projects (including face-to-face engagement within research projects), and more time for the development of the research ‘outputs’ they demand. Most research funding today disincentivizes PAR, and if we want to incentivize this kind of ethically committed approach to knowledge creation (so clearly valuable for studies regarding topics related to social justice), then that needs to be rethought. Similarly, universities must find ways to protect researchers conducting PAR from the temporally restrictive publish or perish race and the research audit cycle. As the literature has made it clear that this form of research takes more time to produce publications and other research outputs, a strong argument can be made that universities should be more flexible with academics carrying out PAR. This argument is even more relevant if we consider that universities claim also to value the ideas embedded in the PAR approach and what it can provide, such as more inclusion, diversity, community engagement, policy relevance, practical impact and decolonized knowledge production.

Second, on the issues of structures and support, universities need to be more flexible in a variety of ways: in their contracts, their ‘supplier’ arrangements, their reimbursement procedures, their ethics approvals, and their expectations about what research is, how it should be conducted, what the ‘outputs’ should be, and how long those should take. This could be in the form of less restrictive contracts, supplier forms and reimbursement processes that might allow academics more leeway in the initial inclusion of partners and the eventual distribution of funds. But simply allowing more time would be helpful as this would offer more flexibility for those doing PAR or other kinds of research that demands a timescale that is incompatible with the current accelerated timelines of academia. However, this would require that administrators and managers trust academics and that they refrain from trying to discipline them through managerial surveillance and auditing, which are currently structured to disincentivize and undermine PAR and related approaches. These are not simple requests in the face of existing neoliberal structures, but if achieved they would go a long way to smooth the terrain.

We also believe that the system needs to transform how research participants are envisioned. Many of the procedures within contemporary institutions of academia are based on a model that sees research participants as passive actors without much to contribute to research processes. As we noted above, some of these administrative procedures, such as those to ensure ethical research, treat participants largely as passive subjects of research. Others, such as ‘supplier’ and reimbursement forms, treat them as market actors, in that they reify research partners as commercial entities selling services to the academy. These models perpetuate passive and market subjectivities among participants involved in research projects and undermine the possibilities for doing PAR, which requires that we see research partners differently. An alternative model would be based in a conception of people outside academia as both having valuable knowledge and having the capacity to participate as equal partners in generating new knowledge. Such a model should allow non-academic actors to participate as equals, instead of being held at arms-length by the academic community.

Finally, even though our analysis has led us to emphasize the negative aspects of the intersection between PAR research and neoliberal academia, we would like to stress also that using PAR led to some positive outcomes. Near the end of the project, one of the organizations commented to one of the academic researchers that the workshops with the local communities were an important space for reflection and that they expect to continue using the data generated to create memories about their work. Further, as noted above, throughout the project the local partners voiced their willingness to engage further in research projects and their desire to see the project grow into something more. While we have used this paper to illustrate the problems within academia today that are disincentivizing PAR endeavours, there does nonetheless appear to be an appetite for more inclusive approaches to the knowledge generation process. Our project was always intended to be a pilot, and we do certainly feel as though we have learned a lot ourselves from partaking in the process. So, maybe the reality is not as bleak as it may initially appear.