Careful knowing as an aspect of environmental justice

ABSTRACT

Environmental justice (EJ) issues commonly include contestation over knowledge claims. EJ scholarship tends to theorize these as issues of participation or recognition. Drawing on a long-term ethnography of community-led air monitoring, I show that frontline communities experience distinctly epistemic injustices, even in situations characterized by participation and recognition. These epistemic injustices, which include exclusion from judgment, inadequate epistemic resources, and denial of status as knowers, point to the need to expand definitions of EJ to include an account of epistemic justice. Departing from virtue-based accounts, I propose ‘careful knowing,’ or practices of empirical investigation and meaning-making responsive to the needs of marginalized knowers, as a remedy for epistemic injustices that occur in EJ settings. Context-specific by definition, careful knowing may include grounding epistemic judgments in the need for community health protection, expanding epistemic resources, and bolstering the confidence and status of marginalized knowers.

KEYWORDS:

• Epistemic injustice
• environmental justice
• epistemic resources
• community air monitoring
• careful knowing

Introduction

Movements for environmental justice (EJ) have long criticized regulatory processes for placing too great an emphasis on scientific expertise, often to the exclusion of the knowledge and values of people most affected by pollution and other environmental harms (Kuehn 1996, Aitken 2009). One strategy of EJ campaigns has been to launch their own scientific investigations, such as health studies, air and water monitoring, and participatory mapping projects (Brown 1992, O’Rourke and Macey 2003, Wylie et al. 2014). For decades, U.S. environmental regulators accepted community-led science grudgingly, if at all. In the 2010s, however, environmental regulators began to imagine that citizen science could play an important role in environmental protection (National Advisory Council for Environmental Policy and Technology 2016), and they developed programs to encourage community participation in environmental monitoring and exposure assessment. Whether initiatives for participatory science actually foster EJ is an open question, however. On the one hand, they stand to address disparities in air quality and build communities’ scientific capacity. On the other, they have been criticized for prioritizing the study of environmental exposures over preventive action and creating ‘data treadmills’ that reinforce the dominant role of science in environmental decision-making (Shapiro et al. 2017, Sadasivam 2021).

Ideally, theories of environmental justice would be able to guide assessments of initiatives to promote EJ through science, much the way that political theory has been used to evaluate the adequacy of different modes of participation (e.g. Gauna 1998). However, existing theories of EJ, which tend to treat disputes over science as issues of recognition and/or participation, leave open the possibility that experts could legitimately claim greater esteem and a privileged place in environmental decision-making on the basis of their scientific knowledge and credentials — contrary to EJ activists’ insistence on respect for the insight of the people most harmed by pollution and climate change, often called ‘frontline communities.’

Theories of epistemic injustice, or wrongs done to people in their capacity as knowers (Fricker 2007), offer a necessary complement to recognition and procedural justice in theories of EJ. They demonstrate the need to be skeptical of expertise-based hierarchies of esteem and participation, because marginalized people are often marginalized as knowers as well. In the EJ literature, theories of epistemic injustice have been used by a handful of scholars to highlight governments’ failures to recognize Indigenous knowledge and ways of knowing (e.g. Tsosie 2012, Weißermel and Chaves 2020). However, theorists have yet to articulate a positive definition of epistemic justice apt for assessing science-oriented initiatives to promote EJ.

In this paper, I offer an account of epistemic justice suitable for EJ contexts: careful knowing, which I define as practices of empirical investigation and meaning-making responsive to the needs of marginalized knowers. This concept, I argue, captures EJ activists’ actual responses to epistemic injustices and subverts the possibility that experts could re-establish epistemic hierarchies by claiming special status on the basis of their knowledge, credentials, or scientific virtues. I contend that definitions of EJ ought to be expanded to include careful knowing alongside recognition and participation, in order to more effectively guard against the domination of environmental decision-making by credentialed experts.

Environmental justice as recognition and participation

Taking his lead from contemporary theorists who have challenged the notion that justice is reducible to a problem of distribution (e.g. Young 1990, Fraser 1996, Honneth 2004), Schlosberg (2004, 2007) first articulated a pluralistic account of what environmental justice means to its advocates, including recognition of the unique cultural backgrounds of communities heavily burdened by pollution, and community participation in decisions about hazardous facility siting and regulation. His definition has come to be used as an analytical framework through which to understand the dynamics of contestation between vulnerable communities, polluters, and the state, and as a yardstick against which governments’ handling of local environmental issues can be assessed.

Drawing on Schlosberg’s definition, scholars concerned with the dominance of experts in environmental decision-making, or the fate of communities’ scientific investigations, have framed these issues as problems of participation, recognition, or both (e.g. Corburn 2003, Holifield 2012). But a closer examination of the theories underlying these aspects of environmental justice shows that they don’t necessarily rule out processes that afford experts a dominant role in environmental decision-making. The next two sections outline their ambiguities and demonstrate the need to expand our definitions of EJ to deal effectively with questions of science, knowledge, and expertise.

Knowledge and participation

The meaningful participation of communities in environmental decision-making is often inhibited by appeals to science and scientific authority (Aitken 2009, Kojola 2019). Empirical research portrays this dynamic as unjust, and positions ‘citizen science,’ or public participation in the making of science, as an important way to increase communities’ meaningful participation (Schlosberg 2007, p. 70, Kimura and Kinchy 2016). The idea that procedural justice would necessitate communities’ involvement in the production of knowledge is reinforced by scholars who justify more robust participation in environmental decision-making by pointing to the uncertainties inherent in science (Funtowicz and Ravetz 1992, Ottinger 2013b) and suggest that expanded knowledge should be a goal of participatory processes (Young 2004).

In contrast, other theories of procedural justice afford scientific knowledge and credentialed experts a privileged place in environmental decision-making. For example, Shrader-Frechette (2005) contends that communities deserve the right to consent to chemical exposures, and suggests that their decisions ought to informed by the best available science. Similarly, Anderson (2011) envisions citizens’ role in democratic deliberation of science-laden issues as one of making second-order assessments of the trustworthiness of scientific claims. Both suggest the possibility that the requirements for procedural justice could be met without allowing people who lack scientific credentials to contribute to the body of knowledge on which decision are to be based. Positing participation as a requirement of EJ, then, leaves ambiguity about the status of communities’ knowledge and ways of knowing relative to the claims of credentialed experts.

Knowledge and recognition

Justice in recognition is a second aspect of EJ complicated by the role of scientific experts in environmental decision-making. EJ scholars and advocates have argued that justice entails not only recognizing cultural differences in how diverse communities interact with and give meaning to their environment. It also demands recognizing differences in communities’ knowledge and ways of knowing, such as the distinctive ontologies and epistemologies of Indigenous people (Schlosberg 2007; Tsosie 2012). Environmental decision-making that acknowledges only Western science as a valid source of knowledge arguably harms frontline communities where people may know differently, by undermining individuals’ abilities to see themselves, and be seen by others, as contributors of judgment and insight to political processes (c.f., Schlosberg 2003, p. 82–83).

At the same time, theories of recognition suggest that justice does not necessarily require equal recognition, especially where it comes to knowledge. Building on Honneth’s (2004) concept of recognition, Congdon (2017, p. 249) makes a distinction between respect for individuals as knowers, which all people deserve equally, and esteem for the particular knowledge or expertise that an individual holds, which would be greater for those who have distinguished themselves by, for example, earning scientific credentials. Fraser (1996, p. 37) agrees that not all knowledge claims deserve to be recognized equally, even when disesteeming certain knowledge systems might inhibit individuals’ participation in political life. Judgments about who is owed recognition, for Fraser, can be legitimately informed by value hierarchies, as long as those hierarchies are just and relevant.

When EJ advocates argue that justice in recognition requires esteem for frontline communities’ knowledge, they imply that the hierarchy that values Western science above other ways of knowing is unjust (see, e.g. Tsosie 2012). This argument is easier to sustain with respect to knowledge systems that are complete and coherent in their own right, such as Indigenous knowledge, than it is when it comes to knowledge claims that resemble those of Western science but do not fully meet its standards, like communities’ first-hand accounts of disease clusters. Following the logic of Congdon or Fraser, peer-reviewed scientific research could rightfully claim higher esteem than a community’s collection of stories; residents would only be owed recognition as people capable of producing research of the highest quality, as defined by prevailing scientific norms.

Epistemic justice through care

Theories of participation and recognition could thus justify dominant roles for science and expertise that EJ practitioners would find objectionable. Theories of epistemic injustice, defined as ‘wrongful treatment and unjust structures in meaning-making and knowledge producing practices’ (Kidd et al. 2017, p. 1), offer an important complement, in that they show how injustice can manifest in purportedly neutral standards for science. The positive notion of epistemic justice articulated by most theorists, however, relies on the cultivation of virtue, an approach that I argue is limited as a remedy for the epistemic injustices experienced by frontline communities. I posit that an ideal of epistemic justice grounded in an ethic of care may be more apt for EJ contexts.

Identifying epistemic injustice

Theories of epistemic injustice, like theories of recognition, acknowledge that some individuals, such as credentialed scientists, might legitimately deserve more credibility than others on the basis of their long study and deep knowledge of a particular topic (Fricker 2007). But as scholars applying epistemic injustice theory in EJ contexts demonstrate (e.g. Tsosie 2012, Temper 2019, Weißermel and Chaves 2020), it also offers a framework for understanding how affording a privileged position to scientists can intersect with and perpetuate injustices in participation and recognition, as well as the distribution of environmental harm.

Epistemic injustice theory reminds us, first, that the identity of an expert knower is inseparable from other aspects of the individual’s identity (Fricker 2007). A white male scientist, for example, may be afforded more credibility than is justified by his expertise, on the basis of his gender and his whiteness.

Second, theories of epistemic injustice show how it is a structural phenomenon, in the sense that there may be gaps within an epistemic system — missing concepts, for example — or epistemic systems may be inadequate to knowing the world as experienced from marginalized standpoints (Dotson 2014, Pohlhaus 2020). Scientific knowledge may be robust and reliable when judged by the standard of its epistemic system, but that does not obviate the need to examine the potential limitations of that system.

Finally, theorists have shown that knowers from dominant groups perpetuate epistemic injustices by using inappropriate frames of meaning to judge the work of knowers with marginalized identities (Dotson 2012) and failing to acknowledge new epistemic resources being produced from marginalized standpoints (Pohlhaus 2012). As I have argued elsewhere (Ottinger 2022), scientists who insist that any knowledge offered to public processes meet their standards for rigor contribute to the silencing of marginalized knowers and help to institutionalize ignorance of marginalized experiences.

The virtue of epistemic justice

Most theorists suggest that epistemic injustices can be remedied by the cultivation of virtue. Justice requires those hearing knowledge claims to do so with curiosity, humility, and open-mindedness, to cultivate a reflexive awareness of the limitations of their own knowledge systems, and to offer a kind of benefit-of-the-doubt to knowers who may come with different frames of meaning (e.g. Fricker 2007, Medina 2012, Goetze 2018). Recognizing that epistemic injustices are often structural, scholars have argued that institutions, too, need to cultivate epistemic virtues (Anderson 2012).

The predominance of virtues as the basis for epistemic justice raises two concerns. First, scientific authority has long been intertwined with claims to virtue (Shapin 1994). Scientists called to cultivate the virtues of epistemic justice could very plausibly do so in a way that re-established epistemic hierarchies on reworked grounds (c.f., Ottinger 2013a). More generally, Dotson (2012) argues that any singular account of epistemic justice is likely to create its own epistemic oppressions and exclusions, because epistemic injustices are diverse and the standpoints of theorists partial. Dotson thus advocates ‘open conceptual structures’ (p. 41–42) that allow for multiple, coexistent accounts tailored to specific manifestations of epistemic injustice.

Care as a basis for epistemic justice

My central claim in this article is that an alternative basis for a conception of epistemic justice can be found in the ethics of care. Defining care as ‘a species activity that includes everything we do to maintain, continue, and repair our “world” so that we can live in it as well as possible,’ Fisher and Tronto (1990, p. 40) stress that, while care may begin as an attitude, it is also necessarily a hands-on practice, grounded in recognition of the needs of the cared-for. Feminist political theorists posit care as a more expansive framework for political judgment than justice, and they reject the idea, common in theories of justice, that obligation and desert should be defined in universal terms. Instead, they see our responsibilities to each other as grounded in our particular relationships and needs (Tronto 1995, Held 2006). Caring relations are necessarily asymmetrical; as a result, thoughtfully navigating power relations between care providers and the cared-for is an inherent part of caring practice. ‘Benevolent domination,’ in Held’s (2006, p. 56) words, is to be avoided, as is coercion, whether blatant or subtle.

An ethic of care suggests itself as a foundation for epistemic justice because caring relations are entangled with epistemic issues. One cannot care adequately for others without knowing what they suffer and what they need, and that in turn requires recognizing the limits of one’s own standpoint and valuing the knowledge of the cared-for (Casalini 2020). Working to generate knowledge through committed attention can be seen as a form of care in and of itself, as scholars calling for care as an ethico-political framework for addressing EJ and sustainability issues have shown (e.g. Tironi and Rodríguez-Giralt 2017).

At the same time, epistemic practices can strive to incorporate care and caring relations (Puig de la Bellacasa 2017). A care orientation shifts the priorities of Western science, calling on scientists to understand themselves in relation to others and mitigate the ways that their power and authority tend to marginalize other knowers (Staffa et al. 2021). In other words, epistemic practices characterized by care incorporate attention to epistemic injustices and offer a potential remedy to them. In contrast to a virtues-based account of epistemic justice, such practices would demand that more privileged knowers not just cultivate better attitudes, but ground their epistemic practices in an awareness of their accountable relations with others. Simultaneously attending to others’ context-specific needs and avoiding benevolent domination would also entail backing off from supposedly universal standards for rigorous knowledge and helping marginalized knowers develop epistemic resources of their own.

In keeping with Dotson’s (2012) call for open conceptual structures, an account of epistemic justice grounded in care need not supplant virtue-based accounts of epistemic justice. If it can be shown to effectively address the epistemic injustices experienced by frontline communities, then it is worthwhile as one account of epistemic justice apt for an EJ context. While existing EJ scholarship suggests potential connections between care and epistemic justice, it has for the most part not developed these systematically.

Methods

Following a tradition of looking to social movements’ definitions of justice as a starting point for theorizing (e.g. Young 1990, Schlosberg 2007), my research focuses on the engagements of communities adjacent to oil refineries and petrochemical plants with projects to monitor toxic gases in the ambient air. Concerted efforts on this issue date back to the 1990s, when activists in Northern California pioneered an inexpensive, homemade air sampler, known as the bucket, as well as a high-tech system for continuous monitoring of pollutants at refinery fencelines. The bucket spread quickly to frontline communities around the world, including petrochemical-dominated areas of the U.S. Gulf Coast (O’Rourke and Macey 2003). The fenceline monitoring system was installed at a refinery in Rodeo, California, in 1996. It remained the only one of its kind until the 2010s, when California regulators began adopting new rules to require similar systems at all of the state’s oil refineries. Around the same time, regulatory agencies, including the U.S. EPA, began seeing frontline communities as potential participants in air monitoring and developing programs to foster their inclusion.

My research on EJ activists’ air monitoring work began in 2001 and included an ethnography of two communities in Southeast Louisiana that used buckets as part of campaigns for relocation (Ottinger 2013a). Since 2001, I have also volunteered at various times in a variety of capacities for three EJ organizations that played leading roles in air monitoring advocacy: Communities for a Better Environment (CBE) in California, the Louisiana Bucket Brigade, and Global Community Monitor. In 2014, I began a project that explicitly examined how activists’ air monitoring work included calls for epistemic justice. Focused primarily in Northern California but including Southeastern Louisiana and nation-wide programs for community monitoring, my research comprised oral history interviews with the activist developers of buckets and fenceline monitoring; participant-observation in on-going deliberations about new monitoring rules and programs; and participatory design of online tools for interpreting air monitoring data. Notably, the activists whose work I studied operated within the paradigms of Western science; Indigenous knowledge systems were not at issue. All interviewees quoted here gave their consent to have their real names used as part of my research. I also use speakers’ real names when attributing quotations from public comments at public meetings.

Air monitoring and epistemic (in)justice

In their interactions with frontline communities pushing for better air monitoring, environmental regulators, and even petrochemical industry officials, often encourage community participation and recognize communities’ distinctiveness. Yet their efforts at participation and recognition still tend to be characterized by epistemic injustices, as I demonstrate below. These stem from treating science as separate from, rather than subject to, deliberation, and reserving a special kind of esteem for those with scientific credentials — neither of which is wholly inconsistent with justice in participation and recognition, as we saw in the preceding discussion.

To respond to these injustices, EJ activists have cultivated a number of epistemic practices that incorporate qualities consistent with an ethic of care. Specifically, they foreground communities’ need to be protected from pollution and the public’s obligation to help meet that need. They create epistemic resources adapted to the need and obligation of environmental protection, rather than imposing universalistic standards for scientific rigor that may be ill-suited to the context, and they encourage people who have been consistently put down by scientists to recognize themselves as knowers. These practices help suggest the contours of careful knowing as a remedy for epistemic injustice.

Epistemic injustice in contexts of participation and recognition

On 31 January 2018, California regulators met with San Francisco Bay area residents to discuss the kick-off of the Community Air Protection Program (CAPP). The CAPP directed attention to air quality in frontline communities, recognizing that they had not necessarily benefitted from programs to reduce overall air pollution, and explicitly encouraged the participation of community members in air monitoring. The kick-off event showcased the efforts of several community groups with a long track record of doing their own monitoring in collaboration with the Bay Area Air Quality Management District (BAAQMD).

Despite the agency’s explicit commitment to community participation, one resident, LaDonna Williams, felt it necessary to ask whether regulators would accept data from low-cost particulate matter sensors that her community had adopted. BAAQMD representative Greg Nudd responded yes, the agency would accept data from the sensors: ‘The data that’s collected by those devices is very useful and it informs us.’ He then qualified his answer, ‘[the low-cost sensor data] is not actionable in any regulatory sense. … I would expect that any regulatory action we take will be followed by a lawsuit, so we gotta be sure that any action we take we are backed up by regulatory-quality data’ (quotations from my transcription of the webcast).

In this interaction, members of frontline communities were welcomed as participants in generating data. However, they were excluded from deliberations about what standards data needed to meet to inform action. Community members’ exclusion from this sort of technical judgment was so normalized, in fact, that Williams began by asking whether her community’s data would be accepted, positioning regulators as epistemic gate-keepers. Nudd, for his part, located these issues of judgment outside the participatory context by invoking ‘regulatory-quality data’ and ‘actionable in a regulatory sense’ as pre-established concepts not subject to discussion.

Similar questions arose at the EPA’s Community Air Monitoring Training on 9 July 2015. The event ostensibly fostered participation by giving frontline communities resources to design and carry out local air monitoring projects. The event’s premises were questioned early on by Brian Butler, then a staff member at Air Alliance Houston. Butler asked the lead-off speaker, EPA’s Ron Williams, in essence, what’s the point of community monitoring if it doesn’t lead to change? Butler then related how his organization had worked with community groups to do a ‘comprehensive study’ of air quality. But when they submitted their research to regulators, they were told that, because they had not used a federally approved method for collecting data, their results didn’t count.

Butler was both challenging the EPA’s standards for evidence and forcing them into the public discussion. But Ron Williams, speaking over applause from attendees who had perhaps had similar experiences, did not even acknowledge the questions of evidence and epistemic judgment that Butler raised. Instead, he suggested he could ‘rephrase’ Butler’s question for him, as ‘How can data be usable, and how can I be taken seriously?’ (from my notes on the web broadcast). In Williams’s framing, it was incumbent on activists to conform to regulators’ scientific methods, and the EPA’s Air Sensor Toolbox efforts were meant to help them do so. Methodology and standards of evidence were not within the scope of participation.

In his dedication to predetermined scientific standards, Williams dodged a second aspect of Butler’s question. Butler wanted to know how air monitoring could lead to change or, in other words, reductions in the pollution that frontline communities were being exposed to. Notably, Williams’s answer did not acknowledge communities’ need for cleaner air, or his agency’s obligation to ensure that they got it. Instead, he suggested that any action on regulators’ part was contingent on ‘usable data.’

These examples show how regulators’ efforts at bolstering participation for frontline communities tend to exclude them from important epistemic judgments—a form of epistemic injustice — while abstracting those judgments from communities’ pressing needs for environmental protection. Regulators may also create forms of participation that fail to recognize the limits of their own epistemic systems and communities’ need for better epistemic resources. At the Community Air Monitoring Training, as well as in the US EPA’s ‘Air Sensor Toolbox for Citizen Scientists,’ training materials stressed that each community air monitoring project should have a Quality Assurance Project Plan (QAPP). Among the required elements of the plan was a schedule for air sampling. A schedule makes sense if the goal is to measure average levels of pollution, which is how regulatory agencies tend to think about air quality — as consistent enough to be modeled adequately by averaging pollutant concentrations over time.

What this epistemic framework misses is the fact that that communities located on the fencelines of industrial facilities do not experience average air quality. Rather, residents are acutely aware of fluctuations, including noticeable surges of pollution resulting from flares, fires, or other industrial releases, which can last for anywhere from several minutes to many hours. Although these episodes are unpredictable and may have unrelated causes, communities argue that repeated releases amount to a chronic problem for their health and quality of life. Accordingly, they object to monitoring methods that average out these spikes in pollution. By insisting that a sampling schedule be included in each QAPP, the EPA’s programs to foster participation effectively denied the significance of unpredictable fluctuations in pollution levels. Not only did it leave little room for community monitoring to be responsive to fluctuations; it also inhibited any effort toward developing methodologies and vocabularies to represent them.

Epistemic injustices thus persist even in regulatory programs explicitly geared toward fostering EJ through participation by frontline communities: community participants are excluded from epistemic judgments and denied access to adequate epistemic resources. These injustices in participation are underpinned, unsurprisingly, by a failure of recognition — in particular, the failure to recognize frontline community members as knowers, even when they may be afforded recognition in other ways.

This dynamic was especially apparent in the resolution of the Diamond community’s campaign for relocation away from the Shell Chemical plant in Norco, Louisiana. African-American residents began in the 1990s to pressure the company to buy all of the homes in their four-street neighborhood. They argued that the proximity of the facility put them in danger and took samples with bucket air monitors to establish links between petrochemical pollution and what residents alleged were unusually high rates of asthma, sarcoidosis, and other illnesses (Lerner 2005. In 2000, Shell announced that, in order to create a greenbelt around its facility, it would offer voluntary buyouts to two streets in Diamond. The initiative split the neighborhood in half and escalated community opposition to the plant. Under pressure from international activists, Shell agreed in 2001 to negotiate with residents and finally offered to buy the remaining two streets in 2002 (Lerner 2005).

Shell executives justified the buyout offer in terms of their newfound recognition of the distinctive character of the African-American community of Diamond. What they came to understand during negotiations with activists, they said, was that familial and care-taking relationships spanned the entire four-street neighborhood in ways that were unfamiliar to them as white people. By initially offering to buy homes on only two of the streets, they had unintentionally damaged the social fabric (Lerner 2005). A joint statement announcing the second buyout invokes the language of recognition: ‘The discussions [between Shell and residents] resulted in Shell having a greater appreciation for the unique social character and interdependencies of the Diamond neighborhood in Norco and how it was uniquely affected by the Voluntary Property Purchase Program [the first buyout]’ (Texas SEED Coalition 2002).

Yet the cultural recognition afforded to the Diamond community in the relocation process was not accompanied by any recognition of residents as knowers, despite the fact that residents had been active in taking air samples and documenting patterns of illness. In the negotiations, Shell officials declared off-limits any discussion of the alleged health effects of the plant. The company had done its own health studies, I was told by David Brignac, a Shell manager involved in the negotiations, and their findings showed no adverse health effects that could be attributed to the plant. When they shared their most recent study with activists, Brignac told me, ‘we said, “our position is, we’re not harming health, and so as a result, we don’t want health to be part of this negotiation”’ (interview, 18 December 2002). By taking this position, Brignac and his colleagues effectively denied any possibility that Diamond residents could have anything to contribute to the company’s understanding of its impacts. Their air samples, lived experiences, and decades of observation were not recognized as relevant knowledge.

In the months following the buyout, Shell launched its own air monitoring program in Norco. They actively sought public participation and included Diamond residents in the process of designing the program. Surprised to find that Norco residents imagined there could be significant variation in air quality across the few square miles covered by the town, the scientists in charge of the program agreed to establish several monitoring sites at representative locations. But although their response to community participation exhibited curiosity — a core epistemic virtue — they stopped short of affording residents full status as knowers, making them members of the program’s ‘Communications Team’ but not its ‘Technical Team.’

Epistemic practices with qualities of care

As the previous section shows, frontline communities face epistemic injustices even in contexts where participation is encouraged and cultural difference recognized. Activists counter these injustices through epistemic practices that show qualities of care. First, exercising epistemic judgment is a key aspect of air monitoring advocates’ work. This practice helps counteract the exclusions from judgment that frontline communities face in government-sponsored participation. Moreover, it exemplifies an ethic of care by framing epistemic judgment in terms of pollution prevention, a core need of frontline communities to which regulatory agencies ought to feel obligated.

The judgments made by frontline community residents and EJ non-profit staff involved in air monitoring programs include where and when to take air samples and how to assess monitoring results. Communities using bucket air samplers purposefully monitor at observably high-pollution times rather than on a predetermined schedule, thereby elevating communities’ experiences of pollution over the demands of techniques to standardize pollution measurements, such as QAPPs. Similarly, activists may interpret their sampling results in ways that diverge from regulators’ interpretive frameworks, in order to highlight the systemic burden that air pollution places on communities.

Community needs and public obligations are also front-and-center in the judgments that activists make about the value of monitoring versus other activities. Howard Adams, a resident of a refinery-adjacent community in northern California, was involved in getting a real-time, continuous fenceline monitoring system established at the refinery in the 1990s. But in 2013, he told me he didn’t support the push to require similar systems at all five refineries in the area, ‘because it’s so expensive and you get so little out of it.’ Instead, he had turned his efforts to supporting industrial safety initiatives, which could more directly protect his and other communities from accidental releases (interview, 19 June 2013). Julia May of CBE, also a force behind fenceline monitoring, told me of a case in which southern California refineries had successfully delayed regulation on emissions from their coker units by saying that more air monitoring data were needed. Although she had been an advocate for monitoring in other instances, she opposed it in this case: ‘it was obvious that it’s a bad engineering practice to just let these vapors dump to the air every day. … you really don’t need monitoring to show that’s a bad engineering practice’ (interview, 6 June 2013). In May’s judgment, rather than expending resources on monitoring, regulators and industry ought to have been focused on addressing polluting practices.

Activists took this same orientation to care into public forums where they tried to challenge communities’ exclusion from epistemic judgment. At the January 2018 kick-off meeting for California’s Community Air Protection Program, Brian Butler, by then at San Francisco-based GreenAction, argued that, ‘there’s no need to identify or figure out everything. We know where the places [where people are suffering from pollution] are. We know where the places are. … What we are calling on the Air District to do is … to protect public health’ (from my transcript of the webcast). His intervention questioned the epistemic judgments of regulatory agencies and tried to establish community members as legitimate participants in making these judgments by emphasizing their existing knowledge. Simultaneously, he tried to shift the basis for judgment, centering the government’s obligation to public health over its interest in gathering more data.

Second, the epistemic practices of activists involve creating resources well suited to knowing air quality in the context of frontline communities. This counters the injustice of limited epistemic systems and, in keeping with an ethic of care, helps to situate polluters’ obligations in the context of relations with neighboring communities. Air sampling with buckets, for example, captures industrial releases that cause foul odors in residential areas. Such releases are generally excluded from regulatory systems for knowing air quality. Air dispersion models, used to predict a facility’s impact on the surrounding area, incorporate routine emissions but seldom accidental ones. Moreover, regulatory responses to unplanned releases vary according their cause: excess pollution released when a facility is in the process of shutting down for maintenance or restarting are usually permissible, whereas a fire or large flare might demand investigation and incur penalties.

From the vantage point of neighboring communities, though, the cause of an odor is less important than its intensity. The practice of bucket sampling, accordingly, begins by assessing the strength of the odor. As a rule of thumb, it is worth sampling when the odor rates at least a seven on a scale of one to ten. In responding to all odors as the same kind of thing, one that can be rated on a common scale, communities using buckets constitute a new category, absent from regulatory logics: releases that befoul residential air. This not only expands the epistemic resources available to express the experiences of frontline communities; it makes a claim about the nature of industrial facilities’ obligations to communities. While environmental regulations may make fine distinctions between different kinds and causes of air pollution, the practice of bucket sampling asserts that facilities need to attend to any release that causes strong odor or disruption for neighboring communities.

Finally, activists’ epistemic practices show care by acknowledging the ways frontline communities’ knowledge is dismissed and denigrated, and helping residents to recognize themselves as knowers. Greg Karras of CBE told me that frontline communities were ‘being encouraged by polluters and the elites and their apologists that, that we’re stupid, we don’t understand, and we have to trust the experts.’ Against that backdrop, community monitoring ‘validated what people already knew and gave them a language to talk to people in power about it’ (interview, 18 March 2014). Denny Larson, a key proponent of the bucket air sampler, had a similar assessment: ‘People know that they’re being poisoned, and they know that when they smell the stuff it makes them sick … but they’re always told that they’re wrong and it’s not the case.’ Bucket sampling, as Larson described it, ‘becomes a really big “aha!” moment for people. “Aha! I knew it! I knew it, see I knew I was breathing this shit and I knew it was bad for me’ (interview, 19 December 2002).

Notably, in this aspect of activists’ response to epistemic injustice, the different positionality of frontline community residents and allied non-profit staff becomes more salient. Both groups participate in practices that open up epistemic judgment and expand epistemic resources. In their attempts to recenter community protection as a basis for judgment and the focus of industry and regulatory obligation, individuals from both groups act to bring care to frontline communities. But non-profit staff, most with college degrees, some in technical fields, are more often afforded status as knowers than residents of frontline communities are; Karras, for example, was invited by BAAQMD to give a technical presentation as part of their public hearings on proposed fenceline monitoring regulations in 2014. When it comes to bolstering community members’ status as knowers, then, non-profit staff are in the position of offering care to residents, meeting needs for epistemic status and confidence – while acknowledging the rich knowledge community members already have. As Ruth Breech, former Program Director at Global Community Monitor, put it, ‘intuitively [community groups] always know something’s wrong and, yes, we help them put that evidence and kind of package it more in a more professional and sophisticated way’ (interview, 21 June 2013).

Discussion: careful knowing in environmental justice

The evolution of ambient air monitoring strategies at refinery fencelines demonstrates that participation and recognition for frontline communities do not preclude epistemic injustices. Regulatory programs to encourage participation in air monitoring tend to bracket technical questions out of the discussion, excluding community members from participating in judgments about how monitoring data should be collected and interpreted, as well as whether there are sufficient data of high enough quality to justify action. Seeing technical questions as outside the realm of public deliberation also masks the inadequacy of regulators’ epistemic systems and stands in the way of developing new epistemic resources that can better communicate the experiences of frontline communities. Further, decision-makers may recognize frontline communities’ unique cultures, the particular environmental burdens placed on them, and their right to participate in decisions that affect them, all without acknowledging their potential to contribute to generating knowledge or making sense of evidence — that is, without recognizing them as knowers.

EJ activists’ engagement with air monitoring issues confronts the epistemic injustices that remain even in settings characterized by significant participation and recognition. They advance their own judgments about where, when, and whether air monitoring should take place, and how it should guide action. Their judgments are grounded not only in the demands of scientific rigor but in the larger need to protect communities from pollution. Activists’ monitoring practices also include the creation of new epistemic resources, like a category of impactful releases, which not only facilitate the creation of new kinds of knowledge, but also make a statement about the nature of polluters’ obligations to their residential neighbors. Non-profit staff who help develop community-centered monitoring practices understand that residents of frontline communities are frequently denied the status of knowers by experts from regulatory agencies and industry. In response, these allies both recognize the situated knowledge of residents and offer tools to bolster residents’ epistemic confidence.

These responses to epistemic injustice differ from those imagined by most theorists of epistemic justice, who call for the cultivation of epistemic virtues such as curiosity and humility to remedy epistemic injustices. EJ activists involved in air monitoring instead develop practices of knowing that respond to the needs of frontline communities and their residents — the need for environmental protection, the need for better epistemic resources to characterize the experience of living on the front lines of pollution, and the need for residents to feel able to assert themselves as knowers in confrontations with people who can claim an identity as ‘experts.’ Their orientation to epistemic injustice, in other words, displays qualities of care, in that it is responsive, practice-oriented, and context-specific. The case thus suggests an alternative account of epistemic justice appropriate to (at least) environmental justice contexts: careful knowing, or practices of empirical investigation and meaning-making responsive to the needs — epistemic and otherwise — of marginalized knowers.

The precise contours of careful knowing will, by definition, vary according to a community’s identity and the nature of the environmental threats they face. We can, however, anticipate common themes across situated practices of careful knowing. Careful knowing would almost certainly require balancing specifically epistemic needs — for more data or better metrics, for example — with overarching needs for environmental protection, health, and economic sustainability. Credentialed experts who develop practices of careful knowing would be challenged to remain genuinely responsive to the needs of frontline communities, as residents themselves understand them, even when epistemic needs are not predominant. As part of careful knowing, we could also expect universalistic standards for doing science to be supplanted, or at least supplemented, by methods tailored to specific, frontline contexts. The question of what data are ‘good enough’ for a given purpose could be expected to become more prominent in practices of careful knowing, as well (c.f. Gabrys et al. 2016). How these general themes may play out in a variety of contexts is a subject for further investigation.

Conclusion

Increasing recognition of frontline communities and expanded programs for their participation have not ended conflicts between residents and regulators over environmental knowledge and its use in decision-making. Indeed, some participatory monitoring projects have themselves become contentious (Sadasivam 2021). One reason, as I have shown, is that epistemic injustices often persist, even as greater justice in participation and recognition are realized: frontline communities are excluded from epistemic judgment, constrained by inadequate epistemic systems, and denied status as knowers.

Overcoming these persistent injustices calls for expanding our definitions of environmental justice to include a positive account of epistemic justice apt for EJ contexts. Following the work of EJ activists to expand and reform ambient air toxics monitoring in refinery-adjacent communities, I have proposed the concept of careful knowing as a necessary complement to participation and recognition in the definition of environmental justice. Careful knowing, or practices of empirical investigation and meaning-making responsive to the needs of marginalized knowers, offers an alternative to virtue-based accounts of epistemic justice. It calls on credentialed experts to situate epistemic activities in relation to communities’ needs for environmental protection and public obligations to meet those needs. It suggests the importance of developing epistemic resources and standards of evidence suitable to the specific circumstances of frontline communities, as well as the need to care for community members’ status and self-regard as knowers. In the same way that participation and recognition have guided EJ scholarship and intervention, the additional concept of careful knowing stands to aid in both the analysis of environmental contestation and the promotion of justice in real-world programs.

Acknowledgments

I am indebted to all of the air monitoring advocates and residents of fenceline communities who have spoken to me over the years. I thank them for their generosity, patience, and insight. My thanks go as well to two anonymous reviewers and Sherilyn MacGregor for their feedback, which has improved this paper considerably.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This work was partially supported by the National Science Foundation under Award #1352143. Preparation of the manuscript was facilitated by an ACLS-Burkhardt Fellowship and a year in residence at the Center for Advanced Study in the Behavioral Sciences (CASBS).