Quantitatively oriented social science occupies an important role in the Environmental Justice (EJ) movement in the United States. Initially, researchers corroborated the claims of EJ activists that communities of color and the poor were in general more disproportionately burdened by hazardous facilities than other groups. However, the contrary findings from other social scientists have sparked debates over the actual extent of environmental injustice. This paper argues that such forms of social science, although useful, can nonetheless present us with an incomplete understanding of social reality. If race, as some empirical studies have reported, is not the statistically best indicator overall, then does that mean that race is never a good indicator of injustice in particular places? The generalizations sought by nomothetically focused social science potentially ignore the spatio-temporal specificity of communities in proximity to noxious facilities. The paper outlines the basis for a critical-theoretic approach that will help to analyze the complex social conditions that generate the environmental inequities experienced among our American communities.
Key Words: environmental justice, critical theory, race, class, methodology.
The Environmental Justice (EJ) movement in America contends in word and deed that a healthy environment is concerned with much more than the protection of nature. Environmental protection also requires that people be safeguarded from the manifold sources of pollution, such as waste dumps and hazardous facilities. The EJ movement argues that the poor in general, and people of color in particular, can experience disproportionate -- and hence, unfair -- environmental risks in their everyday life.
For their part, social scientists as activists and scholars have occupied no small role in tackling EJ concerns (Tesh and Williams, 1996). It is important to study how social scientists have investigated EJ issues because their quantitatively based research findings have crucial consequences for the movement itself and for the public policies issued from Washington and from state legislatures (Camia, 1993; Clinton, 1994; Ferris and Hahn-Baker, 1995; Hacker, 1994). Initially, the analyses uncovered the widespread scope of racially disproportionate burdens. Such conclusions reinforced, and indeed added a measure of credibility, to the claims of activists and EJ organizations in myriad places around the USA. The conclusions of the early studies have not remained unchallenged, for within the social science community other researchers have not reached similar findings. There are various methodological issues as to why the debate rages. EJ activists, policy makers, and concer ned citizens thus face daunting tasks in trying to make sense of the controversies.
It is my contention that reaching the heart of environmental inequity -- people facing unfair hazards -- will require a theoretical approach to complement the prevailing quantitative social science one. I argue herein for the necessity of a critical-theoretic approach to environmental injustices. Certainly, there are many ways to conduct social science, such as via feminist empiricism, critical realism, ethnomethodology, and phenomenology. Critical theory, especially as inspired by the methodological works of the so-called Frankfurt School, is nevertheless an important tool.
Critical theory provides an explicit claim to attack injustice and strive for emancipation (Habermas 1971). While other perspectives, such as many feminist and post-structuralist analyses, explicitly seek to empower the oppressed, critical theory adds a direct focus on the fundamental mediating role of the material relations of society. From a critical theory perspective, demographic categories are influenced by how society is organized to produce the means for sustaining human life and interactions. Such materiality, as it can be termed, does not refer to class defined demographically (e.g., as socio-economic status, or SES). Rather, materiality refers to how the categories of race, class qua SES, and gender (among others) are interrelated in terms of the structures of opportunities and constraints on human actions and life chances. Moreover, a human combines various demographic attributes: the professional woman of color or the white working class man, for example. Our personal levels of education, expertise, and wealth as well as our race and gender all mediate how our individual aspirations can be achieved. Education, expertise, wealth, race, and gender are themselves part of the web of relationships within which we are born, mature, and are (continually) socialized. Some demographic attributes and personal factors are socially valorized and others are not. In terms of materiality, attributes which are socially valued can help us in everyday life, while those which are valued less offer us less of an advantage, or maybe even affect us in oppressive ways.
The materiality that is foregrounded by a critical-theoretic perspective, I contend, is necessary to incorporate into EJ studies. EJ groups face opportunities and constraints that arise from their particular structural positions relative to the dominant relations of production in a social formation. Specifically, many community activists have struggled against coalitions of businesses and local governmental leaders in their attempts to address toxic sites in their localities. The latter occupy a strong structural position because of the hegemonic manner in which capitalism favors certain social groupings over others in the drive for economic accumulation (whether it be termed increasing the tax base or marketing a city as a “name brand”). Material relations can impact on how activist groups struggle within the political-economic power structure of their locality. Overt, even covert, racism creates major hurdles for EJ groups. But consider these quest ions on the compounding dimensions of the materiality that informs EJ issues. Do the activist groups have the technical expertise in engineering, chemistry, or the law? Do they have sufficient money to mount a legal or public relations campaign? Do they face utilitarian arguments on the purported economic benefits of a waste dump or processing facility? Such questions have long confronted community-based organizations. Against wealth and entrenched elites the grass-roots answer often has been fashioned from the strength that arises from community solidarity, at times coupled with resources from national or regional organizations.
This paper performs a meta-theoretical analysis of the quantitatively oriented EJ studies. It continues in the tradition of other critiques made by, for example, Pulido and others (e.g., Pulido 1996a; Pulido, Sidawi, and Vos 1996). The paper’s contribution lies in how it both illustrates the limitations of many extant EJ analyses and enumerates the insights from alternate forms of EJ research. It thus specifies the meta-theoretical aspects that underpin both the conventionally quantitative and the critical-theoretic modes of social inquiry. Meta-theorizing does not supplant the need for theory. But rather than do what theory does -- seeking interpretations of a phenomenon’s meaning or causal explanations of the phenomena themselves -- meta-theorizing examines the ontological and epistemological assumptions on which the theories themselves are grounded. As will be explored below, this paper highlights an analytical framework sensitive to the spatio-temporal differences that undergird the materiality of EJ concerns.
This paper first examines various studies of environmental inequity, noting their several methodological differences. Despite the divergences, there is one area of similarity uniting the various researchers: their shared ontological assumptions as grounded in the quantitative social science tradition. In the next section, such assumptions are examined. A critique is put forth, establishing how such social scientific tenets actually limit research into such spatio-temporally complex phenomena as environmental injustice. In order to sketch the contours of a critical-theoretic approach, the subsequent section discusses three studies attuned to spatial and temporal variations, as well as their lessons for future research.
America has a long history of communities organizing to protest environmental problems (Bullard, 1994b). The event said to have spurred the EJ movement took place in 1982 in Warren County, North Carolina. Citizens from the county and from around the country protested at the proposed site of a landfill for toxic polychorinated biphenyls (PCBs). Through their protests they expressed outrage that such a waste dump was to be sited in a poor county which had one of the greatest proportions of African Americans in the state. The protestors failed to halt the operation of the landfill, yet the protest itself fueled awareness and unity among other groups across the United States (Bullard, 1994a).
Also protesting in Warren County was Congressman Walter Fauntroy (D-DC). The US General Accounting Office (GAO) received a request from him to investigate the association of waste sites with race and income. The GAO examined four commercial off-site hazardous waste landfills in the Southeast (EPA Region IV). The pioneering report revealed that the percentages of poor and African Americans surpassed those of the surrounding areas in three of the four sites (US GAO, 1983).
This section first outlines the trajectories of EJ research. It then delineates the differences in research frameworks.
The “Waves” of EJ Research
The motivation and activism of the EJ movement spurred social science research in the mid-1980s. Such research concluded that the environmental inequities faced by communities of color were national in scope. Perhaps the most widely known study was sponsored by the Commission for Racial Justice (CRJ) of the United Church of Christ. The study, entitled Toxic Wastes and Race in the United States, was released in 1987. Its conclusions echoed those of the earlier GAO study: namely, that the number of persons of color who lived near various types of hazardous facilities (like waste incinerators and landfills) was greater than in those areas without such facilities (CRJ, 1987). Significantly, this study utilized a national scale of analysis, and accordingly substantiated the argument that inequities spanned the entire country. As Robert Bullard indicates, the human costs of hazardous facilities were disproportionately borne by particular communities, while the benefits produced by such firms were dispersed across a much wider area (Bullard, 1994d).
Other quantitative studies conducted by social scientists have followed since the mid-1980s. Generally, they have focused on cities or states as the scale of analysis (e.g., Brueggemann, 1993; Burke, 1993; Mohai and Bryant, 1992; Pollock and Vittas, 1995; Spence, 1995). Included among those analyses was a national update of the CRJ report (Goldman and Fitton, 1994). All have concluded that people of color bear burdens that are disproportionate to their numbers in the population of the study area.
Despite various differences in analytical units and statistical methodologies, those social science reports could be characterized as the “first wave” of research. The focus of the different researchers lay on revealing the country-wide scope of the problem, as well as on conducting various case studies of specific places (Hofrichter, 1993; Westra and Wenz, 1995). They emphasized correlation in their statistical analyses of regional or national data. Almost no attention was assigned to causation, or to the dynamics of how communities of color and hazardous facilities came to occupy the same area. Disproportionate impact for the first wave of research centered on outcome-oriented inequity (M. Greenberg, 1993; also see Bullard, 1994c). The search for outcome-oriented inequity examines whether so-called minority groups face more environmental burdens than other groups. That such burdens might have occurred through discriminatory intent or biased political processes would be secondary to the research process itself (albeit important to political activists seeking redress).
Based on the first wave of research many activist groups concluded that racially disproportionate burdens implicated racism (e.g., Bullard, 1994a and 1994c). They argued for national policies to right the wrongs caused by what they considered to be the racist siting practices of factories and waste management businesses. “Stop environmental racism,” rather than the less evocative “Stop environmental injustice,” became a rallying cry for many. However, such cries did not remain unchallenged.
Debates have arisen within the social science community that question the actual scope of environmental inequity, as well as the causal mechanisms that allegedly had produced it. The conclusions of what we might call the “second wave” of researchers stand in contrast to those of the first wave. To be sure, these researchers agreed that there did exist cases of environmental burdens placed on communities of color and the poor. Nevertheless, analysis of the data gathered at many scales, from individual states and counties to the national level, did not suggest that wide-spread inequities existed.
Several quantitative studies at different analytical scales will illustrate such second wave research. At the state scale of analysis, South Carolina has been intensively studied with reference to Treatment/Storage/Disposal Facilities (TSDFs) and to Toxic Release Inventory (TRI) data (Crew-Meyer, 1994; Cutter et al., 1996; Holm, 1994). Analyses of the association of hazardous waste facilities/sites with race and income found no statistical correlations that indicated state-wide inequity for minorities. A national scale of analysis was used by the Social and Demographic Research Institute (SADRI) at the University of Massachusetts, Amherst. Their project has generated a well known set of findings that contradicts Toxic Wastes and Race. They used smaller analytical units, census tracts, rather than the ZIP Code areas of Toxic Wastes and Race. The SADRI project concluded that there was no correlation between hazardous facilities and race (Anderson et al., 1994; and Anderton, Anderson, Oakes and Fraser, 1994, and Anderton, Anderson, Rossi et al., 1994). Their conclusions have prompted several methodological rejoinders (Goldman, 1996; Mohai, 1995).
The second wave of research generally recommends that all claims of disproportionate burdens should be analyzed in terms of process-oriented inequity, an orientation centering on causation (M. Greenberg, 1993). A process-oriented approach holds that inequitable burdens experienced by one group might be caused by prejudice, procedural unfairness, and/or market forces (Been, 1994; Boerner and Lambert, 1994; Hamilton, 1995). Injustice might result from intentionally prejudicial behavior that is expressed, for example, in the siting process or in residential segregation. Injustice might occur within political processes insofar as meaningful and effective participation by all citizens is not fostered. Or injustice might arise in the operation of socio-economic dynamics: e.g., local market conditions could propel those with lower incomes to live in areas of low residential land values, such as might be found near hazardous facilities. Of course, some combination of the three i s also possible.
Vicki Been provides us with a process-oriented analysis that concentrated on the racial composition of impacted communities over time (1994; for an analysis of the United States as a whole, see Anderson et al., 1994). She replicated Bullard’s study of Houston (Bullard, 1983). She utilized available demographic data from Houston’s past so as to determine the racial composition of the places where at least some of waste sites later were to be located. Been maintained that there was no conclusive evidence of discriminatory intent or of an unfair siting process. Instead, the neighborhoods tended to become predominantly African American after the waste sites were built (Been, 1994: 1405). Bullard contested that finding; he argued that Been’s historical data was not complete enough to draw such conclusions (1994c).
Divergent Research Frameworks
That the conclusions of the two waves of research do not support each other is not surprising. Each uses different research frameworks that vary in important conceptual ways (for insights into the methodology, see Brown, 1995; Goldman, 1994; Mennis 2002; US GAO, 1995; Taquino et al., 2002). There are three major areas of divergence between the first and second waves.
First, no standard unit of analysis exists. The different analytical units of analysis vary from census tracts to ZIP Code areas, and from “neighborhoods” to radial zones drawn with Geographical Information Systems (Bullard, 1983, 1994c; Cutter, 1995; Cutter et al., 1996; Glickman and Hersh, 1995; Mohai, 1995; Zimmerman, 1994). As Taquino et al. (2002) have demonstrated, the different units of analysis can lead to different results even when the analyzing the same state.
Second, the dependent variable -- the hazardous facility or site -- varies. Differing dependent variables abound in the studies, including various types of solid waste landfills and incinerators (CRJ, 1987), TSDFs (e.g., Anderson et al., 1994; Pastor et al. 2001), company emissions found in TRI data (Perlin et al., 1995), and Superfund sites to be cleaned (Lavelle and Coyle, 1992; Zimmerman, 1993). Other types of facilities have been added to the list of dependent variables: hazard waste generators (Atlas 2002); at-risk zones around chemical processing facilities (Pine et al., 2002); air pollution (Couch et al. 2003; Pastor et al. 2002); and hog farms (Taquino et al., 2002; Wilson et al., 2002; Wing et al., 2000). A further factor compounding comparability of studies is the different results obtained when measuring chemical exposures based on different measures of toxicity (Cutter et al., 2002).
Third, the various studies utilize different comparison, or control, populations when determining whether the independent variable (e.g., race or income) faces disproportionate burdens. For example, the SADRI project did not examine the whole country per se, but only those census tracts within the US Census Bureau’s Standard Metropolitan Statistical Areas (e.g., Anderson et al., 1994). As such, the comparison tracts (without a hazardous TSDF) were drawn from a narrower portion of the country. The CRJ study, however, used as the comparison group all ZIP Code areas in the USA not containing a facility (Goldman, 1996; Mohai, 1995).
The debates over (the) correct methodology no doubt will remain vigorous. The stakes are high, not only in terms of the public policy implications, but also in terms of corporate “bottom lines.” To avoid legal liability and a negative public image, the waste management and chemical industries, for instance, have repeatedly stressed that they did not intentionally site facilities in areas of high minority populations (e.g., Bernstein, 1991; Kleiner, 1994; McDermott, 1993). Furthermore, lobbying efforts are put forward by the Business Network for Environmental Justice, formed in 1995 by the National Association of Manufacturers.
This paper focuses its critique on the conventional, quantitative social science of the EJ analyses, regardless of whether they are located in the first or second waves of research. Indeed, both waves share a common set of ontological and epistemological assumptions regardless of their particular findings and conclusions. In this section I will outline a critique of social science via the meta-theoretical procedure of analyzing its guiding assumptions and tenets.
Quantitatively focused social scientists seek uniform regularities in social phenomena, although they may not necessarily discover the universal -- or nomothetic -- laws of social behavior akin to the natural laws of physics. Nonetheless, probabilistic generalizations of human behavior can be pursued via interviews or participant observation, or can be unearthed via statistics using aggregate data. As the sociologist Babbie wrote: the quantitative social sciences could still yield their own type of always provisional, always improvable generalizations, ones which would be attuned to the nature of the complex human subject under study (Babbie, 1992; also see Popper 1965). The nuanced social sciences described by Babbie and implemented by the many EJ analyses discussed above can be labeled as post-positivist (Phillips 1991). Such social sciences nevertheless have their limitations: spatio-temporal variation is excluded by the undergirding assumptions and consequently by the generalizations derived from the studies themselves.
In their scientific goals, methods, and assumptions the quantitatively based social sciences conceive of the world in an abstract way. The social whole is nothing more than an aggregation of its atomized parts, which themselves are the only objects that can be empirically studied (Harvey 1996). In a temporal sense, the atomistic parts of the whole -- such as homo economicus -- consist of a set of timeless attributes whose behavior is determined, at least in statistical terms, by those essential characteristics. In a spatial sense, the social whole is fundamentally nothing more than an isotropic plane “peopled” by those discrete entities (Barnes and Sheppard, 1992; Cox, 1995; Harvey, 1973). Each point on the assumed plane (e.g., a hazardous site) and each entity (e.g., a person or household) are essentially indistinguishable from other points and entities. Each point and each entity lack both historical and geographical specificity.
Time is often a variable in quantitative social-scientific research, yet history is fundamentally absent. This lack of historical specificity contributes to the quantitatively based social science’s abstract notion of the social whole. Insofar as the goal is to seek universally valid generalizations (even probabilistic ones), the discrete entities do not have a past or present. Individual cases are aggregated together in order to extract the similarities that constitute behavioral laws. To the extent that the atomistic entities have any temporal dimension it is with reference to some statistical norm: the entities are concentrated around, or are deviations from, a mean or a regression line.
In their statistical analyses social scientists may situate points on some grid or plane, but geometric arrangements themselves are not nuanced conceptions of space and place. This absence of geographical specificity lends itself to the abstractness of the scientific understanding of the social whole. The points on any isotropic plane are surrogates (or abstract representations) for particular places. But, as such, the surrogate points lack any way to capture the lived experience of the places they represent; they lack any way to grasp some sense of the belonging that real humans feel toward places (Tuan, 1977).
In the notion of the social whole posited by such social science the constituent parts are not conceived in terms of their historical and geographical interrelationships. Those interrelationships themselves are said, according to the critical-theoretic tradition of the Frankfurt School (see below), to constitute the social context in which the distribution of points and entities (e.g., humans) is situated. The social context, or totality, confers meaning on the interrelationships, just as the relationships between the humans themselves constitute the totality in mutually reinforcing ways (Jay, 1984). But for the nomothetically oriented social sciences the social context is external to the phenomena (i.e., the variables) studied, and hence not explicitly integral to the study itself.
The research assumptions of conventional social science have consequences both in a theoretical sense and in a (practical) public policy sense. Theoretically, reification is always a possibility, not only for the variables used, but also for the issues being scrutinized. Here reification refers to the “freezing” of social processes, such that the relations between things are conceived as the things themselves. This is a perennial problem for quantitative social science given their grounding in phenomena, which are operationalized as variables. Variables are snapshots frozen in time, shorn of an intimate connection to the social relationships that influenced them, and that conceivably could help to change them (e.g., as a result of human praxis struggling for a more just world). To minimize reification we can analyze instead the social conditions that eventuate in environmental inequities, as discussed below, rather than simply measuring discrete variables. Furthermore, an understanding of social conditions (and their implicit totality) points to more than the observable. It also points to the emancipatory interest in creating a more humane society. But because such a humane society is only a potential to be achieved, via ongoing and enlightened praxis, and does not yet exist, it cannot be studied by the standard scientific method. Thus, quantitative social science’s variables are only representations of what is, and cannot encompass the potentiality of human praxis to modify social reality.
Further consequences of conventional, quantitative social science’s tenets and assumptions are found in the practical domain of policy making. According to such social science, environmental inequity is a binary phenomenon: it is either present in a study area, or it is not. There are no shades of gray. Related to this is the assumption of the equal distribution of population within an analytical unit, regardless of whether one uses a census tract unit, a ZIP Code area, or a GIS-mediated radial zone. It is assumed that the population therein is equally distributed (i.e., no clustering) and is of equal density (i.e., the number of people makes no difference). Also, it is assumed that no boundary effects exist (i.e., the effects of a waste site do not “spill over” into adjacent analytical units). Obviously, such assumptions enable social scientists to simplify complexity according to the theory’s most salient concepts, and thereby help to spotlight regularities. Such a lack of attention to variation, however, might lead to diminishing the concerns and life chances of citizens. Consider the following fabricated case employing a standard protocol used by quantitative EJ analyses. An analytical unit containing 49% of, say, minority racial group Omicron and 51% of majority racial group Theta will be counted as being a predominantly Theta unit. If that unit also were to contain a toxic site, then it would be analyzed as a case that counts toward the falsification of the hypothesis that toxic sites are more likely to be situated in predominantly Omicron units. Or phrased differently, the toxic site is found in a predominantly Theta unit and not a chiefly Omicron unit, which thereby fails to support the research hypothesis. However, falling through the methodological cracks is the possibility that the facility might be located near a concentration of Omicrons who might be clustered in one portion of the analytical unit.
The policy implications are strong. Conventional social science is immensely useful for its powers of generalization, but its methodological limitations do not merit its uncritical use in policy making. The dichotomous nature of conclusions drawn from empirical research would not necessarily address the “coincidence” of a waste site with a community. If no widespread environmental inequities are found, then policy makers may believe that little or no concern is warranted. I do not contest the view that regularities can be discerned in human behavior and relationships. Rather, it is the privileging of the nomothetic as the central means by which to grasp reality that is contested. Accordingly, we must probe into the conditions that make social reality possible. This will require that we grasp the nomothetic and the idiographic, not as separate “things,” but in a way that unites them in their variations as well as in their regularities. Thus, to get at the heart of environmental injustice -- citizens expressing their deeply felt concerns about hazards in their (particular) places -- would mandate that a different method, like a critical-theoretic framework, should also be used.
Let me briefly outline the ontological and epistemological postulates of the critical-theoretic approach that philosophically underpin this section. They are drawn from the works of the early Frankfurt School tradition of social inquiry (see Jay, 1973; Morrow, 1994; Wiggershaus, 1995). A critical theory can be characterized by several tenets (adapted from Adorno, 1976; Horkheimer, 1976):
(a) The social totality, with its dialectically related constituent and constituting elements, is grounded in the material conditions of how people live.
(b) The elements of a totality -- the human beings -- exist in material interrelationships with each other, as mediated by the societal practices on which the totality is grounded.
(c) Race, gender, class among other facets of social life are material relations between humans, and thus are integral to the constitution of a complex (i.e., non-reductionistic) social whole. As material relationships gender, class, race, etc. affect the life chances of humans.
(d) Such material interrelationships can be fundamentally contradictory (e.g., poverty amidst wealth, environmental ill-health despite government regulations, or the capital/labor relationship criticized by Karl Marx).
(e) The totality may constrain what can be done, but it does not deterministically foreclose options, because humans have the agency (or consciously directed capacity) to change their interrelationships, and thereby change the nature of the totality.
(f) Knower and knowledge are not separate, because the knowers produce the world, and hence knowledge of it, in and through the societal practices of everyday life.
(g) Social inquiry cannot be value-free because researchers must make choices about their objects of study (and critical theorists must tackle directly the contradictions of quotidian life).
An approach based on those tenets permits the interrogation of spatio-temporal specificity because both the parts and the whole are integrated in their complexity, each affecting the other in a dialectical manner. Or phrased differently, phenomena must be studied in terms of the broader, generative social processes which condition them, and by which the phenomena (like the humans themselves via their practices and behavior) might effect changes in society itself.
In this section I discuss three case studies of environmental burdens in particular places. I then draw several lessons that can inform a critical-theoretic framework for EJ analysis.
Three Studies
Complementing the aggregate statistical analyses of environmental justice, other researchers have focused on specific areas and cities over time (e.g., Colten and Skinner, 1996; D. Greenberg, 2000; Gugliotta, 2000; Hersh, 1995; Pastor et al. 2001; Pulido, 1996b; Pulido 2000; Towers, 2000; and Westra, 1999). Due to length constraints, here I will outline three studies of environmental inequities that embody a spatio-temporal understanding of how such inequities arise in a specific place: Krieg (1995), Hurley (1995), and Pulido, Sidawi, and Vos (1996). They are path-breaking studies that tackle social complexity, but in ways which retain the capacity to differentiate the particular in relation to the whole. It is perhaps not surprising that all focus on individual cities; yet their case-study perspective also emphasizes the need to accord theoretical attention to the larger spatio-temporal context in which environmental inequities are situated. By implication, quantitative studies of aggregated data analyzed via statistical regression and scatter plots of residuals are inadequate ways to recapture the “contextual information” lost in a nomothetically oriented analysis.
Krieg (1995) investigated the region known Greater Boston, an area within a 15-mile radius of the city itself. He used statistical tools, but wielded them in a historically specific manner. The different industrializing and racist dynamics of the Greater Boston area pointed to different spatio-temporal patterns of exposure to Superfund hazardous waste sites. Accordingly, Krieg framed the hypotheses to be tested in terms of those historical and geographical variations. In the older portions of the study area, the legacy of racism in residential segregation and in land-use planning was evident in the statistically significant correlation of persons of color with waste sites. The affluent (mostly whites) moved over time from the center of Boston into the suburbs and into the newly industrializing sections of what was to become part of the “Massachusetts Miracle.” Due to housing discrimination, persons of color faced serious impediments when attempting to relocate. However, in an ironic twist, the more affluent whites living in the industrializing sections did not escape environmental hazards, because along with economic miracles came waste sites. Krieg found that in the newer areas of Greater Boston, waste sites were significantly correlated with class (operationalized by Krieg as median income and education).
Hurley (1995) also uncovered varying spatio-temporal dynamics when he studied Gary, Indiana, an area known for its steel and heavy industry. He chiefly used a historical methodology, based on various sources and documents from the city’s past (e.g., newspapers, interviews, public records, etc.). Hurley found that over time neither race nor class was the best single indicator of exposure to air, land, and water pollution sources. In the 1940s pollution affected everyone of all races/ethnicities and income levels. Although neighborhoods were segregated along racial lines, all demographic groups were nevertheless close to the industrial sources of pollution. Thus, residential exposure to dirty air was not associated with class, race, or ethnicity. In the post-war boom, affluent whites moved to areas away from the dirtiest portions of Gary so as to enjoy the amenities that money could buy. Residential segregation, however, prevented even the wealthier African Americans and Mexicans from likewise moving. Hurley noted that poorer whites were also hindered, but for financial reasons, from leaving the more polluted stretches of the city. Polluting industries like USX Corporation eventually did install various pollution-abating technologies. Yet they were able to play different residential sections against one another when the firms sought to reduce their costs of compliance with environmental regulations, complaining that such regulations hampered their global competitiveness. Willing politicians in city hall also helped. Lax enforcement of pollution regulations, as well as the weak or non-existent protest from citizens of color and the poorer whites, combined to keep them exposed to industrial pollution.
Pulido, Sidawi, and Vos (1996) examined two predominantly Latino/Mexican communities within Los Angeles. Their theoretical starting point was environmental racism -- how people of color were either targeted explicitly for, or else received the unfair consequences of, hazardous sites. They argued for a qualitative approach that would explore how, over the course of time, minority populations become exposed to hazardous facilities and waste sites. They studied two areas, Torrance and East L.A./Vernon, using various historical documents and public policy statements. Torrance was a planned community designed to house white workers employed in high-paying jobs in the nearby chemical and metals-related industrial zone. A smaller portion of the community was set aside for Mexicans, who would, following racist ideology, do the low-paying, dirty jobs in the factories. This section was contained within the industrial zones in contradistinction to the white section of Torrance. The racialized division of labor, coupled with residential segregation, kept the Mexicans in “their place.” As time passed, some of the more egregious aspects of segregation disappeared. Nonetheless, those of Mexican origin are still concentrated in the section closest to the industrial facilities. The racialized division of labor still holds nowadays as non-unionized Mexicans compose most of the lowest paying jobs.
According to Pulido, Sidawi, and Vos, mentioning “East L.A./Vernon” may evoke images of heavy pollution and Latino barrios. This imagery, they argued, provided the backdrop for the local government’s attempt to locate a hazardous waste incinerator within the area in the late 1980s. Although the attempt has been viewed as a quintessential example of environmental racism, the authors contended that such racism cannot be reduced to a single locational decision. Rather, it must be understood in its historical context. East L.A./Vernon was historically a mixed Anglo and Mexican area, albeit in segregated communities within the area. Mexicans worked in the lower paying jobs with limited upwards mobility within light and heavy industries. As time passed, the economic restructuring of California became evident in the decline of heavy industry within East L.A./Vernon. Anglos and ethnic Europeans left the area, pollution mounted, and more and more people of color moved into the area. The industries that remained after restructuring, or else arrived later, tended to include low-wage clothing, furniture, and food businesses. By this time, the area was widely perceived in negative terms. Such imagery reinforced the view that it was a waste land, suitable only for hazardous trash. Ignored, of course, has been the environmental health of the Mexican-Americans who live in close proximity to the dangerous materials.
Lessons for a Critical-Theoretic Framework
Here I wish to delineate both the lessons to be gained from the three studies just discussed, and also outline a critical-theoretic framework for future inquiry. Three lessons come to mind from those studies. First, an analysis of the specificity of burdens in time and place does not necessarily imply that a single best predictor is applicable over all. The dynamics of racism and poverty, and the conjunction of the two, can vary over time and space. Also, as Pulido, Sidawi, and Vos show, assumptions like an equal spatial distribution of a population within an analytical unit are suspect because concentrations of people exist within an area, thereby requiring a high degree of specificity to probe into the inequities. Certainly, sets of demographic and toxic data with a fine level of resolution might be able to tap into such concentrations of people in a study area. However, the other objections made here will still exist. While a single best indicator may satisfy the tenets of a quantitative social science, it may not actually capture the fluid complexity of actual communities. Thus, statistical approaches that pursue linear causality -- which came first, the hazard or race/class? -- do not address this insight. Pursuing whether the independent variable (race or class) “caused” the dependent variable (the toxic facility) generates dichotomous research conclusions that do not capture spatial and temporal variations (Pulido, 1996a). It also does not address what injustice means to the people experiencing the problems. Field research and qualitative approaches must supplement any quantitative analysis of aggregated data.
Second, and related to the first lesson, discrete variables (like race distinguished from class) may not fully delineate the dynamics of environmental risks. That both poor white and affluent communities of color have experienced proximity to hazards indicates that theoretical subtlety is required. It becomes difficult to disaggregate them. As Pulido, Sidawi, and Vos indicate, we must not operationalize variables as static categories. Rather, we must apprehend the relationships between the demographic categories of race and class (Pulido, 1996a). Neither race nor class (nor gender) are discrete entities in reality. Such assumptions of a statistically oriented social science simplify to the point of distortion. Issues of class, race, and gender intersect in particular places at particular times (Massey, 1994; Wilson, 1992). Designation by race, class, and gender is as much about the processes whereby they come to be so designated as it is about sociological or biological categories. For example, how race, class, and gender are incorporated into the division of labor is not a given, but rather is a result of various social processes that racialize, even “engender,” who is deemed suitable to do what kind of work. Which group is deemed “acceptable” to live in a certain place is likewise the consequence of social dynamics, sometimes even struggle, over issues like equality and the so-called character of a neighborhood (Bullard, 1995).
The third lesson stresses that we must probe the social conditions underpinning phenomena like environmental inequity. As Pulido, Sidawi, and Vos query (1996), how do some groups become vulnerable to hazards? Such a focus on social conditions transcends framing environmental injustice as a distributional issue. Indeed, much of the EJ debate is cast in terms of the fairness of the process by which the burdens came to be sited in a particular spot. Such concerns over the procedural fairness of the siting process are certainly important, especially given the egregious examples which have come to light (e.g., Cole, 1993-94). Nonetheless, a focus on distribution is fundamentally a focus on the final result of the decision making process, namely, the spatial arrangement itself. Also crucial to understand is how such a distribution was “produced” (Field, 1997; Lake, 1996). Specifically, a critical-theoretical focus on production asks about the structural context in which individuals and businesses, as well as politicians and bureaucrats, make their decisions. In that vein, we must address the issues of power, specifically the loci of power in capitalist economies conjoined with procedurally democratic political apparatuses. Such requires that we pursue the material conditions of production that establish the context in which social activities take place and in which constraints are placed on such actions.
Toward such ends, let me briefly sketch a critical-theoretic framework that can guide our analyses of environmental inequities. It embraces two fundamental themes. First, we should study the structural power that undergirds the dominance of a group, class, or coalition in production decisions in specific places. Second, we must discover how such power engenders social problems like environmental injustice.
Where does structural power reside? According to the Marxian underpinnings of critical theory, structural power derives from a group’s position relative to others within the social relations of production. In those production relations, we spy the fundamental (or structural) power wielded by capitalists who own the means of production and the weakness possessed by the workers who do not. Such a formulation, however, is too vague for use in studies of actual situations. Consequently, we must pose research questions that allow us to delineate the specificity of power in particular contexts. Such questions center on the political economy of particular places: for example, which actors in the city or region provide the linchpin for economic development? Occupying such a role in a city’s or region’s development would confer structural power on some groups to the detriment of others who might protest any problems wrought by those dominant groups (Bullard, 1992; Harvey, 1989). Of course, capital is everywhere not the same, for the sunk costs of heavy industrial investments like massive factories, oil refineries, and mines hinder firms from leaving quickly or too easily (Clark and Wrigley, 1995). We must, therefore, grasp any limitations on the structurally derived power of the dominant group(s). Of course, even those who are not part of the motor of economic development, like citizen groups agitating against toxic facilities, can possess some measure of structural influence: namely, the capacity of social protest and the right to vote so integral to a liberal democratic political system. All such facets must be investigated.
Moreover, we must ask how the asymmetrical relationships of structural power are translated into social problems and the sometimes concomitant political debates. This becomes a necessary question to address whenever we are seeking the material conditions that underpin environmental inequity in specific places. Given its high level of abstraction, the concept of the social relations of production should be made more sensitive to various issues of historical and geographical specificity. Because there is no one-to-one correspondence of general economic categories with political issues (Gramsci, 1971; see R. Williams, 1973; Hall, 1986), this means that we must scrutinize the political realm. Lake and Disch (1992) set the theoretical stage by examining the ways in which the state (in its national and local forms) mediates between the businesses who produce the wastes and the citizens whose communities receive those wastes. All too often, production decisions are not debated in public forums, while complaints from communities have been turned into complaints over distribution (i.e., why are we getting the wastes?). Thus, the capitalist state in its local and national variants occupies a vital position in the creation of environmental hazards. Such state agencies and policies must be studied if we desire to understand the causal factors generating environmental in/equity.
The structurally powerful private groups will often cooperate with politicians and governmental officials on the terrain of the state apparatuses. Such an “urban growth coalition” (Harvey, 1989) points conceptually to a hegemonic bloc, a term akin to the Gramscian “historical bloc” (Gramsci, 1971). Grasping the dynamics of this bloc in the overall policy making process -- e.g., the nature of the issues at hand and how groups become involved in it -- will allow us to focus on the class and racial dynamics that can eventuate in disproportionate environmental impacts. Structural weakness tends to exclude many people, communities, and groups from the bloc. Groups are structurally weak insofar as they do not possess the skills, capital, and so forth to assist in promoting the overall economic growth of the community (or in general terms, the accumulation process). A previously weak group or community might gain some degree of structural power to the degree that it possesses something, like land rights, that become valuable to the economic growth of the area as a whole. (Or such valuable items could simply be appropriated forcefully). Various prejudices also have historically exacerbated the exclusion of some groups from the hegemonic bloc, heaping another burden atop that of structural weakness. Moreover, the issues being publicly debated can revolve around a narrow set of concerns for economic growth and a positive business climate. Other concerns like economic justice or environmental safeguards are often not considered to be debatable by the hegemonic bloc, or else are couched in apocalyptic terms (e.g., jobs versus the environment). Inclusion in the bloc or even inclusion of certain concerns on the public agenda have occurred, of course. Even the decline of some forms of racial prejudice (especially de jure racism) in America can enable previously omitted groups to be incorporated in a bloc. But inclusion is often predicated on whether the group adopts a non-critical view of economic growth and possesses something of structural value. Lacking little if any structural power to force items onto the public agenda, it is no wonder that the poor and the communities of color in the US have historically faced hardships, including environmental hazards, not borne by other more (structurally) fortunate groups in American society.
By grasping how different processes operate at different scales to constitute a given place (Agnew, 1987; Massey, 1994; R.W. Williams 1999b) our attention is drawn to the specificity of communities facing environmental burdens. A place is not an isolated, independent spot on an assumed isotropic plane. A community is a place full of meaning -- cultural, religious, racial, class, gender, sexual -- for the people living there (Bailey et al., 1995; Novotny, 1994; Tesh and Williams, 1996; R.W. Williams 1999a; Wright, 1995). As such, a critical-theoretic framework incorporates an implicit qualitative dimension. By studying structural power and the struggles resulting from the lack of such power, the framework concentrates on how people understand themselves and their communities as expressed in their battles against the “powers that be.”
The critical-theoretic approach as delineated here, however, might raise an immediate objection. Some could assert that it begs a question: it assumes the presence of inequities when the existence of those burdens and their causes are the questions to be posed. As a rejoinder, we can reply that such a response points to a nomothetic understanding of social issues. The idiographic facet -- that there is a community with a waste dump nearby -- is empirically “verifiable,” as numerous case studies reveal. The question for a critical research agenda then becomes what social conditions led to the conjunction of a particular community with the hazard. Popper’s falsificationism, while a cornerstone of post-positivist social science and its often statistical analyses, does not adequately address how a singular case in its specificity arose amidst the more general social processes of racism and capitalism.
Importantly, a critical-theoretic approach prevents the victim from being blamed for the hazardous site. Research into underlying social conditions fundamentally differs from research on individual decision-making that might lead to the juxtaposition of certain communities with toxins. The latter research focus centers on why individuals (or households) would live near hazards, often putting the blame solely on individuals and their discrete decisions (Tesh, 1988). Market-based explanations, as advocated by some in the second wave of EJ studies, argue that individual or household decisions, not discriminatory firms and secretive bureaucrats, have led to the coincidence of wastes and certain demographic groupings (Been, 1994; Boerner and Lambert, 1994; Couch et al., 2003; Hayward, 2003). As the market-based explanations frame it, land is cheaper around industrial zones, so lower-income groups, including among them persons of color, will find it more financially rational to live in such areas. However, such an explanation effectively decontextualizes, even depoliticizes, individual decisions (R.W. Williams, 1999b). The larger societal context (capitalism as well as historically racist practices) is not theorized as constraining individual choice. Ignored in market-based explanations are the ways in which capitalist production decisions lead to pollution and to a downward spiral of wage competition and housing prices in waste areas. Nuanced research into material societal processes is needed, not apolitical explanations.
Much of the extant EJ studies, even those which uncovered broadly based inequities, operate within the strengths and weaknesses of a post-positivist, quantitatively oriented social science. They seek to explain societal phenomena via discerning observed regularities. As a consequence, the EJ analyses of both the first and second waves implicitly conceive inequity in dichotomous terms. It either exists, or not, across the study area. Integral to such a social scientific enterprise is the quest for the best predictor of inequity. Is either race or income statistically associated with nearby hazardous facilities? Does one operationally distinct variable “causally” explain more of the data? Social scientists rigorously test the hypotheses to discern whether they are supported by the data. The scientific method may have been rigorously followed, yet nagging questions remain. For example, if race is not the best indicator overall as some empirical studies have found, then does that mean that race is never a good indicator in particular places? Nomothetically centered social science is thus left with questions that it is methodologically incapable of addressing.
Some political arguments of the EJ movement bear a similar structure. One variable -- usually race -- is cast as the single best indicator that explains why hazards are located where they are. This argument casts injustice as environmental racism, and as a political strategy leads to a binary outcome (Goldman, 1996). Environmental racism charges that businesses, governmental officials, or both, deliberately sited (or let be sited) toxic facilities in communities of color (Bullard, 1994a). The double-edged nature of such an argument is evident in how the industries are responding (Seigler, 1993-94). If burdens are legally actionable only when caused by intentionally discriminatory acts (which recent Federal court cases reinforce (Lawyers’ Committee n.d.)), and if industries can convince others -- say, politicians and the general public -- that they do not site facilities in a consciously discriminatory manner, then the battle to rectify the inequities might be lost.
Pursuing the single-best indicator of environmental injustice across an entire country could backfire against specific communities, or maybe even the EJ movement as a whole. While applicable in some cases as a political strategy, the focus on the best predictor seems less sustainable over time. Certainly, racism has existed, and still exists; indeed, it may explain various siting decisions. Yet in its own way the environmental racism argument fails to account for the complex dynamics of inequities in particular places. Some have argued that it also may hinder efforts to unite progressive coalitions which possibly could tackle the many shared concerns together (Epstein, 1997).
Yet hope for justice springs eternal. The paper has argued the merits of a critical-theoretic approach; it is an approach which concentrates on the spatio-temporal specificity of everyday material existence. To scrutinize the social conditions giving rise to environmental inequity helps us to take seriously the concerns of our neighbors, friends, and fellow citizens. A critical-theoretic approach would provide at least part of the intellectual support needed to attack inequities wherever they may occur.
Adorno, Theodor. (1976) [1957]. “Sociology and Empirical Research.” Pp. 237-257 in Critical Sociology. Connerton, Paul (Ed.). Harmondsworth, UK: Penguin Books.
Agnew, John. (1987). Place and Politics: The Geographical Mediation of State and Society. Boston: Allen & Unwin.
Anderson, Andy, Douglas Anderton and John Oakes. (1994). “Environmental Equity: Evaluating TSDF Siting Over the Past Two Decades.” Waste Age (July): 83-100.
Anderton, Douglas, Andy Anderson, John Oakes and Michael Fraser. (1994). “Environmental Equity: The Demographics of Dumping.” Demography 31: 229-248.
Anderton, Douglas, Andy Anderson, Peter Rossi, John Oakes, Michael Fraser, Elinor Weber and Edward Calabrese. (1994). “Hazardous Waste Facilities: ‘Environmental Equity’ Issues in Metropolitan Areas.” Evaluation Review 18: 123-140.
Atlas, Mark. (2002). “Few and Far Between? An Environmental Equity Analysis of the Geographic Distribution of Hazardous Waste Generation.” Social Science Quarterly 83: 365-378.
Babbie, Earl. (1992). The Practice of Social Research, 6th Ed. Belmont, CA: Wadsworth.
Bailey, Conner, Kelly Alley, Charles Faupel and Cathy Solheim. (1995). “Environmental Justice and the Professional.” Pp. 35-44 in Environmental Justice: Issues, Policies and Solutions. Bryant, Bunyan (Ed.). Washington, DC: Island Press.
Barnes, Trevor and Eric Sheppard. (1992). “Is There a Place for the Rational Actor? A Geographical Critique of the Rational Choice Paradigm.” Economic Geography 68: 1-21.
Been, Vicki. (1994). “Locally Undesirable Land Uses in Minority Neighborhoods: Disproportionate Siting or Market Dynamics?” Yale Law Journal, 103: 1383-1422.
Bernstein, Joan. (1991). “The Siting of Commercial Waste Facilities: An Evolution of Community Land Use Decisions.” Kansas Journal of Law & Public Policy 1: 83-88.
Boerner, Christopher and Thomas Lambert. (1994). “Environmental Justice?” Occasional Paper 136. Center for the Study of American Business, Washington University, St. Louis.
Brown, Phil. (1995). “Race, Class, and Environmental Health: A Review and Systematization of the Literature.” Environmental Health 69: 15-30.
Brueggemann, Martin R. (1993). Environmental Racism in Our Own Backyard: Solid Waste Disposal in Holly Springs, N.C. M.A. Thesis, University of North Carolina at Chapel Hill, School of Journalism and Mass Communication.
Bullard, Robert. (1983). “Solid Waste Sites and the Black Houston Community.” Sociological Inquiry 53: 273-288.
_______. (1992). “Environmental Blackmail in Minority Communities.” Pp. 82-95 in Race and the Incidence of Environmental Hazards. Bryant, Bunyan and Paul Mohai (Eds.). Boulder, CO: Westview Press.
_______. (1994a). Dumping in Dixie: Race, Class, and Environmental Quality, 2nd Ed. Boulder, CO: Westview Press.
_______. (1994b). “Grassroots Flowering: The Environmental Justice Movement Comes of Age.” Amicus Journal (Spring): 32-37.
_______. (1994c). “The Legacy of American Apartheid and Environmental Racism.” St. John’s Journal of Legal Commentary 9: 445-474.
_______. (1994d). “Overcoming Racism in Environmental Decisionmaking.” Environment, 36: 10-20, 39-44.
_______. (1995). “Residential Segregation and Urban Quality of Life.” Pp. 76-85 in Environmental Justice: Issues, Policies and Solutions. Bryant, Bunyan (Ed.). Washington, DC: Island Press.
Burke, Lauretta. (1993). Environmental Equity in Los Angeles. Technical Report 93-6. Santa Barbara, CA: National Center for Geographic Information and Analysis, University of California.
Camia, Catalina. (1993). “Poor, Minorities Want Voice in Environmental Choices.” Congressional Quarterly Weekly Report 51: 2257-2260.
Clark, Gordon and Neil Wrigley. (1995). “Sunk Costs: A Framework for Economic Geography.” Transactions of the Institute of British Geographers 20: 204-223.
Clinton, William. (1994). “Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations.” US Presidential Executive Order No. 12898 (February 11, 1994). Cited in Environment 36:4 (May 1994): 16-19.
Cole, Luke. (1993-94). “The Struggle of Kettleman City: Lessons for the Movement.” Maryland Journal of Contemporary Legal Issues 5: 67-80.
Colten, Craig and Peter Skinner. (1996). The Road to Love Canal: Managing Industrial Waste before EPA. Austin, TX: University of Texas Press.
Commission For Racial Justice, United Church of Christ. (1987). Toxic Wastes and Race in the United States: A National Report on the Racial and Socio-Economic Characteristics of Communities with Hazardous Waste Sites. New York: United Church of Christ.
Couch, Jim, Peter Williams, Jon Halvorson, and Keith Malone. (2003). “Of Racism and Rubbish: The Geography of Race and Pollution in Mississippi.” The Independent Review VIII: 235–247.
Cox, Kevin. (1995). “Concepts of Space, Understanding in Human Geography, and Spatial Analysis.” Urban Geography 16: 304-326.
Crew-Meyer, Kelly. (1994). “Hazardous Waste and Race in South Carolina.” South Carolina Policy Forum 5: 25-28.
Cutter, Susan. (1995). “Race, Class and Environmental Justice.” Progress in Human Geography 19: 111-122.
Cutter, Susan, Danika Holm, and Lloyd Clark. (1996). “The Role of Geographic Scale in Monitoring Environmental Justice.” Risk Analysis 16: 517-526.
Cutter, Susan, Michael Scott, and Arleen Hill. (2002). “Spatial Variability in Toxicity Indicators Used to Rank Chemical Risks.” American Journal of Public Health 92: 420-422.
Epstein, Barbara. (1997). “The Environmental Justice/Toxics Movement: Politics of Race and Gender.” Capitalism, Nature, Socialism 8: 63-87.
Ferris, Deeohn and David Hahn-Baker. (1995). “Environmentalists and Environmental Justice Policy.” Pp. 66-75 in Environmental Justice: Issues, Policies and Solutions. Bryant, Bunyan (Ed.). Washington, DC: Island Press.
Field, Rodger. (1997). “Risk and Justice: Capitalist Production and the Environment.” Capitalism, Nature, Socialism 8: 69-94.
Glickman, Theodore and Robert Hersh. (1995). “Evaluating Environmental Equity: The Impacts of Industrial Hazards on Selected Social Groups in Allegheny County, Pennsylvania.” Discussion Paper 95-13. Resources for the Future, Washington, DC.
Goldman, Benjamin. (1994). Not Just Prosperity: Achieving Sustainability with Environmental Justice. Washington, DC: National Wildlife Federation.
_______. (1996). “What Is the Future of Environmental Justice?” Antipode 28: 122-141.
Goldman, Benjamin and Laura Fitton. (1994). Toxic Wastes and Race Revisited: An Update of the 1987 Report on the Racial and Socioeconomic Characteristics of Communities with Hazardous Waste Sites. Washington, DC: Center for Policy Alternatives.
Gramsci, Antonio. (1971). Selections from the Prison Notebooks. Hoare, Quintin and Geoffrey Nowell Smith (Trans. and Eds.). NY: International Publishers.
Greenberg, Dolores. (2000). “Reconstructing Race and Protest: Environmental Justice in New York City.” Environmental History 5: 223-250.
Greenberg, Michael. (1993). “Proving Environmental Inequity in Siting Locally Unwanted Land Uses.” Risk: Issues in Health & Safety (Summer): 235-252.
Gugliotta, Angela. (2000). “Class, Gender, and Coal Smoke: Gender Ideology and Environmental Injustice in Pittsburgh, 1868-1914.” Environmental History 5: 165-193.
Habermas, Jürgen. (1971). Knowledge and Human Interests. Shapiro, Jeremy (Trans.). Boston: Beacon Press.
Hacker, Barton. (1994). Environmental Justice: Legislation in the States. September. Washington, DC: Center for Policy Alternatives.
Hall, Stuart. (1986). “Gramsci’s Relevance for the Study of Race and Ethnicity.” Journal of Communication Inquiry 10: 5-27.
Hamilton, James T. (1995). “Testing for Environmental Racism: Prejudice, Profits, Political Power?” Journal of Policy Analysis and Management 14: 107-32.
Harvey, David. (1973). Social Justice and the City. Baltimore, MD: Johns Hopkins University Press.
_______. (1989). The Urban Experience. Baltimore, MD: Johns Hopkins University Press.
_______. (1996). Justice, Nature, and the Geography of Difference. Oxford: Blackwell.
Hayward, Steven. (2003). “Environmental Justice: Where Selma and the Cuyahoga River Fire Meet.” Environmental Policy Outlook, American Enterprise Institute Online. November-December. http://www.aei.org/docLib/20031210_%2316111graphics.pdf
Hersh, Robert. (1995). “Race and Industrial Hazards: An Historical Geography of the Pittsburgh Region, 1900-1990.” Discussion Paper 95-18. Resources for the Future, Washington, D.C.
Hofrichter, Richard (Ed.). (1993). Toxic Struggles: The Theory and Practice of Environmental Justice. Philadelphia, PA: New Society Publishers.
Holm, Danika. (1994). Environmental Inequities in South Carolina: The Distribution of Hazardous Waste Facilities. M.S. Thesis, University of South Carolina, Department of Geography.
Horkheimer, Max. (1976) [1937]. “Traditional and Critical Theory.” Pp. 206-224 in Critical Sociology. Connerton, Paul (Ed.). Harmondsworth, UK: Penguin Books.
Hurley, Andrew. (1995). Environmental Inequalities: Class, Race, and Industrial Pollution in Gary, Indiana, 1945-1980. Chapel Hill, NC: University of North Carolina Press.
Jay, Martin. (1973). The Dialectical Imagination: A History of the Frankfurt School and the Institute of Social Research, 1923–1950. Boston: Little, Brown & Co.
_______. (1984). Marxism and Totality. Berkeley, CA: University of California Press.
Kleiner, Richard. (1994). “Quality of Life: Justice Equates Fairness.” St. John’s Journal of Legal Commentary 9: 519-523.
Krieg, Eric. (1995). “A Socio-Historical Interpretation of Toxic Waste Sites: The Case of Greater Boston.” American Journal of Economics and Sociology 54: 1-14.
Lake, Robert W. (1996). “Volunteers, NIMBYs, and Environmental Justice: Dilemmas of Democratic Practice.” Antipode 28: 160-174.
Lake, Robert W. and Lisa Disch. (1992). “Structural Constraints and Pluralist Contradictions in Hazardous Waste Regulation.” Environment and Planning A 24: 663-681.
Lavelle, Marianne and Marcia Coyle. (1992). “Unequal Protection: The Racial Divide in Environmental Law.” National Law Journal 15: S1-S12.
Lawyers’ Committee for Civil Rights Under Law. n.d. “Lawyers’ Committee Files Amicus Briefs with Supreme Court and Court of Appeals in Support of Private Right of Action to Enforce Disparate Impact Regulations.” http://www.lawyerscomm.org/projects/environmentalaccomplish2.html
Massey, Doreen. (1994). Space, Place, and Gender. Minneapolis: University of Minnesota Press.
McDermott, Charles. (1993). “Environmental Equity: A Waste Manager’s Perspective.” Land Use Forum (Winter): 12-17.
Mennis, Jeremy. (2002). “Using Geographic Information Systems to Create and Analyze Statistical Surfaces of Population and Risk for Environmental Justice Analysis.” Social Science Quarterly 83: 281-297.
Mohai, Paul. (1995). “The Demographics of Dumping Revisited: Examining the Impact of Alternate Methodologies in Environmental Justice Research.” Virginia Environmental Law Journal 14: 615-653.
Mohai, Paul and Bunyan Bryant. (1992). “Environmental Racism: Reviewing the Evidence.” Pp. 163-76 in Race and the Incidence of Environmental Hazards. Bryant, Bunyan and Paul Mohai (Eds.). Boulder, CO: Westview Press.
Morrow, Raymond. (1994). Critical Theory and Methodology. Thousand Oaks: SAGE.
Novotny, Patrick. (1994). “Popular Epidemiology and the Struggle for Community Health: Alternative Perspectives from the Environmental Justice Movement.” Capitalism, Nature, Socialism 5: 29-42.
Pastor, Manuel, Jim Sadd, and John Hipp. (2001). “Which Came First? Toxic Facilities, Minority Move-in, and Environmental Justice.” Journal of Urban Affairs 23: 1-21.
Pastor, Manuel, James Sadd, and Rachel Morello-Frosch. (2002). “Who’s Minding the Kids? Pollution, Public Schools, and Environmental Justice in Los Angeles.” Social Science Quarterly 83: 263-280.
Perlin, Susan, R. Woodrow Setzer, John Creason, and Ken Sexton. (1995). “Distribution of Industrial Air Emissions by Income and Race in The United States: An Approach Using the Toxic Release Inventory.” Environmental Science & Technology 29: 69-80.
Phillips, Denis. (1991). “Postpositivistic Science: Myths and Realities.” Pp. 31-45 in The Paradigm Dialog. Guba, Egon (Ed.). Newbury Park: SAGE.
Pine, John, Brian Marx, and Aruna Lakshmanan. (2002). “An Examination of Accidental-Release Scenarios from Chemical-processing Sites: The Relation of Race to Distance.” Social Science Quarterly 83: 317-331.
Pollock, Philip and M. Elliot Vittas. (1995). “Who Bears the Burdens of Environmental Pollution? Race, Ethnicity and Environmental Equity in Florida.” Social Science Quarterly 76: 294-310.
Popper, Karl. (1965). Conjectures and Refutations: The Growth of Scientific Knowledge. NY: Harper & Row.
Pulido, Laura. (1996a). “A Critical Review of the Methodology of Environmental Racism Research.” Antipode 28: 142-159.
_______. (1996b). Environmentalism and Economic Justice: Two Chicano Struggles in the Southwest. Tucson: University of Arizona Press.
_______. (2000). “Rethinking Environmental Racism: White Privilege and Urban Development in Southern California.” Annals of the Association of American Geographers 90: 12-40.
Pulido, Laura, Steve Sidawi, and Robert Vos. (1996). “An Archaeology of Environmental Racism in Los Angeles.” Urban Geography 17: 419-439.
Seigler, Jane. (1993-94). “Environmental Justice: An Industry Perspective.” Maryland Journal of Contemporary Legal Issues 5: 59-65.
Spence, Lisa. (1995). Race, Class, and Environmental Hazards: A Study of Socio-Economic Association with Hazardous Waste Generators and Treatment/Storage/Disposal Facilities in Massachusetts. M.A. Thesis, Tufts University, Civil and Environmental Engineering.
Taquino, Michael, Domenico Parisi, and Duane Gill. (2002). “Units of Analysis and the Environmental Justice Hypothesis: The Case of Industrial Hog Farms.” Social Science Quarterly 83: 298-316.
Tesh, Sylvia. (1988). Hidden Arguments: Political Ideology and Disease Prevention Policy. New Brunswick, NJ: Rutgers University Press.
Tesh, Sylvia and Bruce Williams. (1996). “Identity Politics, Disinterested Politics, and Environmental Justice.” Polity 18: 285-305.
Towers, George. (2000). “Applying the Political Geography of Scale: Grassroots Strategies and Environmental Justice.” Professional Geographer 52: 23-36.
Tuan, Yi-Fu. (1977). Space and Place: The Perspective of Experience. Minneapolis: University of Minnesota Press.
U.S. General Accounting Office. (1983). Siting Of Hazardous Waste Landfills And Their Correlation With Racial And Economic Status Of Surrounding Communities. GAO Report RCED-83-168. Washington, DC: GPO.
_______. (1995). Hazardous and Nonhazardous Waste: Demographics of People Living Near Waste Facilities. GAO Report RCED-95-84. Washington, DC: GPO.
Westra, Laura. (1999). “Environmental Racism and the First Nations of Canada: Terrorism at Oka.” Journal of Social Philosophy 30: 103-124.
Westra, Laura and Peter Wenz (Eds.). (1995). Faces of Environmental Racism: Confronting Issues of Global Justice. Lanham, MD: Rowman & Littlefield.
Wiggershaus, Rolf. (1995). The Frankfurt School: Its History, Theories, and Political Significance. Robertson, M. (Transl.). Cambridge, MA: MIT Press.
Williams, Raymond. (1973). “Base and Superstructure in Marxist Cultural Theory.” New Left Review Nr. 82: 3-16.
Williams, Robert W. (1999a). “The Contested Terrain of Environmental Justice Research: Community as a Unit of Analysis.” Social Science Journal 36: 313-328.
_______. (1999b). “Environmental Injustice in America and Its Politics of Scale.” Political Geography 18: 49-74.
Wilson, Bobby. (1992). “Structural Imperatives Behind Racial Change in Birmingham, Alabama.” Antipode 24: 171-202.
Wilson, Sacoby, Frank Howell, Steve Wing, and Mark Sobsey. (2002). “Environmental Injustice and the Mississippi Hog Industry.” Environmental Health Perspectives Supplements 110: 195-201.
Wing, Steve, Dana Cole, and Gary Grant. (2000). “Environmental Injustice in North Carolina’s Hog Industry.” Environmental Health Perspectives 108: 225-231.
Wright, Beverly. (1995). “Environmental Equity Justice Centers: A Response to Inequity.” Pp. 57-65 in Environmental Justice: Issues, Policies and Solutions. Bryant, Bunyan (Ed.). Washington, DC: Island Press.
Zimmerman, Rae. (1994). “Issues of Classification in Environmental Equity: How We Manage Is How We Measure.” Fordham Urban Law Journal 21: 633-669.
Theory & Science