The troubling and troublesome worlds of urban soil trace element contamination baselines

Introduction

On 16 November 2014, a prominent front-page title of the New York Post warned readers about the toxicity of New York City vegetable gardens, characterised as ‘loaded with lead and other toxic metals’ (Buiso, 2014). The response from local urban community gardens organisations was swift and trenchant. The newspaper had mischaracterised scientific evidence and had engaged in outright deception (New York City Community Gardens Coalition, 2014; United Community Centers, 2014). Among other matters, the public debate revealed the political ramifications of evaluating toxicity levels. There is much at stake, including potential harm to people’s health and access to gardening spaces and fresh vegetables for under-served communities. However, what united the contending arguments was an unquestioning reliance on baseline values as ultimate arbiter in determining whether or to what extent an urban garden soil is to be deemed toxic.

There are other issues that should be considered, such as how toxicity values are determined or what is to be done about settings deemed toxic. These are beyond the scope of this discussion, which is concentrated on the technical challenges and political ramifications masked by baselines in determining whether a site is contaminated, if not polluted by trace elements (TEs). These kinds of considerations call for what has been called a Critical Physical Geography (CPG) approach, which draws from but goes beyond Political Ecology and Science and Technology Studies (STS). However, an ecosocial framework to CPG that is based on ecosocialist politics is arguably more appropriate in challenging existing social arrangements with respect to pollution issues. It widens the scope of analysis on people–environment relations while linking (but not reducing) scientific work to an explicit politics (in this case, an ecosocialist politics).

Using secondary evidence from two cases, a brief discussion follows on two interpretive approaches used by technical experts in estimating potential risk and that have political consequences. One features the disregard for multiple forms of soil use by narrowing the evaluation to a single standard value. This is shown by the case of an evicted community garden in Sacramento, California. The other involves selecting one among multiple possible standards in an environmental impact assessment for a contended college housing project in New Paltz, New York. Such interpretive approaches are technically troublesome, but should also be viewed as politically troubling. The major challenge, to activists and scholars alike, is that merely exposing the existence of such questions – never mind addressing them – takes one immediately beyond one or another field of knowledge. The effort requires technical understandings of soils and contamination processes as well as sensitivity to social context and relations of power. In the concluding section, strategies are outlined that can be useful towards developing critical lines of inquiry on baselines without eliding the possibility of harms to health or wider deleterious ecological effects.

Critical perspectives on baselines

There has been little attention to understanding baselines – even less in the case of soil contamination – in terms of how their determination is intrinsically political, including in the sense that they can be enrolled directly in decision-making processes. There are, however, plenty of precedents for this kind of investigation in different fields that can be marshalled for different purposes. The main fields out of which this research builds are related approaches in the biophysical sciences (soil contamination and ecotoxicological studies), Political Ecology, and CPG, with hints of STS.

Within the biophysical sciences, including physical geography, there have been technically oriented critiques that can serve as a springboard for a deeper inspection into the political reverberations of scientific concepts and technical standards (Engel-Di Mauro, 2014). Examples involving soils can be found regarding historical preconceptions about tropical soils as fragile and infertile (based on assumptions of acceptable levels of stability and fertility), a notion now long refuted empirically within the soil science and physical geography mainstream (Schaetzl and Anderson, 2005: 388–392). Something similar has been accomplished regarding the (not very universal) Universal Soil Loss Equation, based on defining soil loss tolerance thresholds (or T factor) relative to crop yields in conventional agriculture. ¹ Aside from there being no necessary relationship between crop yield and soil loss (Stocking, 2003), the equation has been found to need exceedingly burdensome local sampling campaigns (for calibration) beyond the capacity of most countries (Phillips, 1989). This is an example of technical correctives over the political biases (decision-making processes advantaging the wealthy) that inhere what tend to be regarded as neutral technical interventions.

As useful as they are, technical correctives stop short of critical examinations into the social forces and power relations (including projections of supremacist, coloniser worldviews) involved in human-impacted landscapes (for a related case study on the US soil survey, see Van Sant, 2018). These tasks have been taken up instead in Political Ecology since the early 1980s and more recently in CPG and to some extent in STS. In the former approach, soil erosion was the centre of attention early on (Blaikie, 1985; Stocking, 1987). The T factor in the soil loss equation was one among several soil erosion concepts and methods shown to be both technically untenable and socially prejudicial (see also Stocking, 1996). Notable sets of contributions indirectly related to soil baselines are also those exposing the technical fallacies and anti-smallholder biases of conventional soil nutrient analyses and nutrient budget calculations in determining soil fertility values and crop yield effects (Benjaminsen et al., 2010; Scoones, 2001; Scoones et al., 1996). The contributions of political ecologists of all stripes (and there are many) have been copious, covering many dimensions of people–environment relations and focused on questioning (and sometimes proposing alternatives to) received conceptualisations and constructs (for overviews, see Bryant, 2015; Forsyth, 2003; Perreault et al., 2015; Robbins, 2004; Rocheleau et al., 1996; Zimmerer and Bassett, 2003). Regrettably, being ‘engaged with natural science and ecology’ (Benjaminsen, 2015: 358) still seems to be confused with studying actual biophysical phenomena, which has been and continues to be scarce in Political Ecology. For example, in Political Ecology approaches to soil erosion or fertility there is no theorising of soil formation and development (the works of Michael Stocking constituting a singularity), or explaining soil nutrient availability and mobility relative to evolving soil characteristics. The unit of analysis remains, as was from the start, society in relation to one or another set of ecological variables (and more often just society with ecology as accoutrement). Moreover, baselines have not received the sort of attention that might have been expected, judging from early works questioning how environmental standards are determined and used. However, a few directly relevant studies that exist, discussed below (Brenner, 2011; De Vries, 2005), demonstrate how Political Ecology can be decisive in exposing the problems with baselines, as conventionally derived, and their application’s potentially harmful social repercussions.

STS approaches – also quite variable and with a vast set of subjects broached – should come closer to critical evaluations of baselines, given their focus on the history of science and scientific practices. Much has been accomplished through STS relative to risk determination, including possibilities for wider public involvement (e.g. ‘citizen science’) and for reaching substantive social accountability (e.g. Kimura and Kinchy, 2016; Ottinger, 2017). But the baselines that often constitute the background to risk appraisals remain outside STS’s purview. Moreover, as or perhaps even more than in Political Ecology, biophysical phenomena take a back seat. It is arguably the very emphasis on science history and practice that eludes biophysical phenomena in STS analyses. Hence, investigation outcomes typically fall short of furnishing any clearer ideas or any alternative theories about the workings of actual biophysical processes. Regardless, relative to the purposes of this work, surprisingly little attention has been given to the issue of baselines and soils in general, except as a tangential exercise with soils as pretext (e.g. Latour, 1999).

This is far from saying that STS offers nothing for those interested in soil contamination baselines. An illustration is the fine study contributed by Ureta (2018), likely the only scholar who has, laudably and insightfully, investigated the actual baseline construction process. As would be expected in an STS approach, the case study (carried out in Chile) excluded an examination of actual soil processes or the technical dimensions of the constructed baselines, including any divergence among experts regarding methods and standards. There are therefore substantive limits to the study that arise from a lack of engagement with existing debates among technical experts, especially at the international level. These cannot be addressed here (though some are broached below), but suffice it to remark that what Ureta (2018: 353) has found problematic is familiar to practitioners themselves more widely. For example, the notions that there is no immutable natural background, that site history and non-human organisms must be considered, and that baselines are always multiple are all actually points of departure for, not something foreign to, those involved in studying soils and involved in establishing soil quality standards (Borrowsky et al., 2005; McLaughlin et al., 2011; Morgan, 2016: 74–76).

This lack of immersion in a biophysical science is among the reasons for the recent emergence of CPG. It builds on STS and Political Ecology, but is also a response to the dearth of investigations in actual biophysical processes within those frameworks. CPG combines studies of biophysical phenomena (especially landscapes) with researcher reflexivity and the analysis of the production and politics of knowledge, which tend to be omitted in the biophysical sciences. As in Political Ecology, the aim is to contribute to transforming politics, but this is complemented by a twin objective of creating a truly interdisciplinary science (Lave, 2015; Lave et al., 2014). Within a few years, CPG research has proven a lively field encompassing a great diversity of subject areas, with so far geomorphological and biogeographical issues attracting greater attention. Some substantial work has already been accomplished through a CPG framework, ranging from critiques of soil science to imbrications of social and pedochemical changes (see, for a sample, the collections in Lave, 2015; Lave et al., 2018). Though soil contamination baselines have yet to be addressed, McClintock’s work on soil lead well illustrates how a CPG research programme can effectively capture the ecosocial character of contamination processes and, potentially, the constructs thereto related. As he explains, data generated through conventional environmental science form but an initial foray into unveiling how the scale and distribution of lead contamination (shifts in biophysical processes) emerge through historical and current social relations of domination (McClintock, 2015).

An ecosocial CPG framework with an ecosocialist politics

The above-discussed frameworks underlie but are insufficient for the analysis offered here on soil contamination baselines. In some respects, the present work builds on and contributes to the development of Environmental and Soil Science, Physical Geography, Political Ecology, STS, and CPG. Such is, though, only in part the intent. The other is to combine aspects of all those approaches in the further development of an ecosocial framework, guided by historical and dialectical materialism ² and an ecosocialist politics, where social processes are always placed in historical and dialectical relationships with and within a wider set of ecological and physical processes relative to multiple spatial and temporal scales (Engel-Di Mauro, 2014). It is an approach that entails commitment as much to exploring biophysical processes as to producing knowledge useful towards developing an ecosocialist politics (Engel-Di Mauro, 2015). It is, in fact, a return to and an updating of socialist perspectives that had begun to be developed (but prematurely ceased) for physical geography by the late 1970s (London Group of the Union of Socialist Geographers, 1983; Wisner, 1978).

Relative to social power relations, influential approaches include materialist Ecofeminism, Eco-Marxism, and aspects of Environmental Justice and Political Ecology, among others. Materialist Ecofeminists’ subsistence and eco-sufficiency perspectives are especially important (Bennholdt-Thomsen and Mies, 1999; Salleh, 2009), whereby environmental practices and outcomes are evaluated relative to their contribution to the development of a society in which satisfaction of peoples’ basic needs prevails without undermining ecological and social conditions (see also O’Connor, 1998). This approach overlaps with the world-systems paradigm that initially undergirded Political Ecology, where people–environment relations are viewed as inflected by global capitalist dynamics (Blaikie, 1985; Engel-Di Mauro, 2009).

As such approaches scarcely address biophysical processes per se, an ecosocial CPG framework requires developing a thorough grasp of the study of biophysical (or non-human) phenomena. Such a framework is not limited to physical geography but extends to any biophysical science. More than this, it is a call for direct involvement in a biophysical science, including knowledge production. Finally, though an ecosocial CPG framework, as CPG generally, emphasises social critique and reflexivity as central facets of scientific research, it is not necessarily politically explicit in terms of what sort of society must be overcome and what is to replace it. This is where a commitment to building ecosocialism can be helpful, since it is an outlook that takes both ecological issues and social justice as pivotal dimensions of political struggle.

Ecosocialism, briefly put, is a movement, perspective, and by now even an institutional politics (Engel-Di Mauro, 2019) that draws from and attempts to gather together socialist and environmentalist objectives. It is socialist in the sense of identifying capitalist relations as ultimate and systemic cause of structural inequalities and environmental destruction. Politically, this means struggling for social equality by establishing the social control of the means to life. This includes cross-generational justice as well as respect for differing knowledge systems, striving to combine them (where reconcilable with socialist goals) to the benefit of all concerned. It is environmentalist in calling attention to the ecologically destructive character of currently conventional ways of living and in premising the understanding of biophysical processes on diverse forms of systematic knowledge and inquiry, institutional and otherwise. Ecosocialism, in other words, stands for the development of ecologically sustainable egalitarian communities worldwide (Kovel, 2014; Löwy, 2011; Turner and Brownhill, 2006).

An ecosocial CPG intervention into baseline development or use offers an example of developing a socially critical and scientific basis for an ecologically constructive and politically sensitive relationship to soils (or biophysical processes more generally) that is consonant with the promotion of ecosocialism. The assumption is that socialism, for it to be feasible, requires the concomitant achievement of ecological sustainability, ³ which cannot be practically verified without studying and understanding biophysical processes. In this case, critically evaluating baselines is intended as a contribution to formulating context-sensitive, ecologically sound, and socially egalitarian alternatives in our relationship to contaminated soils.

Situating TE contamination baselines

Soil TE contamination is now a widespread and centuries-lasting phenomenon in largely industrialised settings, within and beyond cities (Kabata-Pendias, 2011; Meuser, 2010; Wortman and Lovell, 2013). Long-term or high-level ingestion or breathing in of TEs in soil particles and possibly in city-grown food can eventuate into permanent health damage. Actual health effects are contingent on whether TEs can be accessed and assimilated biologically (Menefee and Hettariachchi, 2018). TE contamination sensitivity also varies considerably between as well as within species, so that multiple safety levels are formulated according to specific situations or questions (Brown et al., 2016: 28; Centeno et al., 2005; McLaughlin et al., 2011: 13–14; Morgan, 2016: 73). The recalcitrance and therefore cumulative nature of TE contamination makes for the sort of problem that is not resolvable except by containment and prevention. What complicates matters is that TE sources can be geogenic and geogenic contamination may or may not be amplified by human intervention or prior impacts in combination with actions by soil-dwelling organisms (Alloway, 2013; Bourennane et al., 2010; Howard and Olszewska, 2011; McBride, 2007; McKone et al., 2011; Pouyat et al., 2007). It is therefore not straightforward to pinpoint sources of TE contamination or neatly separate anthropogenic from geogenic effects, much less come up with universal standards, as specialists have long known (McLaughlin et al., 2011). Place could not play a more salient role in this instance.

Nevertheless, the degree of soil TE contamination is typically understood through comparisons with baselines (maximum allowable levels or thresholds), as illustrated by the above-described debate over urban garden soil contamination in New York City. These levels are derived ultimately from background levels. As a standard institutional procedure since at least the 1970s, establishing baselines undergirds assessments of environmental toxicity or risk to health, as well as environmental justice. Many people accept and rely on baselines, whether explicitly or not, across the most disparate kinds of communities. The process of constructing baselines is in large measure shrouded in technical experts’ meticulous and detailed work, which is inaccessible to most. The outcome of such work, though, frames subsequent deliberations over and actions on the status (e.g. relative health threat) of an environment (e.g. an urban garden or an abandoned orchard). This alone should pose some concern over the political consequences of technical experts’ knowledge production and its implementation. As some have already pointed out, including among soil contamination experts, who regrettably envision only conventional categories of politics and political frameworks (e.g. Farmer et al., 2009; Provoost et al., 2006), there are value judgements (or political ideas about what the world ought to be) in environmental risk assessment and decision-making processes. Moreover, these may be at odds with the worldviews or priorities of those whose lived environments are being associated with toxicity (Ball, 2002; Holifield, 2012). But as argued here, the consequences of relying on technical knowledge production have even more profound consequences than this. As Sebastián Ureta (2018) has ably put it:

Polluted entities cannot be risky as such. It is only through establishing connections with a certain kind of baseline that they become a subject of risk. So if we do not have insights into how such baselines are produced, we are blind to a key aspect of how pollution-related regulation works. (344)

Still, concerns over environmental degradation imply constructing a means by which to evaluate whether an environmental change is destructive or not, providing an understanding of which conditions are conducive to well-being and which are not (and for which organisms or sets of organisms or ecosystems, and at what scale), or whether an intervention for conservation, restoration, or remediation has attained its intended effects. Awareness of change implies comparing different sets of conditions, but also normative standards of evaluation that are undergirded by values of acceptability. The very credibility of claiming the occurrence of environmental destruction rests on establishing standards upon which one is to make judgements. Baseline values can be useful towards this, but it depends on how they are built and the criteria used to build them. In restorative or conservation efforts, baseline values continue to be used towards ecosystem recovery or improvements in their functioning. These schemes may or may not have people as a priority and it is not always clear whether ecosystems are predicated on human well-being. In fact, people (except the technocrats that devise conservation or restoration schemes) are often still treated as if we were not part of the making and functioning of ecosystems. Such worldviews are directly challenged by the virtual planetary ubiquity of human presence. To make themselves relevant, ecologists have been compelled to come to terms, often quite awkwardly, with what might be described as their own, inescapable humanness (see McDonnell and Pickett, 2012; Martzluff et al., 2008).

Baseline values, as currently devised and applied, pay no mind to relations of power in society and can result in disempowering the very communities that are supposed to benefit from baseline analyses (De Vries, 2005). Hence, baseline values contribute to effacing the social relations out of which they are created, to reinforcing society–nature dichotomies, and, when premised on the absence of people, to contributing to the sometimes violent eviction of entire communities from areas of conservation or restoration. Furthermore, ecosystems change as relationships among organisms, including humans, shift over time. Change is intrinsic to ecosystems and is not necessarily predictable (Botkin, 1990; Zimmerer and Young, 1998). To think otherwise would be like expressing surprise that Manhattan is not a mixed deciduous forest biome or that currently recognised native species were once invasive (Yelenik and D’Antonio, 2013). Baseline values, when taken as fixed targets, can contribute to stifling the development of effective measures in conserving ecosystems or people’s health.

Baseline values also depend on who determines them and how. Much has been made of the shifting baseline syndrome (reference ecosystems being contingent on individual scientist’s past experiences), when the fundamental problem lies in not admitting that decisions over ecosystem characteristics imply value judgements (Campbell et al., 2009). Some of them are rooted in nostalgic longings for or Arcadian imaginaries of bygone worlds (Alagona et al., 2012; Cronon, 1992). Some of them may even be to some extent invented for institutional expedience and with little empirical support (Robertson, 2006). Regardless, when people are considered as part of ecosystems, as they should be, what constitutes an ecologically important species can also depend on prevailing livelihoods (Brenner, 2011), and this may contradict baseline values for ecosystem species composition. To insist on certain kinds of species composition can mean the destruction of local economies. This is reason enough for baseline values to incorporate social context sensitivity and, even better, to state social commitments explicitly. They should entail the consideration of histories, local context, and the perspectives of those directly affected, who may also have developed their own (possibly conflicting) baseline values.

Challenges intrinsic to soil contaminant baseline determination

Unlike most environmental conservation work, the processes and entities to be identified in soil contamination are not directly sensed or measurable. Environmental agencies and scientists, as well as environmentalists and critical or radical academics, often rely on standardised baseline values to determine whether a soil is contaminated, and to what extent. Although it may be sensible to take such standards for granted or as points of departure for environmental justice and other kinds of more overtly political work (e.g. Mohai et al., 2009; Pellow, 2014; Schlosberg, 2007), there are some problems ensconced in the ways in which baseline values are determined. To some extent, the problems are specific to the complexity of soil TE contamination analysis.

It necessitates estimations of typical or background levels for each TE in soils. Soil TE levels vary geographically and over time in the same areas (Kabata-Pendias, 2011). Background levels become the baseline values that serve to derive soil quality standards and threshold values, which are soil TE concentrations surpassing a soil quality standard or baseline. Such thresholds are typically named maximum allowable levels (or some such terminology) in public discourse, where the allowable is relative to potential human health effects. Levels are calculated by applying mathematical models estimating the likely extent (statistically significant) of TE transfer from soils to organisms. Hence, a threshold is expressed as a single number that functions as a baseline that, in turn, rests on a soil quality baseline. Potential or likely health effects are established through ecotoxicological findings, which provide the basis for what can be considered as another set of baselines (e.g. relative to bioavailability). Beyond soil TE concentration thresholds certain uses are institutionally discouraged if not legally prohibited, or beyond which environmental agencies (and other state organs) are legally bound to intervene, start a risk assessment process, and possibly engage in remediation or clean-up operations. In this manner, the baselines (threshold values) used in soil contamination assessment form part of much broader social processes than scientific research and technical applications, as acknowledged by environmental scientists themselves (McLaughlin et al., 2011; Provoost et al., 2006).

Even from a narrow technical perspective, there are several problems to be confronted when using baselines, whether as soil quality standard or maximum allowable contaminant level. One is presenting soil quality standards as single-figure baselines, rather than as a statistical descriptor (e.g. an average) that oscillates within a range of values (e.g. having standard deviations). A similar problem exists for maximum allowable levels. Employed without model parameters disclosure and explanation, they hide the multiple variables selected and that can shift substantively over time and place. That is, the same maximum allowable levels (or index value, more broadly) can result from very different combinations of values for the variables used in the soil contaminant model (a problem of equifinality). It might be thought that representing a baseline in the form of a gradation between lower and higher acceptable values might create a potential opening for contestation over the instance at which a site is deemed a health risk. However, ambiguity can also help exonerate a polluter or an institution from responsibility for clean-up operations and/or reparations by taking advantage of different portions of a range of values according to political expedience. Given enormous disparities in political power, flexible interpretability in itself is no guarantee of greater democratisation in contamination risk determination. This is even more evident when other, deeper issues are considered that are typically obfuscated when data are divulged to non-experts. They include insufficient data and/or monitoring (implying inadequate institutional investments), locality-specific ecological complexity, and shifts in actual background levels. Another set of problems occurs by way of political practices (i.e. decisions over what sorts of conditions we are to live, how we are to live our lives, etc.), which subtend both the concept and application of baseline values.

The above are all major challenges about which the wider public is rarely, if ever informed. First is a simultaneously technical and ideological difficulty in deriving baselines for comparisons because parsing out the geogenic from the human-induced is a complex procedure, due to possible overlaps in source levels and the fact that human intervention may attenuate or exacerbate pre-existing high levels of geogenic contamination. Second, the TE levels to be regarded as contamination vary according to scale of analysis (ecosystem/community, species, population, individuals), the kind and number of species selected, and whether the focus is solely human health. Third, as discussed above, the presence of a contaminant in soils does not guarantee contaminant transfers to organisms in and beyond those soils. This is due to the many factors involved in making contaminants more or less mobile in soils (hence the recourse to modelling to calculate thresholds).

Because human impacts have become virtually global in character, cases of ‘natural’ background concentrations are rare. Among technical experts, therefore, one speaks of ambient background concentrations. This may account for human-induced levels of contaminants, but separating ambient from geogenic sources is not necessarily self-evident. This is because, to reiterate, of overlaps in concentration levels between human (or, better, social) and geogenic causes (McLaughlin et al., 2011). This is regarded as a problem because the technical approach is subtended by scientists’ subordination to nature–society dichotomies embedded in state regulatory frameworks. Such dichotomies are enshrined in laws, at least in liberal democracies, that require ascription of individual responsibility or of harm to individual humans. Geogenic contamination problems are treated as a separable category, even when there is lack of clarity. For instance, pozzolanic ash in Rome has high lead content and local travertine deposits have high amounts of arsenic (Engel-Di Mauro, 2018). However, clean-up operations are not necessarily triggered legally when causes are geogenic. In other words, if each place were studied on its own merits, on the basis of specific characteristics, and with people’s health in mind, it should not make any difference what the main source of contamination is, relative to intervention. But, of course, rock formations cannot be brought to trial and capitalist institutions only recognise human labour, as Karl Marx pointed out long ago (Postone, 1993).

TEs themselves vary in their potential effects but also in the form that they take on once in soils, to make matters even more complex. They can be useful (e.g. iodine, iron) or harmful (e.g. chromium VI, mercury), depending on quantities ingested and whether they remain within organisms and for how long. It is difficult to assess whether and to what degree TEs in the environment are transferable (bioavailable) to people or other life forms. In other words, even establishing whether, to what extent, and how organisms, including us, may be exposed to unhealthy or toxic levels of TEs remains a major challenge. There are many sources, factors, and processes involved; they are environmental and social, as well as pre-existing and evolving (for a related discussion focusing on soil lead, see McClintock, 2015). Each location can have different sets of sources, factors, and processes in variable combinations and varying over time. What is more, each life form differs in health thresholds in that some organisms, even at the individual level, are more tolerant than others to TEs and not necessarily to the same suite of TEs. In spite of such complexity and place-specificity, government agencies have developed and enforced universal standards for contaminant level acceptability.

Technical interpretations of contamination levels and their political implications

The above contingencies should make anyone pause merely on technical grounds about how soil TE baselines are to be decided. Yet even if the matter could be easily resolved, there are also political consequences associated with differing interpretive approaches to soil contamination data. Two are discussed here to exemplify the issue, relying on secondary information from two different US cities to illustrate the processes involved. The first case, the denial of multiple baseline possibilities, is from a study by Cutts et al. (2017). The second, the selection of only one of multiple possible baselines, is from an environmental impact assessment applied to a planned housing project in New Paltz, New York. This second case involved no formal research, strictly speaking. The author was made aware of the assessment by the institution (the author’s employer) promoting the project and was asked by some of the activists opposing the project to comment on that assessment. These cases are also selected because they are both within a common national context (the US) under broadly similar environmental policies and social conditions. Furthermore, they involve the uses of baselines – one explicit and the other hidden – in deliberations over frequently occurring urban land use decisions (gardening and housing).

As for the first case, Cutts et al. (2017) recount how, near Sacramento, California, soil contamination was used in the 2004 local government’s destruction of Mandella Community Garden. The local authorities’ justification for the draconian measure was the presence of high amounts of lead, which effectively called into question the benefits often claimed about urban gardening. What was removed was not only all that the gardeners had built over several decades, but also between 30 and 120 centimetres of the soil they had cultivated (Cutts et al., 2017: 14). The community garden was replaced with another one and on virtually the same site, accompanied by an adjacent housing unit. The complex was renamed the Fremont Mews, endowed with 52 garden plots accessible for a nominal fee. What is of special interest in this case is that local government and construction businesses appealed to soil lead contamination as a major justification to relocate the original community garden and, after much resistance, to destroy that garden altogether. They managed to succeed, as the authors put it, in ‘reframing the garden as dangerous dirt in need of purification’ (Cutts et al., 2017: 14). The authors conclude that contradictions are inherent to urban gardening projects because

urban gardening is dependent not only on the function and form of a particular garden and how and where it arises (location), but also on how its functions and form change over time (duration) and its connection with other proximate and distant garden sites (interconnectivity). (Cutts et al., 2017: 16)

The politics of soil lead in the Sacramento case brings up two issues involving baselines. One is that activists were not in a position to counter government discourse about contamination with technical studies of their own and on their own terms. They apparently were unaware of the contingencies and flexibility of baselines. For instance, if activists had been privy to studies on soil lead, they could have stated from the outset that lead can be contained and they would have been able to formulate the preventive measures that are given by extension agents regarding contamination. This could include raising pH levels and maintaining a permanent vegetation and mulch cover to restrain soil particles’ entrainment into the air (see Menefee and Hettariachchi, 2018). ⁴ Activists, if only they were aware of the lead re-deposition problem (well-known to technical experts), could have gone on the offensive and gathered air pollution data. They could then have alerted the public to a much larger problem and threat to health, thereby channelling attention away from the community garden as problem and refocusing attention on the environmental problems of the city as a whole.

The second case was much less destructive, but insidious nonetheless. In 2012, in New Paltz (New York), a large construction firm joined the local college administration in an ultimately unsuccessful bid to turn an old abandoned orchard into a college residence. Town hall meetings were convened to invite interested parties to voice concerns, as legally required (NYS DEC, 2006). Environmental impact assessment reports were duly made available to the public. There was much at stake financially, for the construction company, the college administration, and the private consulting companies hired for the assessments and building process. Environmental benefits were enrolled to legitimise the residential project by claiming a reduction in greenhouse gas emissions from car-dependent commuters (Kimble, 2015). However, according to one of the consulting firms’ findings, the soil on which the residential area was to be built retains substantial levels of 4,4 DDT and dieldrin, as well as high amounts of arsenic (Ecosystems Strategies, Inc., 2012). Nonetheless, the overall assessment of the site was deemed positive, in spite of concerns raised by a hydrogeology consultant about groundwater availability and quality (Miller Hydrologic Incorporated, 2012). There were several points of contention raised in the ensuing (at times heated) town hall debates, such as the impacts on local sewage treatment and potential strains on the municipal water supply. Given the reassurances from the environmental impact assessment, soil contamination was effectively excluded from discussion.

Yet the technical report was riddled with faulty methodologies and questionable assumptions. Soil sampling was biased (favouring minimum contaminant detectability), most TEs were excluded from lab analysis, and dust-borne contaminant dispersal issues were disregarded (for more details, see Engel-Di Mauro, 2014). Of greater relevance to this discussion is the consultancy firm’s selection of less stringent critical values for acceptable contaminant levels. In this manner, the technical consultants hid some assumptions about what counts as relevant information and even what kind of activities should be allowed on the premises. In the interpretive part of the analytical report, the consultants conveniently omitted the much more stringent ‘Unrestricted NYSDEC SCO’ standards, opting instead to include only the Residential version. This selectivity cannot but go unnoticed by the majority of locals, who are unfamiliar with such documentation or how to interpret it. But by using the critical limits set for residential land use, the consultants effectively imposed a policy decision on what kinds of activities would be permitted. For example, establishing a garden to grow food, which is of interest to many students, is foreclosed as an option. If the contaminant limits used had been for unrestricted use, the developer would be forced to decontaminate the site so as to enable other uses besides conventionally defined residential use, an official definition that also presumes that people do not grow food where they reside. In this manner, they were able to reassure the public that TE levels found on the site are to be regarded as safe.

These two cases, one in Sacramento and the other in New Paltz, exemplify technical interpretive practices that can, in effect, conceal political decisions. The first, from Sacramento, reflects how baseline values can be used by local government authorities to dismantle an urban community garden in favour of construction and real estate businesses. Baselines were taken as given by all contenders, to the detriment of the politically disempowered (the community gardeners). Whether lead levels were treatable never came up for discussion, as the parties involved seemed unaware of the options available to mitigate soil lead mobility. The New Paltz case, where baselines were also taken as given, exposes subtler processes of land use decision-making by means of technocratic default. While in the first case, the existence of a plurality of technically legitimated standards is effaced, in the second case that plurality is reduced to a single standard by expert selection, without explicit justification. Either interpretive scheme, depending on context, can be marshalled to favour the interests of those hiring the technical experts, at least for the cases considered here. In assessing soil contamination, baseline values can therefore be rife with largely implicit or veiled politics. This should not be terribly surprising, because when it comes to soil contamination the matter is not only about ecosystems and health, but also or especially about struggles over land access and use, among other quintessentially social issues.

Towards an alternative approach

The conventional application of baselines is highly problematic, but it is especially pernicious when there are pretences of neutrality or objectivity in their use and when discussions are clouded in technical jargon unintelligible to communities affected. In Sacramento, where urban community garden activists were aware of soil lead exceeding safe levels, baseline values were used effectively in evicting activists and replacing the community garden with expensive housing and adjacent garden allotments for new residents. Specialist knowledge in TE contamination processes and in the technical feasibility of attenuating lead mobility could have helped mount a more effective campaign to save the community garden. An inability to question a particular use of baseline values is part of the reason for the failure, although one may suspect that the local government would have likely used other means to legitimate the bulldozing of the community garden. Nevertheless, the process would have at least made evident the authoritarianism involved: liberal democracy in its utmost splendour. The New Paltz situation points to related problems in that technical experts’ choice of a baseline subcategory acted as default reference for all contending parties. This effectively narrowed the scope of permitted land use in favour of building promoters. Even if the housing project bid ultimately failed, the intended prospective residents (students, faculty) were effectively denied any say on matters of potential health and safety issues within and around the housing structures.

Baseline values, in some respects, can be pervaded by commitments to the prevailing social order when they are regarded as sheer technical matters, voided of local ecological and social histories and multiple perspectives, treated as ultimate scientific authority, and rendered incommunicable to the most affected people. In fact, baseline values run the risk of creating fictions (past or present idylls devoid of toxicity), promoting what can become a false sense of certainty and health safety. In policy use, they can even get detached from actually existing environmental processes and enter a fanciful world where nature can be amended at will. The implications of baseline values and their derivatives may or may not be intelligible to most people. What can happen, as illustrated by the two case studies, is that political contestation is effectively pre-empted by a veil of baseline values and their spin-offs, used as part of a screen of technical expertise behind which decisions are made and imposed without meaningful consultation, if any, with a community affected. These political impositions are difficult to detect without immersing oneself in the study of environmental contamination processes and of related data-gathering and interpretive methods. What institutional approaches simply cannot grasp is that baseline values are at the same time embedded in social relations, including environmental politics.

In sum, baselines cannot be conceived as absolute or neutral standards without ultimately masking the political processes that call them into being. They are relational in that they must rest on constant comparisons. They are enmeshed in decision-making processes, in often unaccountable politics, redolent as they are with regulatory mandates established in the absence of participatory practices or even frameworks. The bases undergirding baselines serve as dynamic markers resting on shifting ecological conditions now largely shaped by human action and thereby social power relations. Technical expertise, though essential, is inexorably political at the same time as it forces engagement with what lies beyond society. This realisation, however, does not lead seamlessly towards a democratisation of knowledge production relative to soil TE contamination. In this connection what is suggested by Ureta (2018) is commendable, which is to invite people living in an impacted site ‘to discuss with them different possibilities regarding the conditions that they think a baseline for the site should include’ (353). But there are some crucial steps missing as well as an implicit assumption about wider social conditions. Understandings of the dynamism of the bases in established baselines could and should be opened up to everyone, but with due technical preparation and skilling as well as analytical instrumentation made accessible to all. Without a diffusion and commoning of necessary technical expertise and instrumentation on matters that often directly impinge on our health (and differentially so), opening up baselines to all is akin to land redistribution to a majority who, in the US case, know not how (or is otherwise unable) to grow crops, and perhaps know not even where food comes from. It would be another way of reproducing, if not exacerbating social inequalities.

An alternative approach could be developed that brings into the open, for public purview, the currently largely tacit decision-making process underlying baseline determination and implementation. What needs to be taken seriously is the ecological and social context and the possible transformations involved in local people–environment relations that can prevent harm. The latter is the laborious and long-term monitoring work required in accounting for ecological change, which implies understanding people–environment relations as mutually transformative processes (a quintessentially dialectical materialist understanding). If context were taken as a primary means to assess TE contamination, it would already by-pass many difficulties created by an insistence on subjecting every place to the same baselines and derived maximum allowable levels or concentrations. Such a practice already exists and is known in the expert literature as a geochemical regression technique. It can account for locally specific soil parameters, such as clay mineralogy, levels and type of soil organic matter, and other soil properties that can lead to TE retention instead of mobility. In other words, TE concentrations are normalised to local ambient background concentrations and soil parameters by means of statistical regression models, which function best in small areas (McLaughlin et al., 2011: 10–12). ⁵ For example, an initial study may find close associations between cadmium and certain types of local soil texture (Zhao et al., 2007). In this manner, the methodology lends itself to context specificity and can be useful in reducing the number of parameters (and therefore lab tests) necessary to estimate TE contamination baselines.

The methodology can be used as part of a much larger discussion involving political decisions and ramifications. A geochemical regression technique, for instance, does not address the social processes that eventuated into TE contamination and that shape its subsequent treatment. Understandings of biophysical processes and associated risk calculations help identify the ecological or physical environmental dimensions or processes of contamination, a necessary building block for what is ultimately a social struggle. Decisions over what counts as dangerous contamination and what to do about it must be made within the communities affected by the health consequences of institutional inaction or by threats of misplaced action (such as community garden closure, given that adequate safeguards can be introduced). How such past or contemporary decisions come about is contingent on pre-existing relations of power and social histories. In the case of New York City community gardens, where there is some public awareness of and gardeners’ concern over contamination problems, some key issues include not only sorting out pollutant exposure potential, but also, among others, gaining or retaining cultivable land (especially by racially minoritised communities), access to education on contamination processes, and the means of environmental monitoring, as well as overcoming histories and current practices of environmental racism.

In other words, the use of baselines in themselves does not necessarily disempower or empower anyone. The matter hinges on the social processes and power relations leading to differential levels of awareness regarding baseline determination and context specificity, as well as knowledge of technical alternatives to circumvent a contamination problem. This is also because no environmental or health risk calculation – whether relying on baselines or not – can resolve what is ultimately a social justice problem. The resolution of contamination problems requires political organising and struggle on many fronts at once, from the challenge of even arriving at the means of identifying contamination and its processes to ensuring health protection and informed inclusion in the use of a contaminated resource.

The participatory and democratising potential of ‘citizen science’ (or even, to some extent, substantive institutional inclusivity of ‘local knowledge’), as pointed out and studied by some in STS and Political Ecology, provides an initial and promising entry point, but there is a risk of tokenism and co-optation without at the same time a commitment to grasping and fighting against structural inequalities (Kimura and Kinchy, 2016) and, one could add, a clearly articulated set of shorter and longer-term political objectives of overcoming capitalism (such as what ecosocialism can offer). What could also assist in such struggle is the development of an understanding, among all concerned in an affected area, of the scale of analysis to be used (a politics of scale, since contamination in an individual garden reflects outcomes of much wider social relations, from neighbourhood to international levels), along with clarity about whose health is most affected and must be prioritised within a community struggling with contamination problems. This can help clarify objectives and build effective strategies (e.g. deciding over which actions to privilege, legal or otherwise; which institutions to target; and where to concentrate the most effort).

Satisfying such aims necessitate not only an explicit politics (especially on the part of technical experts), but also a set of conditions where social equality prevails, if contamination problems are to be justly treated. As this is arguably impossible in liberal democracies and other kinds of political formations under capitalism, the promotion of ecosocial or ecological justice (Pellow, 2014; Pulido, 2017), which is what such an alternative approach would require, entails a struggle to overcome capitalist relations altogether, even if one bit at a time when it comes to democratising the making and use of baselines. The relative success of such an endeavour towards that end, if this line of work is even considered by a broader than academic public, ultimately resides in the outcomes of quotidian struggles.

Highlights

Soil contamination levels are typically assessed using baselines