Environmental imaginaries and the environmental sciences of antimicrobial resistance

Abstract

The environmental dimension of antimicrobial resistance has expanded the scope and scale of policy concern and research interest in antimicrobial resistance to include not just clinical and agricultural settings but a wide variety of environmental spaces and places. This article examines the ways in which environmental scientists researching the environmental dimension of antimicrobial resistance produce culturally specific forms of environmental imaginaries as a means of stabilising complex, uneven and open-ended environmental, human and microbial relations. These imaginaries work to structure the gaze of scientific enquiry towards particular places, objects and scales, and justify particular decisions and practices over others. Drawing on the imaginaries literature from Science and Technology Studies and Cultural Geography, our analysis examines the spatial and temporal dimensions of environmental imaginaries. In doing so, we identify four imaginaries, the environmental hotspot, the pristine environment, the fluid environment and the environmental reservoir. These distinct but interconnected imaginaries produce a constellation of ideas and assumptions that shape scientific practices, the ways and places in which the environmental dimension of antimicrobial resistance becomes known, and the types of interventions and actions that are made apprehensible as a result. In opening these imaginaries to interrogation at this relatively formative stage, we aim to identify ways in which social science contributions can complement and enhance the ways in which the figure of ‘the environment’ is brought to bear on responses to antimicrobial resistance.

Keywords

Antimicrobial resistance antibiotic pollution environmental sciences imaginaries

Introduction

Rising antimicrobial resistance (AMR) amongst bacteria, viruses, fungi and parasites is recognised as one of the most serious global threats to human health because it erodes our ability to treat life-threatening infections (O’Neill, 2016). In particular, the threat of resistance to antibiotics is said to be on par with that of climate change (Harvey, 2019). Public policy has previously approached AMR as either an agricultural antibiotic use or human clinical problem. However, the One Health concept has more recently been mobilised in policy to draw attention to AMR as a problem with interlinked human, animal and environmental dimensions (European Commission, 2017; World Health Organisation et al., 2016; UK Research and Innovation (UKRI), 2019). As a result, the ‘environment’ has acquired new significance in relation to AMR with calls for more research to identify how interventions in the environment might mitigate and manage AMR (Pruden et al., 2013). In this paper, we employ the concept of ‘imaginary’ to explore how environmental AMR scientists represent the environment and environmental interactions. These imaginaries, we argue, are significant not only for scientific practice but for the ways in which environmental policy interventions around AMR are being conceived. At present, the scope of these interventions is bounded and shaped by how the emerging community of environmental AMR scientists imagines the environment. By elucidating and interrogating the most prominent imaginaries within the field of environmental AMR science, we aim to identify ways in which social science contributions can complement and enhance the ways in which the figure of ‘the environment’ is brought to bear on responses to AMR.

Expanding the AMR policy approach to include the environment is informed by a significant body of scientific research that has already been conducted, collectively highlighting two major environmental threats contributing to the rise and spread of AMR. The first threat is from antimicrobial pollution exacerbating the conditions under which bacteria become resistant to antibiotics. Antibiotic residues have been detected in a range of environments beyond the clinic or the farm including wastewater streams, surface and ground water, sediments and soils (Kümmerer, 2004; Rosenblatt-Farrell, 2009). Some of these can be linked back to clinical or animal agricultural waste, but others encompass a wide variety of sources (Gothwal and Shashidhar, 2015) including antibiotic manufacturing, bioethanol production, domestic waste, horticulture and aquaculture. The second threat is linked to conditions favouring the transmission and selection of AMR. This includes selection for resistance due to antimicrobial exposure in the environment (Kümmerer, 2004); spread of resistance between organisms/strains through horizontal gene transfer; emergence of novel resistant bacteria which could spread; and emergence of new, mobilisable resistance genes and/or new combinations of existing resistance genes (Martínez, 2008; Singer et al., 2016; Wright, 2010).

Yet, by comparison with more established accounts of AMR in clinical and agricultural contexts, the environmental dimension poses a distinctive set of challenges for the problem and solution framings of AMR. The clinic and the farm have both been widely represented as discrete or bounded sites where antimicrobials are prescribed and used in ways that contribute to the rise of AMR. In these contexts, research and policy practice on AMR have drawn attention to failures of public and professional knowledge and behaviour regarding the use and prescription of antibiotics, in turn presenting as a solution the promotion of ‘judicious’ (Hulscher et al., 2010), ‘prudent’ (McNulty et al., 2007) or ‘rational’ (Holloway, 2011) antibiotic prescribing and use behaviours.¹ In contrast, research on the environmental dimension of AMR draws attention to a broader set of open-ended processes, flows and interactions between the environment, antimicrobial chemicals and bacterial ecosystems implicated in the rising prevalence of AMR (Bengtsson-Palme et al., 2017; Ding et al., 2014; Greenfield et al., 2018; Larsson et al., 2007). How these complex interactions between existing human and natural systems might be rendered apprehensible and manageable is a significant challenge. This requires decisions about which environmental and AMR processes, relations and flows are significant and can be the subject of scientific analysis.

The recent coalescence of this field of scientific study alongside the lack of mobilised publics, or controversy (Harvey, 2019) means that the ways in which scientists make sense of or, as we frame them in this paper, imagine the environment are particularly significant. As human geographers have shown, natures are produced through material and social processes (Castree and Braun, 1998, 2001). As we will demonstrate, scientists’ environmental imaginaries, although emergent and unsettled, work to stabilise the complex, open-ended and uneven interrelationships that characterise the environmental dimensions of AMR. These imaginaries emphasise certain objects, places and spaces as significant to scientific practice over others. This not only has implications for the forms of knowledge science produces and does not produce about the environment (Murphy, 2006) but also the types of interventions and actions that are made apprehensible as a result. This paper critically examines these environmental representations and their implications for action. In doing so, it draws upon a growing body of empirical and theoretical work on imaginaries and imagination (Fortun and Fortun, 2005; Jasanoff, 2015; McNeil et al., 2017; Watkins, 2015).

Distinguishing certain types of imagination from fantasy and illusion, the imaginaries literature recognises imagination as a cultural resource with performative power that materially shapes social practices and order. Our empirical analysis draws on human geographical investigation into environmental and spatial imaginaries (Davis, 2011; Watkins, 2015) and Science and Technology Studies (STS) literature that explicitly engages with the role of imaginaries in future-making (Groves et al., 2016; Jasanoff and Kim, 2009). Environmental imaginaries represent a constellation of ideas developed by social groups about a given landscape or environment (Davis, 2011) producing how these groups envision, experience and reshape the world (Daniels, 2011). Consequently, environmental imaginaries act to stabilise complex, uneven and open-ended environmental, human and microbial relations, structuring the gaze of scientific enquiry towards certain places, objects and scales, and justifying certain practices within them. Furthermore, analysis of imaginaries requires consideration of the social and material infrastructures which accompany their (re)production (Groves et al., 2016; Grove-White et al., 2000) through the implicit and explicit modes of intervention they promote and enact. Whether explicitly or implicitly, through bounding and structuring complex relationships imaginaries encode assumptions about the future that need scrutiny.

Informed by this body of thinking the paper aims to first, identify and examine the dominant environmental imaginaries within the field of environmental AMR research, and second, elucidate the implicit and explicit visions of the future encoded within them and in particular of how AMR and the environment might be addressed. In doing so, the paper seeks to make the following contributions to the nascent body of social science literature on AMR (see Palgrave Comms special issue Antibiosis; (Begemann et al., 2018; Brown and Nettleton, 2017; Landecker, 2015; Lee and Motzkau, 2012; Lorimer, 2017; Morris et al., 2016)) and its environmental dimensions. The first contribution is empirical, drawing attention to the environmental sciences of AMR and how its constituent scientists make sense of the environment both as key to their knowledge producing practices but also what this implies for the actions that need to be taken to address the problem of AMR. To date, no published social science work has engaged with how scientists make sense of AMR in the environment. The second contribution is conceptual as the paper seeks to develop the existing literature on imaginaries by bringing together the spatial and environmental emphases and concerns of geographers and the temporal emphases, specifically the future visions, of STS insights into imaginaries. Third, we seek to highlight opportunities for social science engagement and contributions to this emerging field.

The rest of the paper is structured as follows. In the next section, we discuss the literature on imaginaries from human geography and STS. The qualitative methods employed to produce our empirical material are then detailed. We then unpack four dominant environmental imaginaries in the environmental sciences of AMR, each of which draws attention to and aims to bound the environment and its risk in different ways. Despite these differences, all have in common a tendency to encode a future that assumes existing infrastructures of antimicrobial use and pollution will endure and that the risks of and vulnerabilities created by environmental AMR can be located and contained – to protect human health – by suitable end-pipe management interventions. We end by emphasising how social science can make contributions to this emerging field through attention to the social, political, economic and historical, as well as ecological, microbial and material contingencies configuring these environmental imaginaries and implicated in producing different vulnerabilities to the environmental dimensions of AMR. We elaborate these arguments in the final section.

A multitude of imaginaries

There is a growing literature examining imaginaries in diverse contexts, including work on social imaginaries, socio-technical imaginaries, environmental imaginaries and scientific imaginaries. However, there are shared themes which we carry forward in our own understanding. A key turn in work on imaginaries centres on a move from a mentalist notion of imagination as fantasy and fiction, to imagination as performative, entwined with material worlds, organised work and practices (Jasanoff, 2015; Watkins, 2015). Imaginaries are shared and collective endeavours, tying groups together through shared values, ways of knowing and understandings of social and material order, including how it was, is and ought to be. Imaginaries do not exist in isolation (from other imaginaries and forms of representation) or are uncontested, but rather are co-produced through ongoing interaction between cultural, political, scientific and material worlds and resistances to dominant understandings. Our use of the concept of imaginary designates a space of material and speculative interaction that shapes scientific practices and creates room for scientific action in society. The literature on imaginaries serves to interrogate these collective forms, representations and practices, how they emerge and shape material practices.

Due to our interest in science, our use of imaginaries sits alongside a broader body of work on scientific representations which shares a similar conceptual space and similar properties of being collective and systemic including work on framing (Morris et al., 2016), metaphors (Larson et al., 2005; Nerlich and James, 2008) or narratives (Kamenshchikova et al., 2018). However, these allied concepts tend to focus on language and communication giving more limited attention to action and materialisation. Equally, frame and metaphor analysis emphasises the strategic, instrumental and self-conscious use of language (Morris et al., 2016), whereas imaginaries share but are less defined by these elements. Instead they attend to the more ambiguous role of imagination in producing shared representations structuring material action and practices (Rhee, 2018). Given the diverse literature on imaginaries, we will now clarify our conceptual approach drawing together work on environmental imaginaries from cultural geography, and STS work on the future-orientated aspects of imaginaries.

The concept of ‘environmental imaginaries’ was introduced by Peet and Watts (1996) to characterise ‘regional discursive formations’ of nature, which they describe as co-produced through the interaction of place-specific social, institutional and political factors and the character of natural contexts themselves (McGregor, 2004; Peet and Watts, 1996). In doing so, they extend work on social imaginaries, specifically that of Castoriadis (1994), by elucidating the ways in which societies endow the natural world with significance and how the natural world shapes social imaginaries. Both their work, and that of McGregor (2004) and Davis (2011) position environmental imaginaries as representing ‘constellation[s] of ideas that groups of humans develop about a given landscape, usually local or regional, that commonly includes assessments about that environment as well as how it came to be in its current state’ (Davis, 2011: 16). Environmental imaginaries therefore encode assumptions about how an environment was, is and should be intermingling the past, present and future.

Work on ‘environmental imaginaries’ is situated by Watkins (2015) within a broader umbrella category of ‘spatial imaginaries’ that encompasses a range of scholarship examining socially held worldviews, representations and ways of talking about place and space developed within human geography. In doing so, three types of spatial imaginaries are distinguished: imaginaries of place (such as ‘Detroit’), idealised spaces (such as ‘global city’ or ‘ghetto’) and spatial transformations (such as ‘deindustrialisation’ or ‘globalisation’). This typology usefully draws attention to the multiple scales at which spatial imaginaries can operate, for instance representing an urban district, a city, a nation state or a broader geo-political region. In looking at how environmental interventions are imagined in response to AMR, such spatial differences are potentially important as is the relationship between different spatial imaginaries.

Research on spatial imaginaries, however, largely neglects the future (Watkins, 2015). Where the future does figure it is largely rendered in a linear form. In other words, the future is either represented as a blank that is completely separate from the present or the future is a telos towards which the present is heading (Anderson, 2010). Some work within human geography and particularly in STS has explicitly engaged with imaginaries as performative discourses with significant normative and futuristic dimensions (Jasanoff, 2015; Jasanoff and Kim, 2009; Martin and Simon, 2008). STS literature engages with imaginaries as more than representations, but as materially embodied ways of acting and doing (Watkins, 2015). Imaginaries therefore co-constitute material action and are thus performances of ‘the future’ in the present (Martin and Simon, 2008). Crucially by recognising the performative, future-making capacity of imaginaries, space is created to consider the explicit and implicit configurations of regulatory forms, political, market and social relations, alongside the technological innovations and material infrastructures that accompany their (re)production. The purpose of such analysis is not simply to describe the futures encoded in the imaginaries but also to open up the possibility of assessing their desirability or considering alternative futures (Groves et al., 2016; Grove-White et al., 2000; Macnaghten and Szerszynski, 2013).

Building upon these conceptual insights we aim to consider both the spatial and temporal dimensions of the environmental imaginaries produced in AMR environmental sciences. In summary, we analyse environmental imaginaries as representing performative discourses that are collectively co-produced through situated material action and practices within specific contexts. The (re)production of these imaginaries through material action represents performances of the future in the present. Equally, in opening these imaginaries to interrogation at this relatively formative stage, there are opportunities to avoid becoming ‘locked-in’ (McNeil et al., 2017) to specific ways of bounding the environment that might preclude alternative ways of responding to AMR.

Methods

The work reported here forms one element of a larger programme of UK based research undertaken by embedded social scientists within a large multi-disciplinary, Natural Environment Research Council (NERC) funded project led by computational biologists and microbiologists investigating AMR and its environmental dimensions. This involved co-location with scientific team members enabling in-depth observation of the project’s development and research practices, formal interviews, recorded group discussions as well as numerous informal discussions over a period of two years. This embedding was vital in providing insights into scientific assumptions and ways of working that inform and influence the analysis of data gathered during the 2017 Environmental Dimensions of Antibiotic Resistance (EDAR) conference which we detail below. These interview data were intended to supplement core insights gathered from the day-to-day embedded research.

EDAR is a bi-annual international conference which, in 2017, took place over four days and presented a number of opportunities for data collection, including eight semi-structured interviews with a range of environmental microbiologists, observations of scientific presentations and other interactions with participants, and a questionnaire survey of participants which produced 15 returns representing a 13% return rate. Potential interview participants were identified through our knowledge of the field and discussion with project colleagues. In total, 18 scientists were contacted via email (12 prior to the conference and six post conference). The eight participants who agreed to be interviewed represent several prominent scientists involved in the field at present and includes scientists from the US, Australia, Germany, Sweden, China and the UK. Each interview lasted roughly 30 minutes. Six interviews were conducted during the conference with a further two being concluded shortly afterwards via Skype. These scientists were from diverse backgrounds including microbial ecologists, marine biologists, molecular biologists, toxicologists, epidemiologists and microbial bioremediation. Disciplinary backgrounds are highlighted in selected quotes to provide additional context to who is speaking and the disciplinary nesting of their perspective. Despite their disciplinary differences, interviewees were broadly united by method, specifically in the use of genetic techniques, genome sequencing and metagenomics. Mathematical modellers and pharmaceutical chemists are also important contributors to this emergent field; however, they are not represented in the interview data nor to our knowledge in the questionnaire data, nor were they particularly prominent in the conference proceedings.

Interviews and the questionnaire were structured around four topic areas: (1) research background, (2) understandings of and approaches to measuring AMR as a phenomenon, (3) perceived important aspects and risks of the environmental dimension of AMR and (4) key messages for policy and future research areas. All interviews were audio recorded and transcribed. The questionnaire used open questions to produce qualitative data and invited respondents to annotate and complicate a diagram that represented AMR in the environment taken from the literature (see Supplemental Appendix A). The textual responses from the questionnaires were consolidated into a single document and analysed in the same way as the interview transcripts. Hand-written notes were made during conference observations detailing important themes and discussions resulting from presentation materials, poster presentations and numerous informal conversations, when feasible. All effort was made to transcribe these shorter notes into documents as soon as possible which allowed for further expansion. These different methods of data collection at EDAR produced a significant volume of textual material which, together with the data generated through working as embedded social scientists within an environmental AMR project, enable us to feel confident that our findings generate rich and meaningful insights into the issue of environmental AMR and specifically the imaginaries of the scientists constituting this field.

All the data collected, including notes from conference observations, interview audio and questionnaire responses which included visual information were transcribed, and treated as a text for the purposes of analysis. Once transcripts were available, coding was undertaken using the MAXQDA software package. The coding protocol included a top-down element which drew on the key themes introduced through the interview guide and anticipated as being present through pre- and post-interview preparatory work. It also included a bottom-up approach in that the semi-structured interviews created space for the interview to progress into unanticipated directions, further recurrent themes were identified as a result. These approaches were used to initially code the empirical material into high-level codes around topics such as ‘risk’, ‘interventions’ and ‘environmental descriptors’ and identify their nuances through sub-coding. The environmental descriptors code was a key starting point for the interpretive analysis presented here. However, the aim of the analysis was to draw together and identify the constellations of codes relating to the different ways in which the environment is being imagined, the consequences of those representations for scientific practice and the envisaging of environmental risks and the interventions to address those risks. We have identified four main environmental imaginaries which we will now outline.

Environmental imaginaries within the environmental sciences of AMR

Our empirical findings are organised as follows. We present four imaginaries evident in this field: ‘environmental hotspots’, ‘pristine environments’, the ‘environmental reservoir’ and the ‘fluid environment’. These environmental imaginaries reflect a spatial imaginary that is close to what Watkins (2015) describes as an ‘idealised space’. As certain spaces become idealised, they signify or come to be imagined as possessing distinctive universal characteristics. In our case, the four imaginaries constitute a shared repertoire guiding scientific practices to certain tested places, material and genetic objects that personify some characteristics imagined to be significant over others. Each imaginary aims to locate sites and spaces in which different environmental AMR risks can be isolated, quantified and thus assessed. Although each was supported and opposed to varying degrees, the dynamics of ‘resistance’ took the form of an ongoing process of critique, including from otherwise supportive scientists, that perhaps reflects the ongoing evolution of these imaginaries in this emerging field.

Our analysis differs from much of the wider imaginaries literature that attempts to present a unified means of representation. Instead, the environmental imaginaries we identify do not represent a singular or cohesive means of imagining the environment. Rather what we describe here represents four distinct but interlinked imaginary repertoires, each of which is entangled in producing different types of scientific practice. They individually represent sometimes quite different means of conceptualising the environment, emphasising different perceived qualities of environmental spaces and different spatialities. Two of the imaginaries, the ‘environmental reservoir’ and the ‘fluid environment’, represent overarching visions of environmental qualities across space, whereas the ‘pristine environment’ and ‘environmental hotspot’ are rooted more closely to specific places representing idealised characteristics. The four imaginaries we have identified and detailed are summarised in Table 1.

Table 1.

Summary of the four imaginaries and their key components.

	Hotspot	Pristine	Reservoir	Fluid
Type of spatial imaginary	Idealised space	Idealised space	Idealised space	Idealised space
Characteristics	Environment is a generative space. Hotspots are the sites of greatest concentration of bacteria and antimicrobials.	Environmental pollution is uneven. Pristine places are free of antimicrobial pollution.	Environment contains a distinct genetic community.	Environment is dynamic, physically and genetically fluid.
Spatiality	Place	Place	Macro scale	Macro scale
Significance for scientific practice	Identifies sites to be tested: wastewater treatment plants, polluted rivers, animal manures and slurries, soil amended with sludge and animal waste.	Identifies sites to be tested to establish a ‘baseline’ for comparison: Wilderness or uninhabited spaces, archive soils, control soils, upstream river sources.	Justifies and guides genetic prospecting to describe this genetic community.	Identifies objects to sample (water) and genetic material of most significance (mobile elements).
Risks	Environment is a space in which selection and transmission enable the accumulation of AMR genes (ARG) and bacteria.	Environment is an originator of ARGs independent of anthropogenic sources.	Environment is an originator of ARGs that may be a latent or nascent threat. It may incubate ARGs from clinical settings resulting in re-exposure.	Environment connects us and enables the spread of AMR bacteria and genes.
Tensions	Ambiguous results from tested places imagined to meet the idealised characteristics.	Different visions as to the extent of past pollution and its implications for finding sites to test.	Different visions as to the nature of the reservoir.	Temporal tensions between continuous material and genetic flows and slower or periodic cycles.
Interventions	Interventions to disrupt hotspot dynamics: provision of new or improved sanitation and wastewater treatment plants, manure composting, longer slurry storage.	None immediately evident.	Research funding to support further genetic prospecting and surveillance of reservoir. Interventions that prevent mobilisation of the reservoir (same as hotspot).	Interventions to disrupt and reduce the flow of antimicrobials, resistant bacteria and genes (same as hotspot).

AMR: antimicrobial resistance.

Environmental hotspots

If microbial ecosystems and antimicrobial pollutants are distributed unevenly across the environment, then the most significant places for generating the selection, accumulation and transmission of AMR genes and bacteria are anticipated as being the places in which they are both expected to be highly concentrated, where potential pathogenic interactions are most ‘dense’ (Brown and Kelly, 2014). AMR environmental scientists routinely describe these as environmental hotspots. Hotspots are idealised spaces in which the dynamics of AMR selection and transmission in the environment, due to present and past pollution, can be made visible through science. In this context, the environment is imagined not just as a recipient of antimicrobial pollution and resistant genes, but as a generative space, or reactor, actively contributing to the increased prevalence of AMR genes and bacteria.

The hotspot as a conceptualisation of environmental pollution is hardly unique to the scientific field of AMR and has a long lineage. Although not made explicit in interviews it draws on well documented disease and pollution dynamics, which emphasise areas of high concentrations and interactions which are correlated with increased negative impacts on certain populations (high disease incidence or carcinogenic effects for example). Its replication in this context is unsurprising, given that antibiotics are characterised as environmental pollutants and the backgrounds of some of the scientists in toxicology and bioremediation. However, unlike its use in epidemiology the environmental hotspot imaginary is not necessarily about the presence of specific pathogens entangled with human and animal life, but abundances of microbial life more broadly.

The hotspot is also motivated by an implicit assumption that where there might be high concentrations of antimicrobials and resistance bacteria and genes, the relationship might be inferred as causal, even though the research itself is only able to establish associative or correlative relationships. The performative work of this imaginary was in identifying highest priority spaces from which to start scientific analysis as the following quote expresses:

Scientist 3 (Environmental biotechnology): ‘I think we can start with some hotspots, so called hotspots, like the agriculture [waste], the hospital [waste], the pharmaceutical factory [waste], and the wastewater treatment plant’.

Amongst these ‘hotspots’, the most prominent was the wastewater treatment plant, followed by polluted rivers and waterways, particularly those downstream of hospitals and pharmaceutical plants, liquid animal slurries and soil amended with human and animal waste. The prominence of wastewater treatment plants as hotspots was linked to several other factors, most notably the quantities of wastewater discharge.

Scientist 8 (Aquatic microbiology): ‘in the UK we discharge 11 billion of waste a day into our rivers, … each of those litres has 1 million or 10 million or 100 million bacteria in there and … you have also got this complex mixture of chemicals so you’ve got all of your metals and your biocides and your antibiotic residues … ’.

This same interviewee contrasted the constant flow of wastewater with manure applied to soil which, because of the periodic nature of manure spreading, was not deemed as significant because the ‘hotspot’ was assumed to dissipate between each application of manure. Additionally, wastewater treatment plants provide a conveniently bounded passage point at which diverse chemical and bacterial inputs from a variety of sources within a catchment are contained and mixed for a period of treatment. An otherwise dynamic environment of flowing wastewater is stabilised for a time within the spaces of the wastewater treatment plant where it becomes an inflow sample, able to be compared with an outflow sample. As a result, they were particularly prominent in the discussion of hotspots.

However, the environmental hotspot was not limited to this relatively small-scale site. The global South was also positioned as a problem space:

Scientist 2 (Microbial ecology): ‘the treatment is actually pretty good [in Europe] … it is really countries that have very poor treatment or no treatment at all which … applies to large parts of India, large parts of Africa, parts of South America … ’.

Scientist 7 (Microbial ecology): ‘I would say that the risk pathways are maximised in countries that are still developing, low income, … when you are thinking about environmental drivers of AMR don’t think western, northern Europe, thinking sub-Saharan, India … ’.

The status of this diverse region as a hotspot for AMR selection and transmission was a consistent theme across the conference and is also evident within the wider academic and policy literature (Grace, 2015; Zellweger et al., 2017). The global South as a hotspot was linked to certain perceptions of social, economic and political factors across this diverse geo-political region. Lax controls on antibiotic sales, poorly regulated pharmaceutical production, and poor to non-existent wastewater treatment and sanitation infrastructure, were represented as enabling the ‘hotspot’ to break the bounds of a particular river or waste stream and come to represent an otherwise large and diverse geo-political region.

The interventions most associated with the environmental hotspots aimed to disrupt the processes of antimicrobial concentration and the capacity of a hotspot to generate increased AMR prevalence. Specific interventions deemed to meet these aims included the provision of wastewater treatment and sanitation where they were absent and improvement in existing infrastructures to remove antimicrobials and reduce AMR bacteria and genes in wastewater streams. Also identified was manure composting and longer timeframes for storing liquid slurries prior to application to soil to achieve the same.

Over the course of the conference, the notion of certain places as hotspots was challenged by some scientists due to conflicting empirical results. Rather than confirming the provenance of a hotspot, the presentation of empirical findings from imagined ‘hotspots’ often complicated such ready characterisation with ambiguous results (Caucci et al., 2016; Flach et al., 2018) and the presence of confounding factors (Bengtsson-Palme et al., 2016). Whereas other work has been more explicit, arguing that wastewater treatment plants are not necessarily hotspots (Quintela-Baluja et al., 2019).

These challenges, however, did not destabilise the dominance of the hotspot imaginary as it does valuable work justifying and directing analysis to certain sites assumed to contain the highest risks. Equally, there was no clear effort to develop alternative conceptualisations of how and where interactions between bacteria and antibiotics in the environment might lead to significant selection and transmission of AMR genes and bacteria. For example, the exposure of bacterial communities to low-level concentrations over time was not being foregrounded as an alternative, despite the potential for such concentrations to enrich resistant populations over time (Gullberg et al., 2011). This raises the question: what would disrupt the hotspot’s dominance, assuming ambiguity persists, and what implications would its decentring have for the practices of environmental AMR sciences?

Pristine environments

Environmental bacteria were routinely situated by environmental scientists as the evolutionary origin of many known resistant genes circulating in human pathogens. The presence of resistance genes in the environment has been shown to be widespread (albeit often at low levels). Given that antibiotics were originally derived from chemical compounds produced by environmental bacteria, it is unsurprising that such bacteria carry the mechanisms to detect and protect themselves from their more toxic effects. In some situations, this may be seen as a relief.

Scientist 2 (Microbial ecology): ‘a lot of the resistant mechanisms are there in natural communities … maybe we shouldn’t be scared about finding it in the environment but we should be scared about finding a lot of it’.

However, the extensive presence of naturally occurring resistant genes produces something of a problem for environmental microbiologists.

Scientist 7 (Microbial ecology): ‘if we don’t know what the background should have looked like we will never be able to say what an elevated level is and so it’s all relative’.

Comparators are key to being able to untangle the resistance baseline from the human produced effects (past and present).

‘Pristine’ environments are therefore imagined as site that can function as a comparator because they are sufficiently untouched by antimicrobial pollution. Examples of pristine environments include river sources distant from downstream pollution, and controlled soils, particularly those that have not been amended with manures containing antibiotics.

Scientist 4 (Environmental microbiology): ‘[name] River as an ideal model for looking at the pristine to urban and agricultural anthropogenic input gradient’.

However, the pristine environment is not just imagined in terms of its spatial distance from anthropogenic effects it is also a matter of distance in time.

Scientist 8 (Microbiology): ‘or we have looked at some of these park grass[es] at [research station name] that has been untouched since 1840 or something you know you can try to look in semi-pristine sites’.

Certain archived soils stored prior to the start of antibiotic use on farms were particularly attractive, promising to offer a glimpse of bacterial communities prior to the amendment of soils with antibiotic residues in animal manures. In each case becoming ‘pristine’ involves a moving away, either spatially or temporally, from human impacts associated with antibiotic use.

In contrast to the other three imaginaries we detail in this paper, the pristine environment does not appear to be associated with any specific interventions. Although the imaginary carries the hallmarks of longstanding pollution control and remediation logics that aspire to return environments to states of imagined material purity, scientists did not express this as a goal of potential interventions. Instead the pristine imaginary was related to scientific practice and data interpretation, specifically how to establish proper comparators and identify control or blank samples.

The ability to claim any environmental place or samples as pristine was routinely contested on the basis that environments free of human impacts were at best few and far between.

Scientist 1 (Molecular biology): ‘ … there is no such thing as a pristine environment on Earth’.

All environments, it was claimed by these dissenters, have been changed in some way as a result of human activity and pollution, a point that echoes work familiar to social science on the social and economic production of nature (Castree and Braun, 2001) as well as being a core contention of defining the current geological era – the Anthropocene (Steffen et al., 2007). Notably, Scientist 1 explicitly connected AMR to the Anthropocene when making a similar point during a conference presentation. Distance upstream of, or away from polluting sites was therefore insufficient given the pervasive distribution of chemicals across the face of the earth.

The ‘natural’ quality of other supposedly pristine sites or materials was also contested. This was most clearly expressed in relation to archival soils:

Scientist 8 (Microbiology): ‘Well people have tried to do it in archive soils … but you know I am a bit suspicious of those air-dried soil samples that are 50 years old’.

Such suspicion foregrounds how archival soils are made ‘pristine’ only through human interventions that fundamentally change the conditions in which the soil and its microbial community exist in preserving them. The same applies to ‘untouched’ parkland grasses mentioned by the same scientist in relation to their work, which are also the result of explicit human decisions and interventions.

In contrast to the hotspot which was re-established in new places in response to ambiguous results, the pristine imaginary faces a more fundamental challenge to its providence, namely the difficulty of finding a pristine place in a polluted world. Even where a pristine environment could be found, doubts arose about the consequences of shifting spatially, temporally and contextually to produce samples unimpacted by anthropogenic effects.

The environmental reservoir

The environmental reservoir, like the pristine imaginary, is rooted in the recognition of environmental bacteria as the evolutionary origin of resistant genes, including those circulating in human pathogens. These environmental resistance or resilience genes (a distinction to which we will return to later in this section) were explicitly positioned as constituting a reservoir of resistance determinants that had been or could be shared amongst bacterial ecosystems and human pathogens (D’Costa et al., 2006; Forsberg et al., 2012; Woolhouse et al., 2015). Unlike the more site-specific nature of the hotspot and the pristine imaginary, the environmental reservoir represents an overarching vision of the environment as a space containing naturally occurring pool of resistance genes. Consequently, hotspots are also reservoirs, while analysis of pristine environments aims to reveal the ‘natural’ levels of the reservoir. Unlike the active, generative potential of the hotspot, the reservoir is a latent environmental threat existing independent of human interventions. It therefore works to justify the significance of the environment in terms of the (unknown future and present) risk it poses to human health.

There are two slightly different dimensions to this reservoir. First, and most dominant, the environmental reservoir is imagined as containing a distinct community of environmental resistance genes or a ‘resistome’, which may be multiplied and mobilised, especially when these communities were exposed to antimicrobial pollution, into human pathogens (Gaze et al., 2013). The major risk posed by the reservoir is that it represents a nascent or latent environmental hazard consisting of unknown resistance genes that at any point could be mobilised into human pathogens. The environment is therefore a space from which future AMR threats will emerge and thus needing scientific attention to make visible and characterise this unknown threat.

Second, it suggests that the environment may act as a space of ‘holding’ or ‘collection’ for resistant genes and bacteria which occur in clinical settings but move into the environment when excreted in human and animal waste. The environment, in this way of thinking, effectively acts as an incubator before re-perpetuating human health problems.

Scientist 7 (Microbial ecologist): ‘is wildlife a reservoir of resistance that will just re-perpetuate the problem that we currently have’.

This is a conceptualisation akin to certain representations of infectious zoonotic disease reservoirs in epidemiology whereby certain populations in which pathogens naturally circulate act as a reservoir from which other populations can become infected. The main difference between the two relates to the direction of transmission of resistant genes into and out of the ‘reservoir’. Subsequently, certain genetic elements, particularly plasmids, with their capacity to transfer resistant genes between bacteria, are key objects of scientific analysis. Their mobility is key to allowing the sharing of resistance genes between non-pathogenic environmental bacterial and human pathogens (Forsberg et al., 2012).

The need to make visible this latent genetic threat justifies the practice of genetic prospecting.

Scientist 2 (Microbial ecologist): ‘a lot of the resistant mechanisms are there in natural communities and that is actually quite an important first step to just describe that’.

This practice uses genomic techniques to characterise this reservoir, in turn solving the ‘problem’ that this genetic reservoir is largely unknown at present, both with regards to the prevalence of known clinically significant resistant genes in the environment, and the presence of unknown resistance mechanisms. These data limitations are in part linked to the inability of traditional methods of agar culturing to make visible many environmental bacteria, but also the previous expense of genetic techniques. Therefore, the future interventions associated with this imaginary are primarily focused on the need for research investments to continue genetic prospecting, alongside innovating new and improved genomic technologies and techniques to enable the complete characterisation and ongoing surveillance of the environmental reservoir and the genetic risks it harbours.

Such knowledge is positioned as enabling us to anticipate and identify the spaces where interventions can be made to mitigate the possibility that resistant genes will be mobilised from the environmental bacteria into human pathogens, as well as reduce the prevalence of known resistance genes. Given the overlap with the hotspots mentioned previously, these interventions were the same as those associated with the environmental hotspots. Specifically, they entailed the provision of new or improved wastewater treatment and sanitation infrastructures, manure composting and storage of liquid slurries. All of these interventions are envisaged as minimising the possibility that genes from the environmental reservoir might be transferred into pathogenic bacteria.

However, this conceptualisation of the environment as a reservoir for AMR genes, and the way in which it represented the community of environmental genes and bacteria, was disputed by one scientist.

Scientist 1 (Molecular biologist): ‘The other word that colours that thinking is reservoir. Which is also used a lot and that word implies a great big storage vat … where resistance genes kind of accumulate and sit quiescent waiting to pounce on us and that’s completely wrong as well, those genes are not a reservoir, they are not in a reservoir … they have a function in the natural world, that function is not to provide protection against the concentrations of antimicrobials that humans use’.

This quote from a molecular biologist rejects the notion of the environment as a store which implies both stasis and containment and instead emphasises a more active, more-than-human vision as to what these genes constitute. With regards to this latter point, the distinction between resilience and resistance genes was important as it similarly challenged the otherwise anthropocentric character of the ‘reservoir’. While no other interview participants nor participants in the conference singled out the reservoir in such explicit terms as outlined above, there were repeated instances where AMR genes were instead defined as environmental resilience genes. This subtle shift in emphasis works to decentre the human and implicitly draws attention to the broader function of these genes for the bacteria that carry them. Regardless of the nuances of its composition, this shift does not challenge the basic premise that the environment is a space in which genes with implications for clinical treatment might be mobilised into pathogens. Indeed, the broader epidemiological analogy remained unchallenged as eventual impacts on human health are central to the significance of the environmental dimension.

The fluid environment

The environment was described by the scientists as being significant because it was a key space through which antimicrobials, resistant pathogens and genes were being dispersed (Larsson et al., 2018). Equally, the importance of physical movement was encoded in scientific practices where the dominant emphasis was on physical objects (liquids) and genetic objects (mobile genetic elements). The environment as a material and genetic space is therefore imagined in more dynamic terms than the reservoir imaginary might initially suggest. We describe this imaginary as the fluid environment.

Water has become a particularly significant material sampled by environmental scientists as a result of its physical fluidity, its capacity to carry pollutants, simultaneously making them available for interaction, the historical importance of water within toxicological research and it being an essential component of life.

Scientist 1 (Molecular biology):‘ … water is the thing that binds, agriculture, soil, oceans, humans together and so flowing water and the things that it carries is the key. … that together with the natural kind of tendency of humans to put all sorts of things that they shouldn’t in waste water’.

Scientist 4 (Environmental microbiology):‘ … to me water makes sense it’s such an essential part of our lives whether we are drinking it or bathing in it or using it to prepare our food and grow our crops’.

Therefore, the significance of water emerges from a range of qualities imagined to be especially important: that it binds us by flowing between, through and into bodies and environments; and all the while carrying the microbes and antimicrobial chemicals that we release into it through our practices of waste disposal. It thus functions as a dynamic container mixing the elements perceived to be necessary to select resistance genes and bacteria, while its fluidity and flow move them both into proximity with humans and animals and potentially significant distances.

Water is juxtaposed with environmental elements considered less dynamic, ‘slow’ or ‘episodic’ which are in turn marginalised. Soil, for instance, was singled out in this way while elements like air were situated as carrying insufficient bacterial life to be significant. The impact on scientific practice is that liquids and liquid systems, in particular wastewater and polluted rivers, were positioned as priorities for scientific attention over what might be deemed more ‘static’ systems.

Yet, while none denied the significance of water, the imagined insignificance of soil AMR was challenged by two scientists who positioned it as indeed being directly connected to humans through food.

Scientist 5 (Molecular microbial ecology): ‘our recent research on the fresh produce (packaged lettuce) showing that … there can be readily a link from the soil to the gut microbiome’.

While the physical properties of water are what make it significant, it is genetic fluidity which is important in situating soil as important as it ensures that the boundaries between soil and the gut microbiome are more fluid than the physical properties of soil might suggest. Furthermore, unlike water, people not imagined as being directly and routinely exposed to soils in the same routine ways. However, certain food production and processing chains create connections between soil and humans. Hence the emphasis by Scientist 5 on fresh produce has additional significance as it is not cooked before eating and thus carriage of bacteria and genes from soil to humans was not interrupted by other practices. The cultivation of crops in soils fertilised with animal manures and human sewage sludge has been another space in which concern around soil has emerged principally in relation to the transmission of AMR pathogens from human excreta onto plants (Rahube et al., 2014) – a longstanding food safety concern.

This genetic fluidity is also key to emphasising the broader significance (and threat) arising from the environmental reservoir outlined in the previous section.

Scientist 3 (Environmental biotechnology): ‘[a] major concern [is] the mobile genetic elements [plasmids], because they can … shuttle between the pathogen and … the environmental non-pathogenic bacteria’.

In both cases, soil and the environmental reservoir, it is not necessarily the physical fluidity of the environment but the genetic fluidity of bacteria that is important. This enables scientific practice to expand to more environmental objects and systems.

For others, the significance of soil came about exactly because of this ‘slowness’, positioned as perhaps facilitating the sedimentation or ‘sequestration’ of certain antibiotics rendering them unavailable for interaction with bacteria, or enabling their degradation. This is not to suggest that sedimentation is a permanent or homogeneous process. Rather, it gestures towards the latent potential of antimicrobial chemicals (Čermák et al., 2008), and the threat of future disruption reactivating the significance of the past, akin to how 19th century CO₂ emissions are understood to be present in the climate of today and into the future.

Furthermore, soil with its different temporal management practices emphasises different cycles of life, for example recurrent blooms caused by manure applications. In this context, an initial flourishing or addition of resistant bacteria diminishes over time as soil bacterial communities return to pre-application states (Heuer et al., 2011) although perhaps subtly changed (D’Alessio et al., 2019). Equally, it has long been noted that once established resistant genes seldom completely disappear, and this residuum can rebound rapidly (Salyers and Amábile-Cuevas, 1997). Or more significantly, soil might prevent the onward passage of resistant bacteria and genes from manure into water systems (Muurinen et al., 2017) perhaps acting similarly to reed beds that filter out pollutants from water and highlighting the capacity to co-opt environmental systems to reduce the risk from AMR. However, these are uneven processes, as the heterogeneity of soil impacts the processes of sedimentation, subsequent bioavailability of antimicrobials, and blooms in resistant bacteria and their persistence (Čermák et al., 2008; D’Alessio et al., 2019; Heuer et al., 2011).

Key to the fluid environments imaginary is the collapse of spatial and temporal distance through an emphasis on connection that is most readily embodied by water and mobile genetic elements. The interventions associated with this imaginary include the provision and improvement of wastewater treatment infrastructures, and manure composting or slurry storage which disrupt the ready and onward flow of antimicrobials and resistant bacteria into and through water and soil. Yet the open-endedness of these fluid processes gestures towards another future, one in which the futility of containing or disrupting these dynamics is exposed. Although both place an emphasis on a need to survey and characterise the prevalence of resistance, the open-endedness and connectivity of the fluid environments imaginary contrast with the environmental hotspots and the reservoir’s gesture towards the possibility of a (human-made intellectual) closure of the environment.

Unpacking embedded futures

The analysis presented so far has engaged principally with the spatial dimensions of different scientific representations of the environment and their consequences for scientific practice and interventions to address the risks posed by the environmental dimensions of AMR. In doing so, we have highlighted four imaginaries all sharing a concern with locating the threat of environmental AMR in places and spaces defined by transmission, contamination or conversely its absence. In this section, we move from the spatial to the temporal and examine the futures explicitly and implicitly embedded in the environmental imaginaries we have detailed.

With environmental AMR primarily imagined as being located in waste streams, the future that is most prominently invoked envisages the development of existing and new socio-material infrastructures for waste management, such as new and improved wastewater treatment plants, slurry storage and composting. These solutions are positioned as offering the capacity to ameliorate the generation and accumulation of resistance genes and bacteria resulting from antimicrobial pollution and their transmission through and within the environment. Associated with these changes is the need to monitor certain environmental objects, in particular water but perhaps also soil, adding additional layers of testing and managerial requirements around antimicrobial pollutants which may extend beyond the antimicrobial chemicals themselves to include genetic elements (Gillings et al., 2015; Li et al., 2018). Whether the goal is to return such environments to pristine baselines is less clear.

Emphasising opportunities for control and containment, this embedded future associated with environmental hotspots and reservoirs is arguably in tension with the more radical implications of what we have called the fluid environmental imaginary of AMR. By emphasising the intrinsic fluidity of environmental systems and genetic material, the fluid imaginary evokes flows and connectivity. The fluid imaginary, through emphasising connectivity due to the physical flow of water and genes, also provides legitimacy to descriptive scientific analysis that does not link directly to present clinical concerns. It also draws attention to the global connectivity of the problem. What happens in the ‘hotspot’ of the global South has implications for the global North. But the fluid imaginary potentially goes further than this, emphasising environmental flux as both an intrinsic genetic and physical property over time and drawing into question claims that this can ever be a containable, controllable and predictable phenomenon.

Some scientists also acknowledged a different set of processes and temporal rhythms, notably, slowness and sedimentation. While the register of slowness draws attention to the longer timeframes over which change takes place and environmental processes operate, sedimentation and the environmental reservoir bring into play the more troubling implications of latency. ‘To be latent is to be not yet: a potential not yet manifest, a past not yet felt’ (Murphy, 2013: 1). Through latency the future is already altered, further questioning the capacity to control and contain AMR in a world already marked by an extensive profusion of antimicrobial chemicals and the effect of their traces. Furthermore, imaginaries such as the hotspot follow established environmental and human health wisdoms where higher concentrations are understood as being highest risk, whereas experiences with Asbestos and Sick Building Syndrome suggest exposures to minute quantities can also have highly toxic consequences impacting human health (Murphy, 2006). This suggests that AMR might represent a slow disaster (Littmann and Viens, 2015), rather than an acute form of toxic violence, a characteristic that has been noted as presenting numerous challenges for how societies represent and respond to them (Fortun et al., 2016; Liboiron et al., 2018; Nixon, 2011).

Furthermore, the absence of a catastrophic imaginary akin to those associated with biodiversity loss or climate change suggests that the environment itself is not considered at risk from antimicrobial pollution and rising AMR. Arguably this is because the environment is being positioned as a dimension of an otherwise human health problem. A major consequence is a limited engagement with what antimicrobial pollution might mean for the microbial ecosystems that it disturbs and to the functioning of which it is vital (Amábile-Cuevas, 2016). Considering environmental AMR in these ways potentially reframes the ‘environmental reservoir’. In contrast to the dominant representation, where the reservoir contains a pool of resistance genes posing a potential threat to human lives, the more-than-human qualities of these genes are brought to the fore. The reservoir instead becomes a latent future asset. The natural roles of these genes in enabling bacterial resilience in the presence of otherwise toxic chemicals considered a positive property of environmental bacteria allowing them to overcome the disturbances and perturbations resulting from antimicrobial pollution now and in the future. It may also present opportunities for utilising ecosystems and bacteria to mitigate AMR risks and contamination, similarly to existing efforts at bioremediation.

Linked to the above is the absence of futures that might denote altogether different socio-material infrastructures, what Watkins (2015) describes as imaginaries of spatial transformation. Although spatial transformations are evoked in relation to globalisation facilitating the transmission of resistant bacteria (and genes) across borders, altogether absent is the potential role of globalisation in producing, concentrating and dispersing environmental AMR. Existing waste management infrastructures, the exit point for antimicrobial materials produced and consumed by established systems of production and consumption, are envisaged to be the main site of intervention. Tacitly, the socio-economic, political, material and historical contingencies that configure the ‘input’ into these waste management infrastructures are expected to continue producing pollution that will flow into and through environments. To some degree this highlights the limitations of scientific knowledge based on locating AMR in certain flows, places and spaces. Through imagining different idealised spaces as the principal sites in which certain environmental AMR might be a problem, it configures certain types of solutions that are based on cleaning, containing or mitigating contamination and transmission in those places.

In the next and last section, we expand on this final theme through identifying opportunities for social science and humanities contributions to this emerging field that seeks to foreground the configurational nature of environmental AMR, its generation, transmission and the uneven vulnerabilities to its effects.

Opening up the field of environmental AMR

The hotspot, reservoir and the fluid environment imaginaries reference sites and spaces with generative and transmission potential primed through contamination by antimicrobial pollutions, and in the case of the pristine environment their absence. These imaginaries mirror other representations circulating in the human health and AMR domain that seek to ‘locate’ risks and pathogenic potential within specific sites, places and spaces (Brown and Kelly, 2014; Craddock and Hinchliffe, 2015; Hinchliffe, 2015). The dominance of environmental scientists as the most prominent group within the research field of AMR and the environment means that AMR interventions that are understood as ‘environmental’ are heavily shaped by how these scientists imagine the environment. Some social science and humanities scholars would likely focus more attention on broader systemic processes implicated in producing (and addressing) the contamination of environments with antimicrobial chemicals and the generation and transmission of environmental AMR. The relatively few social science and humanities scholars within the field at the time of study may help to explain the lack of consideration of such processes within the imaginaries we have identified.

In conclusion, we seek to identify some possible questions for future social science engagement with the environmental dimension of AMR. A starting point for such a task are questions that elucidate the uneven geographies of AMR and the environment (Craddock and Hinchliffe, 2015). In particular, attention needs to be given to how social arrangements are entangled with changing material conditions, producing vulnerabilities for particular people and animals in particular places and times (Giraud et al., 2019). This is what Hinchliffe (2015) describes as the patchwork that creates (or not) the generative and transmission potential for AMR. Such approaches would aim to foreground the social, economic, ecological, political and historical contingencies configuring hotspots, reservoirs, fluidity and the pristine, and perhaps in doing so shifting these idealised spaces into new localities.

In turn, there is a need to explore alternative means of representing the dynamic and complex phenomena of AMR and its environmental dimensions through opening up existing concepts that inadequately reflect complex interaction between social, political, material and ecological dynamics. We have started that work here by examining some of the different environmental imaginaries circulating in this domain and which condition the means of understanding and responding to the ‘problem’ of the environmental dimension of AMR. However, there is a need for further analysis of the discourses that are being mobilised, their longer histories, cultural significance and specificity, as well as further examination of how these representations are co-producing diverse ways of knowing AMR amongst environmental scientists, practitioners and policy makers. Such analysis is crucial for enabling dialogue between alternative means of imagining the environment as a problem space in the context of AMR.

Our hope is that through exploring these and other social science questions possibilities are created for tying the generation, mobilisation and transmission of AMR genes and bacteria through and within the environment to human and animal vulnerabilities via different types of epistemic engagement and inter/multi-disciplinary approaches. Bringing other forms of expertise and experience to bear on the environmental dimension of AMR is therefore an important task within future work. In doing so, work on the research field of AMR and the environment is ‘opened up’ (Morris et al., 2019). That a move towards this is already happening is evidenced in UKRI’s AMR research programme which requires a multi-disciplinary approach in the projects it funds.

By elucidating the environmental imaginaries circulating in this field, we have begun the work of opening up this field and have proposed lines of social science interrogation which aim to make visible human, environmental and animal vulnerability, difference, unevenness and injustice, thereby broadening the capacity to apprehend and act on the environmental dimension of AMR.

Highlights

➤ The increased profile of the environmental dimension of AMR has created a new political and scientific space around AMR.

➤ We examine the environmental imaginaries produced by environmental scientists studying the environmental dynamics of AMR.

➤ We detail four different imaginaries: the hotspot, reservoir, pristine and fluid environment.

➤ These spatially dynamic imaginaries anticipate limited future change restricted to ‘end of pipe’ interventions that leave systemic relations undisturbed.

➤ Bringing other forms of expertise to bear is necessary for expanding this spatial and temporal imaginary repertoire.

Supplemental Material

sj-pdf-1-ene-10.1177_2514848620950752 - Supplemental material for Environmental imaginaries and the environmental sciences of antimicrobial resistance

Supplemental material, sj-pdf-1-ene-10.1177_2514848620950752 for Environmental imaginaries and the environmental sciences of antimicrobial resistance by Richard Helliwell, Sujatha Raman and Carol Morris in Environment and Planning E: Nature and Space

Footnotes

Acknowledgements

The authors would like to thank Prof. Dov Stekel and Dr Jon Hobman for their valuable comments on an earlier draft of the paper.

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the NERC (grant number NE/N01M9881/1).

ORCID iD

Richard Helliwell

Note

References

Amábile-Cuevas

(2016) Antibiotics and Antibiotic Resistance in the Environment. London: CRC Press.

Anderson

(2010) Preemption, precaution, preparedness: Anticipatory action and future geographies. Progress in Human Geography 34(6): 777–798.

Begemann

Perkins

Van Hoyweghen

, et al. (2018) How political cultures produce different antibiotic policies in agriculture: A historical comparative case study between the United Kingdom and Sweden. Sociologia Ruralis 58(4): 765–785.

Bengtsson-Palme

Hammarén

Pal

, et al. (2016) Elucidating selection processes for antibiotic resistance in sewage treatment plants using metagenomics. Science of the Total Environment 572: 697–712.

Bengtsson-Palme

Kristiansson

Larsson

DGJ

(2017) Environmental factors influencing the development and spread of antibiotic resistance. FEMS Microbiology Reviews 42(1): 68–80.

Brown

Kelly

(2014) Material proximities and hotspots: Toward an anthropology of viral hemorrhagic fevers. Medical Anthropology Quarterly 28(2): 280–303.

Brown

Nettleton

(2017) ‘There is worse to come’: The biopolitics of traumatism in antimicrobial resistance (AMR). The Sociological Review 65(3): 493–508.

Castoriadis

(1994) Radical imagination and the social instituting imaginary. In: G

Robinson

Rundell

(eds) Rethinking Imagination: Culture and Creativity. London: Routledge, pp.136–154.

Castree

Braun

(1998) The Construction of Nature and the Nature of Construction: Analytical and Political Tools for Building Survivable Futures. London: Routledge.

10.

Castree

Braun

(2001) Social Nature: Theory, Practice, and Politics. Oxford: Blackwell Publishers.

11.

Caucci

Karkman

Cacace

, et al. (2016) Seasonality of antibiotic prescriptions for outpatients and resistance genes in sewers and wastewater treatment plant outflow. FEMS Microbiology Ecology 92(5): fiw060.

12.

Čermák

Kopecký

Novotná

, et al. (2008) Bacterial communities of two contrasting soils reacted differently to lincomycin treatment. Applied Soil Ecology 40(2): 348–358.

13.

Chandler

CIR

(2019) Current accounts of antimicrobial resistance: Stabilisation, individualisation and antibiotics as infrastructure. Palgrave Communications 5(1): 53.

14.

Craddock

Hinchliffe

(2015) One world, one health? Social science engagements with the one health agenda. Social Science & Medicine 129: 1–4.

15.

D’Alessio

Durso

Miller

, et al. (2019) Environmental fate and microbial effects of monensin, lincomycin, and sulfamethazine residues in soil. Environmental Pollution 246: 60–68.

16.

Daniels

(2011) Geographical imagination. Boundary crossings. Transactions of the Institute of British Geographers 36: 182–187.

17.

Davis

(2011) Imperialism, orientalism, and the environment in the Middle East: History, policy, power, and practice. In: D

Davis

Burke

(eds) Environmental Imaginaries of the Middle East and North Africa. Athens, GA: Ohio University Press, pp.1–22.

18.

D’Costa

McGrann

Hughes

, et al. (2006) Sampling the antibiotic resistome. Science 311(5759): 374–377.

19.

Ding

G-C

Radl

Schloter-Hai

, et al. (2014) Dynamics of soil bacterial communities in response to repeated application of manure containing sulfadiazine. PLoS One 9(3): e92958.

20.

European Commission (2017) A European One Health Action Plan Against Antimicrobial Resistance (AMR). Luxembourg: Office for Official Publications of the European Communities. Available at; (accessed February 2020).

21.

World Health Organization, Food and Agriculture Organization of the United Nations and World Organisation for Animal Health (2016) Antimicrobial resistance: A manual for developing national action plans. Version 1. Geneva: World Health Organization.

, et al. (2018) A comprehensive screening of Escherichia coli isolates from Scandinavia’s largest sewage treatment plant indicates no selection for antibiotic resistance. Environmental Science & Technology 52(19): 11419–11428.

, et al. (2012) The shared antibiotic resistome of soil bacteria and human pathogens. Science 337(6098): 1107–1111.

24.

Fortun

(2005) Scientific imaginaries and ethical plateaus in contemporary U.S. toxicology. American Anthropologist 107(1): 43–54.

, et al. (2016) Researching disaster from an STS perspective. In: Felt U, Fouché R, Miller CA, Smith-Doerr Let al. (eds) The Handbook of Science and Technology Studies. Cambridge: MIT Press, pp.1003–1028.

, et al. (2013) Influence of humans on evolution and mobilization of environmental antibiotic resistome. Emerging Infectious Diseases 19(7): e120871.

, et al. (2015) Using the class 1 integron-integrase gene as a proxy for anthropogenic pollution. The ISME Journal 9(6): 1269–1279.

, et al. (2019) Abundance in the Anthropocene. The Sociological Review 67(2): 357–373.

29.

Gothwal

Shashidhar

(2015) Antibiotic pollution in the environment: A review. CLEAN – Soil, Air, Water 43(4): 479–489.

30.

Grace

(2015) Review of evidence on antimicrobial resistance and animal agriculture in developing countries. Nairobi, Kenya: International Livestock Research Institute. Available at: DOI: 10.12774/eod_cr.june2015.graced (accessed February 2020).

, et al. (2018) Modeling the emergence of antibiotic resistance in the environment: An analytical solution for the minimum selection concentration. Antimicrobial Agents and Chemotherapy 62(3): e01686–e01617.

, et al. (2016) The grit in the oyster: Using energy biographies to question socio-technical imaginaries of ‘smartness’. Journal of Responsible Innovation 3(1): 4–25.

(2000) Wising up: The public and new technologies. Lancaster, UK: Centre for the Study of Environmental Change.

, et al. (2011) Selection of resistant bacteria at very low antibiotic concentrations. PLOS Pathogens 7(7): e1002158.

35.

Harvey

(2019) Antibiotic resistance as big a threat as climate change – Chief medic. The Guardian, 29th April.

(2011) Antibiotic resistance gene spread due to manure application on agricultural fields. Current Opinion in Microbiology 14(3): 236–243.

37.

Hinchliffe

(2015) More than one world, more than one health: Re-configuring interspecies health. Social Science & Medicine 129: 28–35.

38.

Holloway

(2011) Promoting the rational use of antibiotics. Regional Health Forum 15(1): 122–130.

(2010) Antibiotic prescribing in hospitals: A social and behavioural scientific approach. The Lancet Infectious Diseases 10(3): 167–175.

40.

Jasanoff

(2015) Future imperfect: Science, technology and the imaginations of modernity. In: S

Jasanoff

S-H

Kim

(eds) Dreamscapes of Modernity: Sociotechnical Imaginaries and the Fabrication of Power. London: The University of Chicago Press, pp.1–33.

(2009) Containing the atom: Sociotechnical imaginaries and nuclear power in the United States and South Korea. Minerva 47(2): 119.

, et al. (2018) Complex narratives of health, stigma and control: Antimicrobial resistance screening among non-hospitalized refugees. Social Science & Medicine 212: 43–49.

43.

Kümmerer

(2004) Resistance in the environment. Journal of Antimicrobial Chemotherapy 54(2): 311–320.

44.

Landecker

(2015) Antibiotic resistance and the biology of history. Body & Society 22(4): 19–52.

(2005) Metaphors and biorisks: The war on infectious diseases and invasive species. Science Communication 26(3): 243–268.

, et al. (2018) Critical knowledge gaps and research needs related to the environmental dimensions of antibiotic resistance. Environment International 117: 132–138.

(2007) Effluent from drug manufactures contains extremely high levels of pharmaceuticals. Journal of Hazardous Materials 148(3): 751–755.

48.

Lee

Motzkau

(2012) Varieties of biosocial imagination: Reframing responses to climate change and antibiotic resistance. Science, Technology, & Human Values 38(4): 447–469.

(2018) Tracking antibiotic resistance gene pollution from different sources using machine-learning classification. Microbiome 6(1): 93.

(2018) Toxic politics: Acting in a permanently polluted world. Social Studies of Science 48(3): 331–349.

51.

Littmann

Viens

(2015) The ethical significance of antimicrobial resistance. Public Health Ethics 8(3): 209–224.

52.

Lorimer

(2017) Probiotic environmentalities: Rewilding with wolves and worms. Theory, Culture & Society 34(4): 27–48.

53.

McGregor

(2004) Sustainable development and ‘warm fuzzy feelings’: Discourse and nature within Australian environmental imaginaries. Geoforum 35(5): 593–606.

54.

Macnaghten

Szerszynski

(2013) Living the global social experiment: An analysis of public discourse on solar radiation management and its implications for governance. Global Environmental Change 23(2): 465–474.

, et al. (2017) Conceptualizing imaginaries of science, technology and society. In: U

Felt

Fouche

Miller

, et al. (eds) The Handbook of Science and Technology Studies. 4th ed. Cambridge: MIT Press, pp.435–464.

, et al. (2007) Don’t wear me out – The public’s knowledge of and attitudes to antibiotic use. Journal of Antimicrobial Chemotherapy 59(4): 727–738.

57.

Martin

Simon

(2008) A formula for disaster: The department of homeland security’s virtual ontology. Space and Polity 12(3): 281–296.

58.

Martínez

(2008) Antibiotics and antibiotic resistance genes in natural environments. Science 321(5887): 365–367.

(2016) Framing the agricultural use of antibiotics and antimicrobial resistance in UK national newspapers and the farming press. Journal of Rural Studies 45: 43–53.

(2019) Openness to social science knowledges? The politics of disciplinary collaboration within the field of UK Food Security Research. Sociologia Ruralis 59(1): 23–43.

61.

Murphy

(2006) Sick Building Syndrome and the Problem of Uncertainty: Environmental Politics, Technoscience, and Women Workers. London: Duke University Press.

62.

Murphy

(2013) Distributed reproduction, chemical violence, and latency. The Scholar & Feminist Online 11(3): 2.

, et al. (2017) Influence of manure application on the environmental resistome under Finnish agricultural practice with restricted antibiotic use. Environmental Science & Technology 51(11): 5989–5999.

64.

Nerlich

James

(2008) “The post-antibiotic apocalypse” and the “war on superbugs”: Catastrophe discourse in microbiology, its rhetorical form and political function. Public Understanding of Science 18(5): 574–590.

65.

Nixon

(2011) Slow Violence and the Environmentalism of the Poor. Cambridge, MA: Harvard University Press.

66.

O’Neill J (2016) Tackling drug-resistant infections globally: Final report and recommendations. London: Review on Antimicrobial Resistance. Available at: https://amr-review.org/Publications.html. (accessed February 2020).

67.

Peet

Watts

(1996) Liberation ecology: Development, sustainability, and environments in a age of market triumphalism. In: R

Peet

Watts

(eds) Liberation Ecologies: Environment, Development, Social Movements. London: Routledge, pp.1–45.

68.

Podolsky

(2015) The Antibiotic Era: Reform, Resistance, and the Pursuit of Rational Therapeutics. Baltimore, MD: Johns Hopkins University Press.

69.

Pruden

Larsson

DGJ

Amézquita

, et al. (2013) Management options for reducing the release of antibiotics and antibiotic resistance genes to the environment. Environmental Health Perspectives 121(8): 878–885.

70.

Quintela-Baluja M, Abouelnaga M, Romalde J, et al. (2019) Spatial ecology of a wastewater network defines the antibiotic resistance genes in downstream receiving waters. Water Research 162: 347–357.

71.

Rahube

Marti

Scott

, et al. (2014) Impact of fertilizing with raw or anaerobically digested sewage sludge on the abundance of antibiotic-resistant coliforms, antibiotic resistance genes, and pathogenic bacteria in soil and on vegetables at harvest. Applied and Environmental Microbiology 80(22): 6898–6907.

72.

Rhee

(2018) The Robotic Imaginary: The Human and the Price of Dehumanized Labor. Minneapolis: University of Minnesota Press.

73.

Rosenblatt-Farrell N (2009) The landscape of antibiotic resistance. Environmental Health Perspectives 117(6): A244–A250.

74.

Salyers

Amábile-Cuevas

(1997) Why are antibiotic resistance genes so resistant to elimination? Antimicrobial Agents and Chemotherapy 41(11): 2321–2325.

75.

Singer

Shaw

Rhodes

, et al. (2016) Review of antimicrobial resistance in the environment and its relevance to environmental regulators. Frontiers in Microbiology 7: 1728.

76.

Steffen

Crutzen

McNeill

(2007) The Anthropocene: Are humans now overwhelming the great forces of nature? Ambio 36(8): 614–621.

77.

UK Research and Innovation (2019) Antimicrobial resistance: How research is tackling the challenge. Swindon, UK: UK Research and Innovation. Available at: https://mrc.ukri.org/documents/pdf/antimicrobial-resistance-how-research-is-tackling-the-challenge-panel-brochure/ (accessed February 2020).

78.

Watkins

(2015) Spatial imaginaries research in geography: Synergies, tension, and new directions. Geography Compass 9(9): 508–522.

79.

Woolhouse

Ward

van Bunnik

, et al. (2015) Antimicrobial resistance in humans, livestock and the wider environment. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 370(1670): 20140083.

80.

Wright

(2010) Antibiotic resistance in the environment: A link to the clinic? Current Opinion in Microbiology 13(5): 589–594.

81.

Zellweger

Carrique-Mas

Limmathurotsakul

, et al. (2017) A current perspective on antimicrobial resistance in Southeast Asia. Journal of Antimicrobial Chemotherapy 72(11): 2963–2972.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB