Multispecies hierarchies and capitalist value: Insights from salmon aquaculture

Abstract

Research in animal geographies is increasingly paying attention to hierarchies and inequalities within and between nonhuman animals. The way that animals are valued differently and hierarchically within this growing body of scholarship has tended to focus on a range of biopolitical differences between and within species. Collard and Dempsey’s recent contribution, in contrast, points to the importance of hierarchy and difference in the valuation of nonhuman animals under capitalism. Their framework identifies five orientations of human and nonhuman bodies in relation to capitalist value, which in turn provides a heuristic to explore how capitalist accumulation produces and depends on differentially oriented natures. Our contribution to these debates – and to the Collard and Dempsey framework – draws on our ongoing research in Eastern Canada where salmon aquaculture is a growing yet highly contested industry. We focus on two instances of multispecies hierarchy and difference in and around the salmon cage that are central to this form of ocean-based production. In focusing on multispecies relations, we build on Collard and Dempsey's framework in two main ways. First, we show how valuation and devaluation reflect competing but relational capitalist interests, which rely on and produce different natures refracted through the logic of the nature/culture divide: Atlantic salmon are valued as game fish, and as the key species for Canada's aquaculture sector. Second, we show how capital's valuation of one species, in our case farmed salmon, implicates the valuation of others, namely sea lice and lumpfish. Our case studies extend Collard and Dempsey's framework by demonstrating how capitalist differentiation produces violence through and outside of commodification in terms of multispecies difference and hierarchy; the lives and futures of wild and farmed salmon, lumpfish and sea lice are entangled, and reflect relational and changing orientations to capitalist value over time.

Introduction

The production of hierarchical difference is crucial to the production of value. This is an insight we think must remain at the foreground of theorizing around how natures and their energies are incorporated into the generation of capitalist value (Collard and Dempsey, 2017a, 79–80).

The farmed Atlantic salmon is the basis for a global $20 billion a year industry that generates enormous profits for a handful of companies (Asche and Bjørndal, 2011). Salmon are typically farmed in open net pen cages in near shore coastal zones in countries like Norway, Scotland, Canada and Chile. In these cages – “cities of fish” that contain up to 500,000 salmon at a time (Lien and Law, 2011: 81) – salmon are not on their own. The nets that contain the fish are porous and so they allow many other fish and crustaceans to enter the cage, while simultaneously allowing waste, chemicals and other contaminants to leak out and thereby posing a threat to the surrounding ocean environment (Cole et al., 2009; Wiber et al., 2012). The cages of enclosed salmon, perhaps not surprisingly, also attract many ocean predators including tuna, sharks, seals and seabirds (Barrett et al., 2019), and in response, the industry has developed controversial methods of deterring these threats to salmon production (Callier et al., 2017; Quick et al., 2004). The cage infrastructure is designed to contain fish until they are ready for slaughter, but farmed salmon often find their way out and into the open ocean where they breed with wild salmon with serious consequences for surrounding ecologies and the relatively small number of wild salmon that still spend much of their lives in the open ocean (Schiermeier, 2003; Wringe et al., 2018). More recently, so-called ‘cleaner fish’ like lumpfish and wrasse have been deliberately introduced into the salmon cages in order to control sea lice, which have a devastating impact on both cultured salmon inside the pen (Heuch et al., 2005; Thorstad et al., 2015; Torrissen et al., 2013), and wild salmon outside of it (Costello, 2009). Open net pens for salmon are more than cities of fish: they are complex multispecies infrastructures that simultaneously value and devalue nonhuman animals and other natures inside and outside of the cage for capital accumulation.

Our aim in this paper is to contribute to a growing body of scholarship on the role of nonhuman animals in capitalism. We ask: what can we learn about how capitalism values and devalues nature through salmon aquaculture? To this end, we engage with recent debates on how nature under capitalism is valued differently and hierarchically, and we draw on and extend Collard and Dempsey’s (2017a) recent framework for analysing capitalist natures. Their framework is animated by the question: what are the different ways that nature is incorporated into capitalist production? In emphasizing difference, they argue that natures are not valued equally under capitalism. Instead, natures are valued (and devalued) differently and hierarchically, and it is these hierarchical valuations that are vital to the accumulation of capital.

Our contribution to these debates – and to the Collard and Dempsey framework – draws on our ongoing research in Eastern Canada where salmon aquaculture is a growing yet highly contested industry. We focus on two instances of multispecies hierarchy and difference in and around the salmon cage that are central to this form of ocean-based production. First, we examine how farmed Atlantic salmon production relies on yet also devalues wild Atlantic salmon that live outside the confines of the cage. Through this first example we contribute to scholarship that has examined how capitalism uses logics such as anthropocentrism and the nature/culture binary to devalue nonhuman animals. Yet the valuing of cultured (farmed) animals over natural (wild) animals in this industry is highly contested and complex: valuing and devaluing Atlantic salmon through nature and culture is rooted in local histories and geographies, and dominant actors are intervening in surprising ways. In particular, we show how the category of wild remains “a potent and strategic site in and for environmental politics” (Collard, 2014: 154). Our second example is about the role of lumpfish, a fish that is hailed as a biological solution to sea lice in a context where chemical treatments are becoming increasingly ineffective. Lumpfish are valued for the work they do within the cage in removing sea lice from farmed salmon. While they are increasingly seen as crucial to salmon production, they are nonetheless a devalued nature that is simultaneously wasted and discarded. Through these examples from farmed salmon production, elaborated and discussed in more detail below, we add important insights into how capital differentially devalues nature in a multispecies context.

We constructed our two cases by reviewing reports, articles and other media from a range of sources that focus on salmon production primarily in Atlantic Canada over the period from 2015 to 2021. We reviewed reports, research and media content produced by the aquaculture industry, not-for-profit organizations (e.g. Atlantic Salmon Federation, animal and fish welfare organisations), government (e.g. Department of Fisheries and Oceans), film makers (e.g. the documentary Salmon Wars) and fisheries scientists. Through our research on salmon aquaculture we show how nature under capitalism is valued differently and hierarchically, and the ways in which the organisation of and interests in salmon production produce and rely on hierarchical multispecies relations.

The remainder of the paper is structured as follows: in the next section we situate our analysis of salmon aquaculture in debates on hierarchy and difference in animal geographies. We end this section with a discussion of Collard and Dempsey's recent framework for analysing how animals are valued under capitalism, and we locate their work within broader debates in feminist and eco-Marxist theory. In the subsequent section we briefly introduce salmon aquaculture in Eastern Canada, before exploring – in two separate sections – multispecies difference and hierarchy in and around the salmon cage. We end by drawing out the implications of our analysis for the Collard and Dempsey framework and for broader scholarship on animals, hierarchy and difference in capitalist production. A key contribution of our work is to emphasize the importance of multispecies hierarchies and difference and how these are both produced and contested in the production of capitalist value.

Hierarchy and difference in multispecies relations

A central focus of the ‘third wave’ of animal geographies is how human exceptionalism has shaped the relations between humans and nonhuman animals. Research in this tradition has examined how a dominant Western Cartesian philosophy that separates and orders humans and animals has justified the violence, appropriation and commodification of animal species for human needs. Examples from this large body of work range from industrial farming and the “war against animals” (Wadiwel, 2015), to more recent research that has explored the ways in which animals are valued for their liveliness in spectacles (Barua, 2016) and in the animal pet trade (Collard, 2020).

Human – nonhuman animal hierarchy and difference remain a central focus of research in animal geographies. Yet recent contributions have paid increasing attention to the hierarchies and inequalities within and between nonhuman animals. This important shift is the core theme of Hovorka’s (2019: 749) review of animal geographies where she emphasizes the importance of multispecies hierarchical networks and calls for research going beyond “the well traversed dichotomy of ‘human vs animal’ into explorations of how animals themselves are positioned relative to other animals”. Her review of the field provides examples of “species-based relations of power” (Hovorka, 2019: 2) and the logics that shape and influence these hierarchies. The sites where these hierarchies and differences play out include zoos, farms, and laboratories and in the practices of conservation, species reintroductions, and rewilding.

Hovorka’s (2019) review identifies an important and burgeoning area of research in the field. Indeed, there are several recent studies that provide key additional insights into the larger question of species-based relations of power. Rutherford’s (2018) study of ‘coywolves’ – an animal that is a mixture of wolves, coyotes and domestic dogs – provides an example of how this hybrid is considered a threat not only to domestic species but also to other wild species. Calls for the extermination of this animal that has adapted well to urban environments are justified on the basis of its status as a form of “biological pollution”, and its role in transgressing constructed spatial boundaries between the rural and the urban. Lavau’s (2013) work on different fish species in an Australian river is notable for its exploration of the multiple valuations that emerge for different fish as belonging and nature intersect. In Lavau's river, fish are valued against how they relate to indigeneity, wildness and the role they play in terms of ecological functionality. In this way, Lavau's careful study provides an important analysis of the different ways in which animals are ordered in specific environments. Shukin’s (2018: 98) discussion of what she calls “animals saving grace” provides a final example of what is a growing body of research on animal hierarchy and difference. Her focus is on the role that some animals play in holding together what is an increasingly precarious capitalist system. Animals in this context are valued if their lives can “align with the goal of saving damaged environments of life for a future of capitalism” (Shukin, 2018: 98), but are discarded or ignored if they cannot meet this role. These and many other studies of animal difference share a common concern in first specifying and then disrupting the logics through which different animals are valued and devalued while at the same time revealing the violence that these hierarchies have on animal life.

The way that animals are valued differently and hierarchically within this growing body of scholarship has tended to focus on a range of biopolitical differences between and within species. Collard and Dempsey’s (2017a) recent contribution, in contrast, points to the importance of hierarchy and difference in the valuation of nonhuman animals under capitalism. Their critical insight is that nature is not valued or devalued equally and that these differences are fundamental to capital accumulation. Through a structured framework, they propose five specific ways that nonhuman nature is valued and devalued under capitalism. Nature may be: officially valued as commodities or as paid or unfree labour; a reserve army as potential future value; an underground as useful or necessary unpaid or unpriced inputs; as outcast surplus as redundant to accumulation or as waste; and finally, nature can be a threat to capitalist accumulation. These five modes of valuation are conceptualised using Ahmed’s (2006) idea of orientation. Orientation in Ahmed's work refers to how bodies are positioned in relation to other objects, things and institutions. An orientation is not random. It is instead patterned and follows what Ahmed (2006: 16) calls “a path well trodden”, with some orientations being more conceivable, more probable than others. They are also dynamic and changing. For Collard and Dempsey, orientations capture the diverse yet patterned ways in which capitalist production relies on and incorporates differentially valued nature. A broader aim of their intervention is to contribute critical analysis to a field that may not have “adequately grappled with capitalism as an organizing structure for human-animal relations” (Barua, 2016: 726).

Collard and Dempsey's framework aims in part to complicate and extend eco-Marxist approaches that focus on how capitalism uses nature as material or energy for production (see for e.g. Moore, 2015). Their typology builds on two key insights from feminist and postcolonial scholarship. First, most economic activity required for capital accumulation takes place outside the sphere of waged labour and commodification. Feminist political economists have long argued that capital accumulation depends on subsistence and social reproductive work of women, children and others that remains largely invisible, uncounted, and unpaid (see Federici, 2004; Mies, 1986). Scholars such as Mies (1986) extend this argument to include the material and energies of colonized people, and the work of nature. Together, these insights enable Collard and Dempsey to develop a typology of orientations of nature to capitalism that include and extend beyond those that are officially valued as waged and commodified. The typology builds on a second key insight by feminist and postcolonial scholars, that capitalist value and accumulation rely on human differentiation in sorting bodies – as waged labour, commodified, reserve army, waste -- and that how bodies get sorted is not accidental, but rather depends on patriarchal, colonial, racist and for ecofeminists, anthropocentric logics (e.g. Gaard, 2001; Plumwood, 1994). Collard and Dempsey (2017a: 83) draw on and extend this work to create a typology that accounts for hierarchy and difference in the valuation of nonhuman animals under capitalism: “in order to better understand how capitalist value production actually requires and contributes to producing this difference” (italics in original).

While Collard and Dempsey's framework represents a key conceptual and theoretical intervention in a field that is increasingly concerned with hierarchy and difference amongst nonhuman animals, they provide little in the form of substantive empirical examples. It is, as they write, “difficult to convey the dynamism of orientations, that is, how bodies move or are moved through them and how even the orientations themselves might change” in a single article (Collard and Dempsey, 2017a: 94). More recently Collard (2018) has used the framework to analyse the unravelling of the life of otters on the west coast of North America. Sea otters go from being valued as dead commodities, to outcast surplus linked to petro capitalism and the Exxon Valdez oil spill, and finally to the underground as living spectacles in zoos, aquaria and laboratories. The sea otter case provides a helpful illustration of how a single species can move between several orientations. Yet existing work that has engaged with the framework has tended to focus on how a single animal species fits into one or more of the categories (see Enns et al., 2019) or, as is the case for otters, how an animal's orientation to capitalist value shifts over time (Collard, 2018). The focus on individual animals and their orientation raises the question – in part prompted by Hovorka’s (2019) challenge discussed earlier – of how the framework might be used in a context where several different animals are valued hierarchically in capitalist commodity production. How do the orientations of multiple animals valued hierarchically and differently intersect or clash? Do these hierarchies and differences always support capital accumulation or are there contradictions and conflicts? The purpose of our paper is to address these questions through our analysis of salmon aquaculture and the multiple species that are arranged in and outside the cage with a view to contributing to the framework, and the broader debates on the role of nonhuman animals in capitalist production.

Atlantic salmon entanglements in Eastern Canada

Our analysis of Atlantic salmon's capitalist orientations is situated in Eastern Canada. This is a region that has a very long history of human-salmon entanglements. Indigenous groups, commercial harvesters and processors, recreational anglers and, most recently, a growing aquaculture sector that farms salmon in cages along the coast, have all relied on salmon. For all of these groups, salmon was (and is) highly valued as food, or commodity, or recreational game fish. Yet the history of salmon in this region is also about dis-entanglements. Indigenous groups who depended on salmon were dispossessed of this important resource through colonial policies that outlawed Indigenous methods of harvesting. Parenteau (1998: 1) has written extensively on how Indigenous groups in the Canadian Maritimes were, from the mid-1800s, “systematically excluded from the Atlantic salmon harvest” (also see Kenny and Parenteau, 2014; Parenteau, 2004). More recently, commercial harvesters - including Indigenous harvesters - were excluded from fishing salmon in a process that started in the 1970s. In response to concerns about the sustainability of commercial harvesting practices, fisheries officials systematically limited commercial harvesters’ access to salmon. The powerful angling lobby played a key role by convincing regulators that commercial salmon production was wasteful and generated far less revenue than a rapidly growing network of businesses commercializing recreational angling. Despite evidence that Atlantic salmon was a very important part of small scale fishers’ livelihoods (Felt, 1990), by 2000, commercial harvesting of salmon in Atlantic Canada had been banned across the region, and salmon could only be caught recreationally using a rod, a weighted line and an artificial fly, or by Indigenous groups in the region who are allowed to harvest a limited number of fish for subsistence purposes only.

In the contemporary period, the Atlantic salmon in Eastern Canada is connected to humans in three main ways. First, salmon that spawn in rivers and live in the sea are connected to the powerful recreational angling community. They have played a crucial role in ensuring that Atlantic salmon in Canada is valued as a prized game fish, and these organisations have spent enormous effort demonstrating the economic, cultural and social value of Atlantic salmon to people in the region. Atlantic salmon is valued by these groups for its potential to be captured and released repeatedly through catch-and-release methods of angling. It is also valued for its behaviour of putting up a strong fight against capture, thereby providing wealthy men the opportunity to be the “king of fish” (Hustins, 2010: 6). And it is valued for the role it plays in sustaining economic activities associated with recreational angling (Gardner Pinfold, 2011). The number of Atlantic salmon in Eastern Canadian rivers and in the sea is, however, only a fraction of the salmon that are grown in cages in Eastern Canada. This is the second way in which humans are entangled with salmon. In this industry, salmon are valued for their role in generating export revenues and providing jobs in the remote coastal regions of Eastern Canada. Salmon is also promoted by the industry, unconvincingly, as playing a valuable role in providing protein for a hungry planet (Edwards et al., 2019). The third way in which salmon are connected to humans is through regulations that give Indigenous Nations in Atlantic Canada access to salmon for subsistence purposes. Salmon caught under these regulations are used by Indigenous groups for food, social and ceremonial purposes, and cannot be used for commercial sale or exchange.

Salmon – value and threat through nature/culture

How is the Atlantic salmon oriented to capital in Eastern Canada? There is no straightforward answer to this question because different “blocks of capital” (Collard and Dempsey, 2017a) value different Atlantic salmon along the nature/culture axis. Specifically, the aquaculture industry values farmed Atlantic salmon, originating from local wild salmon stocks but differentiated through decades of breeding and selection practices. Farmed salmon are valued as a commodity raised in cages in the ocean until they reach ‘commodity size’ at which point they are slaughtered and processed for human consumption. The recreational angling sector, in contrast, values wild Atlantic salmon that live in the region's rivers and the adjacent ocean. Wild salmon are valued as a game fish for anglers, and the large network of enterprises that support angling including outfitters, guides, hotels, and retailers, an industry valued at $255 million per year (Gardner Pinfold, 2011). At the same time, the two salmon oriented to different blocks of capital are positioned as significant threats to each other: for the angling sector, salmon aquaculture is thought to be one of the most important threats facing the dwindling numbers of wild salmon that live in rivers and oceans; for the aquaculture sector, the precarity of wild salmon stocks – and the angling sector that champions salmon as a valuable game fish – is also a very important threat to their industry. How the Atlantic salmon is oriented to capital as both value and threat is shaped, then, by its location along the nature/culture axis –wild/natural or farmed/cultured. Yet, as we examine in more detail below, the nature/culture axis is fluid and situated and it plays a complex role in how salmon is oriented as value or threat to capital.

The Atlantic Salmon Federation (ASF) is the politically powerful organization representing recreational anglers in the region and advocating for the conservation of the region's wild Atlantic salmon. The ASF has declared the farmed Atlantic sector as one of the most important threats to wild salmon (Atlantic Salmon Federation, 2019; CBC News, 2021). It draws on a large body of scientific work in which three key issues stand out: first, the problem of disease transmission from farmed to wild fish (Kibenge et al., 2019); second, sea lice infestations (Costello, 2009); and third, the impact of farmed fish escapes (Schiermeier, 2003; Wringe et al., 2018). Disease transmissions from farmed Atlantic salmon to wild fish have been especially well documented in British Columbia, where it has led to phasing out of salmon farming licences in parts of that province and a strong call for the industry to shift to closed containment systems on land (Cox, 2020; Summerfelt et al., 2013). Sea lice is a significant problem for the aquaculture industry. Large numbers of fish contained in cages attract these small crustaceans, which reproduce rapidly and uncontrollably, affecting the health of farmed fish within the cage and wild fish outside these structures (Costello, 2006; 2009). Significantly, as we discuss in more detail below, chemical treatments are becoming less effective in the face of sea lice adaptation. Finally, farmed salmon that escape their cages – which is a relatively common occurrence (Schiermeier, 2003) – are known to reproduce with their wild counterparts creating new hybrids (Wringe et al., 2018). Scientists argue that these genetic ‘introgressions’ dilute the wild gene pool and render these fish less fit to handle the rigours of life in rivers and the open ocean (McGinnity et al., 2003). A recent study following an escape of more than 20,000 salmon on the south coast of the island of Newfoundland found evidence of farmed salmon genes in 17 of the 18 rivers surveyed (Wringe et al., 2018). For the recreational angling sector, the problem with introgression is not just that it renders wild salmon less fit to survive in the ocean; it is also that hybrids and farmed escapes may not produce the celebrated ‘fight’ wild salmon give when hooked at the end of a line (Olaussen and Liu, 2011). These hybrid and domesticated salmon threaten a large recreational angling sector that generates income and jobs in hotels, outfitting enterprises and guiding.

Wild Atlantic salmon oriented as game fish for the angling industry represents an equally significant threat to the farmed salmon sector. This threat is intensified when wild salmon numbers decline in regions with intensive aquaculture production (Ford and Myers, 2008). The decision to phase out licences for farming salmon in parts of British Columbia, for example, has been justified on the need to protect wild salmon from diseases and sea lice associated with salmon cages (Britten, 2018; Keane, 2018; Sapin, 2019). In British Columbia the farming of Atlantic salmon – a non-native species – provides further justification by environmental groups and some Indigenous groups to oppose salmon aquaculture. Yet wild salmon also represents a threat in Eastern Canada where the same species is farmed in cages. In early 2021 Canada's Department of Fisheries and Oceans (DFO) released data on wild salmon returns to rivers on the island of Newfoundland. The most alarming results were from the Conne River, which had previously averaged around 2000 fish returns a year, but in 2020 only recorded 150 fish returning to the river to spawn (Dean-Simmons, 2021; also see 2019 stock status update: Fisheries and Oceans Canada, 2020). The Conne River runs into the South Coast of the island, which is also where salmon aquaculture is concentrated. In reporting on salmon returns, DFO salmon scientists confirmed that aquaculture was a probable cause for the decline in Conne River salmon given the proximity of aquaculture cages (Dean-Simmons, 2021). Other groups noted that the returns for this river were far lower than was the case in other parts of the island of Newfoundland where salmon aquaculture is not present (Moffatt, 2021; SAEN, 2021). In Eastern Canada, as in British Columbia, the priority to protect wild salmon represents a significant threat to the salmon aquaculture industry.

We have argued that salmon is oriented to capitalist value as both a cultured fish for commodity production and as a wild fish for recreational anglers. The basis for this orientation is the deeply entrenched nature/culture binary that divides animals into distinct categories. The role of the nature/culture divide in species valuation has been extensively analysed by scholars interested in the biopolitics of conservation (Biermann and Mansfield, 2014). Guided by the Collard and Dempsey framework, our analysis situates the nature/culture divide in terms of how salmon bodies as farmed or wild map onto capitalist value. Yet nature/culture as a way of thinking about animals like salmon is both situated and fluid (Swanson et al., 2018). The situatedness and fluidity of this divide play out in surprising ways in Eastern Canada as the aquaculture industry attempts to defuse the threat of wild salmon through two separate and somewhat contradictory strategies.

The first way that the aquaculture industry has attempted to defuse the threat posed by wild salmon orientated as game fish involves questioning the genetic purity of wild Atlantic salmon in the region. Here the industry draws on the history of salmon hatchery programmes in Eastern Canada that were aimed at enhancing wild salmon numbers in the region. Hatchery programmes were established well over a century ago: the first salmon hatchery was built in New Brunswick in the 1870s and the number of hatcheries soon numbered more than 21 across Eastern Canada (Cook, 2016). From the 1970s there were renewed efforts to enhance wild salmon stocks through scientific and genetic approaches to hatchery production, which led to the establishment of the Salmon Genetics Research Programme (SGRP), supported by the state and the Atlantic Salmon Federation (Anderson, 2007). Hatchery fish produced under this programme focused on improving survival at sea and by ensuring larger number of returns to the region's rivers.

The aquaculture industry has argued that this long history of culturing has so diluted wild salmon populations that it is no longer possible to distinguish farmed salmon in cages from wild salmon in the ocean. Consider, for example, the statement by a representative of the Nova Scotia Aquaculture Industry in an interview for the investigative documentary Salmon Wars. The representative countered the idea that farmed salmon represented a threat to wild fish by pointing to centuries of hatchery stocking in the region: “This idea that we have pure stocks running in these rivers is simply not true. We have been stocking these rivers for over a hundred years” (Salmon Wars, 2012). More recently, and in response to criticisms about salmon escaping from cages, the executive director of the Atlantic Canada Fish Farmers Association argued that conservationists “conveniently ignore any potential impacts of over 100 years of Atlantic salmon enhancement efforts, including ASF's own sea ranching project in the 1970s and 80s that saw large releases of a variety of salmon strains into rivers and estuaries” (Evans, 2019). The implication of these statements is obvious: if all salmon are cultured there is no need to protect salmon that live in the sea from salmon that are contained in aquaculture cages on Canada's East Coast

Suggesting that salmon in Eastern Canada are all cultured regardless of where they spend their lives is one way of defusing the threat of wild salmon, but it is not the only one used by the aquaculture industry. The importance of protecting wild species, we noted earlier, remains critical to environmental politics, including for Atlantic salmon in spite of the many years of human interventions in salmon lives through hatcheries. The importance of wild salmon as a species that deserves protection helps to explain the second strategy of the aquaculture industry. Recently, aquaculture companies have become directly involved in wild salmon conservation in ways that bear a remarkable resemblance to the processes Enns et al. (2019) have documented in Sub-Saharan Africa; they reveal how companies involved in resource extraction are focusing on conservation and biodiversity protection as a way of “temporarily resolving crises that threaten their productivity and profitability” (Enns et al., 2019: 33). In Eastern Canada, Cooke Aquaculture – together with a group of local state organizations, other industry partners, and corporate funded university researchers – has embarked on a similar strategy. Cooke is spearheading an ambitious attempt to support the recovery of wild salmon in New Brunswick rivers through a new approach called ‘smolt to adult supplementation’ (SAS). While efforts to rebuild wild salmon stocks in North America and Europe have typically involved introducing young salmon or fertilized eggs into river systems (see Harrison et al., 2019), SAS involves removing young fish from rivers that are ready to make the journey to the ocean, and then raising them to adult stage in a specially prepared aquaculture net pen placed in the ocean (Cooke Aquaculture, 2020; Fisheries and Oceans Canada, 2018; Hume, 2011). Once raised to adult stage the plan is to reintroduce these salmon into New Brunswick rivers where they will reproduce and lay the foundation for a new generation of wild salmon. The logic behind the SAS approach to wild salmon conservation is research suggesting that fish that reproduce naturally in rivers, and then spend their early life in these rivers or streams before migrating to the sea, are better able to survive the rigours of a challenging ocean environment (Jonsson et al., 2003).

The aquaculture industry strategy to rebuild wild salmon stocks in Atlantic Canada is an attempt to address its legitimacy crisis, and is consistent with other examples where corporations highlight the role they play in corporate social responsibility through wildlife conservation (Enns et al., 2019). Seen through the Collard and Dempsey framework, the effort to support Atlantic salmon conservation can be understood as an attempt to defuse the threat that wild salmon pose to existing and future capital accumulation in industrial salmon aquaculture in the region. Participating in the recovery of wild salmon, particularly if successful, would go a significant way to removing obstacles for the expansion of salmon aquaculture in the region.

Yet the efforts by the industry to support wild salmon conservation have faced resistance from conservationists, angling interests and government regulators. More recent conservation measures for salmon favour less direct human intervention, including the decades old practice of stocking rivers with hatchery raised fish (Harrison et al., 2019). Instead, these strategies focus on what are considered to be more ‘natural’ methods such as removing dams and other obstacles in rivers, and by rehabilitating river systems so that they are suitable for natural salmon reproduction (Kemp and O’anley, 2010; Limburg and Waldman, 2009; Nislow et al., 1999). The underlying philosophy behind these conservation approaches is the conviction that wild fish that are more independent of human intervention are better able to survive in rivers and the ocean. In this context, the aquaculture industry strategy of raising and feeding ‘wild’ fish in cages in the ocean and then returning them to rivers to spawn a new generation of wild salmon has been heavily criticized by conservationists and government regulators. At the time of writing, DFO has not permitted the company to relocate fish from ocean-based cages into local river systems based on the uncertain ecological impacts that introducing cage raised fish might have on existing populations of wild salmon in New Brunswick rivers (Smith, 2019).

We draw two insights from this first example of multispecies hierarchies. First, valuing and devaluing Atlantic salmon maps onto nature/culture binaries, pointing to the way in which these logics remain critical to capitalist value. Second, and relatedly, when animals that are oriented as value to capital face threats from other animals we should expect and attend to the responses that are aimed at defusing or eliminating these threats, and how these in turn generate hierarchies that differentially value and devalue animals for capital accumulation.

Salmon – lumpfish – lice: Multispecies hierarchies in the salmon cage

Salmon cages secured in the ocean attract a wide variety of other aquatic species including seals, sharks and tuna, and these are a serious threat to industrial salmon aquaculture. Yet it is the tiny sea louse that represents the most significant threat to salmon aquaculture production. Sea lice live outside of salmon cages, but they reproduce very quickly and efficiently within the cages with a devastating impact on salmon (Costello, 2006). Salmon with lice infestations are affected in multiple ways: their growth is reduced, susceptibility to viral and bacterial diseases is increased, and mortality rates increase significantly (Heuch et al., 2005; Thorstad et al., 2015; Torrissen et al., 2013). In severe cases, as has occurred in Canada and Scotland in recent years, salmon producers have destroyed an entire production cycle of salmon due to lice infestation (Gill, 2019). Sea lice in salmon cages also threaten wild salmon populations outside of the cages, and this has been another reason why the aquaculture industry has been under pressure to control sea lice infestations in cages (Costello, 2009).

There are a limited number of chemical treatments available to salmon producers, and sea lice have become increasingly resistant to these chemicals (Igboeli et al., 2012; Jones et al., 2013). These chemical treatments can be applied either by chemical bath or as an addition to the salmon's feed pellets, providing vectors for contamination to the marine environment. Pesticides or ‘therapeutants’ have been detected in waters surrounding net pens as well as in the sediments below and have been shown to cause changes in benthic community structure by negatively impacting behaviour, reproduction and mortality rates in many crustaceans and bivalves (Burridge et al., 2000; Ernst et al., 2001; Urbina et al., 2019). Other non-chemical treatments, including fresh water and warm water baths (e.g. the ‘thermolicer’), are being rapidly developed and used in the industry, but they have many drawbacks including increased tolerance in lice (Ljungfeldt et al., 2017) and animal welfare concerns (Grøntvedt et al., 2015). Mechanical methods usually involve removing salmon from cages, which is stressful and reduces their immunity, and can result in significant mortalities. Production is also compromised as it can take a number of weeks before salmon recover from one of these stressful mechanical treatments (Grøntvedt et al., 2015). A more promising solution has involved the use of so-called ‘cleaner fish’, a ‘pro-biotic’ (Lorimer, 2020) control method that the industry promotes as environmentally sustainable, ethical, and a ‘natural’ way of dealing with the problem of sea lice.

Cleaner fish such as wrasse have been used in salmon aquaculture for several decades, but in recent years it is the small lumpfish that has caught the attention of a salmon industry struggling to deal with the threat of sea lice. There are several advantages of the lumpfish over other cleaner fish species: they can be introduced into salmon cages at a much younger age than other cleaner fish; they continue to eat sea lice at much lower temperatures, which is important for producers in Iceland, the Faroe Islands, and northern parts of Norway, Canada and Scotland; and, most importantly, is the evidence that they can dramatically reduce sea lice infestations in salmon cages (Imsland et al., 2014a, 2016a). The potential value of lumpfish to the salmon aquaculture industry has generated huge investments in production facilities for this fish across the North Atlantic (CBC News, 2019; Holmyard, 2018; Powell et al., 2018), it has led to several new university-industry research alliances aimed at better understanding this relatively unknown species, ¹ and it has led to calls for enhanced veterinarian expertise in lumpfish physiology (Imsland et al., 2016b).

It seems straightforward that in the context of industrial aquaculture, salmon are officially valued as commodities, while the tiny sea louse is an obvious threat to production. But what to make of the lumpfish? How do we make sense of lumpfish that co-inhabit salmon cages using Collard and Dempsey's orientations? We want to say two things about the orientation of lumpfish to capitalist production, in this case salmon aquaculture. First, the value of lumpfish to salmon, and to capitalist production at present, and its value as future potential track across several of Collard and Dempsey's orientations. Second, the role of lumpfish in the value relation illustrates hierarchical multispecies entanglements (in our case, lumpfish, salmon and sea lice) in capitalist production, with implications for living and dying. It is this hierarchical order that makes lumpfish expendable; their value depends on adding value to salmon production.

The lumpfish is oriented towards capitalist value through its relations to accumulation in salmon production. On their own, lumpfish have had limited value as a commodity. They are harvested primarily for their roe in the wild capture fishery, though the future of this fishery is uncertain given a recent Canadian assessment of the species as threatened (COSEWIC, 2017). An undated document produced by DFO describes experimental work in farming lumpfish for roe (Department of Fisheries and Aquaculture, n.d.). However, it appears that these efforts were abandoned. Their official value lies in their work eating sea lice in salmon cages, work that is crucial to successful salmon production (see Barua, 2019). As lively commodities, lumpfish are bought and sold alive for the purpose of eating sea lice, but their ‘labour’ is unfree (Collard and Dempsey, 2017a: 86); it belongs to the salmon producers. And their liveliness is valued insofar as their labour serves to eliminate the threat of sea lice, which in turn allows salmon aquaculture companies to brand their commodity as a more sustainably-produced, ‘natural’ salmon and make claims about improved salmon welfare, irrespective of the welfare of the lumpfish used to produce them.

Given the value relation of lumpfish to the salmon industry, it is no surprise that projected demand of lumpfish for the industry has skyrocketed, with some estimating as many 50 million fish will be required for industrial aquaculture annually (Powell et al., 2018), providing the impetus for the development of ‘cleaner’ fish aquaculture in the sector. Cleaner lumpfish come from harvesting wild stocks that are used to breed a single, new generation of lumpfish for salmon cages (Fisheries and Oceans Canada, 2017; StofnFiskur, n.d.). However, there are concerns about the impact of harvesting wild lumpfish given that stocks are either threatened or unregulated by state authorities, and about the risks of disease transmission from wild and first-generation cleaner lumpfish to salmon (Erkinharju et al., 2021). Lumpfish introduced to salmon cages are vaccinated for several different diseases, and so their value to the industry is matched by their potential threat to salmon health. In response to such threats, companies and universities are heavily investing in the development of broodstock to produce reliable and ‘safe’ generations of cleaner lumpfish (SalmonBusiness, 2018). Compared to the lumpfish doing the work of cleaning sea lice, wild lumpfish and broodstock are oriented differently to capitalist value; they are oriented as underground, providing the invisible labour and the bio-material for reproduction that supports the work necessary for accumulation. In this value relation, wild lumpfish and broodstock are not commodities but their reproductive work is requisite for the production of cleaner fish sold as lively commodities to salmon aquaculture producers.

Paradoxically, not all lumpfish are efficient at eating sea lice. While lumpfish have been very effective in reducing sea lice infestations, the rates at which lumpfish actively eat sea lice from salmon are very low, ranging from between 5 and 15% of fish introduced into cages. In other words, most of the lumpfish introduced to salmon cages play no role in reducing sea lice on salmon. They are a reserve army, valued in their potentiality to eat sea lice. There are intense efforts to improve and enhance their ability to remove sea lice from salmon. In this way, the value of lumpfish is not static, but is instead malleable. Lumpfish may be ‘made’ more efficient cleaners and there are several different ways in which this is being pursued. One approach involves attempts to select for better lice eaters: for example, some research suggests that younger or smaller lumpfish are more efficient (Imsland et al., 2016b). In other cases, the solution involves selecting lumpfish ‘families’ that show a stronger trait of cleaning lice off salmon (Imsland et al., 2016a, 2021). Researchers are also focusing on understanding how to improve the vision of lumpfish, presumably to make them more efficient at eating sea lice (Ahmad et al., 2019). A final strategy involves training lumpfish before they enter the cage (Imsland et al., 2019). Assessments of lumpfish efficiency in delousing have shown that it increases over time. In other words, lumpfish become increasingly more effective in removing sea lice the longer they spend in the sea cage (Imsland et al., 2015). It has led to speculation that lumpfish learn to become better at removing lice from salmon (Imsland et al., 2019; Staven et al., 2019), a theory that is given credence by the lack of knowledge of lumpfish behaviour in the ocean – we don't know whether lumpfish remove sea lice from salmon in the wild. This ability to learn has prompted suggestions that lumpfish be ‘conditioned’ to sea lice removal from salmon before they are placed in cages so that the lumpfish begin cleaning as soon as they are placed in the salmon cage.

Lumpfish are also oriented as outcast surplus. One of the key challenges in the use of this specific cleaner fish is the extraordinarily high mortality rate, an issue that has justifiably caught the attention of non-governmental organisations concerned with animal welfare (OneKind, 2018). According to one report, up to 100% of lumpfish die in the first few weeks after being introduced to the salmon cage (Klakegg et al., 2020). There are several reasons for this extraordinary high mortality rate including diseases, parasites and infection (Klakegg et al., 2020; Powell et al., 2018), and the difficulty they face in adapting to the cage environment and feed (Imsland et al., 2014b; Johannesen et al., 2018). These high mortality rates have prompted the aforementioned research on feeds that are suitable to lumpfish and on how to make the sea cages more habitable for lumpfish. Lumpfish are also literally ‘outcast’: lumpfish that do survive in the cages are not transferred to other cages at the end of a salmon production cycle where they might continue their role in reducing sea lice. Instead, for biosecurity reasons lumpfish are destroyed when salmon reach harvest weight. They are disposed of as waste.

Capital accumulation through multispecies hierarchies

Multispecies hierarchy and difference among nonhuman animals, we have shown, are critical to capitalist production of a commodity like salmon. These differences within and between species rely on logics such as nature/culture, as well as orientation devices such as legislation and science that work with, and against capitalist interests (Collard and Dempsey, 2017a). In the case of salmon, for example, laws that ban the commercial fishing of Atlantic salmon serve the interests of the angler lobby, but measures like these that are aimed at protecting wild salmon simultaneously pose challenges for industrial aquaculture expansion. Our research also shows that multispecies hierarchies are not only used by capital, they are also produced by capitalism. Farmed salmon are the result of breeding and selection practices that produced a cultured species that can grow rapidly in the confined environment of cages suspended in the ocean. Similarly, lumpfish are extracted from the ocean and bred in hatcheries to play a particular role in the salmon cage to sustain commodity production in a context where existing methods of sea lice control are failing. In this case the commodified lumpfish is a species that in Shukin’s (2018) terms represents “animals saving grace” – these are animals that are enrolled to play a role in maintaining a capitalist system facing ecological ruin. The threat that aquaculture poses to wild salmon and other natures is a consequence of a production system that is unsustainable, and facing environmental ‘blowbacks’ associated with mass die-offs, algae blooms and uncontrollable sea lice infestations. The multispecies hierarchies of value that we have documented in salmon aquaculture are a direct consequence of how production is organized in this intensive farming system.

The relationship between multispecies hierarchies and capital accumulation is marked by contradiction and conflict. The idea that nonhuman nature is a threat to capitalist value brings an important element to bear on the relation between capital and nature. The literature tends to see nature as cheap, obdurate, or maybe uncooperative, but less rarely is it seen as a threat to capitalist value (Collard and Dempsey, 2017a, 2017b). Our example of sea lice shows how this particular crustacean represents a significant threat to capitalist accumulation in the same way that avian influenza and other diseases represent an economic threat to poultry and other intensive factory farm-based systems. Yet the struggle between wild and farmed salmon also reveals how complex ‘threat’ can become in the biopolitics of salmon. In this struggle, both sides see the other as a threat to capitalist value and accumulation, and both attempt to devalue the other in a way that reveals competing ontologies of salmon in the context of salmon production. The aquaculture industry's attempt to resolve the threat of wild salmon through the implementation of a conservation program that temporarily contains salmon in pens, along with its subsequent termination by the federal government, illustrates the tension between competing capitalist interests and ideas about the purity of nature.

While the idea of nonhuman multispecies animal hierarchies is well established in the literature on more-than-human geographies (Biermann and Mansfield, 2014), this literature rarely connects these hierarchies to capitalist value and accumulation. Most of this work is informed by Foucault's work on biopolitics and the idea that some lives are fostered and others are not. Hovorka (2019: 4) has written that “some animals are valued more than others as expressed through different and often inequitable attention, resources, and policies focused on particular animals’ rights, needs, and well being.” And she has called for more work on the “differential and inequitable lives of animals”. Collard and Dempsey's orientation lens provides a way to do this by mapping out the ways that capitalist interests differentiate nature, widening the focus and making visible the ways in which capitalism incorporates nature through and beyond commodification. In our case of multispecies relations inside the salmon cage, the salmon is officially valued as commodity, the sea louse is a threat to salmon production, and the lumpfish is oriented towards capitalist value through the work it does for salmon production. It is this hierarchical and relational ordering that is vital to capitalist accumulation. The lumpfish as cleaner fish do this work in two main ways, as lively commodities bought and sold to eat sea lice, however inefficiently, and thereby reducing salmon mortalities; and by providing the industry with an ‘environmentally sustainable’ and ‘natural’ way of addressing the problem of sea lice. Such attempts to frame industrial aquaculture as more sustainable illustrate both the strength of purity narratives of nature and the tensions these produce for the aquaculture industry.

Collard and Dempsey's framework shows how capitalism differentiates nature for accumulation, and how such differentiation produces violence through and outside of commodification. Our case studies demonstrate how such violence must be understood in terms of multispecies difference and hierarchy; the lives and futures of wild and farmed salmon, lumpfish and sea lice are entangled, and reflect changing orientations to capitalist value over time. For example, our analysis of lumpfish shows that as they have become enrolled in the salmon industrial farm, they occupy more than one orientation, and that these orientations are refracted through their value relation to salmon production. A key theme in the lumpfish case is their potential value in the industry's fight against sea lice, and that this potentiality is related to a sense that lumpfish are a malleable product. Currently, there are huge investments of capital being put into lumpfish production. Presumably the value accrued to salmon by using lumpfish as cleaner fish outweighs their inefficiency in eating sea lice and difficulty acclimating to the salmon cage, which results in high mortality rates. New scientific projects are being funded to increase our understandings of this fish, and improve the reliability, safety and efficiency of lumpfish through the development of broodstock and theories about how to condition lumpfish to eat sea lice. These lumpfish are subjects of science, different from the lively commodities doing the work of cleaning sea lice, and are oriented to capitalist value as underground, providing the invisible reproductive labour, bio-material and data that (potentially) supports the work for capitalist accumulation in salmon production. And, if the research fails to improve the cleaning work of lumpfish, then perhaps the industry will search for alternative methods for managing sea lice, and the cleaner fish will be outcast

Conclusion

Barua (2016: 726) has argued that “more-than-human geography pays relatively scant attention to modes of organization centred on accumulation” and has lauded recent contributions that engage with lively commodities that “pay close attention to the inequalities and asymmetries generated by capital, to which some strands of posthumanism and new materialism have been symptomatically blind” (Barua, 2019: 652). Our paper contributes to this literature by examining the valuation and devaluation of nonhuman natures associated with salmon aquaculture in Canada. We are working with a framework developed by Rosemary Collard and Jessica Dempsey that draws on Ahmed’s (2006) concept of orientation. The framework has allowed us to document the different ways in which nonhuman natures, similar to human labour, are differentially valued and devalued in this sector of the economy. We find that the multispecies relations among salmon, lumpfish and sea lice provide examples of officially valued, the reserve army, underground, outcast surplus and threat. The advantage of the framework is that it provides a way of seeing how animal bodies are oriented to capital in different ways. Our contribution builds on the framework by moving beyond identifying how an animal is oriented to capitalist value and emphasizing the importance of multispecies relations among and between animals generated through capitalism. Through this approach we show that animals can occupy more than one orientation depending on how they relate to other animals, and we reveal how blocks of capital act to address the threat that other natures may pose to their officially valued nature.

We have contributed in two main ways to the orientation approach. First, there are intense struggles over the valuation and devaluation of salmon that reflect competing but relational capitalist interests. We show how competing capital interests value different Atlantic salmon along the nature/culture logic. Currently, salmon are valued as valuable game fish, and as the key species for Canada's aquaculture sector, and the struggles over valuation reflect the extent to which groups can claim or contest salmon's claim to wildness and genetic purity. A second contribution of our paper has been to stress the importance of exploring valuation and devaluation in terms of multispecies relations. For salmon aquaculture, this has involved us looking inside the cage, and exploring the differential and hierarchical valuations for salmon, sea lice and lumpfish. In the cage, salmon and sea lice occupy orientations as ‘valued’ and ‘threat’ respectively, and lumpfish are oriented in very complex ways that include commodity, the underground, outcast surplus and threat. This view inside the cage provides insights into the relations between nonhuman natures, as well as the way in which nonhuman natures change their orientation over time and space.

Thinking through capitalist valuation beyond commodities and in multispecies contexts is helpful in better understanding the possibilities for life and death among salmon, sea lice and lumpfish. As Collard and Dempsey (2017a: 78) write, “What our typology suggests is that to accumulate capital, capitalism needs the diverse materials and creative forces of natures ordered in a variety of positions within society, not just as commodities. No such position is without violence and exploitation”. We add that applying Collard and Dempsey's approach to multispecies relations opens up possibilities to see how valuation and devaluation of one species implicates others in capitalist accumulation.

Highlights

• The processes of valuing and devaluing salmon reflect competing but relational capitalist interests.

• Valuing and devaluing salmon implicates other nonhuman natures, specifically lumpfish and sea lice, through capitalism's value relation.

• Capitalist differentiation produces violence through and outside commodification, with grave consequences for salmon, sea lice and lumpfish.