Relations of Power and Nonhuman Agency: Critical Theory, Clever Hans, and Other Stories of Horses and Humans

Critical Theory pioneered the theorization of human-animal relations, helping establish that agency extends beyond the human world. Nonhuman agency is now widely accepted within the “new materialisms” and beyond but there are growing calls for more critical approaches that consider why and how such agency is mobilized. These calls effectively bring together the concerns of “old” and “new” materialisms. I therefore return to Critical Theory, bringing its explanatory analysis, practical framework and future imaginary into conversation with more recent research into nonhuman agency. Together, they reveal how the relations between human and nonhuman actors shape and are shaped by their broader socio-political context. I suggest that paying closer attention to human-domestic animal relations in particular might help us resolve some of the issues at stake. Consequently, for illustrations to extend the analysis, the article turns to horses, a nonhuman group who occupy a unique place in our collective unconscious.


Introduction
Long before the contemporary "animal turn," Critical Theory had begun to theorize industrial society's troubling relationship with its nonhuman members (Gunderson 2014). The Frankfurt theorists-particularly Max Horkheimer and Theodor Adorno-were the first to highlight how parallel and intertwined social processes effectively marginalize both animals and specific groups of people such as women, ethnic minorities, and workers. Similarly, both the first and second generation of Critical Theorists acknowledged the agency and subjectivity of nonhuman animals (Gunderson 2014(Gunderson , 2017. These ideas have been somewhat neglected by contemporary humananimal scholars, for reasons explained below. Nonetheless, this dual focus-on power relations and nonhuman agency-renders Critical Theory of potential relevance not only to human-animal relations but also to wider contemporary debates about "new materialisms." The present article therefore builds on Ryan Gunderson's analysis by exploring how the Frankfurt School's theorization of the animal question might in turn contribute to a more critical understanding of how other nonhuman entities also act in meaningful ways. Growing recognition and acknowledgment of multiple more-than-human agencies challenges us to extend our thinking to "beings, things, and objects previously ignored as active agents" (Dowling, Lloyd, and Suchet-Pearson 2017: 824). Research into these "new materialisms" is philosophically and methodologically diverse, but shares a common recognition that agency is not limited to human beings. Bruno Latour's (2007) Actor Network Theory (ANT) is particularly influential, but researchers also draw on the work of Karen Barad (2007), Karen Cerulo (2009Cerulo ( , 2011, Diana Coole and Samantha Frost (2010), Tim Ingold (2008), and Sarah Whatmore (2002) among others. ANT came out of Science and Technology Studies in the early 1980s, promising a way to rethink the social world and the place of people, animals, and other actors within it. Rather than a theory, ANT is an epistemological position that "offers a novel view both of social interaction and of those who can legitimately participate in it" (Cerulo 2009:534). In common with the other approaches mentioned above, ANT stresses that "the social" does not exist prior to interaction, but rather emerges through interactions between diverse "actants" including humans, animals, objects, ideas, and technology. It is therefore of particular relevance and usefulness to those whose focus lies beyond the human world as it denies any "a priori ontological assumption of human superiority" (Taylor 2011:212). The alternative accounts of agency and ontology that this kind of approach engenders effectively promise to alter our categories and concepts fundamentally. In shifting our attention from individual actors and entities to the relationships between them, they allow us to explore how everything from politicians to mice to fungi play an active and meaningful role in the making of our world. Inherently political in reach, then, these new materialisms can help us question both the politics of nature (Latour 2004) and the politics of culture (Latimer and Birke 2009).
Martin Arboleda (2017) suggests the underlying premise of new materialisms is now selfevident and he urges us to move from tautologous accounts of nonhuman agency toward more critical approaches. Thomas Lemke (2018)

makes a similar observation:
It is not sufficient to celebrate the move from dead and passive to vibrant and active matter; we need an analysis of how matter is differentially mobilized and to what ends. (P. 45) The present paper responds to this call by exploring how Critical Theory's analysis of humananimal relations contributes to these contemporary discussions about other more-than-human agencies. Specifically, I examine how one particular group of nonhuman actors-horsesengages with others and how their agency intersects with wider relations of place and power. That is, I aim to bring together "old" and "new" materialisms in a way that might usefully add to our understanding of both.
A more critical approach to new materialisms requires that we acknowledge agency not as a property of individual entities, but as an affective relationship between them (Despret 2013a;Devellennes and Dillet 2018;Lemke 2018). Evidence of the interconnectedness and inseparability of diverse forms of life abound (Hovorka 2019). For example, Owain Jones (2014) draws our attention to trees, noting how their relationships of affective exchange with other agents like people and technology render them "lead players" in the city and beyond. Through these coconstructed relations, trees become specific active makers of place, often defying, exploiting, or benefiting from human schemes. Similarly, Anna Tsing (2015) focuses on how the global trade in matsutake mushrooms is underpinned by "patterns of unintentional coordination" between multiple actors (p. 23). The mushrooms attach themselves to the roots of pine trees in particular. They grow in abundance in forests that have been disturbed by human activity, enabling forests to flourish in otherwise damaged places. The mushrooms command astronomical prices in Japan, where industrial expansion-together with an invasion of nematodes-has decimated the pine forests and their accompanying matsutake colonies. The mushrooms are now increasingly harvested across China, Europe, and North America instead. For Tsing, this tale of multispecies cohabitation demonstrates the possibility of collaborative survival in a time of massive human destruction.
This shift away from those doing the relating (i.e. the human subject who has the capacity to act intentionally) toward the relational webs and practices that connect human and other actors has been enthusiastically embraced by many scholars in the field of human animal studies in particular (Wilkie 2015). Kenneth Shapiro and Margo DeMello's (2010) insightful overview of that field explain why this might be so. Human animal studies originated in philosophy, which was concerned with questions about why and how we value animals (e.g. Regan 1983;Singer 1975). This gradually gave way to more empirical research. That is, across other disciplinesincluding geography, anthropology, and sociology-scholars began to explore how adding animals into the analysis might extend our understanding of concepts like alienation (Benton 1998) and violence (Cudworth 2015). New materialist approaches, in turn, allow us to further explore the seam between such concepts and the wider relationships on which they rest (Shapiro and DeMello 2010).
For example, in their study of dairy cows and milking technologies, Lewis Holloway and Christopher Bear (2017) explore how emergent technologies are shaped by the agency and subjectivity of cows and people alike. Focusing on the cow's perspective, Holloway and Bear suggest that the technology shapes what they can do and what they experience. But this is also affected by the people who program and operate the machines, and the cows themselves. The consequences for cows are benign but also problematic. Cows gain freedom and choice about when they are milked, but they are more easily disciplined in new technologically mediated ways. Similarly, Jocelyne Porcher and Tiphaine Schmitt (2012) suggest that under such conditions, cows do more than function. Rather they invest their intelligence and affects in their work. That is, they respect the rules laid down by the farmer but also have autonomous behaviors. They avoid conflicts, negotiate, and are polite and conciliatory with others. Lindsay Hamilton and Laura Mitchell trace similar processes in their study of Herdwick sheep and shepherds in the U.K. Lake District. They find that sheep, humans, and dogs are embedded in a complex web of relations, markets, and terrains. By focusing on the agency of the animal as well as human members of this particular story, we are able to develop a more sophisticated understanding of what work is and how it creates value. Imbued with ethical, political, and social considerations, this valuation far exceeds a discussion of who controls the means and ends of the labor process: Rather, it takes us into the very politics of species membership (Hamilton and Mitchell 2018).
Vinciane Despret (2013a) notes that we tend to notice animal agency only when it manifests as resistance. Similarly, Fudge (2017) suggests we dismiss animals' cooperation as mindless. To do otherwise is profoundly unsettling, as it challenges our assumptions about what we might expect of them and ourselves since there can be no agency that is not shared with others. Alongside practical improvements, then, Fudge says we need to engage in a process of "re-enchantment," through which we will recognize that the experiences of cows (and indeed horses and other animals) can be as varied and interesting as our own.
So far, we have briefly considered the agency of trees, fungi. and farm animals. However, our entangled relationships with domestic animals are perhaps especially revealing. That is, their shared evolution with humans means they help make us who we are (Haraway 2008). Cats settled near Neolithic human villages and their accompanying rodent populations, while dogs have provided security and companionship throughout the ages. Horses, in particular, hold a unique place in our collective unconscious. The act of riding opens up distinctive forms of interspecies communication and partnership, enabling humans and horses to connect physically and share spatial, geographical, and emotional trajectories (Dashper 2016;Game 2001;Thompson 2011). Therefore, it is not surprising that horses have "inspired the arts, revolutionized warfare, shaped societies, and conquered continents" (Notzke 2013:402). Our long and intimate association in turn "reveals and illuminates important and symbolic societal transformations" that have unfolded within industrial society and affect both horses and humans (Adelman and Thompson 2017:3). With their overlapping micro and macro significance, then, horse-human relations are especially useful in developing a more critical understanding of the human and nonhuman agencies that underpin the new materialisms.
So let us return now to Critical Theory. As outlined above, the Frankfurt theorists were the first to highlight the embeddedness of animals within the larger institutional framework of industrial society (Driessen 2014;Gunderson 2014). In so doing, they drew attention to the overlap between the exploitation of humans and the domination of other animals. Critical Theory thus offers useful insight into the relationships underpinning human and nonhuman agency, and how these are created and constrained within wider structures of power.
As Ryan Gunderson's (2014) work makes clear, the diverse theories that emerged from the Frankfurt School are linked by their explanatory, practical, and normative ambitions. That is, they explain what is wrong with current social reality, distinguish the actors and forces involved, and identify achievable practical goals for social transformation (Bohman 2016). Critical Theory therefore helps new materialisms take animals seriously in three ways. First, it reveals how animals (and people) are engaged within hierarchical relations and the implications of this for individual animals or animal groups. That is, Critical Theory enables us to understand the marginalization of domestic animals in rich countries like the U.K. and its wider significance. Second, Critical Theory provides a practical framework through which we might begin to challenge the marginalization of particular individuals or groups of nonhuman actors. Third, the normative ambitions of Critical Theory promise a new political imaginary that can advance the idea of this broader, more inclusive moral community.
Given the above, it may seem surprising that Critical Theory has not featured more prominently in sociological debates about the status and agency of animals (and other nonhumans). Gunderson's work (cited throughout this paper) is of course a notable exception (see also Driessen 2014;Meijer 2013Meijer , 2017Whitworth 2000). In part, this may be a straightforward issue of its unpalatability. Critics suggest Critical Theory comprises a "most unhappy literature full of third-person passive voices holding forth with 'Big Thoughts' on 'Big Issues'" (Putnam et al. 1993:229). However, the absence of Critical Theory within human-animal research is also understandable in philosophical terms. At the heart of the microsociological tradition, particularly in the United States, is a focus on social interaction (Cerulo 2009). Traditionally, this was assumed to depend on capabilities possessed only by humans, such as intention, self-identity, and-perhaps most significantly for the present paper-communication via language. Cerulo (2009) notes that new materialisms have been decisive in bringing nonhumans into social interaction, via their emphasis on "action" over "mind." As a result, many scholars are now carving out a more central role for animals and other nonhumans in their analyses. But in so doing, they have not (yet) turned to Critical Theory, not least because-as discussed later-the "communicative turn" within the Frankfurt tradition firmly excluded animals from the analysis and has not yet allowed them back in.
The present article therefore suggests that animals are useful in two ways. First, they bring theoretical insight into an increasingly mainstream concern-namely the material and problematic relationships between the human and nonhuman. Second, and perhaps just as importantly, they also promise a reassuring tangibility, an appealing element of slapstick and anarchy (Mathews 2007;Swart 2007). A secondary aim of this article, then, is to join efforts to rehabilitate Critical Theory among a wider audience. Taking it out into the field, literally, we encounter a variety of equine actors and their human companions: Valuable and much-loved competition horses who live a privileged but lonely life; a grumpy mare who keeps her younger fieldmates on the straight and narrow; and the legendary Clever Hans with his apparent aptitude for arithmetic. Academic analysis often overlooks or devalues domestic animals and our relations with them (Kheel 2008;Swart 2007). However, as Jurgen Habermas (1993) himself points out, their problematic status is especially helpful in discussing human-nonhuman agencies: "We communicate with animals in a different way once we involve them in our social interactions, in however asymmetrical a fashion" (p. 106). Thus, while drawing our attention to individual beings and their concerns, horses and other domestic animals enable us to understand the larger socio-economic context.
While the paper is largely theoretical in nature, then, several examples of horse-human relationships illustrate and extend the analysis. Some of these stories are drawn from existing research (e.g. Despret 2004;Schuurman and Franklin 2015;Zetterqvist and Lundgren 2017). Others come from my own fieldwork within post-productive rural communities in the north of England. This focuses on how particular groups of people and horses experience and sometimes resist their relocation from the center of relations of production to the margins of consumption and leisure. This wider project can be understood as an extended case study, in that it aims to examine macrolevel questions through their everyday manifestations on the ground (Burawoy 1998). It also heeds Kendra Coulter's (2018) call for more animal-centric research approaches. That is, I recognize people and horses alike as subjects, objects, and agents in their own stories (Hribal 2007;Swart 2007). Still ongoing, fieldwork effectively "follows the horses," to use Christoph Lange's (2018) expression. Ethnographic data are drawn from historical and contemporary documents, as well as interviews with people who share their lives with horses, as riders, breeders, or business owners. Participant observation-at stables, equestrian competitions, horse auctions, and elsewhere-focuses on human and equine participants alike. The tales told here came up when visiting participants' yards, at organized gatherings of local riders, or beside the ever-popular water jump at a cross-country event. Other stories were experienced more directly: I watch, listen, and join in as people and horses wait anxiously for the vet to arrive or share an exhilarating gallop across a recently harvested field. In bringing Critical Theory to bear on this particular ethnographic material, then, my wider research served as a starting point for both theoretical insights and practical illustrations for the present paper.
In summary, this article responds both to Gunderson's (2014) invitation to take the animal question seriously and to wider calls to adopt a more critical approach to human-nonhuman agency (e.g. Arboleda 2017; Lemke 2018). Specifically, it explores how the relations between one particular subset of human and nonhuman actors are embedded within their broader socio-political context. As such, it aims to bring together "old" and "new" materialisms, discussing each in turn before considering how paying closer attention to human-domestic animal relations in particular might help resolve some of the issues at stake. Hilary Rose (1998) suggests these relations lie at the very epicenter of industrial society's concerns about animals. It is therefore not an esoteric preoccupation to bring them fully and theoretically into our social analysis in this way, but a necessary and mainstream endeavor (Gunderson 2014;Stuart, Schewe, and Gunderson 2013). The article is structured as follows. First, drawing explicitly on Gunderson (2014), I set out how the early Critical Theorists (i.e. Horkheimer and Adorno) provide an explanatory analysis of animal oppression. I then show how new materialisms help reconcile animals more fully into their analysis. A focus on domestic animals in particular emphasizes how nonhumans can be political as well as social actors, thereby introducing the possibility of challenging that oppression. Second, I discuss how Critical Theory offers a practical framework for analysis, specifically via Habermas' ideas about communication as a source of mutual understanding and change. I then explore how new materialisms usefully complement these ideas by extending them to nonhumans. A focus on domestic animals in particular enables us to reflect on how we might more effectively listen to what nonhumans are "telling" us. Finally, the different strands of Critical Theory-together with recent new materialist scholarship-promise an alternative normative imaginary and an accompanying transformation in human-animal relationships. The article suggests that while inequalities of power may be inescapable, they are not necessarily fatal to our relations with horses and other animals.

Overlapping Forms of Domination: How Critical Theory Explains the Oppression of Domestic Animals
The early Critical Theorists place human-animal relationships at the heart of their analysis of industrial society (Gunderson 2014). They viewed the distinction between human and animal as conceptually foundational to Western, post-Enlightenment thought: The idea of man [sic] in European history is expressed in the way in which he is distinguished from the animal. Animal irrationality is adduced as proof of human dignity. (Horkheimer and Adorno 1969:245) This distinction predates the Industrial Revolution, going back to the Ancient Greeks and beyond (Plumwood 1993). In modern society, however, it culminates in a brutality against animals that is unprecedented in its scale and ubiquity (Adorno 1974;Horkheimer and Adorno 1969). Continued technical and economic progress depends upon the ever more efficient harnessing of nature (Adorno 1974). People seek to learn from nature and its inhabitants only that which will enable them to continue to dominate both the "natural world" and other people (Horkheimer and Adorno 1969). Under capitalism, then, an instrumental approach to "irrational" animals paradoxically enslaves humans too (Gunderson 2014). The early Critical Theorists thereby highlighted that the distinction between humans and animals has had very real material consequences for both. In so doing, they effectively anticipated a fundamental tenet of critical animal studies many decades later, namely that "the fight for the animal . . . is a fight for man" (Horkheimer 2007:228; see also Gunderson 2014).
Within Critical Theory, the "dialectic" between human and nonhuman replaces class struggle as the key to understanding industrial society: The problem with society lay not merely with its relations of production [as asserted by Marx] but its entire foundation on the "rationalizing" premises of the Enlightenment, the appliance of principles of reason and scientific method . . . Both "inner" (human) nature and "outer" (non-human) nature were subordinate to these principles. (Whitworth 2000:146) Any potential solution therefore requires some kind of reconciliation with nature and its nonhuman inhabitants. However, the pernicious and persistent presence of instrumental reason across every corner of modern society leads Horkheimer and Adorno to suggest that such a reconciliation is impossible (Whitworth 2000).
In summary, the discussion so far suggests that Critical Theory pioneered the theorizing of human-animal relationships. While linking the domination of animals in industrial society to our own, however, it was pessimistic about the possibility of overcoming the rationalizing premises that estranged humans from these nonhuman "others." Animals were therefore effectively relegated to the theoretical margins, before being sidelined altogether by the "second generation," as discussed later. This helps explain the lack of engagement to date of new materialists and humananimal scholars with their Frankfurt forebears. We now discuss how such an integration might strengthen Critical Theory's analysis, adding also to our understanding of human and nonhuman agency.
Focusing expressly on nonhuman others, and the worlds they share with humans, the usefulness of new materialisms to this analysis is immediately clear. Research sites comprise interactions between myriad human and nonhuman actors, embedded in turn within broader socio-political networks. All actors-human and nonhuman-are active entities, worthy of consideration: They "matter" precisely because of their relationships with others (Latour 2007; see also Barad 2007;Cerulo 2011;Ingold 2008;Law and Mol 2002;Whatmore 2002). Such "relatings" can take the form of everything from a polite introduction between "two quasi-individuated beings" to complex patterns "more reminiscent of a cat's cradle" (Haraway 2008:41). According to this view, the individual actor cannot be the point of origin but "can only share in the action, distribute it with other actants" (Latour 1996:237). For any actor to act, others must act as well. Or as Barad (2007) suggests, agency is not something that someone or something has, but is "a matter of intra-acting" (p. 178). Thus, human and nonhuman participants alike enroll each other into an ongoing program of action that leads toward a specific outcome. For example, the discovery of pasteurization is reinterpreted as a joint enterprise of rats, bacteria, industrialists, and worms among others (Latour 1988). In the context of human-animal relations, Taylor (2011) suggests this emphasis on the relations between actors rather than the actors themselves is useful in two ways. First, it takes us away from perennial arguments about who (or what) is conceived as an actor. Rather, animals, people, and inanimate objects alike are co-constitutive in those relationships. Second, then, "it is an approach which sees all perspectives as equal-as tied together in networks" (Taylor 2011:210). That is, by focusing on the performative and emergent nature of these relationships, new materialisms overcome our analytical anthropocentrism by creating the possibility of a level playing field for all things human and nonhuman.
The extent of humanity's co-evolution with domestic animals renders those relations especially revealing. Humans and horses, for example, mutually construct their relations with each other. Lynda Birke and Kirrilly  align with Thomas Lemke (2018) and Donna Haraway (2008), in suggesting horses are not agents or actors simply by virtue of "being there" in (our) networks. That is, agency is not a given but is the outcome of relations with multiple others enacted through numerous roles (Hobson 2007). Interspecies relations are thereby simultaneously structural and personal: "When we encounter a horse, we meet someone [with] their own experiences of life, their own expectations, their own subjectivity" (Birke and Thompson 2017:44). As a specific individual and a class of such individuals, animals hold species qualities that predispose them to a particular way of being or behaving, but this is also influenced by the agency of the individual animal (Haraway 2008). For example, horses are herd animals primed for threat (Birke and Thompson 2017). When Rosa, the herd leader, is separated from her fieldmates, she stands quiet but alert, watching the horizon until they come home safe. By contrast, her diminutive friend Bea is more likely to canter round hollering indignantly. Youngster Breeze is no longer left alone since she took to hurdling the fence, much to the delight of the elderly Welsh pony next door. Marti Kheel (2008) suggests focusing on domestic animals like horses in this way enables us to place individual beings-and our feelings of care and empathy for themwithin their larger historical and contemporary context. More widely, Lemke (2018) suggests that this reappraisal of matter is intrinsically connected to a rethinking of the composition of the political collective, and the question of how nonhumans shape and govern political practices and social conduct. He adds that a radical "new" politics requires that we move beyond seeing agency as a property of individual entities, since this will inevitably give undue credit to humans rather than nonhumans. Like Despret (2013a) above, he instead advocates a relational approach, through which we can understand how these different materialities work in concert.
If we focus, by contrast, on what Lemke (2018) calls their "modes of doing," it becomes clearer how horses' individual agency and generic collective qualities are embedded within wider relations of space and place. At a nearby competition venue, valuable showjumpers are kept individually in neat, rectangular paddocks. Owner Jane says, "I wouldn't have it any other way! I'd never have Otto in a herd situation, it's just too much risk." Otto and his neighbors are protected from accidental injury, but they are also unable to groom each other or take turns to watch out for long-vanished predators. This set-up is too much for others. Minnie's behavior became so erratic-crashing through fences, rearing under saddle-that her owner feared she had developed a brain tumor. The symptoms disappeared when she moved to a yard where everyone lived in one big group. As suggested by Despret (2013a) above, Minnie's agency only became visible when she was driven to resist her confined living arrangements. They might not communicate their wishes as directly as Minnie, but perhaps we should also "wonder about" Minnie's erstwhile neighbors (Fudge 2017). They too might prefer to take their chances with a clumsy companion or two, but this option is not open to them because of their economic and/or emotional value to their human owners.
As individuals and classes of individuals, then, horses like Minnie and Otto are embedded within wider human and nonhuman relations. Birke and Thompson (2017) suggest this means we need to think in terms of accountability, or "what's in it for the horse?" (p. 138). That is, in pursuing interspecies engagement, we should also question, "is it enough that the horse can ask, only to be told no? (Birke and Thompson 2017:139). Kennan Ferguson (2004) suggests this is an important question for people too, as it potentially challenges the presuppositions and causalities that underpin our human-centric theories. Rather, in order to be useful, any theory must recognize and accept important human affinities with domestic animals in particular. Otherwise-Ferguson says-it will fail empirically and ideologically. Combining Critical Theory with new materialist approaches takes us closer, but we need an additional step in order to achieve the radical "new" politics sought by Lemke and others. Sue Donaldson and Will Kymlicka's (2011) "zoopolis" is helpful here. While they do not reference materialisms old or new, they suggest similarly that a narrowly defined focus on rationality has marginalized particular groups (both animal and human). However, unlike the Critical Theorists, they are able to overcome the difficulties of contemplating the potential emancipation of animals. Donaldson and Kymlicka use the concept of dependent agency to suggest that individuals who cannot speak will make their views known in other ways to those who are willing to listen. This in turn allows us to see historically subordinated animals (and people) as subjective individuals rather than objective groups, who can participate in shaping and sustaining larger cooperative schemes. That is, we both acknowledge their "relational agency" and render them visible (Despret 2013a). Specifically, the very process of domestication presupposes and reinforces three key criteria that underpin political interaction. For just one example of how horses meet these criteria, we can visit Minnie at her new home, a typical "do-it-yourself" livery yard. At fetching-in time, the herd express their subjective goodtheir desire to trade a waterlogged field for a dry stable and a full haynet-each time they drift toward the gateway come mid-afternoon. They also illustrate how participants comply with social norms. As a scrum ensues, Minnie flattens her ears momentarily and the jostling youngsters alongside grudgingly allow her to step forward first. Finally, these same horses are constantly shaping the terms of their interactions with the human and nonhuman members of their group. Last to clatter into his stable, Minnie's neighbor Freddy immediately flops onto the shavings, flailing his legs and scattering the immaculate bed in all directions. His new owner laughs at his apparent exuberance. Her instructor proffers a less palatable explanation, suggesting that Freddy has developed an effective and potentially dangerous method of ejecting her from his personal space: Either way Freddy is clearly exerting his control over events.
In summary, new materialist perspectives illustrate the active role of domestic animals in shaping our common world(s). By drawing on Critical Theory and more recent political theory, we can place these efforts within a wider context. This suggests we can extend concepts like agency and citizenship to animals, but also that we have a moral and political duty to do so. This enriches rather than undermines the concepts in question (Birke and Thompson 2017;Donaldson and Kymlicka 2011;Ferguson 2004). It also requires new forms of representation and institutions (Meijer 2013). This is where we can usefully return to the ideas of the Frankfurt School, namely the "second generation" and its focus on communication and deliberation. If we can find a way to listen to and act upon-in practice-what domestic animals are telling us, then we might enable Lemke's (2018) "new" forms of politics to emerge after all.

Emancipation through Communication: How Critical Theory Provides a Practical Framework to Challenge the Marginalization of Domestic Animals
For the early Critical Theorists, the assumption that humankind is radically different from other animals was fundamental to the process of Enlightenment rationalization and is therefore theoretically insurmountable. By contrast, the "second generation" suggests this rationalizing process has not gone far enough. According to its leading figure, Jurgen Habermas (1981Habermas ( , 1987, further enlightenment-in the form of "communicative action"-is the only way to overcome the alienation that separates us from each other and from nonhuman nature. In contrast to "strategic action" aimed at influencing specific outcomes, communicative action enables participants to build their mutual understanding of the issues they face through rational dialogue. Two features are of particular interest here. First, drawing on the work of Herbert Mead, communicative action assumes a paradigmatic shift from the atomistic self of Cartesian consciousness to an ego that retains an "intersubjective core" that emerges from our interactions with others. The result is a focus on the process and results of communication between subjects, rather than the subjects themselves. This points us toward the second aspect of interest, namely that communicative action is a messy, socially embedded encounter that-crucially-transforms participants on both sides. Who we are is subject to constant reformulation as the identity of the individual and that of the collective are interdependent, forming and maintaining themselves together (Habermas 1987).
Communicative action facilitates a more emancipated society in which survival increasingly depends on social cooperation rather than technical control (Moss and Pavesich 2011). It might therefore provide a practical framework through which to challenge the marginalization of domestic animals and facilitate a "new" form of politics. Indeed, Jurgen Habermas (1982) specifically highlights that this emancipated society should adopt a more enlightened approach to its animal inhabitants by effectively extending our neighborly circle beyond the human. However, he simultaneously denies nonhumans a more active role in that society, when he suggests that "what raises us out of nature is the only thing whose nature we can know: language" (Habermas 1972:314). Since animals cannot grasp the basic structures and fundamental rules of language, "we cannot come to an understanding [with them] about something in the world" (Habermas 1993:109). Habermas' narrow focus on language thereby relegates to the theoretical sidelines anyone who cannot talk. However, the equine participants we have met so far are reluctant to remain on the bench. For a start, horses draw on a vast range of vocalizations that sound very much like language even to an untrained human ear: They whicker when dinner arrives; neigh when a familiar car pulls up; and call out anxiously when separated. It would be faintly absurd to suggest that these examples of equine speech embody the grammatical structures and appeals to validity that Habermas had in mind. Nonetheless, it is perhaps equally peculiar to hold fast to such a limited definition of language. For example, in partnership with their riders, the "dancing" dressage horses at the Olympic Games perform complicated physical feats based on completely silent and largely imperceptible communicative acts (Zetterqvist and Lundgren 2017). A focus on (human) language risks overlooking this "vast repertoire of gestures, movements and signals" that horses and other domestic animals rely on to tell us-and each other-how they feel, what they are about to do, and what they want (Donaldson and Kymlicka 2011:109). By simply extending Habermas' framework to animals, we risk creating a hierarchy that overlooks these less obvious forms of communication (Driessen 2014). Indeed, Habermas (1993) himself changes his mind, later observing that when animals "participate in our interactions . . . it is of the same kind as an intersubjective relation" and we should therefore not just communicate about animals but with them (p. 110).
Eva Meijer (2013) suggests that "seeing animals as mute does not simply reflect a misunderstanding of their capacities: It is interconnected with the way humans have defined language and politics and has led to rendering animals silent as a political group" (p. 28). As Fudge (2017) says above, shifting that understanding represents a fundamental challenge to our understanding of human and nonhuman animals alike. This is a challenge that Habermas appears to acknowledge in his later work, when he goes some way to expanding his analysis beyond the human world and acknowledges animals as social and moral actors (Habermas 1993). John Dryzek (1985) prefigures this move, as well as that of new materialisms, in suggesting that it is indeed possible to extend rationality to nonhumans: This recognition of agency in nature means that we should treat signals emanating from the natural world with the same respect we accord signals emanating from human subjects, and as requiring equally careful interpretation . . . Thus communicative interaction with the natural world can and should be an eminently rational affair. (P. 20) According to Dryzek, we need to dismantle the barriers that may inhibit such rational, interspecies engagement. Robyn Eckersley (1990) is more explicit about what should be done: "The fact that the nonhuman world cannot participate in human speech should be no barrier to their special interests always being considered and respected by those who can participate in the dialogue" (p. 761). One possible route is representing animals and their interests via human advocates. For example, in their work on wild boars, Erica Von Essen and Michael Allen (2016) suggest human proxies might "approximate and extrapolate from [the boars'] perspectives in order to determine and promote their best interests when making contestations of current policy on their behalf" (p. 74). Advocacy on behalf of animals is certainly a useful approach. However, speaking and acting "on behalf of" animals inevitably means political engagement at one remove. Even if we overcome our theoretical reservations to a more direct approach, the practical difficulties of bringing animals "into the room" will remain. An alternative is to think, conversely, how "we" might go out into the field: Instead of only occurring as part of institutionalized and formalized settings, more mundane and nondiscursive forms of deliberation can be part of material political processes too, at least when participants engage in communicative action oriented to understand the positions of others and building trust. (Driessen 2014:95) That is, deliberation is not only about making (formal) decisions but arriving at some (informal) arrangement of mutual truth, consensus, or cohabitation. This can take place anywhere that humans and animals meet. These everyday situations enable us to build sympathetic relations with animals, which generate knowledge and ethical insights (Gunderson 2017). For example, in their analysis of "natural horsemanship," Nora Schuurman and Alex Franklin (2015) illustrate how expertise is effectively co-constructed between human and equine participants. The trainer's status as "expert" depends in large part on the counterperformance of the horse, and the way this is negotiated by both parties. Horses and people alike thereby become active participants in the communicative world around them: Or, as Latour (2007) says, "all the actors do something and don't just sit there" (p. 128). New materialisms thus promise a radically different epistemological starting point for examining these relations, one that is nonhierarchical and accepts that "knowledge is not abstract, rather it is embodied and enacted and emerges from the interactions of various objects within networks" (Taylor 2011:217). In a later article, Schuurman and Franklin (2018) draw on John Law and Annemarie Mol's (2002) notion of tinkering to describe these individualized forms of embodied practice. Similarly, Steve Hinchliffe et al. (2005) describe this as a kind of fermentation or experimentation. Like Habermas, then, these writers prioritize the intersubjective relationships between actors. However, by extending their focus beyond the human, they show how animals, people, sites, and situations are affected by each other through their engagement in specific spaces and places. Through our emotional and psychological responses, we are engaged in a process of mutual becoming (Lorimer 2007;Thrift 2004), or a "collective endeavour" (Hinchliffe et al. 2005:652).
A colorful illustration is provided by the story of German gelding Clever Hans (Despret 2004). He became an equine celebrity and scientific curiosity in the early twentieth century, tapping out the answers to mathematical puzzles with his hoof. Extensive investigations revealed that Hans was "merely" responding to the body language of his questioners, who would unconsciously tense their muscles until he produced the correct answer. Although Hans could not count, he was doing something much more interesting: He was making human bodies be affected, even without their owners' knowledge: Who influences and who is influenced, in this story, are questions that can no longer receive a clear answer. Both, human and horse, are cause and effect of each other's movements. Both induce and are induced, affect and are affected. (Despret 2004:115) Learning how to address animals is not the result of scientific or theoretical understanding, it is the condition or starting point of this understanding (Despret 2004(Despret , 2013. More widely, it is out of these individualized forms of embodied practice, within complex and diverse sites and situations, that shared forms of life emerge (Haraway 2008;Law and Mol 2002). Only by taking risks and allowing animals and others "to object to the stories we tell about them, to intervene in our processes as much as we intervene in theirs . . . can we hope to learn how things matter to humans and nonhumans" (Hinchliffe et al. 2005:655-56).
To synthesize, Habermas offers an appealing framework through which we might challenge the marginalization of animals, but his narrow definition of language preemptively excludes them from his analysis. Together, political theory and new materialist perspectives enable us to extend to domestic animals Habermas' emphasis on communication as a route to emancipation. Clemens Driessen (2014) suggests that communicative action emerges not only within institutionalized settings but also in the unremarkable, everyday deliberative encounters that unfold between people and animals. However, a new form of multispecies politics requires that we increase our receptivity (Dobson 2010): Rather than making animals speak, we should listen harder to what they already have to say. Together, via an extensive communicative repertoire, we can build trusting relations aimed at enabling consensus or cohabitation. It becomes clear that these interactions are material processes that depend upon context and entail embodiment, affect and experimentation on all sides (Argent 2012;Game 2001). This relational approach differs from bargaining or negotiation in that "identities, ideas and desires are not static input to the process . . . but are themselves open for revision and at stake" (Driessen 2014:101). Horses, humans, and others attach and detach to different others, in what Joanna Latimer (2013) calls a process of "being alongside." We might thereby redefine the very process of domestication as an emotional relationship, in which animal and human alike are effectively constructed by their mutual expectations of and faith in each other (Despret 2004(Despret , 2013a. Kendra Coulter (2016) is among those reflecting on the emancipatory potential of this kind of interspecies engagement: "someone does not need to be 'the same' as you in order for you to feel and foster solidarity" with them (p. 150). It is to this political project that we now turn.

Transforming Human-Animal Relationships: How Critical Theory Promises an Alternative Political Imaginary
Critical Theory always has a goal in mind, namely that of emancipation. Habermas is clearly more optimistic about the potential for transforming industrial societies than Horkheimer and Adorno. He suggests that through rational and transparent dialogue we can build mutual understanding of what is and what ought to be. Hannah Arendt (1970) points out that communicative power is itself a form of solidarity: By engaging with one another in this way, individuals become aware of and consolidate co-membership in a collective form of life. Indeed, Habermas' later work (e.g. 1996) focuses on how individual communicative encounters in turn might facilitate collective deliberation about what kind of world we want to live in. As discussed above, this work also raises the possibility of expanding this analysis to domestic animals. He recognizes them as vulnerable members of human society, with whom we should communicate in order to understand their interests (Gunderson 2017;Habermas 1993). By bringing Critical Theory into conversation with wider political theory and new materialist perspectives, we have demonstrated that domestic animals participate in deliberative processes when given the opportunity to do so. Horses and others as engage in open dialogue with humans and others through embodied and affective forms of communication (Despret 2004(Despret , 2013a(Despret , 2013b. The consensus and mutual understanding that can emerge from these trusting relations potentially enable us to challenge the marginalization of domestic animals within industrial society at both the individual and collective level. By extending Critical Theory in this way, we can potentially put forward alternative visions of what a "good life" might look like for humans, horses, and perhaps other animals too. These visions will be multiple, but are likely to be based on appeals to greater justice and solidarity. For example, Coulter (2016) suggests that alienation and exploitation unfold across species lines, with particular groups of animals and people harmed and oppressed by similar organizations, ideas, and processes. From riding school to racing stable, physical proximity, demanding conditions and the emotional bond that develops between horses and the humans who care for them-usually young, female, and/or migrant workers-results in their shared suffering (Miller 2013;Porcher 2011). Horses and human carers alike work hard from a young age and risk catastrophic injury. Adopting a bottom-up focus on species, as we might on gender or class, enables us to generate a "horse-story" to accompany that of a "her-story or a his-story" (Swart 2007:154). In turning to the emancipatory potential of communicative approaches for both domestic animals and humans, however, we need to pay close attention to agency, place and power and the way they intersect for animals individually and collectively (Haraway 2008).
Both Habermas and new materialists (notably Latour) are reproached for downplaying relations of power and place. For example, those who appear to be sniping irrationally from the sidelines of a given Habermasian encounter may indeed be acting rationally, in refusing to engage with the power relations they detect (Russell and Montin 2015). New materialisms in turn deliberately remove the "'hierarchical' view that pervades social science thinking" by arguing that humans, animals, and objects are equally worthy of consideration (Taylor 2011:215). Power is not absent, but rests within the relations between agents rather than within the agents themselves (Latour 1988). As Alice Hovorka (2019) notes, "while we and others have always been networked . . . we and others have also always been ordered" (p. 752). Hovorka adds that relational webs-whether human, nonhuman, or both-are characterized by hierarchies of position and connectivity. Nonetheless, critics suggest that this approach obscures the very existence of enduring structures and relations of power: "If we do not know which actors are more important than others then we deny ourselves the ability to intervene in the hope of altering the existing balance of forces" (Choat 2018(Choat :1036. While ANT is not a moral theory but an analytical approach, it can potentially enable sustained social and ethical critique by "unpacking that which has been simplified or buried" (Williams- Jones and Graham 2003:290; see also Cerulo 2009 andTaylor 2011). This confirms the usefulness of adopting a critical approach to new materialisms, as advocated throughout this paper. For example, Latimer (2013) suggests that overlaying the "flat ontology" of such approaches with the "ethics of emancipation" enables us to achieve a more stratified view of social reality that acknowledges the asymmetrical relations between humans and domestic animals. Specifically, she characterizes those relations not as "being with" (Haraway 2008) but "being alongside." As Helen Wilson (2017) suggests, "encounter" is not an empty referent for any form of meeting, rather "encounters are meetings where difference is somehow noteworthy" (p. 455). This enables us to sustain regard for both the connections and divisions that characterize human-nonhuman relations, avoiding the otherwise totalizing effect on our less powerful animal partners (Latimer 2013).
Thus, as we reflect on the potential of more deliberative approaches to give rise to a more inclusive moral community, a more species-centric approach enables us to remember that "co-constituted relations may also be those of inequality and oppression" (Cudworth 2010:148). Horse-human relations illustrate this well. At the individual level, despite the mutual embodiment and affect outlined above, the human partner alone can break the relationship at any time by selling the horse to someone else (Birke and Thompson 2017). At the collective level, horses-like people-are subject to uneven social and economic outcomes (Hobson 2007). Recognizing them as political as well as social actors enables us to explore how people and horses, individually and collectively, might resist this exploitation and posit possible alternatives. As above, this requires focusing on agency as an outcome of specific relations, rather than a property in itself (Despret 2013a;Lemke 2018). According to Jason Hribal, for example, individual animal subjects do not "suddenly and without much effort, become actors" (Hribal 2007:102). Rather, they enact their lives, engaging with the power structures around them just as people do. Hribal (2007) draws on James Scott's (1987) notion of the "weapons of the weak" to describe how animals resist their own exploitation within these structures. These weapons are on display within the dusty, echoing confines of an indoor riding school on any given Saturday. Like his friends, Teddy is very effective at Scott's (1987) first tactic of faking ignorance. Ignoring the polite leg behind the girth that asks for canter, Teddy instead trots faster and faster, hurtling inexorably around corners while his rider gamely hangs onto his mane. Likewise, Mary-a deceptively placid bay cob-is well known for taking breaks without permission. Toward the end of lessons, she sometimes turns in, drops to her knees, and rolls in the dust, barely leaving her flustered jockey time to jump clear. By focusing on the actual behavior of individuals and groups in this way, we begin to understand how animal practices are "wholly contextual and relational," embedded within wider structures of power (Whatmore 2002:34). However, more positively, in addition to becoming attuned to their resistance, we need to open up "spaces in which [horses and others] can communicate what kind of world they would like to co-create for themselves, with humans" (Birke and Thompson 2017:136). A new materialist perspective, then, enables us to focus on how horses and others communicate and enact these wishes. Humans and nonhumans alike are enacted and enacting through what Vinciane Despret (2013b) calls their "embodied choreography," which in turn "blurs the clear-cut divide between knowing subject and known object" (p. 69; see also Despret 2013a andFudge 2017).
To summarize, together Critical Theory and new materialisms build our understanding of what a "good life" might look like for humans, horses and perhaps other animals too. Critical Theory points us toward a more emancipated future, while a new materialist perspective suggests progress toward it is likely to be slow and messy, coming about through complex, mundane, and material practices (Law and Mol 2002;Schuurman and Franklin 2018). As elsewhere in the article, horsehuman relationships illustrate and extend the analysis. In this case, they suggest that while we need to acknowledge the relations of power within deliberative encounters between humans and horses (and perhaps other animals), they are not necessarily fatal to our aspirations and the means by which we seek to attain them. Despret's (2004) retelling of the story of Clever Hans and Schuurman and Franklin's (2015) insights into the world of natural horsemanship demonstrate that knowledge is "distributed." That is, knowledge comes about through the way in which animals and people are "attuned" to each other: "Both are active and both are transformed by the availability of the other" (Despret 2004:125). Donna Landry (2011) suggests that these micro-level processes have a broader impact. She traces how a sense of mutual responsibility and service persist in horsehuman relations despite disparities of power. Turning to the distant past of the Ottoman Empire for her analysis, she suggests that among the Sultan's equine and human subjects alike, there was a "dignity in service and in serving willingly" (Landry 2011:19). That is, human-nonhuman relations might be simultaneously hierarchical and ethical. Even within relations of power, then, "justice, fair treatment, and respect for others" are possible (Patton 2003:95). In this sense, bringing together Critical Theory and new materialisms with a focus on horse-human relations enriches both our understanding of oppression and our visions(s) of an alternative future.

Conclusion
The aim of this article was to respond simultaneously to calls to take the animal question more seriously (Gunderson 2014) and to adopt a more critical approach to human and nonhuman agency more widely (Arboleda 2017;Lemke 2018). Specifically, it has shown how the ideas of Horkheimer, Adorno, and Habermas illuminate and challenge the marginalization of domestic animals in rich countries and help us imagine a broader more inclusive moral community for the future. In so doing, the article has explored how human and nonhuman agencies at the micro level intersect with wider relations of place and power. The focus on our entangled relationships with domestic animals enables us to understand how the individual and collective agency of one particular group of nonhuman actors-horses-is constrained within particular spaces and industrial society as a whole and how this might be challenged.
In this way, the article has made two theoretical contributions. First, it has extended new materialist discussions about nonhuman agency by using Critical Theory to show how deliberative relationships both embody domestic animals and those with whom they engage and embed them within wider socio-economic relations. This raises the need for further empirical research that places those nonhuman actors themselves and their points-of-view at the center, perhaps via advances in visual ethnography, multispecies ethnography, or ethology. Second, by focusing on domestic animals in particular, the article has theoretically integrated nonhumans into Critical Theory, thereby helping to paint a clearer picture of society as multispecies. In so doing, the article takes a step toward the radical "new" politics promised by new materialisms by recognizing and accepting the important human affinities with nonhuman others that help make us who we are. As Taylor (2011) suggests, we cannot just apply new materialisms "like a new layer, on top of traditional analyses" (p. 215). Rather we need to start from a fundamentally different point, one that focuses on the processes of relating rather than on those who do the relating. For example, we might usefully explore the way in which knowledge and understanding is distributed between humans and nonhumans not only within individual encounters but more widely. Such research would offer insight into the way in which this "new" politics plays out on the ground.
Throughout the article, then, I have tried to retain a critical perspective that also brings a reassuring tangibility to our understanding of both new materialisms and Critical Theory. That is, in turning and returning to our relations with domestic animals in particular, I hope to have furthered our understanding of what Horkheimer (1933) calls "the unity between us and them" (p. 36)

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) received no financial support for the research, authorship, and/or publication of this article.