Valuing difference: How breed matters for animal lives and relations

Most farmed, working and companion animals come to exist, live and die not only as species in specific human–animal relations but also as members of breeds. Their parentage and birth result from human control of animal reproduction. The nature of their lives and deaths is determined by the function of the breed and, for many animals, by their differentiated value within the breed and potential to reproduce or enhance the breed through their offspring. This is the case whether they live within industrial agricultural systems, less or non-industrialised farming, in traditional pastoralism, or as working or companion animals. Since human–animal relations and encounters, at least with farmed and companion animals, are predominantly between humans and bred animals, breed epistemologies, imaginaries and practices are crucial in shaping the nature of those relations, encounters and animal lives. In this paper, I consider how ‘love of a breed’ (Haraway, 2008), that is, investments in the continued existence, value and vitality of a breed, shapes the lives of animals and their relations with each other and with humans.

In the broadest sense, the question of how breed matters to animal lives could be answered simply. Breeding is a dimension of human domination and subjugation of animals for the products derived from their living or dead bodies, the work they do and the services they provide. Modern agricultural animal breeding underpinned the development of European capitalist modernity and settler colonialism (Woods, 2018). It is integral to the deeply commodified world of companion animal nature cultures and industrial food production. Breeding is also cental to the cultures and livelihoods of traditional livestock keeping in the majority world (FAO, 2009; Köhler-Rollefson and McCorkle, 2004). Ending human control of animal reproduction would entail radical shifts in the diverse economies and cultures that depend upon farmed, working, companion and sporting animals, with differing implications across these domains and geographies. Envisioning better worlds is crucial. At the same time, the significance of ideas of breed for animals in contemporary agricultural, sporting and companion animal domains deserves close scrutiny. My focus on the breeding of horses in Iceland in this paper provides a way of addressing the question of how breed matters in terms of animal lives and relations. This focus is particularly fruitful given the national status of the breed as a whole, the significance of horses within the contemporary equine culture and economy in Iceland (Arnórsson, 1997; Björnsson and Sveinsson, 2006; Sigurðardóttir and Helgadóttir, 2015), the degree of interest in the health of the breed and the well-being of horses within it.

Addressing how breed matters to animal lives and relations means considering breeds as objects of knowledge and imagination; projects of definition, reproduction and differentiation (Derry, 2003; 2014; Grasseni, 2005); and as ‘heterogenous biosocial collectivities’ (Morris and Holloway, 2014; see also Holloway et al., 2009) of breeders, breed organisations and, crucially, the animals who are defined as breed members through their genealogical relation to other living and deceased animals. Breeds are contrived categories of subspecies difference defined morphologically in terms of shared physical attributes and genealogically through pedigree. Foregrounding the animals that constitute breeds as lively and plural collectives entails addressing the relationship between ideas of the individual and the collective in ways, as I will argue, that not only include but also complicate a simple juxtaposition of the breed – as a plural noun that elides individuality – and the singular animal. This paper thus extends geographical research on farmed animal breeds and breeding (Evans and Yarwood, 1998, 2000; Holloway and Morris, 2014; Holloway et al., 2009; Morris and Holloway, 2014; Yarwood and Evans, 1999) and recent efforts to conceptualise breed and breeding more broadly (Guest and Mattfeld, 2018) by bringing it into more direct conversation with work on human–animal relations and interspecies kinship and care (Fuentes and Porter, 2018; Govindrajan, 2018; Haraway, 2008, 2016; Puig de La Bellacasa, 2017). Furthermore, I respond to recent calls to centre attention on horses as subjects in efforts to interrogate the nature and ‘conditions of relational engagement between humans and horses’ (Argent and Vaught, 2022: 4) by addressing the significance of breed to human–horse relations and relations among horses.

This paper draws on English language print and online sources regarding Icelandic horse breeding including breed organisation documents, scientific reports and accounts of Icelandic horse breeding histories and contemporary practices and 23 qualitative semi-structured interviews with horse breeders undertaken in 2016 in south and northwest Iceland, usually on their horse farms, as part of a wider project on horse cultures and interspecies kinship in Iceland. The availability of English language sources reflects the international nature of Icelandic horse breeding (see Nash 2020a) as well as the dominance of English in scientific publishing. The research participants, who are anonymised here, were recruited by contacting them though their websites and following recommendations for potential participants. It is also informed by the wider research project that included observational and interview research with those involved in the culture of leisure and competitive riding and horse tourism in Iceland and oversight of horse welfare.

Two key and entwined concepts emerge from and thread through my engagement with this material. One is value. In this paper, I address how the value attached to potential relations with individual animals – relational value – underpins perspectives on the condition of and care for the breed and its members, and its complex relation to the regimes of evaluation that are central to breeding (Holloway et al., 2011). The second is difference. This includes the making and meaning of breeds as not only categories of subspecies difference but also processes of differentiation within a breed in terms of judgements of the varied quality of animals within it, and in this specific case, ideas of diversity as a characteristic of the breed. Furthermore, as I will explore in what follows, encountering difference, connecting across difference and respecting difference, that are integral to interspecies kinship (Govindrajan, 2018; Haraway, 2008), are worked through in practices of care for the breed as a whole and for individual animals. In this paper I consider the marshalling of difference in the making of horses in Iceland into a geographically and genealogically constituted breed. This also provides a context for exploring the perspectives of those involved in the breeding of horses in Iceland on quality and variety in the breed, the ways in which relations with and among horses are aligned in horse care and care for the breed, and how love of the breed shapes the geographies of care. However, this requires a broader initial consideration of the relationships between love of breed as a singular entity and the care of individual animals, especially those whose value rests on their relational potential, and recent efforts to consider horses as individual relational subjects as well as objects of knowledge.

Love, care and encounter value

Animal breeds are fundamental to capitalist economies but they are also entities that solicit interest and investment beyond their purely utilitarian value. To ‘love a breed’ suggests a preference that is deeply felt and entangled with socio-cultural meaning, rather than a solely rational interest in the profits to be derived from animal bodies, labour and reproductive potential, and from breeds as propriety categories (Ritvo, 1995). Breeds reflect and embody wider socio-cultural beliefs and collective identities at different scales including the national, regional and local (Colombino and Giaccaria, 2015; Eriksson and Petitt, 2020; Raento, 2016). Their bodies are appraised for looks as well as uses. The history of livestock breeding is one of shifting negotiations, alignments and tensions between the aesthetic qualities of breeds and the productive potential of animal bodies (Morris and Holloway, 2014). Farmed animals can be valued as aesthetic elements in rural landscapes and as gratifying embodiments of the ‘human ingenuity’ of breeding (Morris and Holloway, 2014). The so-called ‘heritage breeds’ are forms of bioheritage that physically and genealogically embody local and regional traditions of breeding and the historical geographies of farming and human–farmed animal relations (Evans and Yarwood, 2000; Hall, 2019; Yarwood and Evans, 1998; 1999). They are increasingly subject to efforts to enclose them as national genetic resources or enlisted in rural diversification, conservation and sustainability projects (Hall, 2019; Ovaska et al., 2021; Ovaska and Soini, 2017; Tamminen, 2019). Individual herds and flocks can stand for multi-generational family breeding efforts in smaller and less industrialised farms of the Global North, and both breeds and herds signify familial and collective traditions in pastoral communities of the Global South and for indigenous groups (Bixby, 2007; Köhler-Rollefson and Mccorkle, 2004).

Addressing the implications of this affective socio-cultural investment in breeds for the animals that constitute the breed as a living collective means firstly considering the ‘breed’ is a singular category that subsumes individuals within it. This suggests that it functions like other collective categories that objectify animals and distance humans from ethical questions of power and care that may be more immediate when animals are recognised as individual sentient subjects. In his critical engagements with ideas of the mass and the individual in livestock farming, Buller (2013) argued that viewing farmed animals as groups – the flock, herd or population – in large-scale industrial agriculture has a distancing and objectifying effect that makes mass killing possible. Similarly, the breed as an aggregate but singular entity can have the effect of subordinating the individual subjectivity of animals to the breed as a whole and justifying the reproductive control and selective killing of individual animals for the sake of the breeding programme. Their value as breed members overrides their care as individuals.

However, juxtaposing the individual animal and the collective category of breed, and reading breed only in terms of its subordination of the individual to the undifferentiated collective, only partially helps in addressing how breed shapes the lives and relations of animals. This is because it is not simply the idea of the mass that allows breeding priorities to determine the fate of animals. It is also the practice of differentiation. Like brands, breeds define desirable differences between otherwise similar commodities (Pütz, 2021). In breeding, this is a biocultural process. Producing the category of the breed as morphologically distinctive initially entails practices of differentiation between animals in a species to select the foundational stock and subsequently through selective breeding of their progeny. The genetic variability of animals – their non-identical nature – is what makes selective breeding possible both in terms of harnessing regional patterns of sub-species distinctiveness to initially produce codified breeds and in selecting animals within the breed for its ongoing improvement (Ritvo, 1987). The market for pedigree farmed, companion and sporting animals depends on the comparative evaluation of individual animals, and thus differentiation of animals within the breed. Differentiation justifies breeding decisions that include the continued care of some animals and, depending on the context, the killing of others.

Nevertheless, the underlying purpose of evaluation – breeding for profit – does not foreclose the possibility of addressing breeding in terms of human–farm animal intersubjectivity and exploring animal agency within the circumscribed nature of their lives. Carol Morris and Lewis Holloway argue that the evaluation of farmed animals in breeding is based on a relational practical aesthetic in which animals participate actively, albeit unintentionally, through the expressiveness of how they move, their responsiveness to humans and other animals, and the sense of character they convey, so that their ‘livingness is very much part of what is encountered and visually evaluated’ (2014: 5). The possibilities of animal agency are also central to recent efforts to address the capacity of individual farmed animals to shape what is understood as the temperament of the breed and thus inform the direction of breeding (Petitt and Eriksson, 2022). In both cases, this depends on the flock or herd being a size that allows animals to be known as individuals. The category of breed is of a different order, as it extends beyond a herd or flock, but these studies suggest that the work of breeding does not foreclose the possibility of intersubjective relations.

This is the context of farmed animals who are not, primarily at least, bred for those relations. Many other animals are bred for their potential to offer particular sorts of interspecies relationships and connections. Love of companion animal breeds is bound up with the possibility of particular sorts of relations with individual animals, as well aesthetic appeal and cultural associations. Their worth derives from the potential intersubjective relations associated with breed-specific behavioural traits such as affection or loyalty. As with different species of companion animals, existing and potential relations with people shape the making of breeds and their market value (Collard and Dempsey, 2013; Pütz, 2021). For Haraway, ‘encounter value’ refers to the value of mutually shaping, affective intersubjective ‘naturalsocial relationalities’ that are experienced, recognised and appreciated in lively, embodied and intimate interspecies relations with individual animals (Haraway, 2008: 66). While her insistence on the subjecthood of animals in these relations is a radical challenge to their objectification, relational or encounter value could be subsumed into an argument that the consequences of breeding for companion animals are only their subjugation. This leaves little room for elucidating, as Barua (2016, 2017) argues, how animal liveliness – their bodies and ethnologies – and their non-human labour co-constitute encounter value. It also forecloses the empirical, analytical and ethical questions of how encoutner value shapes practices of multispecies care, not only for individual animals but also more broadly. For Haraway, encounter value is the starting point for rather than conclusion of efforts to explore the ethical complexity of love, responsibility and care for a breed and for individual animals. The imperative, she argues, is not to seek ‘exterminationist nonsolutions’ (2008: 105) which would mean the disappearance of most individuals and kinds of domestic animals, but to explore the possibilities of multi-species flourishing within the entanglements of care, power, love and violence in human–animal relations (Govindrajan, 2018; Haraway, 2008).

These are relations of care in all their complexity: at once ‘affective, ethical and practical’; an embodied and emotional response: an obligation to look after another that requires ‘practical labour’ (Van Dooren, 2014: 291). As recent engagements with care have argued, care is not based in apriori normative principles but worked through in the contingencies of care in practice (Mol et al., 2010; Puig de La Bellacasa, 2017). Nor is care innocent. Both care among people and multi-species care are entangled with relations of power and fraught with ambiguity. The ‘violent-care’ of conservation, for example, involves the ‘sacrifice’ of competing, predator or less genetically ‘useful’ animals within breeding programmes in caring for ‘rare’ or ‘native’ individuals and species (Van Dooren, 2015). Similarly, animals who exist as members of ‘rare breeds’ or those ‘under threat’ are situated in regimes of care and value that are far from those who are members of ‘commercial’ breeds in the mass herds of industrial agriculture. The size of breed – the number of animals it contains – matters to the conditions of their lives. Yet, as Van Dooren (2015: 22) has argued for species, the contrast between care for an individual based on animal ethics, and care for the species or ecosystem based on environment ethics, is complicated by the way in which care for an individual animal is often based on characteristics that it shares with its species, ‘sentience or cuddliness’, for example, and care for the species is often informed by and enacted through encounters with individual animals.

This attention to the relationship between care for the species and care for the individual is useful in addressing care for the breed and its members. But species is also a category to consider alongside breed more directly, since it raises the question of how care may be inflected by understandings of species ethology as well as breed. Breed-specific models of care based upon broader ideas of horse behaviour, as I will explore here, can have significant implications for their lives and relations. This is aligned with new directions in equine studies. Recent injunctions to attend more fully to horses as subjects and relational beings in human–horse relationships (Argent and Vaught, 2022; Birke and Thompson, 2018; Dashper, 2017; Davis and Maurstad, 2016) extend two significant research strands in this field: one on the contributions of horses to human society as cultural signifiers and workers (Greene, 2008; Guest and Mattfeld, 2019a), and the other on the distinctive embodied encounter of humans and horses in riding (Maurstad et al., 2013; Nosworthy, 2013; Zetterqvist Blokhuis, 2018). Gala Argent and Jeannette Vaught argue for the need to situate any consideration of horse agency, personhood and subjectivity and co-constituted relationality within an acute awareness of how human-horse relations are ‘fraught with complex issues of privilege, power and responsibility involving the unequal requisition of bodies and the ability to exercise choice’ (Argent and Vaught, 2022: 4), that to varied degrees ‘supress horse autonomy’ (Argent and Vaught, 2022: 6). They expand what counts as relation beyond the immediate corporality of the horse-rider dynamic since human–horse relationships include but are not limited to the encounters of horses and humans in work, leisure and competitive riding. Addressing human understandings of horses remains important they argue, given their significance for horses’ lives even if the imperative and challenge is to better understand and respond to their species-specific and individual ‘capacities, needs and desires’ (Argent and Vaught, 2022: 5). A focus on breed imaginaries and practices contributes to this, since the breeding of horses is not only a socio-cultural practice of making human relations and collective identities, entangled with wider human social structures and processes (Cassidy, 2002; Derry, 2016; Guest and Mattfeld, 2019b), but also a framework for understanding horses and thus shaping how they are allowed to live, what is demanded of them, and the nature of human–horse relationships.

Addressing the ways in which horses are known both as relational subjects and through breed and species avoids neglecting the continued salience of forms of knowledge and categories that define horses as objects of human interest and use (Irvine, 2022). Rather than only identifying the ways in which horses are objectified and simultaneously or alternatively recognised as subjects-in-relation, the point is to explore their mutual inflection or, as Robin Irvine (2022: 188) puts it, ‘how these different kinds of relationship sit alongside each other and feed into each other’. So, rather than attend to the specific nature of human–horse relations in riding in isolation from the wider question of the breed, I examine the relationship between love of a breed as a whole and the value attributed to intersubjective human–horse encounters. I explore this through the perspectives of those who breed horses in Iceland, but firstly address the question of difference – in terms of horse uses, qualities and breeding geographies – in the national project of making horses in Iceland into Icelandic horses.

Producing the breed

Horses are a celebrated form of national biocultural heritage in Iceland: symbols of power and status in mythological and folkloric accounts; essential working animals in agriculture and transportation prior to the mid-20th century; leisure and competitive riding horses since; and acclaimed or personally ‘special’ horses (Arnórsson, 1997; Björnsson and Sveinsson, 2006; Magnússon 1978). All horses in Iceland are understood to descend from the original horses brought to the island in the Viking settlement period (847–930AD) since it is widely claimed that no other horses were imported in the subsequent centuries. It became illegal in 1882 to import any horses in order to prevent the arrival of equine pathogens. All horses in Iceland are thus members of the breed. Yet this single naturalised geographical and genealogical entity belies the work over the course of the 20th century to define horses in Iceland as Icelandic horses. This depended upon the inauguration of horse breeding in Iceland as a national project from the early 20th century with the associated institutions and knowledge practices – breeding goals, standards, advisors, societies, exhibitions, competitions, judging schemes and studbook records – spurred on by state-funded efforts towards agricultural modernisation, especially at its start, as well as personal, familial, local and national pride and pleasure in the breeding of high-quality riding horses – gæðingur – and good farm work horses (Hugason, 1994; Icelandic Horse History Centre, 2018; Pálsson and Haraldsson, 1997).

Producing the breed entailed working with regional geographies of horse difference and through different perspectives on the purpose of breeding and shifting practices of selectivity and inclusion. The register of horses that began in 1923 initially only included prize-winning horses. It was strongly associated with the rise of leisure and competitive riding which began to be institutionalised in the 1920s through local and regional equestrian clubs and from 1949, the Icelandic Horse Association, and became widely popular from the 1970s (Icelandic Horse History Centre, 2018). However, by 1951 it was agreed that the aim of the national breeding programme would be to cultivate a breed that included both ‘strong, amiable and willing’ farm horses and ‘superior riding horses’ (Icelandic Horse History Centre, 2018), thus accommodating the different breeding priorities of farmers and leisure riders. This put an end to the idea of breeding two types of horses separately – one type for riding and one for carrying loads and draft labour – that had arisen in the early 20th century and resurfaced in the 1940s. Instead, it was agreed that the single breed could contain horses best suited to different uses. The breed register opened in 1989 to include all horses in Iceland.

Furthermore, the institutionalisation of horse breeding did not begin with an ideal of island-wide inclusivity but with initial interests in selective breeding from groups of regionally distinctive horses. The breeding of horses was, and in different ways continues to be, practiced and understood through local and regional geographies of lineage and difference. From the late-19th century horses bred by particular farmers using their own or local stallions became well known within their localities and more widely. A local geography of breed lines known for their particular qualities and produced through skyldleikarækt or ‘family/kin breeding’ (Mende, 2015: 18) was also overlain by a regional geography of well-known key horse-breeding areas and the evolution of regional strains of horses with distinctive qualities (Arnórsson, 1997: 44–47). Indeed, the origins of the national Icelandic horse breeding project included efforts to breed horses from particular regions because of the perceived value of these horses. The pure breeding of East-Water horses at the state horse-breeding farm at Hólar in Hjaltadal, which had been established with this objective in 1962, for example, was only abandoned in 1989 (Icelandic Horse History Centre, 2018). Breeding with regional differences was thus initially part of the national breeding project. Narratives of continuity, purity and singularity, and practices of selection, definition and differentiation, are thus intertwined in the making of the national breed.

The local geography of horse breeding still matters as horses are known by their given name and the name of the farm where they were bred, and horse farms are known for producing horses with particular qualities and for their particular breeding lines – ættlínur – that are not based on the exclusive breeding of local horses but on pairing a farm's mares with stallions from potentially anywhere in Iceland (Mende, 2015: 3). At the same time, the skyldleikarækt lines produced from inbreeding among a local group of horses continue to embody a local and simultaneously national horse breeding heritage. News of the end of well-known old local breeding lines is greeted with regret and there are some efforts to maintain or revive surviving old breed lines (Mende, 2015: 17–21). Rational breeding goals and modern delocalised breeding methods thus sit alongside the historical and cultural significance of old local breed lines. Carefully documented and celebrated local and regional geographies of breeding (e.g., Ingimarsson and Pálsson, 1992) have been reconfigured rather than eroded by new technologies and forms of transport that make it possible to move horses within Iceland and from Iceland for breeding. The breed thus continues to be constituted through the potent geographical imagination of shared origins and island isolation and specific geographies of breeding within and beyond Iceland. Icelandic horses are both members of a national breed (whose Icelandic origins were legally fixed in Icelandic legislation in 1994 [Icelandic Horse History Centre, 2018] in response to this international interest) and an exclusive diaspora of shared and pure descent (Nash, 2020a).

In Iceland, horse breeding is generally not lucrative enough to be a sole source of income (Sigurðardóttir and Helgadóttir, 2015) but is also not confined to the most privileged. Most of those who breed horses commercially do so in varied combinations with other sources of income such as sheep farming, horse training and horse-riding tourism. At the same time, many people who own small groups of riding horses, including those who live in urban Iceland, may undertake some breeding and many like to breed at least one foal from their mares. In rural Iceland especially, this can include those with relatively modest incomes. In most cases, breeding entails ‘the stallion and mares roaming free in summer pastures’ (Thoroddsen, 2016: 13) rather than new technologies of reproduction. Prize-winning horses may thus be bred by those undertaking some breeding of their riding horses, by those breeding horses on modest mixed economy farms or by breeders in the 20 or so larger specialised horse breeding farms in Iceland. In the following sections, I address the way those who breed horses in Iceland reflect on the direction of horse breeding, the relationship between formal criteria of quality and how they value horses and the practices of care that are widely judged to be vital to the well-being of horses, to the character of the breed and the quality of human–horse relations.

Preservation and improvement

The International Federation of Icelandic Horse Associations (FEIF) promotes the reputation of Icelandic horses, facilitates leisure and competitive Icelandic horse riding, addresses horse welfare and sets standards and rules for judging individual horses and the breeding goals for the breed. The overall aim of FEIF's breeding programme is the ‘preservation and improvement of the Icelandic horse breed’ (FEIF, 2023: 28). In contrast to ‘endangered breeds’, preservation, in the sense of the minimum number of animals required to guarantee the reproduction of the breed, is not an issue. Its purity is preserved by maintaining it as a closed breed of horses that directly descend from horses born in Iceland. This is ensured by the mandatory entry of horses with proof of unbroken descent into the international breeding register. However, the reproduction of the breed is not only simply a matter of conservation and continuity but also of difference and change. Improvement means more of the animals within a breed achieving or approaching the ideal of the agreed breed standard or eliminating hereditary characteristics judged to be faults of a breed. Developing agreed upon and effective forms of evaluation is central to breeding. Evaluation makes it possible to adjudicate between horses to define their monetary value, to select breeding horses and to measure the quality of the breed as a whole through the annual and longitudinal analysis of competition and breeding scores.

Breeding and competition judging has thus become a sophisticated and regulated practice requiring extensive formal education, both because it is so crucial to breeding and because, unlike the objective outcomes of horse racing, it is inescapably subjective. This is mitigated by extensive formal training and by the breed organisations’ efforts to draw on and disseminate the results of scientific research on Icelandic horse genetics, the heredity of traits, and since 1986 the provision of statistical measures of breeding potential calculated from the measured quality of an individual horse, those in its ancestry and those of their progeny (Hugason, 1991). In practice, the usefulness of statistics of breeding potential and the evolution of qualitative scoring criteria and their weighting, as well as care and training, are matters of debate and deliberation as well as broad agreement in reflections on the condition of the breed. The formal evaluation of individual horses is based on differently weighted scores for three key attributes: their appearance or conformation; the quality of their gaits; and their temperament. It is generally accepted that the evidence of the achievements of horses in breeding shows and competitions confirms that Icelandic horses have improved in quality since leisure riding emerged most significantly from the 1950s and that this is a result of improved diet, care and training as well as selective breeding. Sigmar, senior breed judge, informed me that the longitudinal analysis of horse evaluations verify the improvement of the breed and this suggests that those breeding Icelandic horses across the international breeding community do follow FEIF's official breeding goals.

Yet, there are also some concerns about the effects of selective breeding on the breed as a genetic population, on the practices this entails, the effects on the aesthetics of the breed, and on the increasing emphasis on particular sorts of gaited ability. Despite the formal acceptance of a range of heights within the breed standard, the breeding of taller ‘elegant horses’ (Hugason, 1991: 231) for the internal competition and the international market (as well as improved care) has increased the average height of horses and led to some concerns that this is having the effect of making Icelandic horses look less Icelandic (Guðmundsdóttir, 2014). A more prominent concern is that the overuse of high-scoring stallions in breeding may lead to narrowing of the gene pool within the breed. Those who can afford the very high costs of mares being ‘covered’ by elite prize-winning stallions thus negotiate their desire for the best outcome possible from the pairing and their consciousness of the effects of the overuse of selected stallions on the genetic diversity of the breed as a whole. These concerns have been addressed in genetic studies which affirm confidence in the variety, quality and broad stability of the national gene pool (Campana et al., 2012; Hreiðarsdóttir et al., 2014) and report on the overall improvement of quality of horses in Iceland. Care for the breed as a genetic population appears to align with individual breeding decisions.

However, improvement can create ethical dilemmas, especially in terms of the effects of the increased proportion of horses in Iceland that are judged to be good horses. These issues are not the subject of public debate but are significant personal concerns. Anna, who had a horse farm for 20 years, reflected on the ethical implications and emotional effects of continued practice of culling in the context of improved breed quality:

Over the last twenty years we have weeded out a lot of bad animals in Iceland because we cull. So with the stock of Icelandic horses today it is an exception that you get a bad horse, a horse that is untrainable, or doesn’t have any gaits or has a bad temperament. But the market is controlling everything. So I think today we are culling more and more animals that are fine. There is nothing wrong with them and they are just not good enough to fit into what we want. So there are a lot of ethical questions like that. You get over production and I felt just in the last few years that I was a horse farmer it was getting more and more hard for me. Like you had 10 horses in a year you had to train and you basically would be forced to choose half of them and the rest would go and they would be perfectly fine animals. But 20 years ago, the five that would go would have been no good anyway.

This practice of selecting which young horses to keep is not on the scale of the massive commercial overproduction of horses and the slaughter of the surplus in other contexts. However, the smaller scale of breeding in Iceland means that young horses are well known to the breeder before decisions are made about their future, thus making these decisions serious, and for some, increasingly ethically problematic. Though the statistical proof of improvement abstracted from practices of care justifies culling to achieve overall quality, known animals are not easily disposable, even if their bodies may have other uses. The cultural acceptability of eating horse meat in Iceland is included in some celebrations of the breed's singularity and versatility. Guðrún who bred her own riding horses explained to me: ‘We only have this one breed. It doesn’t matter if you eat it or compete on it. It is the same breed’. But the consumption of horsemeat does not make these decisions about culling easy. Many breeders balance the challenges of breeding enough horses to potentially produce the high-quality horses that ‘the market’ desires in order to make breeding economically viable, the costs of keeping large numbers of horses and the ethical costs of culling animals that are judged to be good but not of high enough quality. Some do so by breeding fewer horses, others do so by working hard to find buyers for these ‘surplus’ horses in other sectors such as horse-riding tourism, riding teaching or ordinary leisure riding. The improvement of the breed as a whole has thus depended on the past disposal of horses judged to be low value and the goal of improvement shapes the chances of survival and lives of horses born into the context of better overall quality.

High-quality ‘competition horses’ have top scores for the quality of their gaits. The distinctive ability of Icelandic horses (along with only a few other breeds) to move in four or five gaits – the tölt and pace gaits as well as the walk, trot and canter/gallop that are common to other horses – is key to the breed standard and breeding goals. The smooth tölt gait is a celebrated defining characteristic of the breed and central to competitive riding. It also makes it possible for riders of varied ability to ride easily and comfortably. All horses in Iceland ‘have’ the four gaits (of varied quality) including the tölt, but only horses with two pairs of a single gene variant also have the ability to perform the fifth gait, the pace (Andersson et al., 2012; Kristjansson et al., 2014). However, the pace has become increasingly emphasised in breeding shows and competition riding and is a focus for some breeders. There are concerns that the emphasis on the pace gait is creating a degree of uniformity among competition horses that is at odds with the qualities that are valued in good ‘ordinary riding horses’. Sigurlín, an established breeder, expressed her dissatisfaction with an emphasis on the pace gait since 2008:

So all of a sudden because they say you should always go for a five-gaited, that is the Icelandic horse, well of course… but if there is some gait that can be missing it is the pace, because we always want walk, you want to be able to canter a little bit, you want to be able to trot, and you really want to be able to tölt, and if you had to choose if your horse could only have one gait you would choose tölt. But it is the pace that we can be without. Almost no one is riding pace, not on purpose.

The emphasis on the pace means that though four-gaited horses can be entered into competitions, the numbers being entered, she suggests, are very low. This is ‘even though a much bigger group of people are looking for four-gaited horses and it is easier to sell and easier to get high prices for four-gaited horses’ because for many ordinary riders having five gaits to control ‘is one too many’. The scoring system in breed judging shows, she argues, does not reflect what is valued in leisure riding: the slow canter rather than the fast gallop and the slow tempo tölt. The system of evaluation is thus orientated to judging the achievement of horse and rider in performing this gait in competition riding.

This mismatch between what is valued in competition riding and what is valued in leisure riding reflects the tension between the breed being defined and celebrated in terms of diversity – of uses, sizes and sorts of horses – and a system of evaluation that necessarily values what is judged to be the highest level of a characteristic or ability. In the case of Icelandic horses, the single breed standard is combined with the appreciation of a diversity of attributes and uses that was central to the eventual mid-20th century definition of the breed as a whole and is central to the FEIF (2023, 16) description of the Icelandic horse as ‘extraordinarily versatile', suitable for high-level competition riding as well as leisure riding for children and adults. As former horse breeding advisor of the Agricultural Society of Iceland, Kristinn Hugason, was an prominent advocate of the argument that this variety within the breed, or as he put it, the fact that ‘good horses are not all good in the same way’ (Hugason, 1991: 231) should be maintained through breeding to maximise the market for different kinds of Icelandic horses. This tension between a single evaluation scale and the affirmation of variety is particularly acute in the relationship between the formal evaluation of horse temperament and the ways in which people value horses. Ideas of temperament are also key to addressing the implications of love of a breed and the promise of encounter value for horses’ lives and relations.

Diversity and difference

The temperament of Icelandic horses is central to versatility as a breeding goal, to the formal evaluation of individual horses, and to the love of the breed. They are bred for the potential for people to have relationships with horses that are described as friendly, intelligent, easy to handle, energetic, sure-footed and strong. One of the four overarching official breeding goals is for a ‘character that enables the horse to be used for different roles with a special emphasis on a calm, friendly and cooperative character. The horse is supposed to be courageous and reliable, both when handling and when ridden’. This includes a range of horses ‘from sensitive and willing horses to calmer types, but always with an emphasis on willingness to perform’ (FEIF, 2023: 74). But what temperament means and how it is assessed have been the subject of considerable debate. Though temperament is most often described in terms of ‘willingness’ and ‘spirit’, the character of horses is formally assessed as ‘rideability’. A highly scored horse for rideability should be co-operative, alert, responsive, willing, relaxed, supple and light on the reins. Scores for rideability carry a significant weight in the total score for a horse, since poor temperament – aggression or extreme anxiety – is dangerous and thus more problematic than any other faults. But as Sigmar explained, temperament is the ‘trait most subjective and hard to judge’. It is difficult to assess visually rather than through riding; and what is judged to be willingness is partly a product of the interaction between the horse and rider. Training also has a considerable effect. Yet the aim of improved judging is for its assessment to also take account of the extent to which is it an intrinsic hereditable quality rather than a relational outcome of human-horse interaction, so that its evaluation can inform breeding.

It is not only the difficulty of measurement that is an issue but also what is being measured. Though rideability is described in terms of responsiveness in riding rather than speed, there are concerns that willingness has become very strongly associated with the speed, power and energy of ‘spirited’ horses in competition riding, a willingness to go forward fast, rather than willingness to cooperate. The official scoring criteria emphasise that the horse should be relaxed. Yet some argue that though speed and tempo are taken to suggest ‘willingness and spirit’, they may in some cases actually reflect the anxiety of the horse rather than its controlled energy. Moreover, the success of highly spirited horses in competitive riding has become a matter of debate. Professional competition riders only want ‘spirited’ horses whose energy they can harness especially for the potentially prize-winning performance of fast gaits. However, breeders do not always favour the highest-scoring horses in choosing horses for breeding. Áslaug explained that in making decisions in breeding

we don’t go after the highest scored. Although they have top scores maybe they are still not the type of horse that you want to breed. Maybe a much lower scored horse has some qualities that you really want in your breeding. Of course, in some ways if the horse does a lot on the track it has quite a lot of spirit and things that you want. But then there are horses that are not scoring well in the track which are still horses with perfect temperament that anyone can ride. Maybe these horses that are scoring really high are not horses that everyone can ride. They are too difficult. Too much spirit. Too much of everything. And our Icelandic horse, we want it to be for everyone.

Willingness, she argues, ‘is getting to be sometimes a bit of a negative word which is bad because willingness is a really positive word about the horse. But right now willingness is getting to mean “ok I cannot ride this horse because he is too willing”’. Breeders report that some buyers, especially less experienced riders, avoid high-scoring horses. Einar reflected on his experience of customers who do not want to buy horses whose parents have ‘scored 8.5 or 9, and 9.5 is out of the question because they think they cannot handle them. This tells us a lot about the system. You are trying to sell your product and it is a little shocking when people want something totally different from what the system wants’. Breeders cannot operate outside the scoring system, yet as in other livestock breeding contexts (Holloway et al., 2011), many rely on their own perspectives on the qualities of their horses gained from the sustained, embodied and relational work of horse care and training in choosing horses from which to breed, rather than solely on scores or statistical measures of breeding potential. Sigurlín explained that she decides which mare she is going to keep for breeding before they are judged at breeding shows since ‘it is important that the horses you have around you should be horses that you like’. Breeders and buyers thus negotiate a scoring system that inevitably scores certain sorts of horses most highly rather than attribute different qualities across a range of horses.

Although caution over ‘willingness’ does not mean that this system of evaluation is fundamentally undermined, there are concerns about the wider effects of implying that less highly scored horses and those outside the scoring system are of little or no value. Many insist that variety in the breed should be a variety of equality that appreciates the worth of ‘ordinary’ riding horses rather than a hierarchical ordering of variety. This is articulated in terms of the value of particular horses as well as in terms of the need for this wider appreciation. Hallbörg, whose horses worked with her in sheep farming and in teaching riding to children, argued that the category ‘Icelandic horse’ has to be a wide and inclusive ‘open window’:

I have a horse that has allowed children to overcome fear and gain confidence by riding a precious horse. He is old now, fat and slow, but always willing to come forward and has spent his life working and being ridden by children – always safe. He is a precious horse. He may just have a year or less left. I will miss him. People would laugh but he is just as important as the competition horse. He is a precious horse.

This appreciation of the variety of horses within the breed is integral to the official breeding goal to breed a riding horse that has varied roles and is suitable for different types of riders. Yet the celebration of spirited competition horses prompts others to defend the value of horses that enable people with no or little experience of riding, such as tourists taking riding tours or those learning to ride, to do so safely and confidently, but are often disparaged as safe, calm and thus dull ‘rental’ or ‘riding school’ horses. Some of those working with horses feel there are risks that breeding and the breed itself could bifurcate. Hildur who breeds horses to use in her horse tours for tourists, expressed her concern that it may inadvertently lead to the creation of ‘two lines of breeding’ one for ‘leisure riding and horse tours and one for competition’, echoing early 20th-century plans to have separate breeding programmes for farm work and riding horses.

More broadly, understandings of the temperament of the breed within Iceland horse culture reflect both the central relational value of the breed and ideas of horse difference. Temperament is judged according to what horses can offer humans in human-horse relations in riding and their ease of management and training. Rideability is not an inherent quality but an anthropocentric judgement. Yet, ideas of species difference are central to understandings of horse temperament in Iceland. The individual character of a horse is understood as inherited directly and shaped by the evolution of the breed as whole. The character of the individual and of the breed is also understood to reflect practices of horse care that are based on the principle that every horse needs space to be ‘allowed to be a horse’ (Líndal, 2008: 14) with other horses. These care practices include foals remaining with their mother up to the end of their first year, early years spent in a herd of horses of mixed ages with minimal human contact, training from the age of 4 or 5 – later than other horse training practices – and long breaks between periods of training. In contrast to most regimes of horse care, horses give birth in fields without human intervention. Svana explained that letting her mare do so is not just about confidence in the good outcome for her horse and its offspring. It is about willingness to risk a complication in the birth of one foal for the sake of maintaining this commitment to not intervening at birth, and to minimal human contact in the early years, and for periods in the year, for the sake of the breed as a whole. Horses, including competition horses, spend significant parts of the year in horse groups in large ranges with little human contact in an annual cycle of open grazing in groups in the winter months, periods of being stabled with access to paddocks when being ridden regularly, and summer months of grazing in the herd.

In Iceland, horses are thus viewed not just as members of an aggregate category, the breed; nor are they only viewed as individuals in specific human–horse relationships. Instead, they are understood as individuals that ideally live in intersubjective sociality in groups with each other more than with people. Icelandic equine studies are informed by and confirm the view that the ideal environment for horses is a social environment (Sigurjónsdóttir, 2004). The well-being and temperament of individual horses, the character of the breed and the quality of human–horse relations all depend on allowing that sociality. Their social life within the herd, experience of rough terrain and harsh weather conditions are widely thought to produce independence, endurance, intelligence, surefootedness, confidence, curiosity and calmness in horses that then shapes their response to and relations with trainers and riders (Arnórsson, 1997; Líndal, 2008; Thoroddsen, 2016). Valuing horses for what they offer in individual human–horse relationships entails practices that acknowledge the significance of relations among them. As Katrín, a senior equine veterinary officer, explained, care must be based on the recognition that the optimal life for horses would be in free-ranging herds and that horses would prefer that to life with people: ‘if you have taken that freedom away’ horses must be given ‘time off’ for months together and enough exercise and human and horse company when they are stabled. Rather than idealising and prioritising human–horse relations, this awareness underpins horse care in Iceland.

There are, therefore, some concerns that the direction of leisure and competition riding means that ‘horses get less time to be horses’ as increasing numbers of horses spend more time stabled and their ‘time off’ is curtailed, with detrimental effects on their welfare as well as temperament. Vilborg reflected on this:

What is the quality of the Icelandic horse? What is the uniqueness of the Icelandic horse? Yes, partially it is how they were kept but that is changing as well. So now people have these horses that are so expensive; they don’t get to roam free. Their lifestyle is not the Icelandic horse lifestyle. So are we loving it to death now?

This question suggests that expensive horses that are used primarily for competition riding epitomise the best of breed, but not all may fully benefit from the care of the breed as whole, which for many people means continuing to maintain the traditional way of keeping horses. Encounter value may be over-extracted in months of riding at the expense of both the individual animal and the qualities of the breed that are understood to be shaped by allowing extended periods of time away from people. However, encounter value is preserved by a predominant practice of care that includes distance rather than continuous contact. Love of the breed and care for individual horses both depend on horses being allowed to be herd animals relating to each other rather than confined to their relations to people. The geography of these human–horse relations is thus not only that of the close-up embodied relationality of riding or the proximities of kinship in the other times, spaces and practices of human–horse relations (Nash, 2020b), but also one of separation and distance. Curtailing human contact is thus a way of respecting the significance of horse relations with each other. It both ensures the well-being of individual horses and preserves the encounter value of the breed. Love of the breed entails sacrificing time with horses to allow them to have social lives with other horses for their sake and to ensure the quality of human–horse relations.

Conclusion

Breed practices and imaginaries matter to animal lives and relations in complex and sometimes contradictory ways across varied contexts and purposes of animal breeding. A geographically attentive focus on ideas of difference and value helps elucidate these complexities. This includes extending geographical work on breed knowledge practices and geographical imaginaries of origin by considering the making of breeds through the selective enclosure of geographical patterns of subspecies diversity into genealogically constituted categories. Breeds are imagined and managed not only through ideas of their singularity and distinctiveness but also internal difference – the variety of individual animals within the category – that is the basis for selective breeding and the creation of differentiated value. As this paper demonstrates, this engagement with animal breeding can be developed further by exploring the geographies of interspecies encounter and care. This directs attention to the relationship between investments in the vitality and futurity of a breed and the care of and relations with individual animals, particularly in cases in which encounter value is central to the species and, more specifically, breed. In the case of horse breeding in Iceland, the potential of the breed to provide encounter value is incorporated into formal breed standards through the language of temperament and is part of the wider cultural figuring of the breed. Yet, it has a more awkward relationship to modes of evaluation that locate members of a breed on a single linear evaluation system that can be applied to all members of a breed since encounter value is produced through and depends on intersubjective relationality. Diversity, variety and versatility are all about relational value, that is, the different sorts of encounters that different sorts of horses provide as individuals in specific human–horse relationships. In practice, the highly refined and formal judging and scoring system that preserves the breed as a single entity coexists with the ways people value horses of many kinds including those conventionally judged to be low-value horses. Those that might potentially offer high encounter value but are not judged to be high-quality horses in terms of gaits and conformation may be culled to maintain the quality of the breed as a whole. At the same time, the ascription of high value to the highest scoring horses can also co-exist with preferences for those that are not conventionally judged to be the best quality.

The difference that matters to breeding also includes species difference. Limiting the frames of analysis to the breed and the individual animal neglects the other categories and collectives – both larger and smaller than breed – through which animals are known and through which their lives are organised, that is the species and, especially for farmed and sporting animals, the group. While encounter value might seem to privilege ideas of connection, Haraway (2008) insisted on the recognition not elision of species difference and that care and connection are inescapably shaped by asymmetries of power between humans and other animals. The term encounter value suggests an outcome but stands for partial connection across species difference, that depends on the ongoing work of care, and respect for that difference. In this case, those working for the health of the breed as a whole and caring for individual animals work with understandings of species behaviour and needs in which ideas of intraspecies sociality are fundamental. Between the idea of the breed as living, informational and imagined entity and the individual disaggregated animal in intersubjective relation with an individual human, are horses in social groups. In this case, the category of herd does not serve the distancing de-individualisation that facilitates the slaughter of farmed animals. Instead, organising horses lives in ways that allow for some life in a herd reflects the recognition that animals want to be with each other (Buller, 2013). This attention to animal sociality thus avoids counterposing the collective category of breed that subsumes individuality to the singular animal, abstracted, as Buller (2013) argued, from their social relations. So though encounter value foregrounds intimate human–animal relations, in this case, care for the breed and care of individual animals means curtailing contact and decentring the human–horse relationship. It entails recognising that it means more to the humans than to the horses, rather than being imagined as symmetrically reciprocal. Encounter value is thus not deployed in this paper as an idealised intersubjective alternative to love of the breed. Instead, I have explored the nature of these distinct and intersecting registers and traced the ways in which people align and negotiate between investments in and care for the breed and care for its members who are both known through the breed and as individuals.

However, while respect for horse difference shapes the degree to which they are allowed to live in social relations with each other it is also a way to ensure the quality of human–horse relations. Difference is not elided in the hope for interspecies connection. But organising horses and companion animal lives in ways which respect that difference is rarely detached from the desire for some connection across difference. Giving horses time and space to be with other horses enables them to escape from human desire for connection and maintains the encounter value of the loved breed. Horses ‘allowed to be horses’ return more willing to serve that desire. This suggests an ethics of care in practice in which those involved work through their sense of what they judge to be best for their horses and what is best for the breed, and their sense of what horses want and what they want for and with their horses. They negotiate desire for connection and respect for difference.

Thus, counterposing care for the breed (as problematically deindividualising) to care for individual animals (whose subjecthood is respected) in addressing the significance of breed practices and imaginaries for animals, risks relying on an unconsidered model of what counts as ‘good’ multispecies care. Thinking of care as directly intersubjective, involving practices that depend on physical proximity and ideally responsive to the individual may suggest that modes of care that are standardised for the general, in this case breed, and that do not necessitate intimacy, are necessarily poorer and do not ensure the well-being of individual animals. This assumption may also reflect human desire to demonstrate and enact their care through practices that involve direct connection and contact with animals, such that intersubjective encounter is the reward for care. While both encounter value and care as practice imply contact, touch, immediacy, closeness and proximity, care for the breed, group and individual can be both proximate and removed. To care in this case is to allow distance, curtail contact, to not intervene. Care for the breed and care for its members thus involves practices of care that are both immediate and remote. Attending to these spatialities of care and encounter across species difference extends geographical engagements with questions of care over distance (McEwan and Goodman, 2010; Silk, 1998; Smith, 1998). These expanded geographies of interspecies care are not innocent, disinterested nor uncompromised, but do reflect a practice of care based on an understanding that human–animal relations are not the only ones that matter for animals, including those bred for the promise of encounter and connection. This consideration of breed imaginaries, knowledges and practices is thus also a proposition to further pursue the implications of the positioning of animals as individuals and as members of groups and species, as well as breeds, for animal lives and relations.

Highlights