In the following subsections, the themes emerging from all informants’ responses are synthesised and discussed: Naturalising, normalising and downplaying racing practices and their impacts; the thoroughbred as an eager and willing participant versus a horse under stress, anxiety, being agitated and disturbed; the perception of equipment and its applications; the visual problem as a problem of showing too much or not enough; the horse-human relationship and the idea of the thoroughbredness of the thoroughbred versus the horseness of the horse. The themes are discussed within the context of research in relation to impacting factors that are raised by the informants—namely, the bit, the tongue-tie and human handling. Two examples of recent interventions from a well-known racetrack operator in North America and the Australian racing authority are included (see Section 4.5.2
) to support the findings and illustrate the hermeneutic research approach (Appendix A
). In Section 4.5.4
, Bergmann’s Layers of Engagement with Animal Protection [3
] are applied to deepen the analysis of the thoroughbred welfare and protection discourse. Recommendations for further research conclude this section (Section 4.5.5
4.5.1. Naturalness as a Guide Versus Naturalness as a Fallacy
What seems to be a significant factor in the industry informants’ process of naturalising, normalising and downplaying racing practices and their impacts on the horse is that many such practices exist because they have “always been done that way”. In the case of bits, for example, Mellor and Beausoleil [17
] find that most horses “exhibit clear behavioural evidence of aversion to a bit in their mouths, varying from the bit being a mild irritant to very painful” and believe that this in itself is a significant welfare issue requiring attention [17
]. They suggest “the non-recognition of clear behavioural evidence of horses’ aversion to bits in their mouths arises because the indicative behaviours have been and are observed so commonly that, except in more extreme cases, they are considered to be normal” [17
]. Cook and Kibler [20
] (p. 551) suggest that, because bits have been standard equipment for millennia, they “are widely assumed to be indispensable and ethically justified”.
When calling on what is natural, one can be expected to question what really is natural. If naturalness was a guide, a starting point to assess the expressions of the thoroughbreds in the images and elsewhere could be similarity to the “closest wild counterparts” [48
] (see also Section 2
). In the case of the bit, Cook [21
] (p. 256) summarises: “At liberty, the running horse has a closed mouth, sealed lips and an immobile tongue and jaw”. The horse is an obligatory nose-breather, and the application of a bit breaks the seal of the lips [106
]. This has a raft of implications for health, welfare, ability to perform and safety, including bit-induced pain being a cause of fear, flight, fight and facial neuralgia, the bit interfering with breathing and locomotion, the bit being implicated in breakdowns and fatal accidents, and it is hypothesised that the bit causes dorsal displacement of the soft palate, induces asphyxia, which causes bleeding from the lungs (EIPH), and it can cause sudden death [21
]. Moreover, Mellor and Beausoleil [17
] conclude that the bit impacts horses in a way that they experience severe breathlessness.
Instead of questioning the application of the bit, the industry informants saw it as part of a normal and natural system in racing. For example, Image 2, which depicts the head of a ridden-bitted thoroughbred with an open mouth identified by Mellor and Beausoleil [17
] as a sign of aversion to the bit, was described by industry informants as depicting “nothing out of the ordinary” (TBI-6), showing “actually a very gentle bit” (TBI-4), and TBI-4 explained that the mouth opens not because the jockey is “tearing at his mouth” but because “the horse is wanting to go forward”, and, so, “the horse [...] is pulling against his mouth”. Most industry informants also expressed support for the use of added pressure-exerting tools and practices to deal with the problems the application of the bit and training, racing and handling practices cause, such as the use of yet harsher bits and nosebands and the application of the tongue-tie (Section 4.2.2
), despite their welfare implications and lack of efficacy [19
]. Other practices in the industry at large, to address health and performance issues, potentially linked to use of the bit [21
] include use of the contested drug furosemide [111
] and surgery performed at the horses’ upper respiratory tract [112
]. These are common interventions despite the side effects of the drug furosemide [111
] and the potential for complications as a result of surgery, with subsequent health and welfare implications for the thoroughbred [115
]. The central focus of these interventions is generally not to protect thoroughbred health and welfare but for humans to pursue an activity that pushes the horses beyond their natural physiological limits. Indeed, those involved in the care of racehorses identified the overuse of veterinary interventions as a significant welfare challenge [61
The examples discussed above demonstrate how calling on what is natural can be a fallacy when divorced from scientific evidence and from the horses’ interest in their own physiological and psychological integrity. It also demonstrates how naturalness as a guide is relevant for thoroughbred welfare and protection even in an environment and under a handling and exercising regime that controls all aspects of their lives and has significantly compromised their nature, agency and integrity.
4.5.2. Naturalness and the Legitimacy of Thoroughbred Racing
Naturalising and normalising the horses’ emotional and behavioural expressions and the impact of particular racing practices depicted in the images can be seen as an attempt to legitimise racing. There are indications that the industry informants were aware that the thoroughbreds’ expressions can be perceived as compromised welfare, as TBI-5 expresses concern about the visual of the tongue-tie (Section 4.2.3
), and TBI-4 adds, when commenting on Image 2, that the open mouth is “not a pain mechanism”. The industry informants’ tendency to ignore and, thus, conceal potential welfare concerns embedded in common racing practices as a way of addressing the public’s perception of racing appears to be an approach taken throughout the international racing industry. For example, The Stronach Group’s media department reportedly has specific instructions to reduce the use of images showing certain whip actions in racing [119
]. In 2018, the Stronach Group’s Gulfstream Park racetrack even produced and distributed a promotional wall calendar that reportedly contained images with some of the whips carried by jockeys in the racing action shots digitally removed [119
]. In at least one instance, not only had the whip been removed but the bit had also been digitally altered to appear as less severe than in the original photograph (see the original and the manipulated images on pp. 5,6 in the article written by T.D. Thornton for the Thoroughbred Daily News [119
]). The tendency of the industry informants to not put into words the extent of the mental and behavioural expressions of the horses, and the impact of the equipment used or the human handling of the horse (Section 4.2.2
), functions similarly to how digital image editing tools are used as a way of “unseeing” what they prefer not to be seen. The industry informants presenting certain aspects as normal and natural indicates they are consciously and subconsciously participating in the industry’s priority project to change and shape the public’s perception of the racing industry and its treatment of the thoroughbred, a phenomenon that can also be observed in other equestrian disciplines [120
What TBI-5 identified as a visual problem is a problem of legitimacy of the horseracing industry [39
]. With their attention directed at sanitising the visual, the industry engages in censorship and resists transparency. This undermines trust in the industry, and trust is an indispensable aspect of legitimacy [121
]. The industry is aware of the risk to its social license to operate [121
]. Nonetheless, in particular racing in the UK, Australia and the US, the regulating racing bodies are resistant to centre the protection of the thoroughbred over industry interests. In Germany, German Racing banned the use of tongue-ties as Rüdiger Schmanns, then Director of Racing for German Racing, stated “[w]ith growing animal welfare activities, especially in Germany, there was no possibility of allowing the use of tongue ties to continue” [122
]. In 2020, Racing Australia reaffirmed their position that the tongue-tie is acceptable, arguing they have found “an appropriate balance between the welfare of the horse and performance” [123
], despite its disputed efficacy and need [124
] and health and welfare impact [19
The application of the bit and the tongue-tie are but two examples. Butler et al. [61
] identified a raft of welfare issues and challenges that demonstrate how common racing practices put thoroughbred welfare at risk. It can be expected that the racing industry will come under increasing pressure if more details of their common practices in racing—and breeding thoroughbreds, for that matter—become increasingly known to the general public. This is largely due to the implications for thoroughbred welfare and the nature of the horse and the concern people show for naturalness in determining what a good life for an animal is [46
]. Currently, industry representatives take the view that the problem is not the impact of racing practices on the horse but that people do not “really understand what is going on there” (TBI-5, see Section 4.2.3
), an aspect previously discussed by Bergmann [2
] (pp. 127–128). Indeed, many people are unaware of the common handling and training practices in racing, and animal advocates believe there is a need to inform and educate the public. Referring to the tongue-tie in Image 4, advocacy informant AAI-1 did not “expect that most people, either at the track or elsewhere, would [be able to] understand what they were seeing”. However, a lack of public awareness cannot be used as an excuse to continue to harm thoroughbreds, nor as an “excuse to ignore the unrepresentative nature of existing welfare policy” [46
] (pp. 29–30). For welfare policy to have democratic legitimacy, it needs to reflect the public’s view of what it means for a nonhuman animal to fare well [46
4.5.3. The Horse-Human Relationship as an Aspect of a Holistic Notion of Naturalness
In the responses of the animal advocacy informants, the horse-human interaction emerged as an important element for horse welfare (Section 4.3.4
). This echoes Butler et al. [61
], who found that the horse-human relationship was identified by those professionally caring for thoroughbreds as a seminal aspect of good welfare. The participants referred to factors such as the “consistency of routine and carer” and horse and human “getting on”, ensuring continuity and attention to detail and not only well-trained and knowledgeable but experienced staff for a “best-life” scenario. Creating a positive horse-human contact was linked to a potentially higher level of care and observation. Hall et al. [11
] described the link between human handling and horses’ emotional and behavioural expressions:
Horse-human interactions undoubtedly influence both the subjective emotional experience and the behavioural expression of the horse. The influence may be due to the intensive management, handling and focused interaction associated with the process of training, and the physical and emotional demands placed on the animal in relation to performance. Methods of training and handling which provoke negative emotions and states such as fear, or where the individual experiences pain, may lead to short term success in relation to behavioural change, but will also produce fearful horses which are not desirable for the horse or human safety, nor successful for performance in the longer term. When frightened or anxious, horses will show escape responses ranging from agitation involving a raised head and neck to extreme reactions including bolting [11
] (p. 184).
Most industry informants ignored or downplayed the human factor in the images, including in Figure 3
, depicting a thoroughbred resisting to enter the starting gate. This may be a result of the informants interested in conveying to the researcher that there are no welfare issues to be seen. It could also be a case of nonrecognition, as discussed in the context of the bit above, due to the normality of horses expressing fear and resistance at the starting gate. As Miles et al. [25
] found, 71% of the studied 2–5-year-old racehorses entering the starting gate demonstrated “unwanted” behaviours. They also found that gate staff responded by using an “artificial aid”, such as whipping over 40% of the time, which explains why TBI-4 made the downward comparison in relation to Image 3, stating “no one has a stock whip on him, no one is hitting him” (Section 4.2.2
). Moreover, it can be suggested that many of the emotional and behavioural responses of the thoroughbreds in the images may, in fact, be learned or shaped by the human factor and the particular activity of racing as such [24
]. The kind of relationship humans have with the horse shape the nature of the handling and training practices, and vice versa, the handling and training practices shape the nature of the horse-human relationship. It is suggested that the underlying horse-human relationship plays a significant role in how the human and how the horse respond [11
]. The low interest in the human-horse relationship and lack of recognition of its importance for equine welfare is characteristic of the industry at large. The participants of Butler et al.’s study [61
], for example, identified staff shortages and a lack of experienced staff as a challenge significantly impacting thoroughbred welfare in various ways.
For a better understanding of the horse-human relationship, this author suggests contextualising it within the framework of naturalness. This contrasts with Yeates [48
], who believes other animals’ interactions with humans are unnatural, and therefore, human-animal relationships are not an aspect of naturalness. However, humans have lived for tens of thousands of years in multi-species communities, whether in close proximity or not. Therefore, it seems more useful for animal protection in a multi-species world to conceptualise human-animal relationships and interactions as being an aspect of naturalness. A reductionist approach to naturalness and the human-animal relationship would mean to artificially separate the innate connection between humans and other animals that is based in a shared evolutionary continuity, also expressed as kinship [126
]. The argument is based in the binary of humans versus nature and the belief that humans are separate from nature is considered by many one of the root causes of human exploitation of animals and nature [127
] and is counterproductive to advance animal protection. The question is, rather, what human-animal relationships should look like under a framework where naturalness is intrinsically valued. Investigations in, for example, fields such as cognitive ethology [128
] and into the ontological nature of the human-animal relationship [31
] can assist in finding answers.
The welfare impact and the ontological status of the horse-human relationship discussed above speak to a definition of naturalness as a holistic notion. The raft of day-to-day welfare issues identified in the general equine welfare literature and unified by the notion of naturalness (Section 1
), the many aspects of an animal’s life in which people relate to naturalness when thinking about a good animal life (Section 2
), the role of naturalness for many equine welfare issues identified by particular groups of horse people, such as owners/riders and others involved in the care of horses [59
], and the animal advocacy informants’ conceptualisations of naturalness (Section 4.4.2
) all highlight the holistic qualities of the notion of naturalness. It appears that reducing this concept to one or a very limited number of aspects is arbitrary and an opportunistic reconstruction of its generic meaning. When narrowing down the meaning of naturalness to this degree, a different term that more accurately reflects what is referred to, such as natural nonhuman animal behaviour only, rather than naturalness should be used. A reduction obscures and co-opts the notion of naturalness and serves the user of the animal rather than the animal’s full range of interests and needs. Accordingly, industry informants dominantly use the concept of naturalness selectively when it aligns with their economic model (of breeding) and their activity (of racing).
4.5.4. The Layers of Engagement with Animal Protection and Naturalness
Previous research that explored the interface of thoroughbred welfare and sustainability found that the industry informants are, in some ways, the progressives in the industry, and they are situated at the reform level of the industry’s welfare discourse [3
]. This current research, however, highlights that there are few individual cases where industry informants share similar concerns to advocacy informants (for example, TBI-9 responding to Image 3, Section 4.2.1
). In this research, the informants were given the opportunity to defend the horse and reconsider current practices based on the images presented (see Section 3.3.1
). However, when it comes to the handling of horses and the application of equipment, the industry informants appear to be more interested in defending current racing practices and maintaining the status quo (Section 4.2
). This bears significant ongoing risks for thoroughbred welfare and protection.
The framework of Layers of Engagement with Animal Protection [3
] is applied to further analyse and discuss these findings. Figure 6
is a further development of the layers presented previously in table format (see Table 5 in Bergmann [3
]) to incorporate naturalness in more detail.
Eight layers were identified. They range from those layers striving to maintain the status quo (Layers 1 and 2) through reform (Layers 3–6) and to those aiming at transformation (Layers 7 and 8). There is no strict separation between the discourse affiliated with any layer. The discourse on a particular issue can move up and down these layers, and the layers can overlap. The layers are not necessarily exclusive but can be, and any of the layers can be engaged within a discourse concurrently. They can augment each other but, also, be contradictory and difficult or impossible to reconcile. It is important to be aware of at what layer(s) the discourse takes place. The layers were identified in the context of the thoroughbred racing industry, but they can be adapted to interrogate other animal industries, interspecies activities or multi-species communities.
Most industry informants’ comments explaining and justifying racing practices invoking the natural take place at Layers 1–4. At these layers, the discourse focusses on functioning for optimal race day performance, with welfare being a by-product of or equal to integrity measures. The industry informants’ discourse supporting techno-bio-medical control (Layer 4) is prioritised to optimise the commodifiable characteristics of the thoroughbred. At the same time, these interventions were presented as being in the interest of thoroughbred welfare and safety, as, for example, TBI-6 and TBI-7 responded to Image 4, the tongue-tie is for the safety of the rider and horse. Thoroughbred welfare, as such, gains more weight in the industry discourse at Layer 3, where the focus is on the visible and most egregious welfare violations [3
], but the idea of naturalness is irrelevant at that layer, as it is for industry integrity, at least from the industry’s perspective (more on the discourse in the intersection of industry integrity and racehorse welfare in Bergmann [3
]). Concern for naturalness was reduced to the legitimating rhetoric that the horse “loves to race”. At Layers 1–4, the industry informants and the thoroughbred industry at large see nature as a limiting factor to be overcome through invasive means such as breeding (Section 4.4.1
), the use of drugs (such as furosemide), surgery and equipment (see Section 4.5.1
Layer 5 offers opportunities for significant engagement with naturalness with its interest in the day-to-day living, husbandry practices, training and environmental conditions and, to some degree, horse-human relationships and the consideration of the horse’s entire lifespan. Here, the general animal welfare discourse places at least equal focus on the day-to-day conditions while centring the horse, thus potentially preventing many of the egregious welfare violations. Five industry participants (Section 4.4.1
) made reference to aspects of Layer 5 to varying degrees, including interests in retraining and rehoming retired racehorses, thus acknowledging the natural lifespan of the thoroughbred extends beyond their use in racing and breeding and that this should be catered to. This interest in aftercare, however, is largely due to public concerns and animal advocacy campaigning and, at this point in time, appears confined to reaching for “low-hanging fruit” projects, signalling that the industry is responding to welfare concerns of “wastage” [129
]. There is, however, potential for the discourse around aftercare to move beyond Layer 5 as developments in aftercare evolve, as the discourse around human-animal relationship develops and the protection status of nonhuman animals grows.
Where Layers 5 and 6 meet, the horse-human relationship gains relevance in the discourse. When discussing naturalness, one industry informant (TBI-3) related to the horse-human bond in one instance (Section 4.4.1
). Generally, however, at the systemic level, Layers 5 and 6 currently have limited relevance for the industry informants and the industry at large. At Layers 5 and 6, the discourse moves beyond veterinary science and others based in the natural sciences. Layer 6, in particular, is situated in the scholarly discourse to engage with, for example, (noninvasive) research in animal welfare, ethology, equitation science and the social sciences. Yeates [48
], for example, can be said to be engaging with naturalness at Layers 5 and 6, but the limitation placed on his definition of naturalness as relating to natural animal behaviour only and being distinct from species-specific needs [48
] limits its potential for advancing into broader animal interests and the discourse taking place at Layer 7. It can be expected that those in racing engaging at Layers 5 and 6 will inevitably sooner or later engage more with the concept of naturalness. This is confirmed with the description of the “best-life” scenario for a racehorse in Butler et al.’s study [61
], where the discourse of the “best-life” scenario takes place at Layer 5 and, to some degree, at Layer 6, with the study participants emphasising a positive horse-human relationship and aspects of naturalness. In the interest of thoroughbred welfare and protection, there is a need to shift the focus onto the horse-human relationship as a welfare issue in racing while the industry exists.
It appears that, in contrast to industry informants, animal advocacy informants overall had a strong interest in engaging with Layer 5—in particular, with aspects of naturalness. Some also engaged with aspects of naturalness at Layers 6 and 7. How the animal advocacy informants of this study conceptualised naturalness resembles how people in general consider naturalness. Both tend to view naturalness in holistic terms, including a variety of considerations (Section 2
, Section 4.3
and Section 4.4.2
Industry informants did not engage with Layers 7 and 8. These are the layers where a holistic notion of naturalness plays an essential and defining role for animal protection. Naturalness is considered an inherent worth to be protected and preserved. A rethinking of the ontological status of the thoroughbred—to acknowledge the horseness of the horse (telos)—is also a hallmark of these layers. This goes hand-in-hand with recognising the essential status of naturalness based on evidence. Adopting a holistic notion of naturalness is expected to maximise its potential for thoroughbred protection. Furthermore, the recognition of the thoroughbred’s nature has to extend to a recognition of their individual natures. It has to go beyond the species to acknowledge the individual’s temperaments, preferences, abilities and boundaries; as one of the animal advocacy informants (AAI-6) stated, the horses “are not all machines who despite their pedigree and their backgrounds want to [...] race” (see also [3
]). Engaging with Layers 7 and 8 aims at facilitating a fundamental shift in human attitudes, belief systems and paradigms. It moves toward engagement with animal protection on the animals’ own terms and implements structures and processes for animal representation.
It can be expected that sections within society are interested in engaging with the notion of naturalness as an intrinsic value once the discourse at Layers 5–8 advances in society at large. This will have implications for how thoroughbred racing and breeding will be perceived.
4.5.5. Limitations and Recommended Research
A limitation of this research is the relative lack of participation of industry informants from countries other than the US and Australia (see Section 3.2
). A broader international participation would have been desirable. However, most of the informants are active at the international level and all play a key role in racing, with all holding senior level roles. Furthermore, in terms of numbers, the US and Australia belong to the top racing nations internationally [130
]. Future research could aim at recruiting informants from other racing nations. In terms of animal advocacy informants, the number of organisations to contact was limited, and their representation can be considered satisfactory (see Section 3.2
). Two other proposals for further research are presented below. These arise from the issues surrounding the horse-human relationship as it manifests in shared horse-human activities and from the impact of common practices on the thoroughbred as discussed throughout this article and, in particular, in Section 4.5.1
, Section 4.5.3
and Section 4.5.4
The question arises as to how horse-human shared activities should look so that they increasingly align with Layers 6–8 as the thoroughbred protection discourse advances. Interest in the nature of horse-human shared activities is increasing generally [31
]. The starting point for these considerations is the finding that, while some advocacy informants felt a sense of unease and violation arising from the horse-human interactions observed in the images, they still had some tolerance for horse-human shared activities (Section 4.4.2
). This tolerance is conditional on the following: The shared activities should be within the realm of what is considered natural for the horse, they should provide mutual benefit for horse and human and they should not exploit the horse (Section 4.4.2
). Framing research into the nature of shared activities within a naturalness paradigm is expected to assist in articulating what such shared horse-human activities that are ethical, nonexploitative and of benefit for the horse could look like. Re-evaluating the activity of thoroughbred racing within this context is of public interest for the following reasons: Racing’s legitimacy is in question due to the nonrecognition of the welfare impact of common racing practices (Section 4.5.2
). Furthermore, animal welfare is conceived of as a public good by some [131
], and racing relies on the public as gamblers and visitors to fund their enterprise.
The starting point for the second proposal is the welfare implications of tack—in particular, the bit and the tongue-tie—and common handling practices (see, in particular, Section 4.5.1
and Section 4.5.3
). The question arises whether, and if so, to what degree thoroughbreds during and post-racing engagement suffer a form of post-traumatic stress disorder (PTSD). Common physical injuries are often described by those interested in ex-racehorses [132
], but there is also anecdotal evidence that supports the suggestion that ex-racehorses are left with emotional trauma [134
]. The evidence presented in Section 4.5.1
and Section 4.5.3
appears to lend support to this suggestion. PTSD has been shown to occur in other animals [135
]; yet, the condition described as PTSD is generally not used in the literature to describe the psychological state of thoroughbreds showing particular symptoms. Noninvasive research to investigate the status of thoroughbreds in the context of PTSD and strategies to prevent its occurrence are required, as long as racing persists. This study has demonstrated that naturalness as a guide centres thoroughbred welfare and protection. It is therefore recommended to frame the suggested research within this paradigm.