Coalitions of touch: Balancing restraint and haptic soothing in the veterinary clinic.

This article responds to recent calls to further incorporate the study of animal health care into the sociology of health and illness. It focuses on a theme with a long tradition in medical sociology, namely clinical communication, but explores matters distinctive to veterinary practice. Drawing on video recordings of 60 consultations across three small animal veterinary clinics in the United Kingdom, we explore how clients and veterinarians (or "vets") fashion fleeting "coalitions of touch," that aptly position the animal to enable the performance of medical work, often in the face of physical resistance. Building on recent developments in the study of haptic sociality, we analyse how care and emotional concern for animal patients is communicated through various forms of embodied action; thus, how the problematics of forced care and restraint are mitigated through distinctive ways of touching and holding animal patients. Moreover, while prior studies of small animal veterinary work have highlighted the significance of talk within the clinician-animal-client triad, we reveal the fundamentally embodied and collaborative work of managing and controlling patients during sometimes intense and fast-moving episodes of veterinary care.

INTRODUCTION

The sociology of health and illness has historically prioritised human health. While there have been recent calls to incorporate animal health care (Hobson‐West & Jutel, 2020), the analysis of animal clinics in sociology remains very much in its infancy (although see Brown & Nading, 2019; Everitt et al., 2013; Hobson‐West & Jutel, 2020; MacMartin et al., 2014; Morris, 2012; Roberts, 2004; Sanders, 1994, 1995; Stivers, 1998). Building on this seminal body of work, this article explores collaboration during consultations in small animal practice, where there has “to date been little research” (Everitt et al., 2013: 453).

One key theme that has emerged in prior studies on animal health care connects to a long‐standing interest in medical sociology with clinical communication (MacMartin et al., 2014; Stivers, 1998), and the nature of interaction within the clinician–client–patient triad (Grimm et al., 2018). The triad presents a range of communication challenges, which have been charted through a distinctive body of mainly conversation analytic work focused on talk (MacMartin et al., 2014; Roberts, 2004; Stivers, 1998). Also concerning communication, the second set of studies has explored professional perspectives on the role and participation of the client (Clarke & Knights, 2019; Sanders, 1994, 1995). This access the attitudes and narratives, primarily of veterinarians, who often problematise the role of the client, seeing them either as unable or unwilling to control their animals, or make good decisions (Clarke & Knights, 2019: 227). While these are studies of talk, or studies of professional perspectives on talk, there is surprisingly less sociological research concerned with the fundamentally embodied character of veterinary work; how veterinarians and clients undertake body work (Heath, 2006; Twigg et al., 2011) with animal patients.

In our study, we analyse video recordings of 60 consultations in small animal clinics, and it is readily apparent that clients routinely adopt the role of veterinary assistant, supporting the vet in the course of physical examination and treatment. They adopt a tactile and moral “coalition of touch” with the vet, to cope with the sometimes unpredictable and wild movements of animals. Below, we explore these tactile coalitions, drawing upon recent developments in the field of haptic sociality (Cekaite & Holm, 2017; Nishizaka & Sunaga, 2015). We think this work has much to offer the sociology of health and illness more broadly, by drawing attention to the ways in which care and emotional concern for patients is accomplished and communicated through touch and physical intervention.

In our case, we found it fascinating that there would often be three pairs of hands on a particular animal, each moving in response to the contingencies of the situation. Even without verbal instructions, animals are held in ways that are attentive to the medical work at hand, and to the status of the animal as a sentient, and loved, actor. We analyse how clinicians and clients deal with the practical and moral‐emotional problematics of delivering health care in and through different forms of tactile engagement, and often while the animal physically resists. This in turn provides us with some scope to consider the agency of animals (Carter & Charles, 2018); in particular, how they respond—and are understood to respond—not just to talk, but also to ways in which they are touched and held.

Veterinarian–animal–client interaction

Reflecting on the overall structure and distinctiveness of veterinary consultations, previous studies have described a complex triad (Grimm et al., 2018; Sanders, 1994) of veterinarian, animal, patient and client (or animal owner). When taking histories, the animal's inability to provide verbal information positions the client in the role of the provider of lay accounts of the animal's health and behaviour. As consultations progress, the veterinarian and the client cooperate to devise strategies through which the animal is cast in the role of the “virtual” or absent–present patient (Gregory & Keto, 1991). Animals are talked about more candidly and bluntly than their human counterparts. While some studies draw similarities to human clinical encounters involving children or adults less able to speak for themselves (Mitchell, 2001; Sanders (1994); ; , , Sanders (1995) argues veterinarian–animal–client interaction “differs from those typically encountered in human practice by being less clouded with medical ritual and mystery, more likely to entail overt negotiation of diagnosis and treatment, characterised by use of plain rather than technical language” (Sanders, 1995: 200). This is most apparent in cases of euthanasia in veterinarian practice (Morris, 2012), where the animal's death is openly discussed, and, therefore, does not include the ceremonial features routinely observed in human health care (Maynard, 2003). Noting a further difference, Grimm et al. (2018) highlight the particular importance of the client in treatment decisions, as they face economic constraints that might alter those decisions, and also offer an alternative from the professional perspective on treatment options.

Studies of clinical communication in veterinary work have revealed that the structure of the encounter consists of history taking, examination, explanation and closing, but each is accomplished in an “iterative fashion rather than the linear fashion described in several models and frequently taught to veterinary students” (Everitt et al., 2013: 458). And these studies of communication have spotlighted the triadic qualities of the encounters. For instance, Roberts (2004) considers talk by veterinarians directed at pets or on behalf of pets, in the animal's “voice” (Clarke & Knights, 2019). They find that such talk is used to “critique client caretaking of the pet, deflate client complaints, palliate client concerns” (Roberts, 2004: 441). Similarly, MacMartin et al.’s (2014) examination of pet‐directed “I know” receipts reveal how vets vocally modulate their stance towards their patients’ expressions of distress. In these cases, talk seemingly directed to the animal is concerned with managing the client's participation. In relation to diagnosis, Stivers (1998) explores vets’ use of prediagnostic commentary to forecast serious trouble. This creates interactional opportunities for clients to participate in the emerging and inchoate diagnosis. More generally diagnosis in the veterinary clinic could be cast as a “complex dance” (Hobson‐West & Jutel, 2020: 398), where professional authority can be challenged due to the complex ethical, and other, responsibilities that emerge from the triad of veterinarian, animal and client.

A further theme that emerges in the prevailing literature sees clients as a source of problems for vets (Clarke & Knights, 2019). Sanders (1994) argues that vets form typologies of “good” and “bad” clients. Accessing these frameworks, he describes clients that are “annoying,” “ignorant,” “demanding,” “neglectful” and “over‐involved.” The ascription of such categories hinges, in part, on the client's inability to control the animal, and to respond appropriately to rules and definitions of good behaviour. Vets accept that animals will behave badly, but expect clients will accept negative framings of uncooperative animals and take remedial actions. “Good clients” side with the vets accounting, are contrite, and offer reasonable explanations of “‘bad behaviour” (Sanders, 1994: 59). But the client's attitudes to the animals (Clarke & Knights, 2019: 229) and emotional state also pose problems for vets, who have to deal with the clouded judgement of “grief‐stricken” and “guilty” clients (Morris, 2012). Such problems may be compounded by clients’ financial ability to pay for costly treatments.

So, prior studies have explored communication in the veterinary clinic and various dynamics that arise in the veterinarian–animal–client triad. However, we know from Everitt et al.’s (2013: 454) study of veterinary work, that the “largest group of consultations” tends to involve “animals presented for routine or preventive procedures such as vaccination or nail clipping” and that all these consultations “involved some degree of physical examination including visual examination and palpation of the animal” (Everitt et al., 2013: 455). And yet, we lack an understanding of embodied work in the veterinary clinic, how vets and clients draw upon multiple modalities—language, but also gaze, gesture and other forms of bodily conduct—to coordinate the delivery of medical care. From our own corpus of data, we see the client is routinely drawn into assisting the vet in undertaking the examination or providing the treatment. From Sanders’ (1994), Sanders (1995) work, we know that animals need to be variously subdued, cajoled and restrained during these episodes. But to the best of our knowledge, few, if any studies, have explored what is at stake when animals, who are sentient and often loved companions, are touched, manoeuvred and even forcefully restrained, in the context of often painful and distressing medical interventions.

Studies of human health and medicine have explored multimodal resources used in the coordination of medical work, and the distinctive problems that arise when the “object” of collaborative work is the conscious or unconscious body of the patient; how medical professionals work on and around the bodies of human actors (see, e.gHeath, 2006; Pilnick & Hindmarsh, 1999; Twigg et al., 2011). The veterinary clinic differs in important regards. Most obviously, one of the human actors is also the patient's owner and companion, and in comparison to the vast majority of human patients, animals cannot be guaranteed to behave cooperatively.

Exploring embodied work in the small animal clinic, we unpack distinctive “coalitions” (Coe & Prendergast, 1985) that form between human clients and vets, and explore how these are instigated and unfold, typically not as a result of talk or verbal instruction, but apparently spontaneously, through improvised embodied action. Some years ago, Coe and Prendergast (1985) outlined how—in medical consultations involving doctor, patient and a patient's relative or companion—coalitions would form. For instance, the doctor and relative might work together to encourage the patient to agree to some treatment recommendations. In veterinary clinics, these coalitions are not merely verbal but incorporate bodily qualities. In particular, veterinarians and clients come together to use their hands and bodies to encourage the animal to participate in the procedure. These episodes involve the formation of what we term “coalitions of touch”—designed to manage resistance and work in support of the medical procedure. Our analysis unpacks various features of these coalitions.

Exploring such themes, we draw upon and inform recent developments in haptic sociality (Cekaite & Holm, 2017; Nishizaka & Sunaga, 2015) concerned with the social and interactional uses of touch. This work has addressed how others are held and manoeuvred in ways that attend to both the practical contingencies of the situation and also emotional and interpersonal considerations (Goodwin and Cekaite 2018). Our contribution to these studies is tied to the distinctive setting we consider, where the animal is held so that they endure short‐term pain and discomfort in the interests of longer‐term health outcomes. In veterinary clinics, clients and vets endeavour to hold animals in appropriate positions for the upcoming or ongoing medical work, and cope with any resistance. These coalitions of touch are intriguing, not only because they reflect participants’ understandings of what the animal will tolerate but also because animals themselves will often resist how they are held; they flex, recoil and indeed bite in ways that demand the dynamic reconfiguration of coalitions of touch. We showcase an innovative way of accommodating the animal in the veterinary clinic then, as a sentient actor that is attentive, not just to some verbal commands, but to ways in which they are touched and held.

Accessing the veterinary clinic

Sixty small animal veterinary clinics in the United Kingdom were identified online and approached by a letter inviting them to participate in the study. The majority of all veterinary work in the United Kingdom takes place in such “neighbourhood clinics” (Robinson et al., 2019). They differ considerably from a large animal, for example, equine (5.5% of all veterinary work), farm (3.2%) or zoo (0.7%) practices, which are more specialised and fewer in number (Robinson et al., 2019: 11). Three responded positively and were visited to discuss the study, consent procedures and practicalities of filming. Following these initial visits, practice managers discussed the study with their clinical teams, and we awaited a response. All agreed to participate. Terms of reference were agreed with the practice owner, including the principle that employees and clients were free to withdraw without giving a reason. All practice employees were approached individually and consented to participate in the full knowledge that excerpts and images from the video recordings would be used in publications and presentations.

In line with guidance on video‐based research (Heath et al., 2010), we gained the consent of clients through a systematic process described in the terms of reference agreed with practice owners. Of course, “consent based on the autonomy of the patient, as is predominantly the case in medicine, is unsuitable for the veterinary context” (Gray, 2020: 1). So, where possible, the study was discussed with clients when they booked their appointment; and they were advised they may be approached to discuss the possibility of filming their consultation. The researcher was given access to clients prior to their consultations and discussed the project with them in a separate, private consultation room. Clients were given full information about the project. They were asked for written consent to both record the sessions and use images from the recordings in research publications and public settings, such as conferences. They were made aware that they were not obliged to participate and could withdraw at any time without giving a reason. Very few people declined to participate.

Over a period of 3 months, we video‐recorded 60 consultations across the three clinics. These varied in length, but were typically short, lasting less than 5 minutes. Recordings took place in consultation rooms positioned off the main reception area where clients waited with their pets. These rooms were linked electronically to the reception; vets had access to the appointment schedule, which glossed the reason for the visit, and would add notes on the system about various next actions for the receptionists. The rooms were small and fairly simple, each had a sink, soaps and cleaners, a bin, towels, cotton wool, a computer, some equipment, such as weighing scales, pet‐treats, as well as muzzles and stethoscopes. Drugs and treatments were stored in cupboards. In each clinic, a single camera was placed in an unobtrusive position, taking the vet's preferences into consideration. Small, cramped rooms posed numerous problems for recording clean unobstructed images. Often numerous actors were present, who would move around, sometimes blocking lines of sight.

Drawing upon resources from ethnomethodology and conversation analysis, we analyse how actors orient to in situ talk and embodied conduct, and confronted the challenges posed by the “triangular” (Sanders, 1994) nature of interaction in the veterinary clinic, where human actors often produce utterances addressed to the animal (“okay old boy, let's have a look at you”), that are treated by the client as having practical relevance for them. To support our analysis, extracts of the video recordings were transcribed using the notation system developed by Jefferson (1984) which captures details aside from spoken words, such as intonation, prosody, speech timing, overlap and so forth, as well as that of Goodwin (2000) to document material and bodily conduct. While a number of prior studies of veterinary practice have collected video recordings and been informed by conversation analysis (MacMartin et al., 2014; Roberts, 2004; Stivers, 1998), they have tended to focus on the structure, form and functions of the talk, with relatively little consideration given to embodied conduct.

As the analysis progressed, we came to focus on the establishment of sometimes fleeting coalitions of touch, through which clients and veterinarians hold, position and limit the movement of animals. Hands would be on an animal, and would move in relation to one another as the animal responded to being examined or treated. This happens in the absence of any formal plan, and routinely without any explicit verbal instructions. We became fascinated by sometimes intense moments of collaboration, where the action is fast‐paced and even wild, but the work nevertheless gets done through apparently spontaneous forms of collaboration. Our work addressed continual microadjustments (Mondada, 2016) through which clients and vets collaborated, drawing upon embodied tactile resources.

FINDINGS

Recruiting assistance

We start by considering how clients come to play a hands‐on role in the consultation. In many of the cases we consider, clients do not need to be actively recruited by the veterinarian, to participate. Understanding the animal is booked in for a particular treatment, the client lifts them onto the table, and holds them in place, in anticipation. In other cases, the client's embodied assistance is more obviously encouraged and we start with a straightforward case (number 40 in the corpus from clinic “F”). Here, two clients (C1 and C2 in the transcript) are attending with their 10‐month‐old kitten which the vet (V) is examining. We are 2 minutes and 16 seconds into the encounter. Holding the animal in mid‐air, the vet asks about the kitten's condition, which they describe as “about normal” and “wheezy,” before the cat is placed back on the table. Then, the vet explains he is “just gonna get (me) a little head torch” (lines 1–3).

As he announces that he is going to get a head torch, the vet maintains his right hand lightly on the kitten. The animal is not restrained by the vet, his hand is merely in the right place to perform restraint, should a problem arise. But by lightly touching the animal in this way, the vet displays a particular orientation to the kitten—that they should not be left unattended. Quite quickly, the clients come to recognise a particular contingency of the vet “getting a head torch”; he is leaving the room and will soon remove his hand from the animal. C1 looks down to the cat as the vet says “torch” and then back to the vet. The vet immediately starts tapping the cat's back, which encourages the client to reach over to place a similarly gentle hand on the kitten's neck and shoulders. Akin to a “baton change,” both parties have one hand on the kitten for a minimal period; a very brief “coalition of touch” is formed and timed so smoothly that the vet's movement out of the room is uninterrupted. So, the vet guides the client into action, drawing upon embodied resources.

As the vet turns the door handle to exit the room, it makes a pronounced creaking noise (images 4–5) and the kitten recoils. However, C1’s hand is in exactly the right place to restrain them, pressing the kitten briefly downwards to the surface of the table. The character of touch is transformed in an instant. C2 likewise sees the problem and responds verbally, in a caring register (“alright, good boy,” lines 9–11), and also gesturally, by reaching over to stroke the rear of the animal.

It is notable that neither C2 nor C1 picks the cat up. They continue to maintain the animal in position and ready for examination. As C2 begins to stroke the cat, C1 relaxes her grip, lifting her hand from the cat's shoulders to stroke its head. But in this configuration, the cat finds new possibilities for movement and sharply turns his body and head to the right. C1 responds, and once more the character of touch is transformed, as the client re‐assumes a firm hold around the kitten's neck. The vet is not even in the room, but the clients take it that their role is to maintain the position of the animal.

In this initial example, we see the limits of a mere talk‐based approach to veterinary consultation. From the talk or transcript alone, we would be unable to see how the client is enlisted to provide assistance. Various problems are resolved instead through embodied work, coordinating particular ways of touching and patting the cat. Rather than arguing the client's actions are explained by their attitudes, we are showing how their participation is interactionally accomplished. Indeed, we have started to consider how clients engage with the embodied actions of vets and how those actions can be designed to reveal the implications of forthcoming courses of treatment for the animal and, by that token, for clients to support the consultation in progress.

Knowing how to hold animals: seeing the procedure

Above, we began with a straightforward example, where the vet performs embodied work to encourage the participation of the client. More typically, the client's participation is responsive to the unfolding organisation of the vet's work. In the absence of discrete efforts to marshal their participation, the client sees the procedure and anticipates its implications for them, determining how they should provide the vet with timely and unencumbered access to the relevant part of the animal's body.

In extract two the consultation is just beginning. The female client (C2) has lifted the dog, “Alfie” (all animal names below are pseudonyms), onto the table. When the vet returns to the room with a pair of surgical gloves, the male client (C1) suggests a muzzle could be used for the procedure (“ave'y’got a muzzle,” line 1). His account for the request (“last time we came in he weren't too happy,” line 4) is enough for the vet to reach for a restraining muzzle.

The vet returns with the muzzle and starts to unclip it for fitting purposes. C2 holds Alfie whilst C1 strokes him under the ear. In the way the animal is touched and held, there is a close similarity with extract one, as we see a delicate mix of restraint and care. As the vet lifts the muzzle upwards, a hand on each strap, C1 recognises the implications of this for his own conduct by withdrawing his stroking hand to give the vet unencumbered access to the dog's muzzle. So, he recognises the trajectory of the procedure and determines its relevance for his own conduct. Verbally marking the onset of the vet's work, C2 inclines her head towards the dog and says “right alfie, be a good boy” (line 7). They both attend to the onset of the muzzling. Without anything being said, they have performed preparatory activities, assembling the animal in ways that will enable the vet's work.

As the muzzle approaches Alfie's snout, the female client's (C2) left arm is around his shoulders and neck. She then shifts her grip, first fussing below Alfie's ear, before gently clenching her hand to hold the scruff of his neck. In this way, she perhaps anticipates the likely response of the animal. But as the muzzle touches the dog's snout, Alfie ducks his head downwards and away, frustrating the vet's work.

The clients immediately reconfigure, as the female client (C2) re‐adjusts her grip, withdrawing her left hand, to give the vet the space he needs. Anticipating further problems, she instructs C1 to “hold his nose” (line 7). From the onset of these minor troubles, C1 has been moving his hand in a number of tentative motions towards Alfie's nose, recognising the kind of help that might be required, whilst displaying his continuing role in the three‐party coalition. Just at this point, the vet succeeds in hooking the muzzle over Alfie's nose. The request for help (“hold his nose,” line 6) is no longer relevant, and C2 recognises this (“doesn't matter,” line 7). In a further sign of her close engagement with the vet's work, it is C2 rather than the vet that verbally reports the successful muzzling (“that's it,” line 8), uttered in a soft caring register (Cekaite & Holm, 2017), as if produced for Alfie (line 8).

In this case then, the vet does not perform a discrete action to cue the client's participation. Instead, it happens apparently spontaneously with the clients able to recognise the implications of the emerging conduct of the vet, as well as the movement of the animal, for them and the part they should play in supporting the task in hand.

In extract three, the animal's reaction is altogether wildered, the adjustments more fluid and fast‐paced. The vet is performing a general examination of a dachshund, called Max, and has just listened to the dog's heart. Max is facing the client, with his head nuzzled into the client's waist, while the client gently holds Max's front shoulders.

The vet is able to listen to Max's heart in this configuration, but he cannot access his eyes and teeth. So, they need to reconfigure. The vet reaches over, and grabs the animal, twisting Max's upper trunk and the client assists. Maintaining his grip on Max's shoulders he helps to slide Max's front quarters into the new position. And with his right hand under the dog's jaw, the vet tries to examine Max's eyes. The dachshund responds by suddenly shaking his head, a reaction that occasions a rapid re‐adjustment of their coalition of touch as Max writhes and flexes.

The client responds verbally in a caring register (“alright Max,” line 1) and also undertakes embodied preparatory work to enable him to resolve the “postural instability” (Schegloff 1998) that has arisen. He needs to brace the dog, but to do this, he will need to reach over to Max's hindquarters with his right hand and slide Max's rear‐end towards himself. However, he is holding the lead with his right hand, which makes this manoeuvre more difficult. So, as he says “alright Max” (line 1), the client switches the lead to his left hand and rotates the animal through ninety degrees (see images 1–3) and braces him. The client's movement has immediate implications for the vet, who has his left arm around the dog's left flank (see images 2 and 3). With the client now restraining Max's body, the vet can use both hands to conduct the examination. In a finely organised adjustment of their coalition of touch, as the client slides the animal towards himself, the vet withdraws his left arm from the dog's flank.

At this point, Max is now suspended off the table by the vet's right hand, which is underneath his jaw (image 3). The client is again slightly behind in the procedure. He needs to support the weight of the suspended animal but is now holding the lead with the hand that is best placed to perform this work. Once more taking preparatory action, the client drops the lead altogether so he can bring his left hand underneath Max's chest and support his weight (see image 4). As he does this, the vet is able to release the body of the animal, so he removes his hold from around Max's neck and with both hands free, starts to perform the eye examination. With both hands around either side of Max's skull, the vet pulls his skin back (see images 5 and 6) to more clearly reveal the state of his eyes. As he does this, the client once more responds, retracting his left hand, to give the vet the access he needs with this right arm (see images 5 and 6), whilst moving his right hand upwards to better support the weight of the dog.

When these few seconds of video were first played, we saw a formless, chaotic tussle. But for the participants themselves, the interaction does not seem formless at all. When slowed down, and viewed repeatedly, in ways that are characteristic of coalitions formed around struggling animals, we see the vet and client are feeling their way through the examination, supporting one another's conduct at a rapid pace. The vet does not verbally instruct the client. Rather, the client sees the procedure through an immediate situated engagement with the vet's conduct and the animal's body. The vet begins the procedure and the client makes sense of it in its course, to establish appropriate forms of restraint that will support the vet's work. As the action unfolds, nothing is said, and yet problems are resolved through subtle adjustments to their coalition of touch.

Restraining animals with care

We have already seen how clients and vets orient to the animal's (dis)comfort through talk. Clients talk to animals in ways that attend to their emotional state, whilst encouraging them to “behave.” In line with previous studies, the client's position is complex, they align with the accounting of vets, which always positions the animal as the source of any troubles, whilst simultaneously restraining and soothing the animal. In the veterinary clinics we studied, this soothing took on a further dimension. Clients attended to expressions of pain and discomfort in the way they held animals. Previous accounts of “haptic soothing” likewise describe complex multimodal processes, where talk and touch are “laminated” (Cekaite & Hom, 2017: 116) and veterinary encounters provide a further and rather distinctive environment for the analysis of such processes.

We consider this in extract four where a small dog, Bella, is being investigated for a suspected case of ear mites. To perform the examination, the vet will need to hold the right ear back with one hand, to insert the otoscope with the other.

In this case, the vet recruits the client's assistance by verbally instructing them. This was rare in our data and, when produced, such instructions tended to reflect the distinctive social relations of the veterinary clinic. In this case, for example, the vet gently requests (“do you want to,” line 1) that the client attempt to restrain the dog (“see if you can,” line 2). So, he does not orient to an institutional entitlement to direct the client's action (Heinemann, 2006). In addition, the vet guides the conduct of the client through an instructional gesture, as he illustrates the kind of hold that is required by momentarily placing the palm of his right hand over the dog's muzzle.

The client reaches towards the dog and takes hold of her snout, but contra to the vet's embodied instruction, she places the palm of her left hand under the dog's jaw (see image 1). Bella immediately resists, backing away and sharply twisting and flicking her head down. As she first tries to take hold of Bella's nose, the client speaks to her in a caring register (“bella,” line 5), before producing a response cry (“uhh, don't,” line 5) as she backs away. Rather than tightening her grip, the client releases. Restraint is not always a matter of brute force, but of learning what animals will tolerate.

As this happens, the vet is trying to insert the otoscope into the ear canal and place his eye to the viewing window. Through the brief period of struggle, the vet manages to grip the ear but pauses before moving his head towards the window. He is attentive to the clients’ work, waiting for her to re‐establish hold of the animal, before proceeding.

Subsequently, the client adapts her hold in two ways. Firstly, she places the palm of her hand on the dog's muzzle, as the vet had instructed (image 3). The animal accepts the hold. Moreover, without further instruction, the client places her right hand underneath the dog's ribcage, lifting her slightly thereby constraining her ability to back away. The client talks to Bella throughout. As she asks Bella whether she is “a good girl” (line 9), the client discovers a particular affordance of the hold she has established. With the thumb of her right hand, she fusses the dog, stroking her back and forth, whilst simultaneously taking some of her weight. Whilst holding the animal in place to enable the medical procedure, she is soothing verbally and through touch.

Verbally, the client produces a repertoire of responses finely matched, emotionally and temporally, to the dog's conduct. As the ear examination continues, Bella elicits a brief squeal. The client is in the midst of asking Bella whether she is a good girl when this happens, and immediately cuts off from this, to switch emotional register. In caring tones, Bella is told it's “alright” and then, when the squeal finishes, in very soft “whispery sounds” (Cekaite & Holm, 2017: 116) she shushes the dog.

As the soft shushes are produced, Bella emits a louder and more prolonged squeal and tries to back away. The client takes the weight of the dog more fully on her right hand, lifting her slightly to impede her movement. As she squeals, Bella is twice told that it is “alright,” before further soft shushes are produced, which cease only as the squealing comes to an end. In both cases then, the client intervenes until the animal is silent. In this new configuration, the client, who is bearing the weight of the dog on her right hand, is no longer free to fuss her flanks, so different soothing modalities come into play in response to different contingencies.

These soothing displays are well understood as products of the participation framework of the veterinary clinic. They are produced as part of a project of supporting medical work, rather than just attending to the animal's emotional state. For example, following the client's question (“you bein a good girl,” line 20) Bella once more emits a loud prolonged squeal. But the vet has now finished his work and rather than comforting the animal in any way, the client immediately releases her hold and begins to speak with the vet. This expression of pain is not attended to.

The second example of haptic soothing is considered below, where the animal is simultaneously soothed but also braced very firmly in position for a treatment that causes them some discomfort. In this case, the animal is rendered docile through a certain degree of force, but even here, we see different haptic registers engaged. This extract also features Max who, now muzzled, is about to have his swollen anal glands examined through an invasive procedure that will take roughly 30 seconds. We return to this case to illustrate how, within the same consultation, vets and clients move into and out of multiple coalitions of touch, each subtly formed in ways that are attentive to the particular problem at hand.

As the action begins, the male client's (C1) role will be to brace the dog from the side, whilst the female client (C2) fusses and soothes him from the front. The coalition of touch includes three parties, each with a demarcated role. C2 verbally instructs her companion, saying he should “get him around here” (line 3), pointing to Max's right flank, whilst passing him the lead. She speaks to Max saying, “it's alright I’m here” (line 8). Again, without any verbal instructions from the vet, the clients assemble a coalition of touch that gives the vet access to the rear of the animal, while simultaneously restraining and soothing them.

The vet, having sterilised his gloves, verbally marks the beginning of the examination (“alright lad (0.2) be over soon,” lines 10–12). The vet's talk to the dog is treated by C2 as highly relevant to their work as an assistant, as he moves to brace the dog more fully. More generally, clients were often alerted to the onset of discrete medical tasks in this way, not simply by the vet starting the work, but through basic verbal cues, often overtly directed to the pet.

As the vet grabs and lifts the dog's tail, C2 restrains the animal, tightening his grip incrementally. To hold an animal tightly is to invite resistance, and clients tended to brace them fully only when entirely necessary. As the vet prepares to insert his index finger into the dog's anus, the client moves his right arm to clasp his hands together (images 2 and 3). The client anticipates the negative reaction from the animal. The work is so finely coordinated that C2’s left hand firmly clasps his right wrist precisely as the vet inserts his finger. The client is guided by the sight of the vet's movement, he sees the procedure, which allows the dog to be forcefully restrained only when necessary.

As the treatment continues Max emits a continuous low growl. C1’s response is not to brace the dog more firmly, but to begin to stroke him with the thumb of his right hand (image 4). Again, we see clients taking up opportunities to modulate touch in ways that are attentive to unfolding dynamics, in this case, to soothe an animal that is increasingly agitated. As the animal's growl builds, he barks once more and rears up, struggling violently, his front right paw breaking free of the client's hold. The tactile affordance of the brace—which allowed the client to soothe the animal with his thumb—is now lost. The soothing ends and he cannot readjust his hands easily because the dog—who is still being treated—might break free entirely. In the end, there are six hands on this small dog forming a complex coalition of touch.

Limits of restraint

Physical restraint is not always straightforward, as animals and their owners often find themselves fashioning coalitions of touch that are not effective or desirable. We now consider some limits of restraint, noting veterinary work can be dangerous. Indeed, a survey conducted in the United Kingdom in 2015 (Veterinary Record 2015) found roughly two‐thirds of vets reported being injured in the course of the previous year. Across the 60 consultations we considered, on three occasions animals tried, but failed, to bite or scratch vets.

Our final empirical case involves a kennel cough injection, where half a millilitre of liquid is squirted up a dog's nose. Dogs dislike this treatment and often resist intensely, whilst older dogs are understood by all parties to be able to anticipate it, often from quite subtle cues. As the action begins, the client and vet have been discussing the procedure, with the client proffering an opinion that problems begin when dogs see the syringe. For his part, the vet has opened and removed the syringe from the paper packaging some minutes previously. This is part of his professional experience, that even the ripping noise can cue negative reactions from some dogs.

The vet approaches and presents the syringe to the dog, producing talk designed for her (“Yes. up your nose. come on,” line 1). He grips the dog's snout, but covers her eyes with his left hand and lifts her head upwards. He raises the index finger of his right hand ready to compress the plunger (see image 3) should he connect squarely with the dog's nose. However, he does not and the dog starts to struggle, flicking her head away from the vet. The vet's firm “no” (at line 7) is addressed to the dog, and marks her transgressive behaviour. The dog continues to resist and soon becomes suspended off the table, writhing and flexing in the arms of her owner, who emits a brief groan (“arghh,” line 9) as he struggles to hold her steady.

It is common for animals to respond negatively to the way they are touched and, ordinarily, clients and vets then fashion a new coalition in which the animal acquiesces. This case is unusual because the animal cannot be calmed; the vet is trying to hit a moving target. At this point, the dog growls and lashes out at the vet. Flicking her head downwards and to the left, she escapes the vet's grip, and attempts to bite him (image 5). The vet narrowly escapes. The client, who is working hard to hold this large struggling dog, verbally marks the transgression, producing a response cry (“OU‐oup,” line 12), albeit without apologising on behalf of his pet.

The vet's choice now is whether to try again to administer the treatment or to restrain the dog with a muzzle. He briefly starts to try again, raising the dog's head once more. But she remains uncooperative and it is the client that suggests the use of the muzzle (lines 16–19), directing the conduct of the vet delicately and indirectly. Rather than producing a request, or even instruction, he merely describes a possible recollection (“I think,” lines 16 and 18) of what “we” did previously. The dog is muzzled, and the injection is immediately and simply administered.

This extract shows ways in which the coalition between the client and vet is practical, but also normative. The client does not question why the muzzle was not used originally, sparing his dog some distress. In our materials, clients always verbally align with vets in their construction of the animal's non‐cooperation. They typically align with the vet in the way the animal is held too. The client above could have physically withdrawn the animal, but they maintain them in a treatment‐relevant position. In this case, upon successfully completing the procedure, the vet jovially calls the dog a “silly thing” that at “her age” should have “more sense.” The problem was the dog and not the vet's approach. The client, who suggested the use of the muzzle, laughs along.

DISCUSSION

Coe and Prendergast (1985) considered how physicians and family relations establish coalitions in triadic interactions with patients. Rather than being static and fixed, coalitions would ebb and flow around particular concerns. For example, the daughter of an elderly parent might “bid for a coalition” by seeking to elicit the doctor's support against the espoused wishes of the patient (Coe & Prendergast, 1985: 244), which might succeed or be overlooked. In our data, of course, the animal patient cannot speak and these dynamics do not come into play during diagnostic discussions (Sanders 1985; Morris, 2012; Everitt et al., 2013). Instead, they arise through the course of routine examinations and treatments where clients work to support the delivery of health care, often against the apparent wishes of their pet companions, through what we have called “coalitions of touch.” The characteristics and broader importance of these tactile coalitions are considered below.

Firstly, whilst the veterinarian has professional tactile competencies, the client may know better how to hold, calm and soothe the animal. Indeed, only twice across our corpus was a veterinary nurse called to assist the vet. Once, where an elderly woman was struggling to control her large dog, and another, where a thyroid injection was to be given to a disgruntled cat. In all other cases, vets rely upon the labour of clients. Medical care in these contexts would be more difficult or more costly to provide without the client's assistance. In training given to vets and veterinary nurses, the work of holding animals is often described as a taken for granted clinical skill, and Sheldon et al. (2016: 6) argue clients “should never be asked to restrain their animals” because they do not “understand the complexities of restraint.” In our data at least, the restraint of animals is typically not just the vet's work, it is performed in coalition with the client.

Secondly, the coalitions of touch in this article reveal further aspects of verbal communication within the clinic. Sanders (1985) describes good clients as those that “cede interactional control to the vet” whilst being “attentive to instruction” Sanders, 1994: 159). But in our data, we found clients played a much fuller role. On the rare occasion when vets produced direct verbal instructions for clients, those instructions were considerably gentler than those observed in studies of medical teams (e.g., Mondada, 2014). Vets might merely request clients “have a go” at helping. Furthermore, and in line with the argument of Everitt et al. (2013: 457) that the client has a “significant influence on the structure and interaction of the consultation,” we found that clients would even verbally instruct vets on matters pertaining to the restraint of the animal. Whilst this might seem potentially transgressive, both parties seemed well able to manage it. For example, rather than directly instructing the vet, the client in extract 6 merely recalled a previous encounter where restraint was necessary. So, the client's access to the animal's history may inform not only diagnosis (Hobson‐West & Jutel, 2020) but also decision‐making about how to conduct examinations and perform minor treatments.

Thirdly, whilst talk remains important, we have also noted the limitations of a purely talk‐based approach. In some cases, there is no verbal request to form a tactile coalition and the client's participation is organised entirely through embodied actions. Even where vets use pet‐directed talk to guide the client's participation (MacMartin et al., 2014; Roberts, 2004) these cues hardly explicate the forthcoming procedure. As Goodwin (2000) has argued, directed actions are best understood as taking definite trajectories. So, when a vet says, “alright lad, be over soon” (extract 4), this does not wholly elaborate the procedure or when it will begin. The implications of the vet's pet‐directed talk have to be established by the client, through an immediately situated engagement with the vet's embodied conduct. We observed many occasions where clients had to improvise, and draw upon their understanding of the trajectory of the vet's conduct, coupled with their historical and emerging tactile knowledge of their animal, to determine how they might participate. In these interactions, contingencies changed quickly, and parties participated as if spontaneously, feeling their way through the sequence. Without attention to bodily conduct, these fundamental features of veterinary practice may be overlooked.

Fourthly, the coalitions of touch we have explored connect to an interesting body of research concerned with “haptic sociality” (Cekaite & Holm, 2017; Nishizaka & Sunaga, 2015), which considers the communicative functions of touch, through actions such as “stroking, patting, holding, or hugging the other” (Hertenstein & Weiss, 2011: 283). We think this work has much value in coming to terms with the distinctive character of the veterinarian–animal–client triad, certainly in the context of small animal clinics, drawing attention to the way care and emotional concern for the animal patient is communicated through various forms of touch and physical intervention. For example, even quite strong forms of restraint are modulated to exhibit care, and reconcile tensions at the heart of such situations. Indeed, we found that restraint is not a matter of brute force, but of holding the animal so it acquiesces to the imposition of certain physical restrictions. It is possible to hold a cat in place with the lightest touch that nevertheless, when removed, will see them flee (extract 1). Touch is communicative in this sense. Further, we have seen ways in which haptic soothing is achieved in the very course of the act of restraint. For instance, clients make good use of their opposable thumbs to stroke the animal whilst simultaneously holding them steady. Nevertheless, it is evident that the client prioritises restraint, over tactile soothing, and emotional support. When the need arises to tighten or re‐shape a grip, the tactile soothing ceases until the design of the grip offers new opportunities to stroke the animal, revealing an overarching orientation to the integrity of the procedure and, maybe, the limited time available to them in the clinic.

Following on from the above, we have shown that clients and vets alike do not touch and hold animals as objects, but as subjects. Cekaite and Holm (2017) suggest that whilst touch interacts with other modalities, gaining its sense from the way it interacts with forms of talk, gesture, gaze and so on, it should be considered a fundamental and primordial basis for sociality (Merleau‐Ponty 1964, in Cekaite & Holm, 2017: 111). Touch, they argue, forms “compresence,” a “single intercorporeality.” This point is perhaps especially intriguing in the context of the veterinary clinic, raising questions about how far clients, vets and animals can be said to understand one another through touch and forms of bodily engagement.

For many years, people's expectation that animals, such as dogs, understand basic vocabulary, such as “come,” “sit,” “ball,” and so on, has been reported (Mitchell, 2001). In our study, we have explored animals responding to touch. We saw animals were clearly attentive to subtle transitions in the way they were held. This provides one way of gaining analytic traction on the agency of animals (Carter & Charles, 2018), and how this coexists with human corporeality (Cekaite & Holm, 2017: 111). We observed many occasions where animals appeared to recognise—through their conduct—the implications of the embodied conduct of human actors. The animal is able to recognise various affordances of the way that they are held, moving to break free when a grasp is relaxed or hand withdrawn. In this sense, animals are acutely attentive to the implications of different forms of touch. Whilst animals are constructed as faux subjects through talk, in our data, we at least glimpse how they might be considered more fully as able to recognise, albeit within a very limited register, the immediate implications of bodily movements for them (Laurier et al., 2006: 19).

CONCLUSION

Whilst the focus of our article has been the distinctive characteristics of animal health care and the occupational work of veterinarians and their clients, we believe that it also has broader significance for the sociology of health and illness. Knowing how to hold others, in ways that are socially attentive to their discomfort, privacy, embarrassment, and agency is a quite central competence across a range of medical settings. Indeed, the concern with body work (Heath, 2006) reflects the fact that health and social care involve a great deal of physical and hands‐on labour, and yet this work is “too often neglected” (Twigg et al., 2011: 172). Touch is recognised as a critical feature of care work (Bjorbækmo & Mengshoel, 2016; Cacchioni & Wolkowitz, 2011; Emerson, 1970; Parry, 2004; Pedrazza et al., 2018; While, 2021) and indeed, professional touch is known to have the potential to be harmful or misconstrued (Green, 2017). But we believe that our concerns with “coalitions of touch” and haptic sociality, and more broadly, the interest in multisensoriality (Mondada, 2019), offers new lenses with which to explore practices of body work.

These lenses encourage us to interrogate the ways in which aspects of tactile conduct are produced, coordinated and oriented within the moment‐to‐moment, the interactional accomplishment of care work. And they offer the methodological opportunities to better understand the tensions that can arise in the course of this labour: between restraint and soothing; between pain and procedure; between success and failure; between appropriate and inappropriate; and indeed, between professionals, patients and their companions. The focus on the institutional and interactional organisation of touch and handling offers a rather novel perspective on the “body work” that is central to so many fields of health care, whether physiotherapy, nursing, or innumerable activities in emergency care, general practice, dermatology, geriatrics, paediatrics, mental health, and many more. We, therefore, hope that our article offers ways of analysing and conceptualising everyday health care in and beyond the veterinary clinic.

AUTHOR CONTRIBUTIONS

Nick Llewellyn: Conceptualization (equal); Data curation (lead); Formal analysis (lead); Investigation (equal). Jon Hindmarsh: Conceptualization (equal); Data curation (supporting); Formal analysis (supporting). Robin Burrow: Investigation (lead); Methodology (equal); Project administration (lead).

DATA AVAILABILITY STATEMENT

The video data cannot be anonymised and contains much information that is personal and cannot be shared.