Abstract

The lived proximity of sheep (Ovis aries) to their human caregivers allows us an opportunity to directly impact the understanding of those who raise them. The premise of my research is to use both quantitative (distance from conspecifics) and qualitative (embodied empathy) methods to highlight sheep individuality and lay the groundwork for a moral argument analogous to personhood. The first objective of this study is to use size-relevant measurements to observe sheep, in order to determine social interests. The second is to reflect on the use of this method to change the paradigm that nonhuman animals are merely objects for consumption. During the research period of three months I measured two variables: proximity and type of interaction towards conspecifics. Then, using a biocentrically informed method to interpret these behaviours/interactions, concluded that the sheep in this study show a preference for specific other sheep. From the sheep’s preference we can infer “friendship” as defined by choosing to spend time with specific others. Such behaviour suggests a richer emotional life than nonhuman animals, particularly those used in meat production, are normally credited with and thus a rationale for granting these animals moral standing akin to personhood. By seeking to understand sheep, humans can practice extending our moral circle to include nonhuman animals. This awareness can be applied to promote ethical consumerism and not only better understand the welfare needs of nonhuman animals but also their wellbeing. Such studies can be a step toward understanding other-than-human personhood as more research into nonhuman animal perspectives advance in the fields of anthrozoology, care ethics, and applied behaviour science.

Introduction

The New York Declaration of Animal Consciousness (Andrews et al., 2024) was signed initially by 40 world-leading experts on human and animal consciousness, asking and answering:

“Which animals have the capacity for conscious experience? While much uncertainty remains, some points of wide agreement have emerged. First, there is strong scientific support for attributions of conscious experience to other mammals and to birds. Second, the empirical evidence indicates at least a realistic possibility of conscious experience in all vertebrates (including reptiles, amphibians, and fishes) and many invertebrates (including, at minimum, cephalopod mollusks, decapod crustaceans, and insects). Third, when there is a realistic possibility of conscious experience in an animal, it is irresponsible to ignore that possibility in decisions affecting that animal. We should consider welfare risks and use the evidence to inform our responses to these risks.”

With this declaration in mind, those working with nonhuman animals, whether they are researchers, caregivers, work in the meat industry or simply feed ducks at the park, need practical actions to shift from a traditional Western/Kantian paradigm of nonhuman animal exploitation. While the above declaration does not seem radical to many of us who work with nonhuman animals every day, the signatories of the declaration bring the current scientific consensus of nonhuman animal consciousness, cognition, and even emotion into light for the public and policy makers. This author hypothesizes that the way for society to accept this consensus is through personal recognition of these traits in the nonhuman animals they care for. Hence the concept of ‘personhood’ could be helpful in this cultural adjustment. Milton (2005) suggests that the criteria for personhood can include: emotions, intentions, and individuality. The importance of these criteria becomes evident when arguing for personhood in nonhuman animals that have, in traditional western perceptions, lacked these characteristics. As illustrated by Pardo (2023, p. 1) who states, “An analysis of these legal personhood decisions reveals that courts have rejected animals’ complex cognitive abilities as a justification for an extension of the law. Courts have been careful to state that societal-changing legal developments must originate in the legislatures, not the courts.”

Let’s examine how personhood might be recognized in nonhuman animals whose lives are deeply entangled with human systems—particularly those often overlooked. Sheep, as domesticated animals raised primarily for consumption, occupy a paradoxical space: They are physically proximate to humans yet conceptually distant in moral consideration. Their social behaviours, emotional expressions, and individual preferences are rarely acknowledged beyond agricultural utility. By focusing on sheep, this study aims to challenge entrenched assumptions. Kantian ontology asserts sheep exist primarily (or completely) for the use of humanity (Kant, 1963; Singer, 2009) including the biological definition of sheep; specifically, sheep are domesticated herbivores, which centres their “use” to humans. The Kantian approachdefines sheep as instrumental: for meat or wool. According to Sealey and Charles (2013, p.498) the word “lamb” itself “denotes the consumable commodity of animal’s flesh.” This paradigm also defines sheep by their sociality and spatial place inhuman culture: They are flock animals, found in farms and on fields. And finally, sheep can be defined as metaphors: Often the term sheep is used to describe a human with mild mannerisms, a follower, or someone who cannot think for themselves (Mullin, 1999).

An opposing paradigm seeks to include sheep in a cosmological perspective and highlights their presence within the category of nonhuman animals, with all the requirements of personhood (Giri, 2006). In embracing a global perspective, Giri (2006, p.1277–1292) notes that “Cosmopolitanization is an ongoing process of critique, creativity and border-crossing which involves transformations in self, culture, society, economy and polity.” Further she notes that “cross species dignity” is an issue of “cosmopolitan responsibility.” i.e., expanding our moral horizon to ensure that our global perspective encompasses respect and compassion for all species.

Within this cosmopolitan paradigm, sheep are inclusively defined as “people,” as outlined by Ingold (1988, cited in Nadasdy, 2007, p. 32). Here, “people” are understood as beings with intentions and purposes, not necessarily resembling humans but shaped by their unique preferences, intelligence, and social behaviours. This paradigm also identifies sheep as “actants,” portraying them as individuals with subjective perspectives where they are the central protagonist in their world (Nadasdy, 2007). This is demonstrated through their capacity for agency and personal preferences (Law & Lien, 2012). Additionally, the paradigm emphasizes sheep’s social tendencies, highlighting their ability to connect and form relationships. “Bonding” is described as an instinctual drive evolved to encourage parental care: “Mothers form strong and exclusive bonds with their offspring” (Marino and Merskin, 2019, p. 4). Together, these definitions offer a more nuanced understanding of sheep as relational and intentional beings. If we are discussing humans, we would associate this bonding with the assumption of emotionalattachment, an assumption often withheld from nonhuman animals who do not occupy our homes, particularly those in western society. The bond instead is described as instinctual rather than emotional, particularly related to animals who are farmed or exploited (Loughnan, Haslam, & Bastian, 2010). Lastly, Ingold’s inclusive paradigm defines species based on their intelligence (1988 as cited by Nadasdy, 2007, p. 32). We (humans) often compare nonhuman animal intelligence to human intelligence for perspective, as outlined in Chapman and Huffman (2018) and described eloquently by Merino and Merskin (2019); however, this anthropocentric view of intelligence is unhelpful (Broom, 2016). For sheep, intelligent behaviour could be understood as the enhanced ability to determine safety and locate food sources or shelter (Marino and Merskin, 2019).

Suppose we define sheep using this second paradigm, affording them the benefit of the doubt towards emotions and evolutionarily necessary intelligence as outlined by the world experts in the New York Declaration of Animal Consciousness.  It becomes difficult to view them as objects alone. While the Kantian philosophy characterises sheep as so different from humans that they don’t deserve to or are not able to make their own choices or demonstrate agency (Krumer-Nevo & Sidi, 2012), the recognition of choice regarding friendships with other beings demonstrates their potential as “persons” (Latour, 2005; Bekoff, 2001). I propose that if they can form friendships like human-human friendships, demonstrated in ways recognizable to human people due to embodied knowledge, i.e., the behaviours associated with friendship, then this recognition could be held as further evidence that nonhuman animals can have an innate right to personhood and moral consideration. This research may be an early step in providing a path for this recognition.

Observing sheep in an environment that allows them to demonstrate their innate agency, we can then interpret the sheep behaviour observed, using a practice of “embodied knowledge” or the use of one’s own physical senses to gain understanding of another. This approach aligns with growing recognition in affective neuroscience that nonhuman animals experience a range of emotional states (Panksepp, 1998), increasingly reflected in ethological studies of farm animals (Boissy et al., 2007; Dwyer, 2008; Marino and Merskin, 2019), and with companion animal research showing that guardians routinely observe behaviours consistent with joy, fear, and affection (Serpell, 2019). Seeing nonhuman animals must be a prerequisite for knowing, and knowing is a prerequisite for care (Hamington, 2008). I hypothesise that we can see sheep’s preference for conspecifics, or “friends,” by using size-relevant measurements, and in doing so, highlight sheep individuality. In this paper, I propose three ideas:

1. 1. Sheep “intelligence” or cognitive ability should be based on the sheep’s own evolutionary necessities, specifically the need for social bonding. This falls in line with both Burghardt (2007) and Horowitz and Bekoff (2007) and is described by Marino and Merskin (2019, p. 5-14).
   2. We can and should use ideas like Hamington’s embodied empathy to help recognize the behaviours that point towards emotional connections within the social group.
   3. This method is a way to bridge the gap between the paradigm of nonhuman animals as things to “persons.”

Methods: Environment, interaction and observation

From October 7 to December 7, 2015, I observed the interactions and documented behaviour of 20 sheep in a mixed-history flock of individuals on a rural farm animal sanctuary in Lancashire, U.K. (location: 53.8 Lat. -2.4 Long). Within this paper I define “flock” as the individual sheep that live on the same property and stay within close proximity to the larger flock (in relation to other sub-flocks on the property). I use the term “sub-flock” to denote sheep who have broken off from the larger flock. Keeping in mind that these groups form and reform as activities change. For example, Sheep A may be part of the larger flock while grazing but may retreat with a sub-flock while resting. For both these definitions I have outlined the measurements in “sheep length”. I noted these observations in proximity (during husbandry tasks) and at distances of 100-400 feet (~30-150 meters) from myself to the sheep. These distance ranges allowed for observations that were not likely to alter the sheep behaviour. These written or voice recorded observations included notes totalled over 80 time periods lasting from five minutes to one hour. These randomized “check-ins” helped to mitigate observer bias and allowed the possibility for more variety of behaviours to be observed. During each check-in I noted sheep behaviours and what kind of interactions were taking place. To give an objective measurement rather than the subjective terms “near” or “far,” I used the abbreviation SL or “sheep length.” This method was adapted from the Clothier Animal Response Assessment Tool (Clothier, n.d.) and through personal communication (email and direct discussions) (Feb 2025). Clothier’s assessment indicates the importance of the size of the animal observed rather than simple measurement, i.e., a 5-foot distance between two rhinoceroses has a different implication than a 5-foot distance between two mice. At a 5-foot distance, mice would not necessarily even be aware of each other, where the rhinoceroses would almost certainly be interacting. Hence, acknowledging the length and size of the nonhuman animal is a more relevant indicator of social interaction. For this research, five sheep lengths (5SL) are observed as an interaction. Greater than (5SL) was categorized as “alone.”

Observations check-ins included three main categories: disturbance caused by observer or other caregiver (husbandry tasks, i.e., feeding, cleaning, or maintenance of the field), disturbance of the environment (such as deer, weather, or injury), and no observed disturbance. Observations note how closely, how regularly, and in what circumstances flock members chose to interact. At night, resting happened more often as a group but, due to the darkness and the movements of sheep (if disturbed), it was difficult to observe who rested near whom. Therefore, I did not include resting in the figures. I defined the behaviours based on activity and determined interactions based on sheep length. I interpreted and categorized behaviour as affiliative, agonistic, or neutral based on the context of the behaviour or interaction if there was one. Their behaviour was further interpreted using methods described by Hamington (2008) as “interspecies empathy” or “embodied knowledge” as a way to care for and know other animals. Hamington proposes that embodied knowledge is the innate knowledge our bodies hold that leads to an empathetic understanding of others who also have bodies. This analysis also calls on Milton’s (2005) explanation of egomorphism as a process of “perceiving characteristics in other animals rather than attributing characteristics to other animals” (p. 255, original emphasis). Through these tools (sheep length, embodied empathy, and egomorphism) I have attempted to avoid the “sentimental” error of attributing human characteristics to nonhuman animals, also known as anthropomorphism (Rivas and Burghardt 2002 p. 9).  While the bias towards a human perspective should be recognized (Karlsson, 2012), we should also recognize that we may be equally biased towards a mammalian perspective and a perspective of a biological being (a being with a body as described by Hamington, 2008). By acknowledging all of these potential biases, we may be able to come to accurate conclusions about a nonhuman animal perspective (Humphreys, 2023). We will call this “critical anthropomorphism,” a term suggested by Bekoff (2002 p.87) and Burghardt (2007, p.268) to interpret behaviour through thoughtful reflection and understanding of personal bias (Rivas and Burghardt, 2018), species specific behaviour, and personal history of the individual subject (Shapiro, 1990; Sanders and Arluke,1993), in this case sheep.

While I observed a total of 20 sheep, for the sake of clarity and analytical depth, I have chosen to focus on the behaviours and interactions of three specific individuals: Swayze, Endora, and Mojo. These three were previously adopted  by the farm sanctuary, while the remainder of the flock were owned by a local sheep farmer. Noting that a study with a longer duration may want to choose a larger sample size, these three individuals as case studies may point a way towards future research into sheep social behaviours. I will share their biographies here to highlight their individuality, give examples of their behaviours, and demonstrate how this knowledge can alter one’s perception of an individual. The other flock members will also be referenced by their names. Due to the focus on three subjects, only 12 sheep in total are discussed. For the remaining eight sheep not discussed, I either did not observe them during interactions with the three sheep subjects or they only had occasional interactions categorized as “other” such as urinating or limb movements that were not overtly associated with an interaction. Future research would benefit from the use of long-term visual recording so behaviours may be reviewed for accuracy.

The biocentric anthropomorphism model:

To operationalize a species-sensitive form of anthropomorphic interpretation, this study builds on—but departs from—the broader framework of critical anthropomorphism (Horowitz and Bekoff, 2007; Karlsson, 2022). While critical anthropomorphism acknowledges the role of human perspective in interpreting nonhuman behavior, biocentric anthropomorphism focuses specifically on animals as biological beings and emphasizes embodied empathy as a methodological tool (Milton, 2005; Shapiro, 1990; Hamington, 2008). It draws on shared biological and sensory capacities between observer and subject to inform interpretation, without collapsing species difference. This approach resists mechanomorphic reduction (the attribution of mechanistic characteristics to non-human animals) and centers the animal’s own agency, emotional expression, and relational context (Bekoff, 2002; Marino and Merskin, 2019).

The following stepwise method combines spatial observation with empathetic interpretation to explore sheep social preferences and emotional responses:

1. 1. Conducted spatial observations of sheep over a three-month period, using randomized intervals and varying distances (30-150 meters) to avoid altering natural behavior. Proximity was measured in species-relevant units to track social positioning
   2. Defined interactions using “subject lengths” (SL), a metric based on the average body length of sheep. Interactions were classified as within 5SL, while distances greater than 5SL indicated solitude. This standard allowed for consistent, size-sensitive behavioral coding (Clothier, 2010).
   3. Documented affiliative behaviors and recurring proximity patterns, such as resting near specific others or synchronized movement, interpreted as indicators of social preference and emotional investment (Reefmann et al., 2009; Marino & Merskin, 2019).
   4. Tracked categories of disturbance, including observer presence, environmental factors, and undisturbed conditions, to contextualize behavioural data and reduce interpretive bias (Fraser & Matthews, 1997).
   5. Conducted background research on species-level behaviour and ecology, alongside individual sheep biographies, to inform both general and specific interpretations of social dynamics (Chessa et al., 2009; Mullin, 1999).
   6. Applied a biocentric interpretive framework, treating sheep as subjects with agency. This involved resisting reductive categorizations and remaining open to emotional and relational nuance (Bekoff, 2002; Horowitz & Bekoff, 2007).
   7. Practiced embodied empathy and egomorphic reflection, using my own sensory and emotional responses as heuristic tools. Egomorphism—interpreting behavior through the lens of shared physiological experience—was grounded in ethological knowledge and used to enrich behavioural interpretation (Milton, 2005; Shapiro, 1990).]
   8. Focused analysis on specific individuals, while referencing the broader flock to contextualize social roles and relational patterns (Brent et al., 2014; Christakis & Fowler, 2014).

This method integrates empirical rigor with ethical attentiveness, offering a pathway toward recognizing sheep as emotionally complex individuals with moral relevance (Humphreys, 2023).

Results:

Biographies of Three Sheep:

These biographies, along with the photographs of the subjects Swayze, Mojo, and Endora, are supplied here to concretize their individuality and help readers to use their own bodies to extend empathy and moral imagination (Malecki, et al., 2019).

Swayze: Swayze is a Herdwick adult female (ewe). She is reportedly 3 years old and the only adult ewe in the sanctuary flock. She was originally in a small Herdwick flock that were kept as pets before being sold to a farmer and then given to the sanctuary after two years. The farmer has reported that Swayze had three lambs (one each year), and the first two died after a few days. The third (a male) was born late in the season 2015. Swayze and her lamb were the only two Herdwicks at the sanctuary in a flock of Texel Mules, a common breed of sheep mixed for hardiness and meat production. In contrast, Herdwicks are a rugged hill-sheep, and lambs often have a dark coat. According to the farmer, who gave her to the sanctuary, when Swayze was first introduced to the Texels, the other sheep appeared frightened, and they ran away from her if she approached. Consequently, she became withdrawn; in the farmer’s words, “it was a blow to her confidence.” With her lamb at the sanctuary, she spent most of her time among the rushes where she was hidden. Swayze’s lamb was removed (artificially weaned) in early September 2015 by the farmer to be sold for meat. Swayze was visibly distressed for 48 hours after his removal and demonstrated this by trotting back and forth across the field with her head up and vocalizing. Using biocentric anthropomorphism (Milton, 2005; Shapiro, 1990; Hamington, 2008), I would characterize her searching behaviour as highly distressed. This is supported by the discussion of ewe-lamb bonding in Marino and Merskin sheep (2019). During this time and for weeks afterwards, Swayze would head-butt the other lambs if they came near her. At the start of these observations (three months after separation from her lamb), she was allowing many of the lambs (now over 6 months old) to stay within 3SL but became aggressive (head-butting) at other times, mainly when there was preferred food available.

Figure 1. Swayze (photographer: Marika Bell).

Mojo: Mojo was a castrated Jacob’s male, born 2015. Mojo was in the second batch of sheep placed at the sanctuary He arrived with his mother as the only two Jacob’s sheep, a breed considered hardy. According to the original farmer, Mojo and his mother came from an all-Jacob’s flock that was raised with very little human contact and consequently had shown no interest in interacting with human caregivers. At the sanctuary, they spent most of their first few weeks within 10SLs of the Jacob’s cross sub-flock and would actively avoid any members of the main flock or any caregivers. Mojo’s mother was removed for breeding in September 2015. Since that time, he spent more time with the new flock of female lambs and was seen attempting to mount one of them about a week after his weaning. Since the removal of the other adult female sheep, he was observed mostly in the company of Swayze. She seemed more accepting of him than of the female lambs. Mojo was avoidant of human caregivers, as demonstrated by increasing his speed to move away from humans, although he became more confident if other sheep were nearby.

Figure 2. Mojo (photographer: Marika Bell).

Endora:Endora was an adolescent female Texel Mule. During the time between arrival at the sanctuary and weaning she demonstrated inquisitive behaviour towards human caregivers and showed interest in interacting. Although she was with her mother for six months, on the day of weaning Endora intermittently vocalized for a few hours but seemed to resolve her distress relatively quickly. During her time at the sanctuary, she was treated for worms as part of a normal husbandry protocol, as were the other sheep, but she also received antibiotics, a foot bath and hoof trimming for a sore foot that was causing her to become lame. Since these treatments, she remained at least 10SLs away from all humans unless corralled and showed no interest in food lures to increase interaction. Since her stressful time of treatment for lameness, she chose to be left behind by the flock rather than approach human caregivers for treats.

Figure 3. Endora (photographer: Marika Bell).

Behaviour descriptions

I recorded two dozen distinct behaviours by many if not all sheep in the flock. I categorized these into types of behaviours and separated the behaviour of grazing alone and grazing with chosen others (within 5SL). Here I will use the three subjects to demonstrate the interactions identified.

Figure 4 shows behaviour frequency for each subject sheep. Swayze spent 63.6% of the observation time grazing with others, 14.3% of the time engaged in behaviors interpreted as information-gathering, such as “looking out,” “curiosity,” or “searching,” and 5.2% grazing alone. Mojo spent only 46.8% of the observed time grazing with others and 27.7% alone. His other three most frequent behaviours were equal at 4.3% each. Endora grazed with others 55.6% of the time, grazed alone 25.9% of the time, and spent 9.3% of the time avoiding or behaving fearfully.

Figure 4(A). Personality differences are demonstrated by frequency of behaviour
(A) Aside from grazing, Swayze spends the largest portion of the observed time on information-gathering.

Figure 4(B). Personality differences are demonstrated by frequency of behaviour
(B) Mojo’s shows a wider spread of behaviours than Swayze or Endora.

Figure 4(C). Personality differences are demonstrated by frequency of behaviour
(C) Besides grazing still being the predominant behaviour, Endora spends the largest portion of her time on avoidant and fearful behaviours.

Figure 5 shows us who the subject sheep preferred to graze with. I define preference as choosing to spend time with others when there is an option not to. This preference is inferred as all sheep could spread out on 8 acres of land. The acting on preference shows agency (Law & Lien, 2012). Swayze spent her time (60.4%) with four other sheep. The table also shows that between the three sheep; she had more grazing partners total (10). Comparatively Mojo only had five observed grazing partners and spent 37.1% of his time alone. The sheep he spent the most time grazing with were Endora, Swayze, and Berry, total time 45.6%. Endora spent 31.8% of grazing time alone; She spent a combined 47.7% of her time grazing with Swayze, Mojo, or Lambington.

Figure 5(A). The percentage of time our three subject sheep spent grazing within 5SL of others or alone.
(A) Swayze was within 5SL of 10 of the 19 other sheep in the flock during the observation periods—most often with Endora, Trouble, Lambington, and Mojo. Of the three subjects, she was least often alone.

Figure 5(B). The percentage of time our three subject sheep spent grazing within 5SL of others or alone.
(B) Mojo grazed with 5 of the 19 other sheep; he spent the most time with Endora, Swayze, and Berry. However, he spent the most time alone during the observation periods.

Figure 5(C). The percentage of time our three subject sheep spent grazing within 5SL of others or alone.
(C) Endora also showed most of her time with only 5 of the other 19 sheep in the flock, primarily Swayze, Mojo, Vadr, and Lambington. She also spent more than a quarter of the observation periods alone.

I emphasise that Figures 4 and 5 can only show us data points. They do not convey the embodied empathy and knowledge of individual history of the animals. This individual history, along with knowledge of species-specific behaviours, allows us to use embodied knowledge and empathy to further interpret these results. Bekoff (2001) refers to this as “critical” or, as I prefer to call it “biocentric anthropomorphism,” which is more specifically attuned to animal subjects and grounded in shared biological capacities. This interpretive framework is operationalized in the model outlined in the methods (section 2.1), which combines spatial observation with empathetic interpretation to assess social preference and emotional expression.

Humans and animals share embodiment. They are different bodies, but the possession of physical bodies in space is a starting point for shared knowledge and understanding. Biocentric anthropomorphism allows the observer to recognize preferences, understand those preferences as valid, and act on them (Bekoff, 2001; Burghardt, 2007). Broom (2016) showed that preference tests have been long used to clarify in animal welfare science whether nonhuman animals feel pain, contentment, or even awareness of their environment.

To demonstrate that table figures and percentages alone cannot give us all the answers we need, I will use the grazing behaviour as an example of the necessity of biocentric anthropomorphism. I chose grazing as the focus for this category as it was the behaviour I most often observed. The goal of collecting this information was to find out if the sheep cared who they spent time with. If they had no preference, we should see a random spread of time spent with various members of the flock. If sheep prefer certain companions, we would expect to see them more often within five sheep lengths of those individuals. Future research would benefit from long-term video review of behaviour along with current technology of tracking bands to note activity level. Also note that I could have broken down the term “grazing” into different sub-brackets. For instance, grazing information, grazing in a sub-flock, grazing within the full flock, being in a new access field or grazing in an old access field, or grazing within distances to other sub-flocks. However, this was not necessary for the goals of this research.

Example 1: Swayze

Looking at Swayze’s charts alone might suggest that she prefers to spend her time grazing alongside others. However, when considered in the context of her individual history (as outlined in her biography) and through the lens of biocentric anthropomorphism, a different pattern emerges: Swayze does not actively seek out flock members to graze with—rather, they seek her. While proximity data (measured in sheep lengths) cannot capture the directionality of social initiation, this distinction becomes apparent through embodied observation and empathetic interpretation, as enabled by the biocentric model.

Swayze seemed unsettled by this interaction initially. She would regularly try to separate herself from the other sheep (avoidance). As the research progressed, she stopped separating from the other sheep and began more “information-gathering” behaviour, seeming to accept her position as matriarch. I use the term “matriarch” here to emphasise that Swayze’s role within this primarily female flock (traditionally breeding males are held away from the ewes until mating season, although this is not applicable in the sanctuary) was not only as the oldest of the ewes but also the sheep who was followed by the others when a disturbance provoked movement in the flock. This concept was discussed by Squires and Daws (1975) specifically as “leadersheep.” The idea of a nonhuman animal matriarch is long established in other groups made up of primarily female members and their offspring such as elephants, hyenas, and orcas (Cooke, 2022) . Of the three sheep, Swayze engaged in the most of what I termed “information-gathering” behaviour, as illustrated by the activities of scanning, flared nostrils, and looking over her shoulder. As she grew into her role as matriarch, examples of this information-gathering behaviour included often placing herself in a “lookout” position, scanning the environment while the other sheep grazed. On one occasion I observed her out of line-of-sight of all the other sheep and then observed her searching out (head high, eyes scanning, and vocalizing) other flock members. This seemed her intention as she continued her search through three separate fields until she was able to see the final flock member lying in tall grass. She then stopped her search and began to graze. I rarely observed that Swayze was resting or lying down during the day. She also was most often positioned near the front of the flock during flock movements.

Example 2: Mojo

Mojo’s charts suggest that he spent almost equal time grazing alone or grazing with others. However, what the chart cannot show is that many times when he did seek out others to graze with, they moved away from him, and he ended up grazing alone. Mojo’s results showed him to be at least 10-20SLs away from all other sheep during grazing. When he was in the company of other sheep, he was primarily following and/or was moving towards a food source. Mojo’s most notable behaviour was his strategy of burrowing under the other sheep when accessing high-value food, as well as his tendency to hang back rather than lead, when the flock was disturbed e.g., moving fields or interaction with new caregivers. I observed the avoidance of antagonistic encounters predominantly by Mojo, who was the smallest of the sheep in the flock, but also the only male sheep and the only sheep with horns. I never observed head-butting directed at him.

Example 3 Endora

In 14 of 56 of my observations of Endora, she was not within 20SL of any other flock member. I recorded Endora often with two others (Annikin and Lambington) who also remained avoidant of caregivers during the observation period (see fig 2). These three chose (applied their agency) to be left behind by the greater flock when food was offered. The three together did tend to form a sub-flock if the larger flock remained apart for more than 10 minutes; however, they were also observed spending time together in grazing behaviours. Without knowing Endora’s history, her avoidance and caution, mainly of human caregivers, could be viewed by a casual observer as stereotypical “sheepish” behaviour. Knowing her history allows us to put her behaviour into context and adjust husbandry as needed for her welfare and health.

The avoidance of human interaction observed in the three focal sheep is consistent with their limited history of human socialization. All three demonstrated avoidance behaviours, except when offered high-value food by caregivers. Antagonistic interactions between all individuals in the flock—such as pushing and head-butting—typically occurred around access to resources, though I also recorded head-butting in contexts where no clear resource was present.

In addition to these individual behaviours, I occasionally observed the flock—including the three focal sheep—engaging in coordinated group actions. These included “group purpose” behaviours, where three or more sheep acted simultaneously: dispersing into sub-flocks, vocalizing together, or grazing in a regimented line, shoulder to shoulder. One particularly notable behaviour involved a sheep acting as a lookout, resting in the pathway of a bottlenecked area, then looking over her shoulder and vocalizing towards the flock in response to a disturbance. While these collective behaviours were not the primary focus of this study, they offer intriguing insights into flock-level coordination and merit further investigation in future research.

Discussion

Humans have had a reciprocal relationship with sheep for at least 11,000 years of coevolution (Chessa et al., 2009). Humans have kept sheep in flocks partly because of their gregarious nature. As with most group herbivores/prey animals, this grouping together is assumed to be a survival instinct, often referred to as “safety in numbers” (Dwyer, 2008, p. 115). This survival technique can also be viewed as not just instinct, but a preference for spending time with others of a similar mindset or interests, demonstrated in this flock by fearful individuals forming sub-groups. Humans also tend to gravitate towards those similar to them or those they enjoy spending time with (Brent, 2014). So instead of attempting to explain away similar behaviour in sheep as mechanistic (Aaltola, 2008), it is more parsimonious to accept that choosing to spend time with specified others when you have the option not to is a valid definition of friendship (Seyfarth and Cheney, 2012; Christakis and Fowler, 2014).

Let’s revisit the qualitative method of biocentric anthropomorphism I used to interpret sheep behaviour and interactions. Shapiro’s (1990) term “kinesthetic empathy” was described by Sanders and Arluke (1993, p. 383) as “using knowledge about the individual animal’s history and the animals’ social construction of particular type to ‘critically temper and inform’ empathetic understanding of an animal’s postures, movements, and use of space.” Along with Hamington’s (2008) embodied care ethic and Milton’s (2005) practical application of egomorphism, it forms the basis for the biocentric interpretation of the interactions between Swayze, Mojo, Endora, the other members of the flock, and human caregivers.

Postures, movements, and use of space are particularly important indicators describing specific behaviours and preferences, according to Suzanne Clothier (2010), a behavioural assessment expert, who writes and teaches about animal personality testing. Clothier regularly uses the same or similar terms to describe temperament traits, which she developed as a proposed standard for an assessment tool called “CARAT” (Clothier Animal Response Assessment Tool), which allows the prediction of preferences and reactions of animals in different scenarios and helps us answer the question Who are you? when understanding animals (Clothier, 2010).

For those who might argue, as Dawkins (2007) does, that we cannot count personal accounts of nonhuman animal intention as anything other than anecdotal, I point to Wemelsfelder (2012, p.224-228) who suggests that we should endeavour for “Balanced, open-minded scientific inquiry” and not “routinely assume that an individual’s actual experience is inaccessible to others,” and to Bekoff (2001, p. 21) who maintains that “We can be biocentrically anthropomorphic and do rigorous science.” For this reason, I included the biographies of the three sheep under observation in the methods to help readers understand not only the individual sheep’s motivations but also allow the readers to practice their own use of biocentric anthropomorphism and, as Sanders and Arluke (1993) suggested, use our knowledge about the individual animal’s history and their social construction to critically temper and inform our interpretation.

There are three main issues under consideration here, each building toward a broader ethical inquiry. (1) Do sheep form friendships? This question is central to discussions of personhood among nonhuman animals typically considered consumable within traditional Western ontologies. If sheep demonstrate a preference for interacting with specific individuals—rather than distributing themselves randomly—then we are morally obliged to look deeper into their perspective (Boissy et al., 2007; Christiansen and Forkman, 2007; Dwyer, 2008; Marino and Merskin, 2019). (2) Can we interpret sheep behaviour in ways that access their mental states? While we cannot use human language to ask sheep directly, we can apply biocentric anthropomorphism to interpret their behaviour critically and empathetically (Bekoff, 2002; Burghardt, 2007; Sanders and Arluke, 1993; Wemelsfelder, 2012). (3) Do sheep possess personhood, and if so, what ethical obligations arise? If the answer is yes, or even maybe, then we must ask whether our moral and ethical duties extend beyond the bare minimum of animal welfare (freedom from fear, thirst, hunger, and pain) to include opportunities for comfort, pleasure, and personal choice—what we might call wellbeing (Balcombe, 2002; Chessa et al., 2009; Gruen, 2011).

The negative perception of sheep as “other” can be reshaped through research like that presented here, framing sheep relationships as a matter of preference and recognizing them in terms of friendship. Shifting human understanding to see sheep as individuals with personal connections has the potential to transform the interspecies dynamic, moving beyond viewing sheep solely as edible objects or metaphors for undesirable traits like unintelligence or passivity (Marino & Merskin, 2019).

The practice of observing sheep to determine social preferences—such as proximity, affiliative behavior, or selective interaction—can help us recognize them as individuals with distinct relational lives. This recognition invites us to describe their behaviour through biocentric anthropomorphism, a method that interprets nonhuman actions empathetically while remaining grounded in species-specific context. Such interpretation is not merely descriptive; it activates our moral imagination (Hamington, 2008), encouraging us to extend ethical consideration to animals traditionally excluded from moral frameworks.

Furthermore, once we have gathered and interpreted behavioral data, we can deepen public engagement by re-presenting this information in ways that resonate emotionally and ethically with human audiences. This second layer of biocentric anthropomorphism — translating scientific observations into relational narratives — can help bridge the experiential gap between human and sheep lives (Bekoff, 2002; Hammington, 2008). For example:

Swayze is aloof and shows little interest in physical contact with other sheep. Despite her ambivalence, she shows a sense of care for the other flock members, demonstrated in her behaviour of scanning for danger and leading the flock to safety or food. Mojo is isolated by his lack of experience and socialization. Yet, he still attempts interactions and is a problem-solver who knows who to follow and how to get what he needs. Endora is apprehensive, relating to any new experience with the awareness of the potential for danger — remaining wary, and showing anxiety and fear through her avoidance and body language (Reefmann et al., 2009).

Milton (2005) defines a person as one that experiences emotions, intentions, and individuality. Based on this definition and the use of embodied knowledge (Hamington, 2008) to perceive characteristics within the sheep — i.e., egomorphism (Milton, 2005) — I conclude that sheep must be people. However, I cannot and should not assume sheep are likehuman people. Sheep are biologically similar to humans (Chapman and Huffman, 2018): They have a nervous system, a complex brain, and share a (distant) common ancestor with humans, and they have evolved on the same planet for at least 11,000 years alongside humans. However, they are their own kind of people, and we should resist the attempt to humanize them further than is necessary to help us extrapolate the familiar to the unfamiliar—else, we lead ourselves to treat them with the expectations we have for human-people, which will predictably lead to disappointment as they fail to meet those expectations (Dwyer, 2007; 2008).

With additional tools, such as wearable transponders, GPS collars, or AI-assisted video analysis, the biocentric anthropomorphism model has implications for human–nonhuman animal relations. As Hamington (2008) proposes, meaningful relationships with nonhuman animals allows us to exercise our “imagination muscle” towards the extension of care and empathy to unfamiliar others: humans or nonhumans. This research proposes a model for seeing nonhuman animals by taking a short period of time each day and looking at who each nonhuman animal spends time with, to come to know them as individuals, worthy of moral consideration.

For a human engaged in farming nonhuman animals, this choice-test model can allow discovery of preferences that lead to better understanding (Fraser and Matthews, 1997), welfare, and wellbeing of the animal in question. It may develop the moral imagination into an ethic for care leading to overcoming the contradiction of empathy with animals and consumption (Hamington, 2008).

Conclusions

Insights from the Psychology of Human–Animal Intergroup Relations Conference (Edinburgh, July 2-5, 2025) demonstrate a continuing belief among nonhuman animal advocates and researchers that these nonhuman animals have intrinsic value and exist for their own purposes. While this belief extends beyond vegan communities, omnivorous individuals may experience cognitive dissonance and benefit from pathways that help them understand animals as unique individuals. By adopting biocentric anthropomorphism and enabling animals to express agency through personal choices, they can engage in activities that bring comfort, joy, or pleasure (Balcombe, 2002). This approach has the potential to enhance welfare, even for nonhuman animals whose fate is as consumable products.

This research explored three ethical questions: (1) Do sheep form friendships? (2) Can we interpret sheep behaviour in ways that access their mental states? (3) Do sheep possess personhood, and if so, what ethical obligations arise? These questions challenge traditional Western views of farm animals and invite deeper moral consideration.

To investigate them, I used a methodology combining biocentric anthropomorphism with species-relevant proximity and length measurements. This dual approach enabled both empathetic interpretation and structured behavioural analysis. Biocentric anthropomorphism draws on embodied knowledge (Hamington, 2008), kinesthetic empathy (Sanders and Arluke, 1993), and egomorphism (Milton, 2005), while proximity data helps assess social preferences and spatial relationships within the flock.

Findings suggest that sheep form selective social bonds and exhibit distinct behavioural patterns. Swayze, Mojo, and Endora each demonstrated unique tendencies in interaction, avoidance, leadership, and problem-solving. Based on Milton’s definition of personhood—emotions, intentions, and individuality—I conclude that sheep are people—not human people, but their own kind of people. While biologically similar to humans (Chapman & Huffman, 2018), sheep have evolved alongside us and deserve recognition on their own terms. However, over-humanizing them risks projecting expectations they cannot meet (Dwyer, 2014).

Each sheep showed preference for specific others. While this may reflect the study’s short duration, small sample size, or single-observer bias, the preferences remain notable. Future research could expand its scope through the use of wearable transponders, GPS collars, or AI-assisted video analysis. These technologies could enable continuous, non-invasive tracking of proximity, movement patterns, and social preferences across larger flocks. When paired with biocentric anthropomorphism, such tools could preserve the relational sensitivity of this method while increasing its scalability and precision. This integration of observation and empathy offers a promising path for deepening our understanding of nonhuman animal personhood at both individual and population levels, accessible to any sheep carers. These refinements could strengthen the model and deepen its implications for human–nonhuman relations.

By combining biocentric anthropomorphism with proximity-based measurement, this methodology offers detailed interpretation of biologically similar species interactions. While subjectivity is inevitable, the notion that nonhuman animals are merely instinct-driven systems has been widely challenged (Bekoff, 2002; Wemelsfelder, 2012). Companion animal guardians often observe emotional states like joy, fear, and affection, aligning with research in affective neuroscience (Panksepp, 1998) and welfare science (Serpell, 2019). Ignoring these insights risks ethical blind spots. Biocentric anthropomorphism allows us to engage rigorously with nonhuman animal experience while remaining scientifically grounded.

This research proposes a simple model for “seeing” nonhuman animals: observing who they choose to spend time with. For farmers, this choice-test model can reveal preferences that improve welfare and wellbeing (Fraser & Matthews, 1997). For those seeking to expand moral imagination, it offers a tool for extending care and empathy to unfamiliar others. By using biocentric anthropomorphism alongside species-relevant distance measurements, we can begin to see sheep not as passive commodities, but as individuals with cognitive awareness, emotional lives, and friendships — worthy of recognition, respect, and ethical consideration.

References

Aaltola, E. (2008) ‘Personhood and animals: three approaches’, Environmental Ethics, 3(2), pp.175–193. https://doi.org/10.5840/enviroethics20083025

Andrews, K., Birch, J., Sebo, J., and Sims, T. (2024) Background to the New York Declaration on Animal Consciousness. www.nydeclaration.com

Animal Behaviour (2015). ‘Farm Animals/Sheep’. Retrieved from: http://www.animalbehaviour.net/for-kids/farm-animals/sheep/

Anon (2015). Sheep101. Retrieved from: http://www.sheep101.info/flocking.html

Balcombe, J. (2002). The Use of Animals in Higher Education: Problems, Alternatives, and Recommendations Humane Society Press

Bekoff, M. (2001). The evolution of animal play, emotions, and social morality: On science, theology, spirituality, personhood, and love. Zygon: Journal of Religion & Science, 36(4), pp.615–652. https://doi.org/10.1111/0591-2385.00388

Bekoff, M. (2002) Minding Animals: Awareness, Emotions, and Heart. Oxford: Oxford University Press.

Brent, L.J.N., Chang, S.W.C., Gariépy, J.F. & Platt, M.L. (2014) The neuroethology of friendship. Annals of the New York Academy of Sciences, 1316(1), 1–17. https://doi.org/10.1111/nyas.12315

Boissy, A., Manteuffel, G., Jensen, M.B., Moe, R.O., Spruijt, B., Keeling, L.J., Winckler, C., Forkman, B., Dimitrov, I., Langbein, J., Bakken, M., Veissier, I. and Aubert, A. (2007) Assessment of positive emotions in animals to improve their welfare, Physiology & Behavior, 92(3), 375–397. https://doi.org/10.1016/j.physbeh.2007.02.003

Broom, D.M. (2016) Considering animals’ feelings: Précis of Sentience and animal welfare (Broom 2014), Animal Sentience, 5(1). https://doi.org/10.51291/2377-7478.1015

Burghardt, G.M. (2007) Critical anthropomorphism, uncritical anthropocentrism, and naïve nominalism, Comparative Cognition & Behavior Reviews, 2, 136–138 https://psycnet.apa.org/record/2008-07027-009

Chapman, C.A. & Huffman, M.A. (2018) ‘Why do we want to think humans are different?’, Animal Sentience, 23, 1–8. https://doi.org/10.51291/2377-7478.1358

Chessa, B., Pereira, F., Arnaud, F., Amorim, A., Goyache, F., Mainland, I., Kao, R. R., Pemberton, J. M., Beraldi, D., Stear, M. J., Alberti, A., Pittau, M., Iannuzzi, L., Banabazi, M. H., Kazwala, R. R., Zhang, Y., Arranz, J. J., Ali, B. A., Wang, Z., … Palmarini, M. (2009). Revealing the history of sheep domestication using retrovirus integrations. Science, 324(5926), 532–536. https://doi.org/10.1126/science.1170587

Christakis, N.A. & Fowler, J.H. (2014) Friendship and natural selection, Proceedings of the National Academy of Sciences, 111(Supplement 3), 10796–10801. https://doi.org/10.1073/pnas.1400825111

Christiansen, S.B. & Forkman, B. (2007). Assessment of animal welfare in a veterinary context – a call for ethologists, Applied Animal Behaviour Science, 106(4), 203–220. https://doi.org/10.1016/j.applanim.2007.01.004

Clothier, S. (n.d.) Clothier Animal Response Assessment Tool. Retrieved from https://suzanneclothier.com/carat-temperament-assessment/

Cooke, L. (2022) Bitch: A Revolutionary Guide to Sex, Evolution and the Female Animal. Doubleday.

Dawkins, M.S. (2007) Feelings do not a science make, BioScience, 57(1), 83–84. https://doi.org/10.1641/B570114

Dwyer, J. (2007). A non-companion species manifesto: humans, wild animals, and “the pain of anthropomorphism”. South Atlantic Review, 72(3), 73+. https://link.gale.com/apps/doc/A301283587/AONE?u=anon~ef2f44c5&sid=googleScholar&xid=4a44e47a

Dwyer, C.M. (2008) Behavioural development in the sheep, in M. Spooner & H. Fraser (Eds.) Behavioural Development in Farm Animals, pp. 113-134. CAB International/

Fraser, D. & Matthews, L.R. (1997). Preference and motivation testing. In M.C. Appleby & B.O. Hughes (Eds.) Animal Welfare. CAB International, pp. 159-173. https://www.wellbeingintlstudiesrepository.org/valaexp/2/

Giri, A.K. (2006) Cosmopolitanism and beyond: Towards a multiverse of transformations, Development and Change, 37(6), 1277–1292. https://doi.org/10.1111/j.1467-7660.2006.00524.x

Gruen, L. (2011) Ethics and animals: An introduction. Cambridge University Press.

Hamington, M. (2008) Learning ethics from our relationships with animals: Moral imagination, International Journal of Applied Philosophy, 22(2), 177–188. https://doi.org/10.5840/ijap200822214

Horowitz, A. & Bekoff, M. (2007) Naturalizing anthropomorphism: Behavioral prompts to our humanizing of animals, Anthrozoös, 20(1), 23–35. https://doi.org/10.2752/089279307780216650

Humphreys, R. (2023) Critical anthropomorphism, in Animals, Ethics, and Language. Palgrave Macmillan Animal Ethics Series. Palgrave Macmillan, pp. 89–112.

Ingold, T. (1988) ‘Introduction’, in Ingold, T. (Ed.) What is an Animal? Routledge.

Kant, I. (1963) Lectures on ethics. Translated by L. Infield. Harper & Row.

Karlsson, F. (2012) Critical anthropomorphism and animal ethics. Journal of Agricultural and Environmental Ethics, 25(5), 707-720 https://doi.org/10.1007/S10806-011-9349-8

Krumer-Nevo, M. & Sidi, M. (2012) Writing against othering, Qualitative Inquiry, 18(4), 299–309. https://doi.org/10.1177/1077800411433546

Latour, B. (2005) Reassembling the social: An introduction to actor-network theory. Oxford University Press.
Law, J. & Lien, M. (2012). Slippery: Field notes in empirical ontology, Social Studies of Science, 43(3), 363–378. https://doi.org/10.1177/0306312712456947

Loughnan, S., Haslam, N. & Bastian, B. (2010). The role of meat consumption in the denial of moral status and mind to meat animals, Appetite, 55(1), 156–159. https://doi.org/10.1016/j.appet.2010.05.043

Malecki, W., Sorokowski, P., Pawlowski, B. & Cienski, M. (2019). Human Minds and Animal Stories: How Narratives Make Us Care About Other Species. Routledge.

Marino, L., & Merskin, D. (2019). Intelligence, complexity, and individuality in sheep. Animal Sentience, 4(25), 1. https://doi.org/10.51291/2377-7478.1374

Mendl, M., & Paul, E. (2004). Consciousness, emotion and animal welfare: insights from cognitive science. Animal Welfare, 13(S1), S17–S25. https://doi.org/10.1017/S0962728600014330

Milton, K. (2005). Anthropomorphism or egomorphism? The perception of non-human persons by human ones, in Birke, L. & Smith, J. (Eds.) Animals in Person: Cultural Perspectives on Human-Animal Intimacy, pp. 255-271. Berg.

Mullin, M.H. (1999). Mirrors and windows: Sociocultural studies of human-animal relationships, Annual Review of Anthropology, 28, 201–224. https://doi.org/10.1146/annurev.anthro.28.1.201

Nadasdy, P. (2007). The gift in the animal: The ontology of hunting and human–animal sociality. American ethnologist, 34(1), 25-43. https://doi.org/10.1525/ae.2007.34.1.25

Panksepp, J. (1998) Affective neuroscience: The foundations of human and animal emotions. Oxford University Press.

Pardo, M. C. (2023). Legal Personhood for Animals: Has Science Made Its Case? Animals, 13(14), 2339. https://doi.org/10.3390/ani13142339

Reefmann, N., Kaszàs, F. B., Wechsler, B., & Gygax, L. (2009). Ear and tail postures as indicators of emotional valence in sheep. Applied Animal Behaviour Science, 118(3-4), 199-207. https://doi.org/10.1016/j.applanim.2009.02.013

Rivas, J. and Burghardt, G.M. (2018). Crotalomorphism: A metaphor for understanding anthropomorphism by omission, in Bekoff, M., Allen, C. & Burghardt, G.M. (Eds.) The Cognitive Animal, pp. 9-18. MIT Press.

Rowlands, M. (2016). Are animals persons?. Animal Sentience, 1(10), 1. https://doi.org/10.51291/2377-7478.1110

Sanders, C. R., & Arluke, A. (1993). If lions could speak: Investigating the animal‐human relationship and the perspectives of nonhuman others. Sociological Quarterly, 34(3), 377-390. https://doi.org/10.1111/j.1533-8525.1993.tb00117.x

Sealey, A., & Charles, N. (2013). “What do animals mean to you?”: naming and relating to nonhuman animals. Anthrozoös, 26(4), 485-503. https://doi.org/10.2752/175303713X13795775535652

Serpell, J. (2019). How happy is your pet? The problem of subjectivity in the assessment of companion animal welfare. Animal Welfare, 28(1), 57–66. https://doi.org/10.7120/09627286.28.1.057

Seyfarth, R. M., & Cheney, D. L. (2012). The evolutionary origins of friendship. Annual review of psychology, 63(1), 153-177. https://doi.org/10.1146/annurev-psych-120710-100337

Shapiro, K. J. (1990). Understanding Dogs through Kinesthetic Empathy, Social Construction, and History. Anthrozoös, 3(3), 184–195. https://doi.org/10.2752/089279390787057540

Singer, P. (2009). Speciesism and moral status. Metaphilosophy, 40(3‐4), 567-581. https://doi.org/10.1111/j.1467-9973.2009.01608.x

Smuts, B. (2001). Encounters with animal minds. Journal of consciousness studies, 8(5-6), 293-309. https://www.ingentaconnect.com/content/imp/jcs/2001/00000008/F0030005/1213

Squires, V. R., & Daws, G. T. (1975). Leadership and dominance relationships in Merino and Border Leicester sheep. Applied Animal Ethology, 1, 263–274. https://doi.org/10.1016/0304-3762(75)90019-X

Wemelsfelder, F. (2012). A science of friendly pigs… Carving out a conceptual space for addressing animals as sentient beings. In Crossing Boundaries (pp. 223-249). Brill. https://doi.org/10.1163/9789004233041_013