Written by: Antonia Henderson, Ph.D.

Equine psychologist, Antonia J.Z. Henderson, Ph.D., dispels myths about herd dynamics and offers advice to encourage horses to get along.

Thumbnail for Herd Harmony: Why Can’t We all Just Get Along?

Clix Photo

Allowing horses to live naturally in stable social groups is the panacea for a litany of equine ailments. It improves fitness, all but eliminates colic, cures ulcers and reduces or eliminates cribbing, weaving and other stereotypies associated with compromised psychological well-being. Undeniably, there is a robust literature suggesting that keeping horses in stalls benefits only one half of the horse/human dynamic, and that beneficiary is not the horse.

Still, owners are often reluctant to subject their valuable horses to the potential risk of group turnout and the typical life for today’s sport and many pleasure horses is to live in box stalls for approximately 16 hours of a 24-hour day, and the remainder (for the lucky ones) in small, individual paddocks, with no horse-to-horse interaction. Owners and trainers could afford to take a second look at group turnout, however. The research evidence suggests that although equine aggression does happen, the risk of injury may be greatly overestimated, and the possibilities for minimizing that risk are considerable.

How risky is it? Dispelling myths about herd dynamics

Myth #1: Aggression is Common and Leads to Injuries

The literature indicates that aggression is rare in stable herds, and almost never results in injury. Rather, horses are more likely to use “the minimum amount of aggression a situation requires” (Hartman, 2009), usually in the form of threats rather than contact. In 46 tests of introducing new horses to residents in various conditions of pre-exposure, Elke Hartman from the Swedish University of Agricultural Sciences observed only one incidence of actual body contact in an aggressive encounter, and the resulting injury was minor (2009).

One of the key reasons that aggression is uncommon is that it is adaptively costly. For animals living in stable social organizations, behaviours that mitigate conflict rather than escalate it are adaptive, and thus become evolutionarily selected. An aggressor expends energy resources, risks serious injury and potentially forfeits the benefits of group membership should the conflict result in ostracism. Aggression may have equally dire consequences for the winners by threatening group cohesiveness and jeopardizing future cooperation (Aureli, 2001). In the grand scheme of herd living, serious aggression seldom reaps a benefit that outweighs its cost.

Myth #2: Dominance hierarchies make horses aggressive

Horses living in a natural social herd form inter-species social hierarchies that serve to minimize aggression rather than exacerbate it. Filipa Heitor, from the University of Lisbon, Portugal, found that in an established herd of Sorraia horses dominance relationships were clear, linear and seldom contested (2006). Even after the introduction of a breeding stallion, dominance ranks remained relatively stable and served to keep aggression to a minimum. As Heitor explained, once ranks have been established, a dominant horse who wants access to a resource need only threaten a subordinate, who in turn avoids or acquiesces; this ritualized dynamic ensures the stability of the hierarchy and of the entire group.

Myth #3: Horses hold grudges

Franz de Waal first studied friendly interactions after conflicts between chimpanzee opponents, and was able to dispel the myth that opponents stay angry after aggressive encounters (1979). Rather, opponents were in closer contact after conflict than before it and exchanged friendly behaviours such as grooming, embraces and kisses. Looking at reconciliation behaviour in horses, Alessandro Cozi and colleagues from the Institute of Applied Ethology in Saint Saturnin, France, found that horses also appear to effectively kiss and make up. In his observations of horses living in a stable social group, 86 per cent of conflicts were followed by a friendly reunion within 10 minutes post-conflict. Furthermore, although reconciliations between the two opponents were common, reconciliations involving a third horse were even more likely. Third horses acted both as appeasers (initiating friendly contact with the aggressor), or consolers (initiating friendly contact with the victim). Cozzi proposes that these “triadic interactions” may provide a critical social mechanism for conflict resolution and the maintenance of social cohesion
(Cozzi, 2010).

Myth #4: You can’t turn a stallion out with anyone

A study by Sandra Granquist and her colleagues at the University of Iceland lends support to the notion that stallions are unlikely to initiate aggressive encounters with other stallions, or anyone, unless there is a very good reason for doing so. Granquist (2012) observed the impact of the introduction of a new stallion and his nine horse harem into a pre-existing, large, semi-feral herd, divided into four, distinctand non-overlapping harems. Although interactions with the newcomer were more frequent than with other stallions, direct interactions among stallions were rare and, when they did occur, non-violent. Rather than direct confrontation, stallions communicated indirectly through dung and urine markings, and maintained their sovereignty with timely harem herding to prevent interactions between their mares and other harems.

Granquist’s results suggest that housing stallions together may work on the open range, but it is not a feasible practice for fit, fed and valuable sport and pleasure horses. Or is it?

Sabrina Freymond (2013) and her colleagues at the Swiss National Stud successfully pastured five to eight elite, expensive, sport horse breeding stallions in group turnout during their non-breeding season. As a result of their success, the Swiss National Stud continues to pasture a “bachelor band” every year. Freymond noted that stallions, rather than fighting, engaged in ritualistic behaviours (abbreviated, non-contact behaviours that substitute for actual aggression) to establish social hierarchies, and that in each subsequent year even these ritualistic behaviours subsided. To reduce competition, stallions were housed in large pastures away from mares, had multiple feeding stations, met in adjacent stalls before group turnout, and hind shoes were removed before the initial encounter. She found that antagonistic behaviours subsided quickly, there were no injuries, welfare was greatly enhanced and management costs were significantly reduced.

An addendum to this myth is that mares and geldings should be separated for peaceable group turnout. This dictum has not been supported, however; mixed groups are no more aggressive, nor more likely to inflict or sustain injury than segregated gender groups (e.g. Jorgensen, 2009). Interestingly, Jorgensen did find that boys appear to have more fun, with mixed groups and all gelding groups showing significantly more play behaviour than all female groups.

Myth #5: Aggressive horses are bad horses

Anyone who has seen an aggressive horse encounter can attest to the fact that it is unquestionably horrifying. The attacker appears vicious, savage and brutal. From an evolutionary perspective, however, the moral constitution of the aggressor is simply not a part of the equation. Evolutionarily speaking, it pays to be of higher rank – you get readier access to resources (food, water, shelter, mating privileges, etc.), and you benefit from less aggression directed at you. For mares this advantage carries on to their offspring; a higher quality diet results in healthier, thriving, faster growing, offspring that breed earlier, contribute to the mare’s overall reproductive success and maximize her genetic contribution.

According to the theory of “evolutionary stable strategies”, aggression is always weighed in a cost/benefit “is it worth it?” scenario, considering the value of the resource, the risk of injury, and the likelihood of winning (Aureli, 2002). It doesn’t pay to make a bid for a higher rank unless the resources that you want are highly prized, the risk of injury is low and the chances of winning are in your favour. So we tend to see competition in animals that are close to each other on this linear social ladder, where they have a better shot at winning. A horse who is newly disadvantaged (older, weaker, more infirmed) is a likely target for those close to him on the pecking order. Picking on a guy when he is down seems like decidedly poor form to us. This aggression is not morally driven, however, but is simply a natural outcome of a greater evolutionary plan.

Turn them out!

Since 1994, Dr. Sue MacDonnell has kept a herd of approximately 100 semi-feral ponies to study equine physiology and behaviour at the University of Pennsylvania’s New Bolton Centre. With minimal preventative health care, modest supplemental feeding in the deep of winter and almost no other veterinary or human intervention, the ponies thrive. The mares are continually fertile, have few reproductive difficulties and no need for veterinary assistance. Without farriers, hoof health is excellent. They have not experienced a single case of colic or founder or seen any incidence of stereotypic behaviour. Lameness and/or injuries that require intervention are almost non-existent. MacDonnell states on her website that she and her team are “interested in understanding the factors contributing to the ponies’ extraordinary good health and fertility compared to similar stock kept under domestic conditions”.

Although MacDonnell’s team is undoubtedly searching for a more nuanced answer, there is one resounding truth that surfaces from her, and others’, research: herd living works for horses and it works really well. With a little preemptive planning and management, creating more opportunities for herd turnout could provide a huge benefit for our pleasure and sport horses as well.

Developing a happy herd

Start early and mix up the groups

Horses learn to be socially adept with other horses by being with other horses, ideally, from the beginning, and in large heterogeneous groups. Although same age and same sex groups are generally the norm after weaning, research suggests that mixed groups (mixed in age, gender, size, dominance, breed etc.) reduce aggression and are more conducive for teaching young horses how to be good citizens. Danish researcher Jan Ladewig maintains that in heterogeneous groups, horses learn multiple aspects of social communication that diminish aggression. Additionally, in large groups with ample space, horses can more readily find a compatible partner and more easily avoid those who are not (Ladewig, 2013).

Giles and colleagues (2015) also found that more interactions and more “displacements” (where one horse moved, usually threateningly, toward another, resulting in the second horse moving away) occurred more often in herds with less variability in size, age and body condition. These authors suggest that the common practice of grouping similar horses together may be less ideal than mixed groups which lead to fewer interactions and thus less risk of injury.


Research suggests that pre-exposure reduces conflict. How much pre-exposure is still up for debate, but Hartman (2009) reported that as little as five minutes of tactile contact (i.e. letting horses smell, touch, squeal and interact) in adjacent stalls or paddocks before the initial group turnout mitigated the occurrence of aggressive encounters.

Don’t mess with the groups

A key to maintaining safe herd dynamics is to keep the groups consistent – new additions and even new subtractions force the herd to reestablish dominance hierarchies. Although some studies have found that introducing a new herd member does not necessarily change the dominance lineup, it does create greater potential for positions to be challenged (Jorgensen, 2011). If groups must be changed, or when setting up groups initially, removing hind shoes can reduce injury while negotiations are being sorted out. Once challenges have been reconciled, the shoes can be replaced.

Groups work best when resources are plentiful

Haifa Benhajali, from Tunis University studied 44 Arabian broodmares housed at a meager breeding facility with high-density housing, minimum vegetation and virtually no shelter from the 45° C temperature (2007). (The authors note that these conditions were the facility’s norm and not designed for their study!) Ordinary behaviours such as rolling, lying down, resting and even urination and defecation were almost never observed. The mares were continually moving, and the most commonly observed behaviours were antagonistic interactions. However, contrary to predictions, even in this high density, high stress situation, aggression was still relatively rare. Rather, the high stress environment seemed to jeopardize positive social interaction. Mares had few preferred partners and mutual grooming was never seen.

When environments are rich with resources, particularly edible resources, psychological well-being flourishes and aggression is reduced. The provision of multiple, widely dispersed, feed and water sources, several shelter alternatives (from wind, snow, rain and sun), numerous scratching stations or trees, and (in an ideal world) plentiful pasture, makes arguments over resources unnecessary and not easily defended. Where grazing needs to be supplemented with hay or grain, food becomes a defendable resource and herd dynamics change.

To reduce aggression, supplemented feed needs to be plentiful and widely distributed across multiple feeding stations. Where horses require restricted diets due to obesity concerns, consider using grazing muzzles for easy keepers rather than reducing the food source for the whole group.

Give them space

No research has suggested a magic ratio of acreage to horses. Many factors work into the equation of how much space is enough (age, number of horses, length of time together, activity level, individual personalities, etc.) and, as we saw in the Tunisian mares, horses will suffer considerable confinement without resorting to overt aggression. As a general rule, however, the more space you can provide, the more harmonious the herd dynamics will tend to be.

Turning your meager (or non-existent) pasture into an amusement park with the provision of multiple enrichments such as scratching posts, horse toys and edible enrichments such as straw, also reduces aggression.