Back in 2011, in the article “Horse Whispering Demystified,” I questioned the wisdom of chasing horses about in round pens, and critiqued the marketing spin about learning “the language of Equus” to establish a more harmonious relationship with horses. I noted that the natural horsemanship methods were probably not overtly harmful in the hands of a skilled practitioner. However, subsequent reading and research has caused me to revise that perspective. The premise that a dominance hierarchy is critical to all equine social organization, and we must thus insinuate ourselves into that hierarchy as the alpha in order for our horses to respect and perform well for us (e.g. Roberts, 1996, 2007) is not only misleading, but potentially threatens equine welfare.
Horse Herd Hierarchy
First, the notion that feral horses establish a rank order that remains relatively stable has not been supported. Indeed, research that has tried to establish a rank hierarchy by observing aggressive encounters between members has proven difficult because very little aggression occurs. Feral horses show little motivation for dominating others, as conflict is energy costly and best avoided (ISES, 2018).
For domestic herds, the bucket test, where herd members are tested in all possible pairs as to who gets to shove their muzzle in the oat bucket first, has been used to determine a rank order. However, Swiss researchers Marie Roig-Pons and Anja Zollinger (2017) found that the bucket test hierarchy was quite different to the hierarchy seen when the horses were free-ranging, and that hierarchy is clearly not the only factor that describes social relationships within a group.
The International Society of Equitation Science (ISES) position statement on dominance states that there is little support for an overall rank in equine social organization. Rather, horses tend to form bilateral relationships, where a new contract is negotiated separately for each herd member, and each new situation (2018). As equine behaviourist Robin Foster (2017) notes, “dominance rank and hierarchy are useful constructs to scientists, but from a horse’s perspective what’s important – and remembered – are past interactions with other horses, which helps resolve future conflicts without fighting.”
Even if it could be established that herd hierarchies are as solid and unvarying as has been claimed, there is no evidence that horses consider us as one of their social circle, and thus deserving of a rank order (ISES, 2018). Horses comply not because they see us as their leader, but because their human trainers have consistently reinforced that response in the past. A horse that refuses to go forward into a spookier part of the trail is not balking because he does not believe in his leader, but because his training to go forward has been insufficient. He has not yet learned, by repetitions of the go forward command in increasingly more challenging contexts, to override his fear and to go forward into unwelcoming places (McLean & McGreevy, 2010).
Are There Alpha Leaders in Horse Herds?
A major theme in many training philosophies is that the horse must come to respect the trainer as the alpha leader. In most natural horsemanship systems, this is established through chasing a horse in a round pen until licking and chewing are observed, at which point the trainer releases the pressure and offers the horse the option to follow or to “join-up” (e.g. Roberts, 1996, 2007). In 2007, Konstanze Krueger tested 26 horses using natural horsemanship round pen techniques and found that the time until following the trainer decreased daily over 10 trials, no matter how the chasing was done, or by whom – elements purported by Roberts to be critical to successful join-up. Also noteworthy is that when these same “joined-up” horses were given the option to follow the trainer in a small paddock or larger field, with or without other horses present, not one of the 26 horses did so. If these horses had come to see their trainer as another dominant horse, and one worthy of respect, surely this relationship would extend beyond the parameters of the round pen.
A creative study by Henshall and Padalino (2012) found that horses could be trained to join-up to a remote controlled car that chased them around a round pen, but would stop when the horse stopped and faced the car, thus reinforcing their approach behaviours. This research certainly questions the notion that join-up results from a trainer learning equine language in order to gain a horse’s respect.
Furthermore, although licking and chewing has been marked as the pivotal point indicating that the horse is ready to show respect and join-up, in reality, licking and chewing originates from a surge in saliva output as the body switches from the dry mouth typical of the sympathetic nervous system (pain, fear or confusion can all turn on the sympathetic system), to the return of saliva associated with the resulting body relaxation of the parasympathetic system. These behaviours are nearly always indicative of a horse being released from a higher to lower state of alarm or distress (McDonnell, 2016). Studying 200 feral horses, living in Ecuador, Norwegian equine researchers Margrete Lie and Ruth Newberry (2018) discovered that in threatening encounters, both the aggressor and the submissive horse licked and chewed, although, surprisingly, aggressors did so slightly more often.
Being Fair to the Horse
Equine scientists almost universally agree that what we are seeing in the round pen can be explained by the principles of negative reinforcement, which forms the basis of almost all equine training under saddle: we apply an aversive stimulus, until the horse offers the desired response (or the first baby steps toward the desired response), we release the pressure, and the horse is more likely to offer this response in the future. Much round pen training, however, is a particularly egregious application of negative reinforcement because it sets in motion the horse’s flight and fear response. Unlike other responses that require numerous repetitions to become established, fear responses can be learned in one trial, are notoriously difficult to eradicate once learned, and can return sporadically and unpredictably with little provocation (Le Doux, 1994: McLean, 2001). As McGreevy and McLean (2010) note, “this research has monumental implications for horse trainers: avoid fearful reactions at all costs.” Horses’ typical round pen responses have very little to do with respecting a leader; rather horses flee because they are frightened and eventually learn that fleeing doesn’t work, but stopping and facing the trainer will end the aversive chasing.
In Kruger’s study, three of her original 19 horses had to be withdrawn on humane grounds because of atypical profuse sweating and an eventual failure to respond to the experimenter doing the chasing.
When trainers believe that horses’ undesired behaviours stem from a defiance to their authority, the more plausible and parsimonious explanation – that the horse does not understand the question, that there is no correct response, or that the horse is psychologically or physically unable to comply – is often overlooked. The dominance training model has very little to do with the complex social organization of herd living. It is, at best, unfair, at worst, inhumane, nearly always counterproductive to training, and may ultimately jeopardize horse welfare.
To view the ISES position statement on dominance, click here.