After seeing a post warning of the dangers of 'New age behaviorists refusing to punish dogs, and even claiming dogs do not have pack animals in their DNA' I wanted to give an evidence based insight into the subject, sorry its a long one!
As pet owners and canine professionals, it is vital we understand how our dogs communicate and interact socially in order to help resolve behaviour problems. The idea that dog’s form packs with
strict hierarchies, maintained by displays of
dominance and submission, is one of the most prevalent theories among dog owners and professionals alike. However, many are now starting to question its accuracy. Some will argue that dog’s form strict linear hierarchies and we must be our dog’s alpha or pack leader in order to prevent every behaviour problem from lead pulling to severe aggression.Others argue that dogs are not pack animals and are instead social opportunists and so we do not
need to be the `pack leader’ in order to prevent behaviour problems, instead, we must simply
understand how to motivate our dogs and how they learn.
Out of interest I wanted a better understanding of the view on this topic from the public compared to professionally educated dog trainers. I completed a short survey which I shared on social media.
In this survey I asked participants their age, if they had ever had any dog training related education, whether they had worked with dogs and if they believed in hierarchy/pack structure and the alpha
dog. My survey revealed interestingly that the older generation were much more likely to believe that pack structure existed and dogs would challenge humans and other dogs in order to become
‘alpha’. Many answered that they believed dogs lived in packs but did not believe they were in a battle for dominance. All of those who had studied canine behaviour at university didn’t believe
dogs formed packs or had alphas. Many who said they did believe in alpha dog’s and pack structure said they had learnt this information from tv shows and interestingly, from professional dog trainers too.
This makes sense considering that during the 1900’s the main theory regarding the social behaviour of the domestic dog was that, when living in groups, canine social relationships are managed
through formation of a linear dominance hierarchy. Pack structure / dominance theory suggests that each individual in a pack holds a rank and that higher-ranking individuals have priority access to
resources, have the right to breed and initiate social interactions ahead of those lower in the ranks . Lower ranking dogs are kept `in their place’ by displays of dominance by higher ranking individuals. Lower ranking dogs may also challenge higher ranking individuals in an attempt to improve their social status within the pack. All domestic dogs are said to have an innate desire to achieve the ‘alpha’ status.
The concept of social ‘dominance’ relationships in animals began with Thorleif Schjelderup-Ebbe’s studies of captive groups of domestic chickens in the early 1920s (Schjelderup-Ebbe, 1922). He found that chickens had a ‘pecking order’ in which the most dominant individuals in the group had the best access to resources such as food, and would maintain their status with ‘dominant’ behaviours such as pecking other birds (Schjelderup-Ebbe, 1922).
The idea that wolves and therefore dogs form packs with strict hierarchies came from the then ground-breaking paper by Rudolph Schenkel (1947) “expression studies on wolves”. At this point,
there was very little / no information regarding the behaviour of wolves the did not live in captivity.
Schenkel studied captive wolves in the Basle zoological garden in 1934, he then extended his research in 1939 to include observations from other canine species in the zoo such as raccoon dogs, and finally domestic dogs (Schenkel, 1947). His studies concluded that the wolves lived in a pack ruled by one male (the alpha) and his alpha female. Violent rivalries occurred between the wolves which existed to maintain the social structure. The alpha dog maintains his status in the pack my regular displays of dominance over the other males in the group (Schenkel, 1947). The
subordinate wolves in the group will fight to raise their social rank and are sometimes, but rarely, successful in taking the position of alpha (Schenkel, 1947).
Scott and fuller (1965) continued with this behaviour model when investigating the behaviour of domestic dogs in captivity, raised independently from humans. They stated that these wild dogs
lived in a pack and defended their territory just like wild wolves would, and that pet dogs display this behaviour too when they defend their home (Scott and fuller, 1965). In their book, they state
“Within a natural social group, agonistic behaviour is reduced to a relationship of dominance and subordination. This may take several forms, depending on the degree of dominance. Some dogs
simply growl at each other and move apart. More typically, the dominant dog places his feet on the back of the other, growling as he does so, while the subordinate one keeps his head and tail
lowered. A still more subordinate animal may roll over on his back while the dominant one stands over him, head-to-head; the subordinate animal rapidly snaps his teeth and yelps. “(Scott and Fuller, 1965). They also noted other dominant or submissive behaviour such as a submissive dog avoiding eye contact with a more dominant dog (Scott and Fuller, 1965). Many other studies reported similar findings. For example, Rabb, Woolply and Ginsburg (1957) Studied captive wolf behaviour and noted dominance and submission behaviour patterns between same sex individuals and the exitance of an alpha male who was the most dominant individual in the pack.
Dr David Mech, citing Schenkel’s work, published his book “The Wolf: Ecology and Behaviour of an Endangered Species,” in the 1970’s, plus hundreds of scientific papers which support the idea that wolves form packs with strict hierarchies, led by an alpha male wolf (Mech, 2005). It was then thought that dogs were likely domesticated when humans captured and raised wolf pups, who were then gradually selectively bred into the domestic dog breeds we have today . Therefore, wolf behaviour was a good indicator of dog behaviour and should dictate how to raise and train our dog’s.
. Supporting this, Bruce Fogle argues that by giving in to our dog’s demands, we may be showing them
that we are weak and this may lead them to challenge our position in the pack (Fogle, 1992).
Dominance aggression, may therefore be a result of not maintaining our position as the pack leader. The alpha roll, first popularized by the monks of new skete in the 1978 book How to Be Your Dog's Best Friend (Monks of New Skete, 1978) is one of the ways we are taught to deal with dominance in
our dog’s, based on the idea that a more dominant dog will force another dog onto their back.
This theory is still very prevalent today. As Ceasar Millan states on his website, dogs are pack animals
who form packs with humans and other dog in their home (Millan, 2020). . He claims behaviour problems in dogs are caused by owners who fail to be their dog’s pack leader. While he is by far not the only dog trainer who used dominance-based training techniques such as alpha rolling, he is the most famous, and one of the main reason's dominance theory is so widely accepted by pet owners.
However, Milan actually has no formal education in canine behaviour as stated on his website (Millan, 2021). He attributes his success to owning dog’s his whole life, like so many others. So, should we really be listening to a man with no qualifications?
In recent years, more and more professional dog trainers are claiming that there is no need to be your dogs ‘pack leader’ and that dominance-based training methods are actually damaging to our dogs.
The first problem modern dog trainers see with dominance theory is that dogs are not wolves. As Shenkar states” when we want to study cats, we look at cat behaviour. When we want to study horses, we look at horse behaviour, but when we want to study dogs, we look at wolves”.
Wolves and dogs diverged from their last common ancestor between 11,000 and 41,000 years ago (Pomery, undated). Although they are distantly related, dogs and wolves are different subspecies, and the two have very different behaviours and physical characteristics. For example, while wolves have a breeding pair in their group, dogs are promiscuous and opportunistic breeders, any male has the chance to breed when the opportunity arises so there is no need for an alpha male (Shenkar,undated).
Studies into wild dog behaviour has not produced the same results as studies into wolf behaviour. For example, Roberto Bonanni et al (2010) looked at free-ranging packs of dogs in Italy and found that leadership between dogs was fluid and varied from situation to situation. For example, in one pack, which had 27 members, there were 6 dogs that took turns leading the pack, but at least half of the adult dogs were leaders, at least some of the time. The dog’s that led the pack were usually the most experienced, rather than the most dominant. Furthermore, Coppinger and Coppinger (2001)
studied feral dogs and found most will live independently or in small, fluid groups. Unlike wolves, when a bitch comes into season, any male is free to mate with her, and she will raise the pups on her own (Coppinger and Coppinger, 2001).
Furthermore, even if wolf behaviour was representative of dog behaviour, the information we had on wolf pack structure was based on captive wolves, and not representative of wild wolf behaviour. The famous study on this subject conducted by Rudolf Schenkel (1947) that we discussed above was done on a pack of up to ten wolves kept in a 200 m2 pen. These wolves were not representative of a wild wolf pack because they were unrelated and in captivity meaning that understandably, tensions would be high and rivalries common. Were we looking, for example, at two captive wild animals fighting over a high value resource such as food or sleeping space, or, were we looking at an alpha wolf maintaining his position through violent displays?
We know that wolves, unlike dogs, are not friendly with unrelated wolves, so surely this behaviour would be expected. As Mech states in the introduction of his study on wild wolves “Attempting to apply information about the behaviour of assemblages of unrelated captive wolves to the familial structure of natural packs has resulted in considerable confusion. Such an approach is analogous to trying to draw inferences about human family dynamics by studying humans in refugee camps. The concept of the alpha wolf as a ‘top dog’ ruling a group of similar-aged compatriots is particularly misleading” (Mech,1999).
Moreover, Mech, who previously supported Schenkel’s findings and published his book about wolf pack structure, has since doubled back on what he once stated as fact and now claims he was wrong. On his website he states that although his book is still available to buy, he is desperately trying to remove it from circulation (Mech,2021). He has learnt a lot since then, and the reality is that wolves do not have strict linear hierarchies maintained by displays of dominance, but rather wolf packs are simply family units, with one breeding pair and the other members being their offspring (Mech, 2021). There is rarely any need for displays of dominance as the other wolves naturally follow the oldest and wisest members. All members eat together, hunt together and sleep together as a family. He also states that young members of any wolf pack eventually disperse to form their own family, therefore forcing them to stay together for many years while studying captive wolves, is bound to cause conflict (Mech, 1999). Mech also challenges the idea that some wolves are naturally dominant or submissive, as is claimed by many dog trainers today, stating any wolf has the chance to lead a pack, all he has to do is breed (Mech, 1999).
If dogs are not wolves, and wolves don’t form linear hierarchies, then why can’t we move on from dominance theory as dog trainers? As the Pet professional Guild states on their website “It is the position of the Pet Professional Guild (PPG) that dominance theory is an obsolete and aversive method of interacting with animals that has at its foundation incorrect and misinterpreted data which can result in damage to the animal-human relationship and cause behavioural problems in the animal”. The dominance model of dog training has its foundations in a number of concepts which
have now been proven to be false (Bradshaw, 2011).
However, the problem is dominance type training can work and can seem to make total sense. While many believe this is because the theory is correct, I believe this is often due to a phenomenon called “learned helplessness”. Learned helplessness was discovered by Seligman in 1967. In short, he discovered that when he repeatedly electrocuted dogs and gave them no way to escape, when they did eventually have a clear escape route, they didn’t choose it and instead stayed put and suffered electrocution again (Kelsing, 2010). The dogs had learnt that there was nothing they could do to improve their situation so even when given an escape, believed they could not take it. Learned helplessness can be seen in dogs who have been trained with harsh dominance type methods. For example, a dog who barks at other dogs to communicate that he is scared may be pinned to the floor by a dominance type trainer. After repetition the dog learns that even though he is scared, there is nothing he can do to improve his situation, so he stays quiet. While many would claim the dog has learned he is no longer ‘alpha’ and so is ‘behaving’ it would be clear to me that the dog is suffering intense emotional distress and is failing to respond to a scary situation because he believes there is no way out.
As stated in Barry Eatons book (2008) much of the behaviours we claim as dominance in domestic dogs are actually human social constructs. For example, at work it would be rude not to open a door for your boss, to allow them to walk through it first. Therefore, we believe that when a dog walks through a door before us, they are challenging our dominant position. However, there are no doors in the wild for dogs to open to allow their ‘alpha’ to walk through. In reality, this behaviour is simply a dog excited to leave the house. Furthermore, others are simply based on misconceptions. For example, the idea that the alpha wolf always eats first, and therefore we should always eat before our dogs, is a myth. In reality, most of the time the whole wolf pack eats at the same time, and, if food is scarce, it is the pups who eat first (Eaton, 2008). Another example would be the idea that we shouldn’t let dogs on furniture, to ensure they are lower down than we are, because a high-ranking wolf is always higher up than lower ranking wolves. This, again, is a misconception. As Eaton (2008) explains, the adult wolves often sleep in a high position so they have a good vantage point to warn off intruders, and the younger wolves would be lower down, where it is safer.
As discussed above, the alpha roll is a common method of ‘training’ used by dog trainers who believe in dominance theory. In his book ‘The intelligence of dogs’ Stanley Coren states “You should deliberately manipulate and restrain your dog on a regular basis, placing it in a position that, for wild canids, signifies submission to the authority of a dominant member of the pack” (Coren, 2006).
However, as Eaton (2008) states, this behaviour in wolves is actually voluntary and can be seen when a wolf rolls onto his back when approached by another wolf, at no point is a wolf forced onto their back by another wolf. Furthermore, while many believe dog’s do in fact display this behaviour McConnell (2002) states that no well socialised dog will pin another dog to the ground. McConnell goes on to say that forcing your dog into a submissive position is a great way to cause them to become aggressive towards you as in their social framework, you are acting like a lunatic (McConnell, 2002).
Unfortunately, the idea of pack structure and dominance has many to believe that physical punishments are acceptable when training our dogs, since this is how wolves would discipline lower ranking wolves. However, according to a new veterinary study published in The Journal of Applied Animal Behaviour, if you use physical corrections to train your dog, you are likely to cause your dog to become aggressive (Yin, 2009)
The study looked at how owners corrected dogs for a number of issues, such as aggression towards other dogs or guarding resources, and how many of these dogs then became more aggressive, or bit their owner. The results are below:
“The highest frequency of aggression occurred in response to aversive (or punishing) interventions, even when the intervention was indirect:
• Hitting or kicking the dog (41% of owners reported aggression)
• Growling at the dog (41%)
• Forcing the dog to release an item from its mouth (38%)
• “Alpha roll” (forcing the dog onto its back and holding it down) (31%)
• “Dominance down” (forcing the dog onto its side) (29%)
• Grabbing the jowls or scruff (26%)
• Staring the dog down (staring at the dog until it looks away) (30%)
• Spraying the dog with water pistol or spray bottle (20%)
• Yelling “no” (15%)
• Forced exposure (forcibly exposing the dog to a stimulus – such as tile floors, noise or people –
that frightens the dog) (12%).
In contrast, non-aversive methods resulted in much lower frequency of aggressive responses:
• Training the dog to sit for everything it wants (only 2% of owners reported aggression)
• Rewarding the dog for eye contact (2%)
• Food exchange for an item in its mouth instead of forcing the item out (6%)
• Rewarding the dog for “watch me” (0%)” (Yin, 2009).
I personally have found this with my own rescue dog too. When I adopted nell at 5 years old, she
was being trained using alpha rolls and physical corrections, and had been for many years, due to her lack of socialisation leaving her reactive. Subsequently, she had gone on to bite her owners, and her behaviour got worse, rather than better. Faced with a very nervous, damaged and unsocial dog, I attempted to train her using positive reinforcement, confidence building obedience and negative punishment such as walking away. Her behaviour in general dramatically improved.
Some would argue against this evidence, and claim that dog’s do form hierarchy-based packs, since many multiple household dogs will fight over resources, and some can see a clear ‘pack leader’ who isn’t challenged. However, this behaviour can be explained differently if you look at it with new eyes.
It is a dog’s natural survival instinct to protect something that is of value to them, and some dogs have been bred to have a stronger desire to want certain things, for example, a spaniel will be much more likely to guard a tennis ball than a great Dane would, due to innate drives.
The Resource Holding Potential model explains why some dogs will guard an object from other dogs and some will not (Bradshaw, 2011). According to this model, when a conflict-of-interest arises, the dog is said to make a decision based on two questions: how much do I want this resource? and, how likely is the other dog to beat me in a fight if we fight for it? (and how much the other dog wants it/ is likely to fight for it) (Bradshaw, 2011). This model can be applied in the place of the dominance model to explain why fights break out regarding resources in the home, or how dogs keep the peace.
Therefore, dominance cannot be a personality trait, but rather something fluid that changes from situation to situation, as it is in humans. There are many situations like this, that could be looked at in a different way, and therefore be explained without dominance theory.
I believe that dominance-based training methods, are often harmful to our dogs, and do not address the causes behind behaviour problems, they merely attempt to control symptoms of behaviour problems.
Study (Vieira de Castro et al., 2020) after study (Blackwell, Twells, Seawright and Casey, 2008) after study (Schilder and van der Borg, 2004) has shown that dogs trained with dominance based techniques are far MORE likely to display aggression, experience more stress and have a lower score on problem solving and cognitive ability tests than dogs trained using positive reinforcement.
Canine behaviour science is new, which means theories are going to rapidly evolve and we owe it to our dogs to keep up. I too, grew up strongly believing in dominance training and have had to unlearn much that I took as gospel.
Bibliography of all studies quoted is available on request.