Part 1: Dogs are special

Dogs are a unique species

Dogs are unique as a species in being able to bond closely with humans, despite living with conspecifics (others of their species). Dogs just need minimal exposure to humans by the age of 9-14 weeks (1). Wolves, on the other hand, can only bond with humans if they are separated from conspecifics before their eyes are open, and to develop a bond, they need a lot more contact with humans than dogs do (2). 

Dogs are geared for a partnership with us, and this potential to bond can be 'switched on' with surprisingly little contact, just a few minutes a day. Without this minimal contact before 14 weeks, dogs are extremely wary of humans, and have to be tamed as though they were a wild species (3). Contact makes bonding possible. Bonding gives dogs the motivation to learn how to decipher humans and understand our social rules. Right from the start, dogs learn to communicate with their owners, their mother and their littermates. Later, they meet strange humans and dogs. They learn the meanings of a wide range of human vocalizations and body language, from coos of adoration and gentle strokes, to angry shouts and waving arms.

Humans are obsessed with language. True, it's useful, we can teach dogs to find objects by name, like 'keys' or 'mobile' - handy if you've lost them. Verbal encouragement helps dogs find objects. Sometimes canine intelligence is assessed in terms of how many human words a dog understands. Border collies, for example, can learn a surprising number of words for objects, and can also categorize them (4). Some dogs have a wide range of expressive vocalizations, while others are taciturn. Curiously, even non-dog owners can distinguish between the bark of a lonely dog, and that of a dog alerting to a stranger (5). This ability to understand barks seems to be built into humans. However, we aren't as good at understanding canine body language as we are at interpreting their barks, and children often misinterpret canine signals (6).

Dogs tend to be much better at understanding our body language than we are at understanding theirs. They look at us for cues, like pointing, or the direction we're looking in. Dogs from any breed can do this, in spite of the variations between breeds in levels of ability. Wolves, however, are less likely to look at humans than are dogs, especially when trying to solve a problem, a key difference between wolves and dogs (7). Dogs get around our deficiencies by doing whatever works. Some sit on their owners' laps when it's walk time. Others take their leashes to their owners, or stand by the door and look hopeful. It's worth keeping a diary to record which of your non-verbal signals your dog is picking up, and how your dog tries to communicate with you.

In addition to understanding humans, dogs also learn to communicate with one another. This isn't always easy. We've created breeds of dogs which often communicate in different ways. Spitz dogs tend to be expressive. Their body language is easy to 'read', while other breeds may be much less expressive. There's potential for misunderstanding when a dog first meets another with a different 'body dialect'. Yet most dogs which have the chance to meet a wide range of breeds come to understand them well. In this potential to accept strangers, even those who are different from us, dogs are more like humans than wolves, which tend to be intolerant of strange wolves on their territory (8).


Tilly and Conor taught Toby how to sit-stay.


Dogs can also learn social rules, both human social rules (like 'don't jump on strangers') and canine social rules (like 'don't jump on bad-tempered arthritic dogs'). The process of learning social rules is called 'socialization', and as Miklosi remarks, dogs are unique as a species in going through a double socialization process, 'A puppy is expected to learn the rules of social life of dogs, as well as many of those of the human community' (9). A fully socialized dog is socially competent in both human and canine societies.

Socialization: the key to developing canine social intelligence

Socialization is critical in allowing dogs to develop their social intelligence. It's one of those words we all think we understand, but which we often misuse. Partly this is because ethologists don’t always use the word to mean the same thing. Sometimes they mean simple exposure (10), other times exposure plus interaction during the 'sensitive period', when a pup is approximately three to 12 weeks old (varying from one dog to another) and is particularly open to new experiences. Now, with wild species, youngsters stay within their social group, so exposure plus interaction does mean learning social rules from socially competent adult conspecifics. However, pups are more likely to meet incompetent teachers, like kids who encourage them to bite, or owners who let large-breed pups bully smaller pups. Here, pups aren't being socialized, but instead are learning bad habits. It's a sign of the social intelligence of dogs that they can learn new social rules as juveniles, but it does handicap a dog to be taught bad habits as a pup. Sometimes people refer to 'oversocialization' meaning 'inappropriate interactions'. This confusion stems from seeing 'socialization' simply in terms of interactions, forgetting that interactions have to be appropriate if pups are to learn social rules.

Most species other than dogs don't interact intensively with a wide variety of humans and conspecifics. Dogs face a difficult challenge. To be fully socialized, they have to learn to be socially competent with a wide variety of individuals from both species, so pups need good teachers. The best teachers are socially skilled adult humans and dogs. Some dogs are particularly talented teachers. They may turn away from pups indulging in rough play. Feisty small dogs are often very good at scolding much bigger large-breed pups.

Though expert advice on socialization tends to focus on the 'sensitive period', clearly a pup isn't fully socialized at 14 weeks. Dogs need to carry on learning as juveniles, to become fully socialized adults, and if they spend much of their juvenile period alone, this tends to stunt their social intelligence. Expert advice also stresses the need for contact with a variety of humans. Scott and Fuller showed that dogs are resilient, in that they can later develop bonds with different humans despite having only minimal contact with just one human during the sensitive period, even without handling (11). Dogs have a strong, innate, desire to do things with humans, so can build on even very minimal contact. However, pups do benefit from intensive appropriate contact with humans at levels they can cope with, during this period.

So, much more happens during the sensitive period than bonding. Pups are learning how to communicate, for example signals that mean 'I want to play'. They're also learning social rules, such as that it's OK for a human to stare at them, and they can safely stare back. They learn about their physical environment too, for example, which sounds mean danger, and which sounds, like washing machines, can safely be ignored. These experiences help them cope with new situations. Pups can even begin an apprenticeship in high-level skills during this period, for example pups watching their mother doing police scent dog work later learnt the task more easily (12).

Learning to co-operate

Communication and learning rules are important for co-operation. Both humans and dogs have the potential to co-operate with conspecifics, and with one another. We're also capable of getting involved in serious conflict when things go wrong. Conflict can be costly, both for the individual, and for the social group, so we put a lot of effort into co-operation and avoiding conflict. One way we learn how to co-operate is through imitation, small girls, for example, sometimes pretend to be teachers of their dolls. Dogs also like to imitate humans (13). Individuals from both species also tend to enjoy play, an activity that involves participants signaling 'I don't mean any harm'. Play can help to develop mutual trust, better communication, and a deeper understanding of social rules (14). Some dogs are especially skilled at moderating their play styles to suit their partner. Socially competent dogs make concessions to physically weaker partners, because they want the game to go on.

Pups that don't have suitable teachers can act like bullies. Mickey, a mixed breed littermate of my older dogs was retained by his owner. At six months, he endlessly chased his smaller mother and sister round the yard. They just wanted to get away from him. Yet when he came to visit me, and met his uncle, older and much bigger than he was, he was very deferential. There was no aggressive posturing on either side. Uncle Rugby showed friendly curiosity, and Mickey simply accepted his uncle's dominance. The ability of dogs to size up a social situation and adapt rapidly is a sign of their social intelligence.

Mickey was a brat, a poorly socialized youngster, rather than 'dominant'. The word 'dominance' is a neutral term in ethology, though many non-ethologists confuse it with imposing one's will through initiating conflict (15), so see it as a 'bad thing'. Generally today it's accepted that the wannabe tends to initiate conflict, not the dominant animal (16). The model that early writers used was based on groups of captive wolves. Later writers, like Packard (17) found that wild wolves, which are usually related to one another, are much friendlier to others within the group than are unrelated captive wolves. This makes sense if you think of the behavior of humans in groups where people have a common aim, and contrast that with the behavior of prisoners confined in a high security prison. 'Dominance' has a different meaning in each social setting. Going beyond crude 'dog as wolf' models can help us realize just how smart our dogs are, and how to give them more opportunities to use their intelligence.

Humans are very adaptable, and can operate in social situations with a clear externally imposed hierarchy, for example, a university, and in situations where it is less clear, like a group of friends. We also vary in terms of our personalities. Some people are bolder and more confident, or more easy-going than others. Furthermore, we vary in terms of skills and knowledge. Hierarchies may emerge without being externally imposed, simply because one person stands out as bolder, and inspiring more confidence. Groups of dogs living together also vary. Sometimes they form clear hierarchies, other times they don't. Their ability to live peacefully in groups also varies. Scott and Fuller noted that the Wire Haired Fox Terriers they studied weren’t well equipped for group living (18). It's a talent of dogs, that, like children, they can decipher the rank of adult humans. Csanyi, head of an academic department in Hungary, noted that colleagues´ dogs deferred to him because he was the boss of their owner (19). Humans, however, aren't as skilled at understanding the subtle signals that denote the social structure of a group of dogs, as is shown by the work of early writers!

Curiously, dogs that are dominant over others in multi-dog households aren't as good at learning a task demonstrated by a canine experimenter, as are subordinate dogs, though both perform equally well when the task is demonstrated by humans (20). Maybe humans are often too obsessed with, and distracted by, language to 'read' dogs. Certainly, people with aphasia tend to be better at reading body language (21). Perhaps dogs are smarter than us at understanding body language because their survival depends on it, while our dominant position, as controllers of the key resources that dogs want (like food and walks), can foster laziness. We too often focus on telling dogs what we want of them, rather than listening to them. When our lives depend on our dogs, for example guide dogs, we have more of an incentive to pick up on dogs' messages.

Whether or not hierarchies are a 'good thing' depends on several factors, like the personality of the dominant animal, the availability of resources like food, or space, and how and why the group was set up. Wolves tend to live in family groups with one breeding pair. Wolf parents provide food for youngsters, and are affectionate with them, as well as enforcing discipline. Dominant social animals in the wild, whether parents or not, tend to want to keep the group together as far as possible, which they can't do by physical force alone. So, owners bringing up dogs are in a similar position to parent wolves. However, much of the early advice that owners should act like 'alpha wolves' is suspect, partly because we've come to understand wolf behavior better, and also because dogs are smart enough to know we are humans, though they may interpret some of what we do in dog ways. A key difference between dogs and wolves is that dogs want to understand us, which helps when, like parent wolves, we teach impulse-control and social rules.

Some rules are critical for co-operative living in multi-dog households. They include ensuring that everyone gets to eat. Toby, my ex-hunting mutt, was emaciated and scarred from having to fight to eat. He'd been kept with bigger dogs, and was abandoned for fighting - the only way he could get to eat. He became more relaxed under a rule of 'everyone eats their own dinner, and no-one messes with anyone else's'. If I was absent-minded, my other two dogs, Tilly and Conor, would walk up to Toby's bowl, and he’d back off, so I learnt to pay attention. All went well until I went away for 10 days. Toby was thinner, snappy, and snarly when I got back. He also no longer deferred to the other two dogs. I'd forgotten tell the dogs' carer that mealtimes had to be supervised, so Toby had reverted to fighting to eat. What's remarkable is his versatility and adaptability. He quickly learnt new rules, and quickly reverted when the rules changed. Today he eats peacefully with the other two dogs, who no longer try to muscle in on his dinner.

Why did Toby initially defer to Tilly and Conor, when he was better at fighting? Perhaps because they understood the rules of the household. When Toby first saw me eat, he tried to scoff my dinner. Under the rules he knew, food put down was available for the first dog to eat it. I explained that this was unacceptable by using a low power water pistol. He jumped off the table, and checked out Tilly and Conor, sitting patiently. He joined them, and the three were rewarded at the end of my meal. Tilly and Conor also taught Toby to sit and stay. He checked out what they were doing, and imitated them.

From wolves and ‘furkids’ to partners and friends

So humans enforce rules, because it's in the interests of the dog-human group (eg 'no fighting'), or in the interests of the dog (eg 'don't run across a busy road'). Yet it's a mistake to extrapolate this 'caring discipline' into a 'dog as child' model. Crude models of 'dog as wolf' and 'dog as child' both fail to perceive how smart dogs are, and the difference between a youngster, and a mature, well-mannered dog encouraged to develop its skills. Dogs can perform feats that no human can do, like warning us of dangers which we can’t sense, or finding objects we can't locate. Blind people and their dogs are partners from different species who have developed high levels of communication and trust. Humans pass some leadership decisions (such as where to go) on to dogs, recognizing that their canine partners have superior skills in some areas. The dog is aware that the human doesn't perceive everything, and has the desire and ability to share its knowledge.

Csanyi notes that humans and dogs are two species capable of alternating leadership (23). Alternating leadership can increase 'group intelligence'. A group is better able to take intelligent decisions when the most skilled and knowledgeable member can take the lead. Well-mannered, mature dogs willing to help their owners can become valuable partners, albeit junior partners, and when they have mutual respect and enjoy one another's company, friendship between human and dog is also possible.

Service dogs have to be polite with both human and canine strangers. Generally, pet dog owners want the same of their dogs. Wolves in packs tend to be hostile to outsiders, but are quite versatile. Packs may take in strangers, especially if an adult wolf is killed by human hunters (24). Dogs are potentially well-equipped to learn to get on with strangers. They take their cues from us, through our body language including tone of voice, so a friendly 'good day' to a human stranger tells a dog 'this person is OK'. Dogs can help us achieve better manners!

We can learn a lot from dogs' social intelligence. In many ways they’re smarter than we are. Their social intelligence has helped them to be remarkably successful as a species. But the question 'what can a socially competent dog do?' has many different answers, since dogs' social environments and inbuilt potential skills of dogs vary so much. It is these differences between dogs that I will explore in part two.
Alison Lever, 2014


Thank you to Greta Kaplan, Heather Houlahan, Tiffani Howell, Janeen McMurtrie, and Melissa Starling for interesting insights into this topic, and to Tiffani and Janeen for comments on earlier drafts of this article.

1) Scott, J.P. and Fuller, J.L. (1965) Genetics and the Social Behavior of the Dog. University of Chicago Press, Chicago
2) Klinghammer, E. and Goodman, P.A. (1987) Socialization and management of wolves in captivity. In Frank, H., ed. Man and Wolf. Junk Publishers, Dordrecht.
3) Scott and Fuller, op. cit.
4) Pilleya, J. Reidb, A.K. (2011) Border collie comprehends names as verbal referents. Behavioral Processes, 86, (2) pp184-195
5) Pongracz, P. Miklosi, A. Molnar, Cs., and Csanyi, V. (2005) Human listeners are able to classify dog barks recorded in different situations. Journal of Comparative Psychology, 119, 136-144.
6) Meints, K., Racca, A. and Hickey, N. (2010) Child-dog misunderstandings: children misinterpret dogs' facial expressions. in Proceedings of the 2nd Canine Science Forum, Vienna, Austria, p99
7) Miklosi,A., Kubinyi,E., Topal,J., Gacsi,M., Viranyi,Zs., and Csanyi,V. (2003) A simple reason for a big difference: wolves do not look back on humans, but dogs do. Current Biology, 13, 763-766.
8) Miklosi, A. (2007) Dog: Behaviour, Evolution and Cognition. Oxford University Press, Oxford p82
9) Miklosi, op cit p22
10) eg ibid., p.87
11) Scott and Fuller, op.cit.
12) Slabbert, J.M, and Rasa, O.A.E., (1997) Observational learning of an acquired maternal behaviour pattern by working dog pups: an alternative training method? Applies Animal Behaviour, 53, 309-316
13) Topal, J., Byrne, R.W., Miklosi, A. et al. (2006) Reproducing human actions and action sequences: Do as I do!' in a dog. Animal Cognition, 9, 355-367
14) eg, Horowitz, A. (2009) Attention to attention in domestic dog (Canis familiaris) play. Animal Cognition, Jan: 12(1) pp107-118.
15) eg, Fogle, B. (1992) The Dog's Mind. Pelham, London
16) McGreevy, P., Starling, M., Branson., N.J., Cobb, C.M., and Calnon., D. An overview of the dog-human dyad and ethograms within it. Journal of Veterinary Behavior (2012) 7, pp103-117
17) Packard, J.M. Wolf behavior: Reproductive, social and intelligent. In Mech, D. and Boitani, L, eds. Wolves: behaviour, ecology and conservation., pp 35-65. University of Chicago Press, Chicago.
18) Scott and Fuller op. cit.
19) Csanyi, V. (2005) If Dogs Could Talk. North Point Press, New York
20) Pongracz, P, Vida, V., Bankegi, P., and Miklosi, A. (2008) How does dominance rank status affect individual and social learning performance in the dog (Canis familiaris)? Animal Cognition, Jan 11(1): pp75-82
21) eg Sacks, O (1985) The Man Who Mistook His Wife for a Hat. Picador, London
22) Csanyi, op.cit
23) Packard, op.cit.