Do Dogs Form “Real” Friendships?

I had an entirely different blog written and about to be posted, but there’s a swirl of discussion going on right now about an article that came out in Time Magazine by Carl Zimmer about “friendships” in animals. He has lots of good information from researchers who argue that true friendships are formed in many social species, including horses, dolphins, and baboons. I was a tad irritated at suggestions that “we” (scientists) haven’t accepted that friendships can be found in other animals until just recently…” look at the writings of Jane Goodall and Frans de Waal for example for exceptions to that…  but in general it’s a truly good article.

But imagine my surprise when he writes that evidence of true friendships can not be found in dogs.  He says: “.. most scientists think [they…dogs] fall short of true friendship….. noting a lack of evidence in dogs of constancy, reciprocity and mutual defense…” found in other species. In other words, dogs can’t form true friendships with us, or other dogs. It is true that dogs are less studied than many other species regarding their social relationships and that we can’t use anecdotal observations as substitutes for good data. But the article doesn’t say that there is not enough evidence to make conclusions about the social relationships of domestic dogs. It says “…most scientists think they [dogs] fall short of true friendship.” It also says “… dogs have become capable of being sweet and loyal to humans, but it’s likely that they treat us more as guardians than friends.” It is one thing to say that we don’t yet have the evidence, and yet another to make conclusions based on a lack of data-driven research.

If you want to read more about, this, go to my Facebook page. Carl has been good enough to join the conversation, which I greatly appreciate.  My argument is that, if as the article states, true friendship requires “constancy, reciprocity and mutual defense,” then at least observationally we have ample suggestions that social relationships between some dogs are as strong as social relationships between some horses. I suggested on FB that the reason many scientists are hesitant to attribute “friendship” to dogs is not just the lack of data, but rather a scientific bias against domestic, familiar animals. “Familiarity breeds contempt” is as true in science as it is elsewhere.

As someone who has observed and worked with domestic dogs for over twenty three years, it is hard for me to imagine arguing that they can’t form social relationships analogous to friendships in our and other species. Lassie appeared to fall in love with Luke the day she met him, followed him everywhere, groomed him daily, paid attention to no other dogs and sank into what looked like a depressions when he died. That’s not data, but neither does it suggest that dogs aren’t capable of forming true friendships. I find myself somewhat amazed that we are even arguing this point. Of course we need more data, those of you who know me know I’ve been encouraging good research on canine behavior for 23 years, but concluding without data that dogs probably can’t form “real friendships?”  Given that there is no data behind that either, I don’t see it as a supportable or reasonable conclusion. Especially given our observations, which may be anecdotal, but still are valuable (as were Jane Goodall’s observations of chimps.) If the article had said: “Although it appears that dogs form strong friendships, we need the same kind of data collected on them as we have of other species,” I’d have no problem with it.

Jump onto Facebook if you want to read Carl’s and others comments. I expect he is busy answering no small amount of email and comments on the subject!

MEANWHILE, back on the farm. WILLIE IS WORKING SHEEP!!! Be still my heart, that’s about all I have to say. Granted, the sessions are short, painfully so, but he looks sound so far and he’s working as if he’d never had a year long break (it was a year ago last week he was injured) and it’s so much fun for both of us that we can hardly stand it. When I say “That’ll do” he runs to me, spins in happy circles and I clap my hands and we grin at each other and somehow my heart gets bigger in my chest and we float back to the house and life is good. I’ll never be able to work him for very long, and I’m sure we’ll have set backs, but just being able to work a little bit is more wonderful than I can say.

No work this weekend though, we’re off to the Big Apple for the Dog Writer’s of Association of America annual awards. Love Has No Age Limit is up for an award, as is a column that co-author Karen London and I wrote for the APDT Chronicle.  FB readers have suggested I not wear jeans and my usual plaid shirts from LL Bean or Land’s End. Okay, I promise I’ll get the straw out of my hair, but I’m not wearing black. I just can’t understand why all black is so chic when it’s the “color” that oppressed women are forced to wear all over the world. I’m wearing orange, and NYC will just have to deal with it. The banquet is Sunday night, Saturday we’re going to see the play Memphis on Broadway. All very fun and exciting… but when do I get to work Willie on sheep again?

And apologies for no new photo today: I’ve spent all my time on the first blog (next week!) and the article about friendship and now have to pack, clean the house for the sitter, write out my lengthy set of instructions, etc etc.  Just too much to do today! I’ll make up for it next week!




  1. says

    I just recently witnessed how a friendship between dogs can form. A dog, newly adopted from a shelter, previous street dog from Romania , meets my dog Rani and insists of becoming my dog

  2. Susan G. says

    I saw your FB post and enjoyed the comments. My dog can be reactive with unknown dogs, but he whines, wags his tail and wiggles for his friends. He’s a completely different guy. Even dogs he befriended once upon a time are remembered and greeted with affection even in their rarest of encounters. It is so obvious to me he enjoys and looks forward to seeing his friends. He just waits and watches sometimes in the hopes a friend will appear (and reacts much differently if it’s an unknown or unfriend). It’s so strange the things people think, or don’t think, about dogs (and other animals) when so much obvious evidence is right in front of many of us. Congratulations on the latest recognition of your valuable work!

  3. Liz says

    The cover of this issue of Time is two dogs, and it strikes me as an interesting choice for an article where their friendship-making abilities are challenged. I suppose the image could have been to further challenge our beliefs, but …when speaking about her cover image, the photographer states that she “had to make sure the dogs that were coming were actually friends.”

    And as to how widely held this belief of dog friendships might be, they are commonly accepted enough that in Time Specials, they feature an elephant and dog pair in The Adorable Dozen: Bizarre Animal Friendships. They refer to the pair as “best friends” in the title of the video (when viewing video #1, the title for video #2 appears as “Elephant and Dog, best friends”) and mention it several times in the video itself.,28804,2066018_2066006_2066037,00.html

    I haven’t read Zimmer’s article, but if dog friendships are more or less a myth, then it seems more scientific studies are necessary to straighten the subject out for everyone.

  4. Carole L. says

    I think you’re right about familiarity breeding contempt. We tend to be arrogant enough to think we know everything about the animals we’re closest to, yet we ignore all the anecdotal evidence we see in front of us. With researchers like you and Frans de Waal and many others on the case, I’m certain the data will catch up with what we all know to be true.

    So glad to hear about Willie getting back to work! I’ve just recently been letting my young aussie-doodle, Maisie, off leash again after 3 months of restrictions as she healed from a broken leg. Your posts about Willie’s recovery gave me both encouragement and ideas for keeping her entertained and challenged throughout her healing process. Her happiness at running and chasing her ball and playing with her pals is clear and infectious, and makes all the tough times worthwhile. Can only imagine how joyous you and Willie feel after such a long hard road! Thanks for sharing!

  5. Beth with the Corgis says

    Hmm. I think I need to think about this one.

    I will say this: horses were my first love and are my true passion, though I can no longer ride due to health concerns. I love dogs, cats, and all animals, but horses are my true love.

    I know dogs like other dogs and form attachments. Dogs definitely form attachments to people. But do dogs have dog-friends the way horses have friends? That’s what I need to think about. I’m thinking of all the dog pairs I’ve known, and while some are friends I know of many, MANY houses where two dogs live together and more or less co-exist. They might play and show occasional affection. But not in the way horses do. Horses almost always pair up. Horses need a companion, to the point that horses that are not allowed pasture time with other horses are often paired with a goat or at least a cat.

    Horses will get so “barn-sour” that they refuse to leave the area of their buddies, if you don’t carefully guard against it. How many dogs do you know that won’t go for a walk without their dog-friend? Or their human, for that matter (though we no longer give them that choice).

    Horses are so inclined to form pair-bonds that if you want to introduce a new horse to a group, the typical thing is to stable them next to one “nanny” horse and then turn out the two together to bond, before introducing the now-pair to the herd.

    Horses spend hours in mutual grooming. There have been many cases of horses going blind and the owner didn’t notice, in part because the buddy horse would lead him around the pasture.

    My dogs have other dogs we call “friends.” They do seem to enjoy some dogs’ company more than others. They are definitely capable of forming attachments. But I guess in comparing them to what I know of horse friendships, and chimp friendships, MOST dogs come up short. You do hear stories of two house-mate dogs loving each other to pieces long past the “we are young together and we both have energy so we play constantly” stage. But I think those stories tend to strike a chord BECAUSE they are relatively rare.

    With horses, friendships are so expected that careful owners go to great lengths to prevent them from getting too close, lest the horse become unmanageable in his desire to stay near his pal. It’s the rule, rather than the exception. As it is with people— put a group of people together and they will rapidly break off into small groups or pairs of friends. I’m not so sure it’s quite that way with dogs, though I do see dogs gravitate towards other dogs with similar play styles.

    So while I would tend to say dogs can form friendships, I would say there are many individual dogs (probably the majority) who never do, in the true sense of the word. But for horses and chimps and people, not being able to make friends (and do so quickly) is seen as a sign of an underlying problem. I suppose I’d like to see more research.

  6. Beth with the Corgis says

    I also wanted to add this: the other animals you listed above all form long-lasting bands that may stay together for life. The mares in a band will stay together forever unless forcefully stolen by a stallion, and they will usually try repeatedly to break away back to their home band for some time before giving up. Dolphins form long-lasting groups. So do many apes.

    But the ancestral wolf forms a pair bond, then raises pups which disperse upon reaching sexual maturity + achieving adequate hunting proficiency. Those young adults then find a partner and set up their own pack, which is again a pair-bond and pups.

    And there is some belief I think that the early dog was more of a loner/scavenger; certainly many male domestic dogs don’t show the kind of food-bringing behaviors that male wolves do.

    Point being that there was not much evolutionary benefit to forming friendships outside of pair bonds. I can’t think of any wild canids who are known for their friend-forming: coyotes will pack up loosely or hunt singly and form pair-bonds at breeding time. Foxes form pair-bonds at breeding time. Perhaps some of the wild dogs form looser packs that are more inclined to stay together.

    But I believe that pair-bond behaviors fall more under the “procreating” column than the “mutually beneficial friendship” column. And we certainly haven’t selectively bred dogs based on their ability to befriend other dogs.

    I imagine my view will be the minority, though.

  7. Rebecca says

    No comment about dogs and friendships, but I do have one about sheepherding. I have always heard that one of the reasons that sheep respond to dogs, especially border collies, is because the “eye” is supposed to remind them of predators and thus the flock falls back into basic defensive modes, which can be influenced by where the dog is in relation to the flock. And then I see this:, all about a _rabbit_ herding sheep. Surely sheep have seen rabbits, and know that they are fellow herbivores and not a threat? So why do they act like it’s a predator, going so far as to charge it in some cases, the way they do the dogs? Just curious, since I would think that spending this much energy avoiding a rabbit would be counter-productive in the wild.

  8. FJM says

    Good to hear Willie is able to do what he loves best again!

    I know the plural of anecdote is not data, but I have just seen too many examples of dogs forming friendships with dogs, humans and other animals to believe that there is no evidence! Of the three other dogs living close to us, two are polite acquaintances, one is a BFF – the dogs visit her every day, and when she is away they check each day until she comes back. And then let joy be unconfined! Our family poodle had a close friendship with my Siamese cat – the cat washed the dog, and would climb up to steal food, and drop bits down to the dog waiting below. She took better care of the dog than she ever did of her own kittens!

  9. says

    Well, as I posted in FB, we HAVE the research. It’s a Master’s thesis: Trisko, R. K. 2011. Dominance, Egalitarianism and Friendship at a Dog Daycare Facility.

    When my dog was young, we met a boxer pup. They played everyday in the park, with each other, ignoring all other dogs. We changed homes, and a few months later met with them in another park. It was awesome! I have never seen my dog so happy. Same with my parents housekeeper. She sees her once every two or three years, and still pees herself each time! She does that just with her, she loooves her.

  10. Susan Mann says

    Congratulations on your awards, and on Willie getting back to work! Arie is now also back to work (agility) after more than six months of hobbles and other restrictions.

    The idea of other animals being capable of forming friendships, and dogs not being able to do so, is laughable. The first thought I had when reading your FB post this (alright, yesterday!) morning was your description of your relationship with Luke, and how he saved you from an angry sheep, and how pissed off he’d sometimes get at you (aren’t we more likely to get annoyed or angry with our friends than with strangers?) So how can an animal have an inter-species friendship if the capability for an intra-species friendship isn’t already present?

  11. trisha says

    Thanks for the research reference, will look it up when I get back. And Beth your well thought out argument is well taken, but we can get into a lot of trouble forgetting that dogs aren’t wolves, certainly not behaviorally anymore. In addition, there are many references from credible scientists of strong differences in relationships between individual wolves. Many of these, granted, are between breeding pairs, but some pairs appear to adore each other, others not so much. Some personalities match up, others don’t. Would write more but would miss my plane!

  12. Beth with the Corgis says

    Thanks for your response Trisha. I realize dogs are not wolves, which is why I mentioned other canids. And the point about evolution is valid: dogs are not evolved to spend many years with a constant set of companions the way apes, horses, elephants, and dolphins are. To throw in some more ideas, it is generally considered cruel to keep a horse (or elephant, or dolphin) without the companionship of its own kind, as it is so necessary for their mental well-being. I don’t think the same can be said of dogs. Many dogs seem to enjoy dog company, many others seem to prefer living alone or can take or leave the company of other dogs.

    I think it is hard-pressed to define a “preferred play partner” as a friend. The definition listed above— reciprocity, constancy, and mutual defense— is really not captured by “I like to play, and your play style is similar to mine, so we will play together.” Two-year-old children (hardly capable of developing true friendships) have preferred play partners. Friendship can not take hold til sense of self/other is more well-developed.

    So let’s look at the three:

    Constancy: I think there is the most evidence of all for this aspect in dogs. It is clear that most dogs like the familiar and will prefer to spend time in the same routine, with the same larger group, given the chance- at least once they hit adulthood.

    Mutual defense— this one is tricky. Dogs, as social animals, will often guard against any threat to their territory or their social group. It’s one of the reasons we domesticated dogs, after all, so it’s impossible to argue they won’t defend. But how is “mutual defense” defined? I need to read the article so I may need to stop by the bookstore. I get the vague idea that it involves protecting your friend WITHIN the social group, because I know that chimps form alliances where a group will all watch each other’s back. One of the reasons to buddy up a horse before introducing it to a herd is so that if the rest of the herd bullies it, it has a friend to help keep it safe. Monkeys and dolphins and elephants will act as “nannies” and nurses for their friends when they are in labor or caring for young.

    Does the territorial behavior of dogs rise to this. It MAY but it’s not certain. Trisha gives a great example in one of her books about Luke coming to her aid when she was trapped by an angry sheep. My female Corgi is more inclined to run off if she’s scared. Jack will run towards us and jump and bark if we seem in distress. Seemingly a clear sign of “mutual defense.” But wait! I have seen him, at the dog park, run up to a strange dog (one he’d not yet met, not even once) when it yipped loudly when another dog got too rough. I’ve seen him run up to strange humans who make sounds of distress. Is it friendship if the behavior is generalized to “anyone in distress.” I would be hard-pressed to say it is. Low tolerance for distress and trying to stop that is not the same as giving aid to one or two favored few out of a sense of friendship. Again, anecdotes are not data, but it would be interesting to see an objective study if dogs are more inclined to help their friends than their acquaintances or strangers. Will a dog prone to guarding help guard any person who feeds them (resource-protecting)? Or will they not care one way or another about one person who feeds them, but guard another out of mutual friendship.

    Reciprocity: Here is where dogs fall very short. How many dogs, without being trained to do so, will voluntarily bring us food? A blanket or pillow? Surely they must know that we need to eat too, right? The excerpt I read from the article you refer to included adult male apes of some sort voluntarily sharing food. How many dogs will offer to share food with their dog-friends, when it is not a part of pup-rearing or courting behavior? I’m trying to imagine two adult, intact male dogs being given a great treat and voluntarily saving some of it for their pal. Remember, courting behavior is not the same as platonic friendship; MANY animals exhibit all three behaviors when courting. Dogs don’t volunteer to help us or their other dog friends (remember, the examples of those who do make the internet because they seem so rare and hard for us to imagine). Dogs give us a lot– companionship, protection— but mostly they are doing those things for themselves and we are the innocent recipients. It feels good to pet a dog and look at his furry face, but he comes to us to be petted because he likes it, not because we do. A cat will show affection for its human by trying very hard to groom it back (biting, licking, kneading). Dogs rarely try to pet us back. Some will lick us but that is a whole other topic. It’s not grooming.

    Much as I hate to say it, on reflection I would characterize the majority of dog-human interactions as “symbiotic” rather than “friendship” in the true sense of the term. They get benefits from us (food, shelter, attention, play) and we get benefit from them. They get attached to us, very much so, but affection is not the same things a friendship.

    Perhaps I have just spent so much time over decades reading about the behavior of the animals that do form friends (elephants, dolphins, apes). And honestly I see little of that behavior in dogs. I think it can happen, but again I think it’s more the exception than the rule and I still am staying with my argument that evolution did not put pressure on dogs to form friendships— they don’t live in that type of dynamic –so when it does crop up it’s due to individual personalities and not programming.

    To put it another way: stable a horse alone for a few months. Stable a second horse alone for a few months. Put the two horses together and they are almost guaranteed to be friends in a heartbeat, the desire for platonic attachment to their own kind is so strong. Can we say the same about dogs?

  13. Kerry M. says

    This conversation is definitely asking me to challenge my assumptions. I have never doubted that dogs have doggy friends. I had two different pairs of dogs that essentially lived together their whole lives growing up as puppies/adolescents a year apart or so and then living together until old age. I would say that they had a real connection, but did they have constancy, reciprocity, and mutual defense? Hmm. Not sure. They did play together for many years and it was very cute. I have had minimal interactions with horses and have been comparing my own experiences with what Beth wrote up and I do see a difference. I have never had a dog turn down a walk because only one was going. Of course, the dog who gets left behind has a very bad day. Me going out by myself is one thing. Me going out with the other dog… tragedy of the greatest order. I suspect the friendship is there between dogs but it’s getting lost in observation because the friendship between dogs and humans is just so much stronger.

    Which brings me to my main point, anecdotal or not, I think it’s kind of ridiculous to say that there aren’t strong human/canine friendships. A clear case of not seeing the forest for the trees. You can observe this in almost any house with a dog where we feed and care for them and they curl up beside us on the couch. Or by looking at behavioral problems, how many dogs have separation anxiety at the thought of being separated from their family? Or what about working dogs where they will put themselves in harms way to help a blind person cross a street or find a missing person who may be under dangerous rubble. It’s just… well, I’m going back to saying it’s just ridiculous to say that this doesn’t count.

  14. em says

    I admit it. My initial reaction upon hearing the suggestion that dogs don’t form friendships was, “Oh, how ridiculous! Somebody’s been hanging out with the wrong kind of dogs!”

    I figured that this was the same kind of pseudo-scientific assertion that results from the confusion of technical terms with colloquial usage- the same kind of thinking that insists that because domestic cats are not social animals in a technical sense (no cooperative hunting, social hierarchy, etc.), all domestic cats are aloof, solitary, anti-social animals by nature, even despite the contrary evidence of the millions of households containing affectionate, attention-seeking, dead-mouse delivering lap-warmers.

    On a more serious note, I will say that while my personal experience can be heaped onto the pile of anecdata supporting true peer-to-peer, nothing in it for them dog friendships (with each other, and sometimes with people or other animals), I suppose I HAVE known dogs that probably don’t have friends. But then, I know people with no friends, too, but I doubt anyone would say that that’s an indictment of the whole human species. I suspect that the very strong bond between humans and dogs, combined with the relative isolation from one another that many pet dogs live in, likely does interfere somewhat with their ability to form friendships with one another.

    I’ve also casually observed that dogs who are very strongly focused on people often have the hardest time forming friendships with other dogs. I’m not sure whether relationships between people and dogs count as friendships because they are not peer-relationships (though if anyone has never met a dog who demonstrates constancy, reciprocity and mutual defense toward their human guardians, I’m very sorry to hear it). Certainly some dogs forge friendships more readily than others, but Otis has relationships that I can only characterize as friendships with dogs. He looks for them eagerly in places he expects them to be, greets them joyfully, happily passes his toys back and forth with them, cooperates with them to coordinate a chase, and will jump to their defense if they get into trouble. Constancy, reciprocity, and mutual defense? Check, check, and check.

  15. Rebecca Rice says

    The other thing that I think we need to be aware of when discussing this, especially anecdotally, is that there are huge differences in how people define “friends”. I have been in some heated net debates about how people should/do treat friends, and after much discussion, determined that at least part of the heat came from how people like me, who have few “friends”, versus people with many “friends”, defined the term. Because what they call friends, I would call “acquaintances”. People who are fun who hang out with, who I like to have in my life, but not who I would call to help me move a body. Friends, to me, have to be more than just fun to be with. They have to know all my deepest, darkest secrets, and still want to be around me afterwards. While other people in that discussion were describing the coworker that started last week as a friend. Which makes a huge difference in when you are discussing whether you will then “let a friend move in with you”, “tell a friend that their significant other is cheating”, etc.

    So, keeping in mind that there is a difference in how people describe relationships amongst humans, it is going to be hard to determine from anecdotal evidence whether dogs have friends, because two people could look at the same behavior, and one could say “friends”, the other “acquaintances”. Neither would be wrong, according to their world view, but it would make for completely different interpretations.

  16. mungobrick says

    I can understand that dogs who live together may not be friends – it’s like your siblings, it’s not like you had any choice about living with them! My dog is very dog-savvy, and as she gets older she plays with fewer dogs, although she is always friendly. But the boxer she has played with since she was a year old – he is definitely her friend. She won’t go on the walks we go on with him until he arrives. She just sits and waits for his car. On his part, he recognizes our car from inside his car and cranes to watch it anxiously until we or they arrive. Despite months of long separations when he had a pacemaker put in and she recovered from a a breakdown from severe anxiety, the joy of reunion and the pleasure they take in each other’s company is palpable on both their parts. I cannot see what they have as anything other than friendship.

    Why do scientists insist that animals can’t have emotions until they’re dragged kicking and screaming to accept the notion?

  17. frans de waal says

    It would be interesting to hear what Barbara Smuts (U Michigan) has to say. Although not mentioned in Carl Zimmer’s article, she was the first to publish a book (mid 1980s) entitled “Sex and Friendship in Baboons.” There was enormous resistance to this terminology at the time, but now it is wholly accepted. Barbara is a great dog lover and I bet she’d say dogs have friendships. I even remember her having written about it.

  18. Margaret McLaughlin says

    Query–How much does the amount of dog/human or horse/human interaction time affect the need both animals need for friends of their own species? Many dogs are with their people much of the time, even if they are not actively interacting, & it’s a rare horse who gets to sleep on her human’s bed. I would love to see some studies of domestic dogs & their relationships with each other if they live in a group & spend less time with humans; foxhound packs & sled dogs might be possibilities.

  19. Larry C. says

    @Bonnie, Jeanne Robertson and Fannie Flagg are responsible for my fascination with southern beauty queens.

    As for canine friendship, dogs are social animals that form a remarkable variety of relationships with people, dogs and other animals. I think the whole discussion uses loaded terms that anthropomorphize dogs. Dogs are not people. Trying to paste words for human emotions onto a dog is a mistake. I just take them at face value.

    One of my dogs likes to visit the neighbor’s horse, and they play. The horse will chase the dog, then the dog will chase the horse. Are they friends? I don’t know, but they have a good time and exercise each other, which they both need. I resist the temptation to read any more into it. They have worked out a game that they enjoy.

  20. Beth with the Corgis says

    Em, I think that is interesting with Otis. I have seen very few dogs (off-hand I can’t think of any) who happily share toys. I have seen lots of dogs say “You have it and I want it.” I have seen dogs tease other dogs into playing “catch me if you can.” But I have not personally known dogs who will happily offer a much–loved toy to another dog, AND let them keep it for any length of time— part of the literature with chimp friends involves willingly handing over food, and I was trying to think of something similar in dogs and came up empty.

    So again, the reciprocity is one area where I have seen little evidence in dogs who are not courting/rearing pups. I would be interested in hearing others. Of everyone’s anecdotes, yours is I believe the only one that directly gives an example of what I consider reciprocity.

    As for the relationship with humans, again in any other animal or human interaction we would not describe one that is totally dependent on the other as a “friend.” Dogs are bred to be perpetual adolescents in their demeanor (compared to wild canids). If you look at their behavior objectively, if anything they treat us as parental figures. And thank goodness they do; I would not want something with those teeth and that bite strength seeing themselves as my equal….

  21. Beth with the Corgis says

    I forgot to add that I looked for the issue of Time and it’s off the newstands. I’m hoping to find it at the library. In the meantime, it might be helpful if someone who read it could give examples of relationships the author describes as friendships, and see if something similar exists between dogs.

    For instance, searching online I have found many articles and studies that refer to elephants, dolphins, and bats (so not only super-intelligent animals) routinely assisting other females in birthing. I’ve read of various apes voluntarily handing over food to some of their troop-mates (but not others). I have read of dolphins who will assist an injured dolphin over many hours by bumping it to the surface to breathe. I have heard of higher-ranking horses protecting a lower-ranked friend from being bullied by middle-ranking horses. These are all examples of altruism: animals helping other animals with no immediate benefit to themselves (as opposed to, say, cooperative hunting). And in the animals listed above, these actions are routine, the rule rather than the exception.

    Are there many good examples of similar behavior in dogs? I mean actual helping. I have heard and witnessed over the years many, many examples of mutual AFFECTION between dogs and other dogs, dogs and humans, dogs and cats, dogs and any number of animals. I have read of very few examples of dogs assisting other dogs in any of the routine aspects of life. For instance, when one dog is sick, do we have good examples of a dog “buddy” bringing food or a favorite toy or blanket? Do we have good examples of dogs helping other dogs give birth? Of willingly foregoing a much-desired treat to let a favorite friend have it?

    I am quite certain dogs are capable of some level of love and affection. But scientists are apparently defining friendship in terms of specific actions. Is there much anecdotal evidence of this?

    That is what I’d like to hear. I think (and it’s probably not a popular thought) that a dog protecting its primary food source (us) from assault would not fall under the category of altruism, since there is a clear and immediate benefit to the dog.

  22. Kat says

    So my first question is what happens to a dog that grows up in isolation, all physical needs met, toys to play with, everything necessary for mental and physical health except social contact. I’d guess that the evolutionary path taken by domestic dogs is such that dogs can and do form friendships across species lines much more easily than other animals. And I suspect that dog/human friendships are more common than we might predict.

    I consider my dog Ranger my friend and partner. Constancy, yep. We want to spend as much time as possible together looking out for each other and just being together. Reciprocity, I’d say so. He doesn’t give me his food but if I asked for a share he’d share it. If his crazy sister gets too wound up and starts being mouthy the instant I yelp he’s there to distract her and keep her away from me until she’s less wound up. When we’re walking through a slippery patch he’ll put tension on the leash in a way that aids my balance (something he figured out for himself not something I taught him) and of course I share my food with him, I come to his aid if he’s in difficulty and I look out for his best interests. Mutual defense, yes there is. We met someone on a walk one day whose whole demeanor screamed gang banger. I was a bit nervous and Ranger was immediately between this person and me never taking his eyes off of them until we were well past. Another time we met someone who was staring inappropriately at Ranger (a very creepy person) and I was in front of Ranger in a heartbeat.

    I’ve seen Ranger dig up one of his bones buried in the yard and give it to another dog and when his best canine pal moved away I watched him grieve for months. When another dog pal moved away he pretty much shrugged it off. The one dog was what I would describe as his friend and the other was a preferred playmate; the emotional attachment was different. I’ve also seen Ranger trick his friend out of the very desirable stick and once Ranger had control of it I watched him chew it in half moving a bit away with his half so his friend could have the other. I’m at a loss what else to call it if what I’ve seen from him isn’t friendship both to me and to select other dogs.

  23. Annie R says

    I just read Jennifer Arnold’s new book, (the founder of Canine Assistants in Atlanta) and it makes me wonder if there’s any literature on the bond between service dogs and their humans — in some cases, the dog is actually more wordly, capable, and a better communicator with the outside world than the human they serve. Yet they still bond to, and spend all their time and energy doing what their human needs, and don’t drop their role when other (more capable or sociable) humans fuss over them.
    It seems to me a great example of constancy, reciprocity, and even mutual defense if you count the dynamic of the pair moving through the world together and providing safe space for one another. Such as seizure dogs, dogs who help a Parkinson’s patient “unfreeze”, etc.
    If no one has done a study on this, maybe there’s one in there for a PhD student to make their mark with, eh?

  24. FJM says

    I’m very interested in what Beth with Corgis says – particularly as I have witnessed Reciprocity from a cat to a dog – and cats are not usually renowned for forming deep friendships with other cats, let alone other species. It would be interesting to know whether there is research that shows whether food sharing can be completely separated from mothering/nurturing behaviour – my (spayed) female cat definitely attempts to teach my dogs to hunt, using the whole range of behaviours that she would have employed had she ever had kittens. The dogs, on the other hand, would no more voluntarily share food with the cat than they would fly, although they have been taught to let her take her turn when playing games for treats. Do human children – or other baby apes – voluntarily share their sweets before they have been taught it is the “right” thing to do? Or learned that it means they get to share their friend’s sweets next time?

    But do horses display reciprocity? I know my pony’s life was hugely improved when I found him a companion, and they were very happy together, but I still had to ensure buckets of pony nuts were well separated to avoid arguments!

    I think there is probably more evidence of Constancy and Mutual Defence in dogs?

  25. Patricia McConnell says

    Great to see you too Fran’s. fantastic talk, as usual. I,m in NYC, (and learning to use my new iPad), but will contact Barbara S when I get back. but I wish ZImmer had spoken to you!

  26. Beth with the Corgis says

    I wanted to point out that the author of the article says his original source was this piece:

    I can only read the abstract. But some notes from the abstract:

    “In species where males disperse, friendships are more likely among females. If females disperse, friendships are more likely among males.”

    In dogs, and nearly all canids, both males and females disperse.

    “Male allies have superior competitive ability and improved reproductive success; females with the strongest, most enduring friendships experience less stress, higher infant survival, and live longer.”

    How does that relate to a species where a male/female pair bond, and sometimes their grown offspring, are responsible for rearing of young?

    “Not all friendships, however, depend on kinship; many are formed between unrelated individuals.”

    Wolves rarely live with unrelated individuals who are not mates. Dogs, throughout their history with us, have been asked to live in many situations, but large groups of unrelated individuals who are expected to routinely sort out their own interactions and help take care of each other was never part of the equation.

    I suppose I am a bit surprised, Trisha, that you seem to have glossed over the heart of the science. What scientists now call “friends” are animals where large social groups (not consisting of a primary male/female pair bond) live together for long periods. The group generally consists of related and unrelated females, young adult males, one or multiple breeding-age males, and does NOT have a mated pair at its core. The animals in question are all examples of ones in which the father is not directly responsible for raising the offspring. So instead of developing pair-bonds, these animals have a stronger tendency to develop same-sex, or platonic opposite-sex (in the case of dolphins specifically) relationships to help look out for each other, help rear young, take part in grooming routines, etc.

    Dogs come from a large family of animals where the reverse is true: social interactions are primarily centered around the nuclear family. Canids engage in mutual defense of territory, cooperative hunting, and mutual (usually) raising of young. We have taken great advantage of all these traits, to the benefit of both us and dogs.

    No where in that evolutionary tree is there a place for peer-to-peer friendships to develop. There is room for mated pair affection. There is room for parent/offspring cooperation, relatively extended child-rearing, and some ongoing interaction for a time between young adult offspring and their parents and their parent’s next litter.

    Honestly, being very upset that dogs are not held as shining examples of exhibiting a behavior that their evolution would not expect them to exhibit is sort of like being upset that a bat is better at echolocation than your pet parrot.

  27. says

    We demand a lot from our dogs. Our relationship is imbalanced because humans ‘own’ dogs, therefore I find cannot call, the unflinching dedication our dogs have towards us and us towards them, a friendship. I do strive for that though ie within the boundaries of urban life, respect care joy contentment understanding and acceptance of our differences/possibilities/abilities.
    I have witness undeniable friendships between dogs and cats and dogs and dogs where indeed there was protection, care, pure joy at being together etc.
    Dogs have all the qualities friendships demands. It’s the humans who need work.

    Wonderful news about Willie!!! Would love to see pictures.

  28. says

    Beth with the Corgis has made some excellent, thoughtful comments. Frans de Waal mentions Barbara Smuts in his comment above — here are some links.

    Does your dog need a BFF?

    In the Company of Dogs

    Play and the “50/50″ Rule

    In general, I’d say that Smuts’ articles tend to support the points made by Beth w/ Corgis.

    I suspect the Chilean “hero dog” was thinking of dinner. On the other hand, this dog/elephant relationship is special:

    For decades I’ve shared my home with multiple dogs: working border collies, rescued pit bulls, mixed breeds. I’ve seen many instances of affection and preferred play-partners, but none of two dogs forming a deep pair-bond.

    Another thought, for what it’s worth: I imagine that Barbara Smuts, in her articles, and others commenting are not talking about sexually intact dogs. What role does neutering play?

  29. Deanna in OR says

    Anecdotal, but two dogs we used to own, a Collie, Kashka, and a Beardie/Golden, Maile, met when they were 2 years old (both female). They were 5 days apart in age, and after dating for a year or so, my (now) husband and I would laugh and say, well, we are stuck with each other for at least 11 or 12 more years, because we couldn’t possibly split up Kashka and Maile. They were truly best friends, always playing together, mopey when the other wasn’t there on rare occasions, and when Maile died of a stroke at 13-1/2, Kashka definitely acted depressed. She died 5 months later–physically, of bloat, but I think really of a broken heart.

    Anecdotes aren’t data, but observations (like this, over many years), lead to the questions that lead to collecting data, if someone bothered to look for it.

    As far as judging human/dog relationships as not being friendships, due to power differences—most all human/human relationships have power differences! We rarely like to admit it, but many, if not most, human friendships have differences in power but are still a relationship we call friendship. Anyone who has had a “heart dog” would laugh at the idea that dogs and humans can’t be friends.

  30. FJM says

    I am finding this discussion fascinating – and coming round to thinking that my first reaction – “Of course dogs have friends!”- was based on a quite different definition of friendship. I am also thinking again about my own friends, and about the number for whom I would freely give up something I value as highly as my dogs value food … like my independence …

  31. says

    Beth w/the corgis,
    wolves have nuclear family groups, but they can get up to 40 wolves, so sometimes it’s not that “nuclear”. And wild dogs (feral, dingoes…) form groups with multiple breeding pairs, usually 3-4 pairs per group. The rest of the group usually is not related, and “hang around” each other. But they defend their territory, and each other.

  32. Patricia McConnell says

    Will respond when back from traveling, lots of excellent discussion here, but I will still argue that yes, dogs can form friendships.

  33. says

    It seems to me that we would have to define “real” friendship before we can test it. Is forming pair-bond for mutual protection or to raise children real friendship. Does friendship have to look like human friendship to be valid?

    But even though the concept seems nebulous, if we define friendship as looking after one another, affectionate interactions, wanting to be with one another, being loyal to each other, sharing, etc, we can find many examples of dog friendship. We had a fox terrier when I was a child. One day, upon leaving, my mother told the dog, “take care of the babies” referring to our kittens. We came home to find him sitting over a dead rabbit and watching the kittens eat it.

    And I too have seen dogs barely tolerate each other or want to be with each other and grieve for the one that dies first.

    But I suspect friendship looks very different between prey and non-prey animals.

  34. Lisa W says

    This is fascinating. Why would we think dogs don’t form friendships? Or why are some people resistant to the idea they do? That’s almost as interesting to me as the original question.

    I would say the articles linked by Luisa Do Not support Beth w/Corgis’ hypothesis. Here is an article Smuts wrote for the Journal of Consciousness Studies, first piece on baboons, second on her dog, Safi. In my (humble) opinion, all her work cited here has clearly upheld the position that dogs Do form friendships.

  35. em says

    Yep, Otis does something that I would objectively call “sharing”. He willingly gives his toys to other dogs in the interest of play and good social feelings. Interestingly, my first humorous response, that people who doubt dog friendship may have been spending time with the wrong sort of dogs, may have been more spot-on than I really thought.

    Individual personality clearly plays a role (I find it interesting that Kat with Ranger reports many of the same behaviors- like Ranger, I would describe Otis as a calm, confident dog on whom other dogs seem to rely for guidance and protection. He seems naturally suited to a leadership role, oddly enough because he is not at all bossy, snarky or grabby about resources.)

    But though Otis is my personal wonder-dog, I see others (different breeds, different levels of energy and confidence ) display similar behaviors every day. What they often have in common, however, is size. Until I owned a really big dog, I never observed or fully appreciated how giving and tolerant a dog can be with other dogs. This stands to reason, if balance and parity are seen as important to a friendly relationship. When dogs play with equals or with bigger dogs, they have no reason to pull their punches. They need to be assertive to be competitive in games and to gain access to food, toys, and attention. I’m sure that they are just as capable of reciprocity, but it’s not as easily visible in their actions because they are not in a position to be easily dominant. If I were to play golf with Tiger Woods, you had better believe I would not be giving him any mulligans.

    Otis easily could use his size and strength to dominate the other dogs and claim all the toys and food for himself. Yet day after day, in one interaction after another, I see him pulling his punches, slowing his running speed, letting the other dog catch him, have the toy, win. He doesn’t want the toy more than he wants to play, and he doesn’t want ALL of the dead rabbit more than he wants to get along. He might not bring another dog food unsolicited because he wants them to have it, (that’s REALLY complex empathetic behavior) but he won’t keep them from it when he easily could. I really don’t think that he’s afraid of a fight he could easily win (if it even came to that-a threat would likely be enough). I think he wants to get along, be happy together– be friends.

    To that end, he’ll drop his ball whenever Sandy (not even half his size) comes near him. She’ll pick it up and toss it around a bit, then drop it herself. He’ll gleefully pounce on it, toss it around for a few minutes, bring it to her and repeat. When she first came to live with us, he spent weeks ‘flopping’- batting her once and flopping to his back without being touched- to encourage her to wrestle and play more confidently (he no longer does this with her, because she more than holds her own, but she wasn’t the first or last dog to get the flop as a confidence-booster).

    His relationship with Sandy is special, but not singular. His golden retriever friend from the park will flabbergast newcomers (human) by reaching INSIDE his mouth to take his ball. He doesn’t drop it and give it to her as he does with his other friends because unlike them, she won’t ever drop it or give it back. (How’s that for reciprocity?) Still, he won’t threaten or punish her over it. She takes it with no growling, no resistance, no hard feelings on his part. Would he allow this behavior from a non-friend? I can’t say for sure because none have ever dared, but I doubt it. He doesn’t bring them his toys. He and his best buddy will not only play the toy- trading game, but they’ll both settle to gnaw on opposite ends of the same stick.

    I might think of all of these behaviors as evidence that Otis is very passive and submissive, except he’s not. It seems that his friends-dogs he loves and trusts- can do no wrong in his eyes. He doesn’t get irritated at their roughhousing, he doesn’t guard his resources from them, but he never shows them (or anyone else) signs of submission (low head, low tail, rolling over, face licking, stepping aside) either. And if danger rears its head? Otis is the one out in front.

    If a newcomer or a casual acquaintance gets into a serious snark (I’ve only seen one true fight at the park in three years of daily visits) with one of his friends, not necessarily even a very close friend, he will confront the outsider 100% of the time (Usually with a growl and a hard eye- on rare occasions he’s gone as far as a tackle-and-pin). He never targets his friend, even if they were the aggressor. The other dogs not only benefit from this, they are AWARE of it, and their behavior will visibly change in his company.

    Even after as little as ten minutes in his company, shy, insecure dogs will become noticably more confident when approached by a new dog. One little terrier at the park greeted Otis “just like he usually does”, cowering, tail tucked, trying to hide behind his owner’s legs. After a cursory sniff and few minutes comfortable walking together, he wagged and stood his ground for the next big dog, then astonished his already incredulous owner by running happily out TOWARD a third 100+ pounder. Sandy, who never barked and once hid under the bed from a maintenance man in her former home, now races down the stairs to fearsomely warn off meter men and DPW workers. Out in the yard, she will actually pause and glance over shoulder to make sure that Otis is available to back her up before dashing out to bark at a stranger. Much as she loves and respects my husband and I, when the Fourth-of-July fireworks start, Sandy runs to find Otis, her protector. It’s not all one-sided, though. When small dogs try to mount or harass Otis (he deals with rude dogs under 40lbs or so by pretending that they don’t exist), Sandy will correct them.

    Touching as it can be, this mutual defense is not always a good thing- I had to enforce a “break up” with a female great dane friend, because she was getting increasingly erratic and aggressive toward smaller dogs and Otis would charge across the play field to her defense no matter what, despite the fact that a) she was starting it and b) she had 80+ pounds of weight advantage. Even if I was totally confident that Otis wouldn’t bite, the last thing a giant aggressive dog needs is help.

    So that’s my long-winded anecdote in support of dog friendships. Otis is a special dog, but he is not unique. Many of the dogs who regularly frequent the park show some or all of these behaviors, and most of them have what their owners would describe as ‘friends’. They may be a self-selecting group, but they are not rare specimens. The majority of pet dogs, though, do not have the chance to form friendships, even if they have the inclination- their interactions with other dogs are either limited and tightly controlled (two leashed dogs passing on the street are unlikely to forge a friendship, while ‘sibling’ relationships might be friendly or not) or confined to chaotic dog-parties (most dogparks, daycares) with many dogs engaged in energetic play in a confined area. Combine that with poorly developed canine social skills and a highly emotionally absorbing bond with humans, and I can understand why many dogs do not form friendships, but I strongly believe that most can.

  36. em says


    On the contrary, I would argue that the Barbara Smuts articles you reference make many eloquent arguments about not only the existence, but the importance of friendship bonds and social interaction in dogs. Perhaps you are referring to the social inequality sometimes found in these friendships (one dog consistently assuming the “dominant” role in play, etc.)? If so, I would argue that reciprocity is not the same as equality, and that many human friendships involve habitually more/less assertive behavior, as determined by individual personality.

  37. Beth with the Corgis says

    Lest I sound cold, I wanted to say that I think good dog/dog relationships more resemble parent/child then friend/friend, and I think I’ve seen some (limited) research to back that up.

    Trisha, you have spoken before about Lassie “Falling in love” with Luke. I think that Luke was an intact male, yes? And Lassie obviously a female; not sure if spayed or intact and if so when.

    But your description of her arrival, and how she “bumped” down Luke’s previous preferred playmate, sounds like pair bonding more than friend-making to me. Especially including your gut reaction “she fell in love with him the moment she saw him.” Our instinct sometimes says things our minds have processed but not yet articulated.

    I do think, personally, that SOME dogs are capable of forming recipricol friendships. But I don’t think it’s been selected for in terms of evolution or selective breeding, and I think it’s an aberration more than the norm in dogs.

    The links above about the myth of 50/50 play are a great example; If I recall the article from reading it previously, there is not 50/50 play, one dog wins most of the time, and that matches what I have seen. Thanks for posting them, Luisa!

    I believe researchers classify pair-bonding and kin selection (helping your mom raise her next litter, as do wolves, some foxes, some coyotes, and crows) as something different and for good reason. Even some insects show great “devotion” to their mates. It is hard not to anthropomorphize that sort of bonding as “love” and in some higher animals it may be, but it’s frequency across all sorts of species points to something a bit more basic. Remember that “basic” doesn’t mean it’s not lovely. Our own strongest feelings often come from deep instinct (protecting children, jealousy, etc).

    My understanding of the scientific example of reciprocity has to do with animals “keeping track.” I.E., if I’m a chimp sharing meat, I might not share with the chimp that just groomed me this morning. But I may share with the chimp who has groomed me most often in the past three months. Or something like that. Of course food-sharing is just one example; there are animals who DON’T food share and use other forms of “payment,” so to speak.

    I know my dogs don’t willingly share any valued objects. I have seen a tiny number of dogs in my life who do seem to “gift” things to people. What fascinates me as I’m discussing is cats, who seem to reciprocate more even though they are supposed to be solitary animals.

  38. Beth with the Corgis says

    A few responses:

    I think the idea of reciprocity does not need to mean “equality” but if one animal almost always is the one making offerings/grooming/feeding the other, etc, it’s not defined as reciprocity. In most social animals, you see solicitations of subordinates towards the higher ranking. And you also see cases of displaced parenting; if one animal is always the one grooming the other and gets nothing for it in return, it’s not reciprocal. It’s part of a dominance ritual or a solicitation to mate, usually.

    Em, your descriptions fascinate me because I too live in a situation where I regularly see large groups of dogs (of all different sizes, from 100+ pounds to a Yorkie) and I have never really seen anything that you describe. These dogs are playmates, they have favored play partners, some will self-handicap to continue a game with another. I have never in my life seen a dog who really wants a tennis ball allow a less-dominant dog to take a tennis ball willingly. I have seen my submissive bitch back off from a ball she really wants because a more dominant dog (read: Any Dog) wants it more. I have seen dogs play tug and “let’s carry this stick together.” I have seen dogs lose interest in a toy and let someone else take it. I’ve never seen a dog hand over a toy to another dog, and again this is a large, well-socialized group that sees each other regularly and has ample off-leash time to forge bonds. So if you see it not just in Otis, but in plenty of other dogs, I find that interesting and in all seriousness, you might want to start filming it because it also doesn’t match up with a lot of what I have read over many years on dog behavior. It might make for an interesting study as to why this particular group of dogs has started behaving in this way. Even Trisha’s own books clearly indicate that the dog who is more dominant and wants an item more almost always gets it. Your theory that it might have to do with size is interesting; perhaps dogs that nearly always have to self-handicap just to interact with other dogs form a self-handicap “habit” that just extends to other areas?

    teresavet, I have read quite a few of the wolf/wild dog studies (the latter are harder to come by). I am aware some packs have multiple litters. I just read a fascinating study where they did scat DNA analysis of 5 packs. They found some interesting things:

    1) Around half (I believe) of the packs had multiple litters, a much higher percentage than anticipated. The point was made that unless multiple den sites are used, it can be hard to determine how many litters there are. But here come the kickers:

    2) Interrelatedness was high, with no “adopted unrelated wolves” in any of the packs. This also surprised them, since there are many reports of loner wolves joining packs. Two suggestions they made were that their sample may simply have not been large enough, or unusual pressures might cause unrelated wolves to try to join a pack. A third theory I thought of was that “lone wolves” joining strange packs may in fact be grown offspring returning home after an unsuccessful attempt to start their own pack.

    3) Some of the extra breeding pairs were incestuous, others seemed to be cases of full siblings of the “alpha” pair, though I have trouble interpreting the data:

    As for feral dogs, there are few studies. Most of what I’ve seen indicates not even pair bonds are formed, all bitches raise pups alone with almost a zero success rate, individual dogs may form loose, temporary associations which are quickly dissolved.

    I’ve seen some of the actual studies but can’t track them down now.

    I have seen repeat examples of two things in dog/dog or dog/other animal interactions, both of which are interesting. One is displaced mothering: a dog, usually female, takes a smaller animal under her wing and cleans it, cuddles it, and otherwise treats it in much the way (sans nursing) that a mother will treat her puppies. Two is dogs behaving the way puppies treat each other; playing together in a “jockeying for position” way that both find mutually enjoyable.

    I can’t wait til Trisha gets back and explains her theory.

  39. says

    in the 50/50 play articles, they are not evaluating “winning” per se, but self-candicapped behaviors. They found that these behaviors were found in the individuals who did most of the “wanna play” gestures. And these animals were also the youngest. So they themselves say they don’t know if it is that “dominant individuals win and don’t get self-handicaped” or if it is “older individuals don’t like to play so much and young individuals have to convince them with play-signs and self-handicaped gestures to play more often”. Because leadership in dogs usually comes with age.
    Some articles tell us that feral dogs don’t form packs, but others show they do:
    Cafazzo, S., Valsecchi, P., Fantini, C. & Natoli, E. 2009. Social dynamics of a group of free-ranging domestic dogs living in a suburban environment. Journal of Veterinary Behavior: Clinical Applications and Research, 4, 61-61.
    Font, E. 1987. Spacing and social organization: Urban stray dogs revisited. Applied Animal Behaviour Science, 17, 319-328.
    In these articles the dogs formed pair bonds, males helped raise pups, and they formed stable groups of 3-4 unrelated pairs.
    So some dogs CAN form alliances, groups and bonds, other don’t. Different breeds, different ecologic situations? we don’t know, but dog behavior is much more complicated than we expected (and that’s why it’s so fun to study!)
    By the way, Trisko is working in the Smuts workgroup.
    If anybody needs the texts of the articles, I can get them to you.

  40. says

    @ Marilyn I liked your question:” does friendship have to look like human friendship to be valid”

    What is most disturbing to me are the words used in the article mentioned ie …dogs fall SHORT of real (human defined) friendship.

    Why do we have to rate other species in comparison to ours? What is friendship anyway. Humans have such diverse friendships. I think it is impossible to answer this question in a global manner.

    Dogs are loyal beings– I think we can all agree; loyalty is an important quality of friendship therefore certainly they form strong bonds with us and other species.

    I still hold on to my initial post which asks the question: must we be free and on equal footing to form ‘real’ friendships. And with 48 hours to think about it I think not. I believe that I have been a good friend to a few of my employers for example.

    So at this point I really believe that the word ‘friendship’ encompasses many definitions–from the occasional ‘landing a hand’ to ‘companionship’ to ‘I have your back no matter what’.

    …and of course, dogs and cats fit one or more of those criterias.

  41. jackied says

    I have always thought that our relationship to our dogs is more like ‘parent’ than anything else. On forums I notice that people frequently refer to themselves as ‘mum’ and ‘dad’ in relation to their dogs, and I’ve heard people do it in person too – my own parents do. I try not to do it but it still slips out!

    Perhaps as a consequence I tend to think of my own two dogs relationship as more like siblings than like friends. Sometimes playing, sometimes jealous or ignoring each other, and without any choice as to whether they get to live with each other or not. Plus I think we may have inadvertently bred domestic dogs to relate more closely to humans than to each other. It makes them more traineable.

    I do think dogs can have doggy friends – my reactive dog has one or two friends that he is a completely different dog with – but I suspect that this friendship is on a lower level to that of horses, or humans.

  42. Liz says

    I agree that this is a fascinating subject with many enjoyable questions. I mostly wonder about possible differences between attachment and friendship, and the definitions of each, after skimming the thesis by Trisko, the articles by Smuts, and comments here. Staring at my dogs, wishing they could tell me definitively- in their own canine ways- I have yet to will them to an answer. (The relationship between them varies, but they are often quite close.) How do I describe this closeness, and is it harmful if I think of them as friends?

  43. Beth with the Corgis says

    Chloe, I agree with what you are saying. I was trying to decide why this whole blog had me a bit flustered and I think it’s a couple things:

    One, this issue, I think, is an upcoming issue. I don’t think ANY of us, except Trisha have read the article. WE don’t know how the author defines friendship. My understanding from his own blog is that this arose from a study he read that in part is looking at the evolution of HUMAN friendship and looks at animals who live in similar social structures to humans (large group that splits off into smaller sub-groups, families living within larger society, the types of animals that tend to practice what we might call “politics” where social relationships can translate to power. However, I don’t know and either do most of us because none of us has had the opportunity to read the article. I think that the reason it was looking at friendship in human terms is because that was the point of the science. For instance, baboon males will respond to distress calls of females who have groomed them and befriended them, but not to distress calls of other females LIVING IN THEIR GROUP who they are not friends with. That’s just one example I found, but again I don’t really know because the Time article, to the best of my knowledge, has not even hit the newstand. Their website says it’s the Feb 20th issue. Which leads me to why I have finally decided I was a bit flustered (and still am):

    Trisha, I have tremendous respect for your work and research. You have always made great pains to differentiate the behaviors of primates (us) who lived in large, complex social groups made up of different familes, how we interact, and how that (sometimes negatively) impacts how we understand our dogs, who come from a totally different evolutionary backgroud, where they DON’T interact much with unrelated strangers, where they form tight nuclear familes and defend “the pack”, etc etc etc.

    So I think I am a little surprised that this blog would question another scientist without defining the terms for us or giving us any examples. I think that there is confusion between human-type friendships (based on all sorts of things, but evolved from a complex society where unrelated adults with their own families come and go within a larger social structure and then break off into core groups of closer units—a fission/fusion society like primates and dolphins and some of the other animals mentioned have) and nuclear-family type loyalties (as is found in most canids).

    I think we define friendship as mutual affection and loyalty, in the vernacular, but this is clearly not the way the author defined the term (at least I’m supposing, again not having read the article). So we say “Of course dogs have friends!” because of course they can show mutual affection and loyalty, but we are not talking about the same thing the author is talking about.

    I don’t know. Other blogs on this site that have questioned science have done so on the defining terms of the original study. This one seems not to have. And so none of is is talking about the same terms.

  44. em says

    Beth with the Corgis said: “So if you see it not just in Otis, but in plenty of other dogs, I find that interesting and in all seriousness, you might want to start filming it.”

    I was thinking that exact thing! Though I cannot believe that the dogs at my park are particularly unusual in their capacity to form friendships, the idea that you and Kat suggest, that friend-bonding may have a ‘cultural’ element to it- meaning that it is a complex of learned behaviors and interaction styles common in some groups of dogs and absent in others- fascinates me! Do a few ‘domino dogs’ with strong natural inclinations instruct and inspire others to share, cooperate, and form deep bonds, or is it the opposite- can a few hostile or domineering dogs block this process in their play community?

    To my untrained eye, it seems that certain settings and circumstances are particularly conducive to friend-making for dogs. The deepest friendships I currently know of are between dogs who a)are preferred playmates, b) see one another regularly c) walk (especially off-leash) together in addition to play. I agree that a preferred playmate is not necessarily the same as a friend, but a place where dogs walk with preferred playmates on a regular basis seems likely to be a breeding ground for friendships.

    At my local park, the dogs have a chance to interact with many different dogs, but usually only a few at a time, and relationships that go beyond playmate and into friendship often form when people adapt their behavior so that dogs who show a strong preference for one another can meet and interact more regularly.

    People want their dogs to get the most exercise and fun out of the park as possible, and since it is so big and so sparsely populated, they synch up schedules so that the dogs who like each other can walk and play together, while those who are indifferent to one another generally interact briefly and go their separate ways. Most days it seems that the park is full of small groups of bonded dogs (2-5 animals) who cross each other’s paths occasionally and come together in bigger groups in a couple of meeting places (the field, the creek). The dogs all generally get along well but typically show marked preference for their friends. I do notice that the energy of the dogs starts to shift when the group gets to around ten or more- more excited, less relaxed, less focused- like a night club instead of a cafe. Still fun and positive, usually, but much more distracted.

    This format may help foster friendships, but the actual relationships are initiated at the dogs’ inclination. We often walk with dogs whose owners I like, but who are not really preferred by my dogs, despite being well-socialized, perfectly nice dogs. Otis and Sandy seem to feel comfortable with them (Sandy’s likely to play with any affiliative dog, while Otis is choosier) but do not show any special enthusiasm in greeting them, reluctance in parting from them, or any particular sign of attachment like sharing behavior.

  45. says

    @jackied : You bring a whole other subject which I hope to see discussed some day. The parent/child relationship between humans and dogs (actually now that I think about it it’s more between humans and their pets not just dogs..)

  46. says

    For a definition of friendship, I like the one from Silk (2001): “friendship should be reserved for relationships among nonkin that are characterized by frequent participation in afiliative interactions (often, but not necessarily, including grooming); involvement in coalitionary aggression, particularly in defense of the partner; high rates of association; mutual responsibility for maintaining proximity; high degrees of reciprocity in directional, nonaggressive activities such as grooming and food sharing; continuity across time and context; high degrees of tolerance (co-feeding), loyalty, and compatability; and low degrees of stress when together”.
    Behaviour 139, 421-446

    I still think that SOME relations between SOME dogs are of this nature. Some dogs would be more capable of forming friendships (with each other, or with humans) than others (breed, age, background…). We don’t have the literature to confirm or to deny this form of social relations between dogs. I’m beginning to think I should study it!
    Tomorrow I meet with my thesis director, maybe he’ll like the idea 😉

  47. Kat says

    I suspect Em is on to something with “relationships that go beyond playmate and into friendship often form when people adapt their behavior so that dogs who show a strong preference for one another can meet and interact more regularly.” Thinking of it from a parent’s perspective I think my children learned how to be friends with others by having the opportunity to interact with others especially with their preferred playmates. It’s why we arranged play dates for them. And it is definitely true that at one of our dog parks, the one where I see more friendships, the people do arrange to be there at the same time so that their dog has lots of time to play with and bond with preferred playmates. At the park where there are fewer doggie friendships I haven’t observed the same level of coordination among the people; the culture there seems to be more of a you show up at the same time as yesterday and hope the dogs your dog liked are there again today. I’m left wondering how much opportunity in general people give their dogs to form canine friendships outside their family and what role it plays to be regularly given the chance to form friendships.

    And, a bit off topic, a puzzling interaction between Finna and Ranger from this morning. Finna has become a fetch fanatic and is a bit of a resource guarder especially when it comes to her balls. Ranger has never been much interested in fetch and makes a point of not being near her balls. In an effort to encourage Finna to learn patience and impulse control I chuck the ball into the bushes and do something else while she rummages around hunting it. I won’t throw the ball again until I’m finished with whatever (raking a small pile of leaves, transplanting volunteer lemon balm…) Today I was working with Ranger on weaving between my legs and saying his prayers. At one point while Ranger was standing five or six feet away from me waiting to be asked to “through” again Finna brought him the ball. She carried it up and dropped it between his front feet, backed away, then looked at him (anthropomorphism warning) as if she wanted him to join the game. Ranger smiled at her and took a couple steps back as if he was saying thanks but you go ahead. I scooped the ball up and chucked it again. Was Finna making a gesture of friendship? She was giving him her most prized resource. Or did she think he might throw it for her? I so wish I could read dog minds.

  48. Julia Little says

    Just a few years ago to get to the office where I worked in Sorrento, Italy I had a daily ten minute walk from the station along the main road through the town centre and along to the far end of town. Most mornings I would find myself trailing an elderly man and his tiny, off the lead Yorkshire Terrier companion. The dog trotted along at the man

  49. trisha says

    Thanks again for all these comments, fascinating stuff. So much so that I’ll be writing a second blog tomorrow about this issue. And I also want to follow up on the post about clickers versus silence. Next week I’ll be interviewing the authors of the study I mentioned, and I’ll let you know what I learn. I had some questions after reading their methods. But tomorrow I’ll write more about friendships… doing some research now before I write more.

  50. says

    In my humble opinion, Willie herding sheep and the two of you taking great joy in it is a perfect picture of friendship — two friends doing what they love, taking joy and pride in it, and equally sharing the success of the moment.

  51. JJ says

    What has disturbed me most about this conversation is the definition of friendship. I’ll tell you right now, if I get a particularly prized chocolate truffle, I’m not going to be sharing it with my friends. Nope, it is all mine. I might even growl at you if you try to take it from me. Does my unwillingness to share *my* food mean I’m not capable of having friends? Am I the only human who fails in the definition of being able to have true friends?

    I could go on and on with examples on how I think either the definition of friendship is sorely lacking or the examples “proving” dogs don’t make friends do not reflect what I see all the time in dogs.

    I think that this discussion of friendship capabilities in animals is as important as discussions of emotions in animals. It is an extremely important discussion. I also think that humans will end up with much the same answer as we now have concerning the emotional life of animals–an answer that most dog owners already know the answer to.

  52. Annie R says

    @meganwf: Thank you so much for the David Lee Roth video. WOW!!! I can’t believe I learned more about herding from a 6-minute video by a rock star than I have from watching videos of competitions on TV! His metaphors are hilarious (I hung out with musicians in my youth and he is SO right on, and the imitation of “sheep speech” had me laughing so hard I almost peed! And I so appreciated that he calls it an “art form”, such respect for the dogs and so entertaining at the same time.

    @JJ: Yeah, I agree re: the chocolate truffle! Friendship can’t be evaluated by any given moment that’s for sure! I also think that friendship is encouraged in our dogs by the fact that we spay and neuter our pets, thus taking away the mating motivation for them hanging out together and enjoying each other’s company; in a way it seems it would free them and to make friendships even more possible between dogs (just a theory . . . ).

  53. JJ says

    Here is part of an e-mail that I got from a friend on a different topic, but it speaks to the discussion here:

    “here is a picture of my cat washing my dog. I can also get you a picture of my dog washing my cat as it is a ritual they perform several times a day.”

  54. says

    Having witnessed “constancy, reciprocity and mutual defense” in dogs personally, I have little doubt that they form genuine friendships but I can see the difficulties in studying the phenomenon and generating real data. Where I have seen these expressions most commonly is in the pack environment and the social structures most people encounter with dogs aren’t packs.

  55. says

    I work at a doggy daycare and I without a doubt i believe that dogs form friendships. My own dog has two best friends that he plays with if they are there. I see dogs pair off all the time and it is almost always the same dogs everytime. When they see each other enter the play yard they bound towards each other, greet, and are off playing. They are truely excited. I’ve seen dogs wait for their buddy to arrive at the gate and as soon as that dog gets there they are off. My own two dogs are friends I know. They play, groom each other, wait for each other to go outside etc. So yes, in my opinion dogs form friendships.

  56. New Guy says

    Forming a “friendship” with ones dog… same as bonding with him/her?

    We have a new dog from rescue. Very nice with a few minor issues, but after a few months we seem not to have yet bonded.

  57. Lisa says

    I think dogs do make good friends. My family has a beagle, he’s 14. Up the road from my house, a family had two dogs. They used to play all the time. They would come to our door and just wait for us to let our dog out, and my dog would do the same at their house. About two years ago, the other dogs had passed away due to old age and disease. However, our dog still goes to their house and waits hoping that they still might come out to play. It shows that our dog misses them. If you didn’t have a bond with another dog, it would simply be impossible to miss them.

  58. says

    What is incredible to me are all the documentation of inter-species friendships(bonding) that take place between dogs with primates, cheetahs and other species. It seems to me dogs are the most adaptable friends in the animal kingdom.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>