What Do Words Mean to Dogs?

One of the segments on the BBC show The Secret Life of Dogs that generated several interesting comments was the segment that showed “Betsy,” (not her real name, you gotta love that some dogs now have ‘nom de plumes’) retrieving an object after being shown just a photograph of it. Wow. That truly is amazing. It got me to thinking about the trouble I had a year or so ago when I tried to teach Willie to discriminate between toys based on a name. For over a year he’d been told to “Go get your toy,” and he’d pick out whatever toy he liked best that was nearby.

When I tried to teach him to touch or pick up a toy based on a different word (“ball” or “rope”), he became hopelessly confused. He became so stressed over the entire operation that I dropped it and went on to teach him other things. My best guess is that he had categorized all signals as requests for action, not as labels of an object. That’s a pretty big frame shift to make. And yet, here is Betsy, able to pick out 250 different objects by name and by just looking at a 2-dimensional photograph (which dogs, by the way, were not supposed to be able to do). And of course, there’s Rico, the other famous BC, who could not only pick out hundreds of objects, he could get the larger of two versions, AND could ‘fast map,’ or be asked to retrieve something he’d never heard of or seen before, and deduce that it was the only unfamiliar object in a group of familiar toys. Wow. These are some smart dogs.

Will now has learned labels of living things: he knows “Jim” and “Trisha” and “Sushi.” I can say “Go to the Barn” or “Go up the hill” and he runs the right way, but I suspect he doesn’t think of the “barn” or “hill” as a label of the object itself, but rather of an action and direction. The BBC segment got me rethinking about what Willie understands as an action and what he understands as a label. He knows “get your toy”… and I suspect understands that “toy” relates to the objects he plays with, but does he see it as a label or as an action? Is it more about playing with anything in front of him, or the objects themselves?

I’m inspired now to start testing this out with Will. I’m going to write down every word and movement I make that I think might be a cue to him, and what I think it means to him. This, needless to say, is going to take some time, so stay tuned for what I come up with. Meanwhile, I’d love to hear from you. How much of what your dog understands relates to an action, and how much is it a label of an object? Is there any difference between learning to label a living thing versus an non-living object? Willie definitely knows “sheep” and “Sushi,” and I wonder if it matters that the ‘object’ of a label is living or not? He knew all the names of my other dogs, in that he’d turn and look at them if I said “Where’s Lassie?” He gets excited when I say “Dinner,” but does it mean the actual food itself, or the action of eating them? See how interesting this can get?

Meanwhile, back at the farm: We are all continuing to heal. Willie played for hours with his new friend, Max, an adolescent Border Collie who lives not far away. Watching them run and run and run huge circles around the Orchard Pasture cheered me up immensely. The best part of it was the way they played, because Willie didn’t always play as well with others. Willie would start running with another dog, then get in front and air snap toward their face to stop their forward movement. Then he’d stand still with a satisfied look on his face. Look what I accomplished! Boy, am I a good herding dog! Eventually the other dog would just stop running, because what was the point? However, recently the lovely Dobberman Mishka seemed to have taught Willie to stop playing by herding. Mishka put up with his “herding” two times, and then growl-charged at Willie. Totally appropriate, very controlled; I remember saying “Good girl!” I was so impressed. Willie backed off with that confused, silly look that male dogs get around females who discipline them, and never tried it again. They began racing instead and now it’s Willie’s favorite game.

It helps that Max is fast. Really fast, and so is Willie, but neither can really beat the other, so they run and run and run in the snow, stop and get their breath and then run some more. And you’d think that would have tired Willie out enough to sleep through the evening? Of course not, he kept dropping toys in our lap all evening long. I think it just warmed him up. Here’s some photos of Max and Willie playing. Not the best photographs, but I love how it shows the world we live in right now. Black and white (dogs). Black and white (everything else!)

Comments

  1. says

    I’ve seen a dog who was trained to “read”. You could hold up a flashcard with various words written in block letters, like Sit and Down. Of course, the dog had just been taught to associate the shape of the word with an action, not to actually read, but it was still pretty cool. I haven’t watched the special yet, but it seems like Betsy could easily have learned to associate a particular picture with the act of retrieving a particular toy, without understanding that the picture was *of* the toy.

    Corrie knows some toy names…”ball”, “blue toy”, “frisbee” (although we have to call it “the magical flying disk” while we’re in the house, lest he start bouncing off the walls). And he knows some locations, like the kitchen and the bathroom. And he can even tie them together sometimes, like “get your ball from the bathroom”. But I think you’re onto something with that being words for an action in his mind, not for the actual place or item.

  2. Frances says

    I am glad that the healing has started for you all – and what a wonderful time the two BCs are having!

    Like you, I am fascinated by the subject of dogs and language. Mine must have a hell of a time trying to pick out meaningful terms and actions from the constant flow that comes their way, though! Starting agility has shown me just how imprecise (to be kind!) my communication can be with them – it really would help if I could remember the names of the pieces of equipment, and my left from my right, and run at the same time.

    I was interested to come across Sean Senechal’s books on canine sign language – although more than a bit sceptical, too, especially when I realised the most glowing review of the book was written by the author. Has anyone any experience of teaching their dogs to communicate with them through signing, beyond the scratch-at-the-door or kick-the-water-bowl level? It would be amazing if it worked, but I was rather put off in the first few pages, when she used “literacy” to mean capability for language – much as I would like my dogs to settle down quietly with a good book, I think they are more likely to want to chew it than read it …

  3. Amy W. says

    Betsy was amazing!

    The first time I really took notice and was amazed at my dog’s ability to understand what I was saying and take action on it was when I realized he understood the phrase ‘LET’S GO HOME.’ I take him to the park daily, and somewhere along the way, without realizing it, when it was time to leave I would say ‘let’s go home,’ and walk to the car. Then one day at the park, I said ‘let’s go home’ and my dog, Axle, stopped turned around and ran to the car and waited for me to open the door.

    After thinking about it, a lot of things that he has learned are actions rather than labels. Even if I formally train him, it’s still usually an action that I want from him: sit, down, come, etc.

    The pictures are great, I especially like the top one.

  4. Mary Beth says

    How fascinating! I agree…my dogs understand the action not the word. I hadn’t put two and two together until you said that.
    I suppose scent articles in competitive obedience are the same way. The dog’s aren’t focusing on the objects in the pile, they are “going” and “scenting”.
    My dogs have a huge vocabulary, but I’m pretty sure you’re right, they are all actions, not labels.

  5. JJ says

    I can’t think of anything I’ve taught my dog that wasn’t an action or can’t easily be interpreted as an action.

    Duke’s bag of action tricks: sit, down, lie flat as a pancake, turn left, turn right (his left and right), get in the car, go potty, back, off, drop it, find it, catch (the food that I’m throwing in your general direction), etc.

    OK, here’s one for you: Is “stay” an action?
    I know it’s still more of a body thing than an understanding of a noun. It just struck as me as funny to include it in the list of action tricks above.

  6. says

    What a fascinating subject! I never though of it before, but it seems obvious now why my dogs are so easily confused when I ask them to get a specific toy. They dash of at the “GO GET YOU…” part and then grab whatever is near, look at me with question marks in their eyes and seem to wonder that sometimes I am all excited about (when they got what I had in mind) it and sometime just look back expressionless (when it was not). However, they seem to be better with living beings, like a person or another pet. I wonder why that is the case?

  7. Kristen says

    My dog Belle does understand the difference between her toys. She can bring back (with reasonable accuracy)her: ball, tug, chewie, chickie, birdie, frisbee, blue cheeseburger, frogball, and raccoon. I feel pretty certain that she understands the nouns differentiate which toy she is supposed to bring. Otherwise I think most of the rest of what she understands is likely tied to action… “go to (back yard, flyball, Grandma’s, etc.)” The only one I’m not sure about is “Roxy”, her best dog friend. I don’t know if she knows that “Roxy” means “the dog” or “playing with the dog”. Not sure how to test that one.

  8. JJ says

    Here’s some thoughts this discussion has fired in my brain:

    A poster above talked about being able to distinguish when her dog is barking at a person coming to her gate vs a squirrel in the yard. I would say that the dog is communicating about different nouns/things/persons as opposed to actions. So, if a dog can communicate/use his/her “words” to distinguish between nouns, why wouldn’t they be able to understand our words to distinguish between objects?

    Alternatively, could we speculate that the dog is communicating that a person is “coming” vs a squirrel is “running”??? More about actions that objects? Perhaps the dog is communicating that “I want to *greet* a person” vs “I want to *chase* a squirrel”. Would it be the same bark for all desires to chase vs all desires to greet? In other words, is the different barking words really about an action the dog wants to do?

    As speculated above, perhaps it is a matter of training. After all, isn’t this whole area of science about “behavior modification”? About training behaviors? We’ve been focused for years on verbs. So, it is any wonder that that is what our dogs exhibit in response to our words? Even our experiments testing a dog’s understanding of a word revolves around an action. For example, *point* to the object whose word I just said. Perhaps we need better experiments?

    Here’s an honest question: How much does it matter? What would we learn if we find out that X% of dogs can indeed understand that different words mean different objects. Or what if we learn that dogs can only pick up actions from our words? What would that change? I’m not being critical of the discussion. I’m just trying to understand, “what’s the big picture”? Why do we care beyond academic curiosity?

  9. says

    “Willie would start running with another dog, then get in front and air snap toward their face to stop their forward movement.”

    this happened several times with the horses over the weekend. We have 3 horses, 2 mares, 1 gelding. In this weather when the footing is too slippery they don’t get a lot of .real. excercise so they are a little wound up. Lucky for us, we have a beautiful indoor arena to turn them out in. When the 3 are out together, my mare seems to be the “fun police” and she will get in front of the other two, bounce to a stop and either buck or kick at them. I am sure this is to control forward movement, which worked most of the time. Once they slowed down, she would trot forward and do it again if they sped up.

    My dogs are not so advanced as to understand words, but if I remember right the Cattle Dog, Skid boot, knew a huge vocab.

  10. says

    I have tried to teach my dogs “go find” a certain person, but so far they have not got the names down. If I say “go find Michael” they will just go to the nearest person they find, or if we’re playing hide and seek they will search for whoever is missing.

  11. says

    I lived with Labs for years. Big-hearted and brave, lovey and loyal, sloppy, goofy Labs. Purebred or mixed, some were smarter than others, but they’d all do ANYthing for food, even pretend they knew what I was talking about (well, maybe not don’t pull on the leash). If food was the endgame, any of them would and did do most of what I wanted. Like dogs do, each of them gave me generous gifts in return.

    The last one, beautiful brown Gracey, had a horrible beginning and came from our humane society loaded with neuroses and wildly reactive behavior. Her amazing gift was to teach me what it is to bring a dog back from the brink with patience, consistency and love. After dying of bone cancer 5 years later, she sent me back to the shelter to get someone who is arguably the best gift of all.

    From the minute he looked up at me from under his lashes, I felt Garrett was sent to me as a sort of reward for everything I’d ever done right for dogs. Not a Lab (surprise!), but an alien, some sort of handsome herder, he’s a dog so willing and smart that even at the age of 18 months, I never had to teach him anything. I just let him know which things he does that I like. Really.

    He came as a rezdog (reservation dog) stray from the Oklahoma Cherokee, so I made sure he understood this is his forever home, and he fit in easily. Yes, he somehow knows all of the usual obedience (sit, stay, come, heel, cross with the light, how to be welcome anywhere, pretty much anything at all), what all of his toys are individually called and where they are. As a matter of fact, he usually keeps them all in a neat pile and all of his bones in another. As part of our arrangement, he does everything as long as I ask for nothing. Sort of. He charms everyone he meets, I don’t ask him to love cats or squirrels, and we both know how lucky we are. I am not making this up.

  12. Trisha says

    Re a few of the comments: I would never say that dogs can’t learn to label objects. I think it is overwhelmingly true that they can, and that many do. But every dog is different, and it seems to be much harder for Willie than for some other dogs. I am just interested in how dogs perceive our cues, and how their are ‘translated’ from person to dog. Regarding why would we care…? Well, partly it IS ‘academic curiosity’ for me. One of my first memories is lying on the living room floor, wondering what it was like inside the brain of my dog, Fudge. How much was it like mine? How different? That question has driven much of my life, so I guess I must think it’s important!

    But beyond that, I think there are some important practical reasons to ask those kinds of questions. One is that people often assume their dog understands a cue exactly as the person does, when in fact the dog has an entirely different definition of the word or action. (At first, “Where’s Jim” was defined by Will as run to the window and get excited about something wonderful about to happen. I didn’t realize this until I said “Where’s Jim?” when he was sitting beside Jim.) Secondly, I think animal cognition is important relative to welfare issues. The more we discover about the internal states of sentient animals, the more some people will be might be willing to consider their welfare more seriously.

    And, on another note: so interesting that Holly’s horse ‘played’ the same way as Will!

  13. says

    This is all very interesting to me. My dogs have always have a good vocabulary. My current boy, Tag (chocolate lab – 8 years old today!) knows much more than most people give him credit for. The best is when he brings me a shoe/slipper (without being asked…and he does bring it to me…that retriever thing going on, ya know), I tell him to “go get the other one” and he does. He has done this since a wee pup. If I’m sitting in my office and ask him to go get my slippers, he searches around the house until he finds them. If I say, “in the living room”, and point down the hall, he goes to the living room. Yeah, I know, cues given. He takes direction from pointing very well, too. Anyway. He knows the names of several of his toys…warthog, kong, little kong (I find it amazing that he can differentiate between the two…course, am sure I emphasize the word ‘little’), & ball. Those are the main ones. He also knows food dish. He will pick up anything from the floor that I ask him to (paper, phone, remote, book, fabric, etc.), though that is probably not pertinent to this conversation. ha. Just sayin’.

  14. says

    Although Merlin will not “go find [object]” I have figured out that he knows at least 100 words. Most of them are food/prey/fun related. Mention “Tastee Freeze” in a sentence and he will head towards it on the next walk (even if that walk is 2 hours later). The neighbors know to call squirrels “The S word” if he is within hearing or he will start scanning all the trees and telephone poles in the area. Most recently I realized he knew the word “soup” when I was talking to my mother on the phone and she asked what I was going to do with the leftovers I had. I gave the one word answer “Soup.” and he jumped up, looked at me, licked his lips, ran to the kitchen and sat by the stove.

    He can also read the words “SIT” “DOWN” “STAND” “UP” (ie beg) and “SHAKE” in both English and a symbolic language I made up, but if we don’t practice it at least once a month he loses the ability and we have to start over from scratch. I worked on reading in the hopes that we could move on to words like “COOKIE” and “OUTSIDE” so that he could communicate with me better, (no matter what he wants his method of telling me is to stand by me, stare and whine until I guess the right thing) but he couldn’t get beyond the basic five commands.

  15. Shalea says

    I have tried to teach a dog labels for objects (his different toys), but the dog in question (greyhound) was really, really not interested. He seemed to appreciate the thought when I brought him a new toy, but that’s about it.

    My current dog is very nearly blind, and so I attempt to apply labels to sounds he reacts to (“that’s a squirrel/truck/heat pump/etc.”). Not sure if it’s working (if he’s applying the label I give to the sound he hears) or if he takes my acknowledgment of the sound as a sign that it’s nothing he needs to be concerned with. It’s worth noting that if I tell him “that’s a person” his reaction is entirely different than it is to other random noises (because people just might pat the dog). But he could just be associating the word with the action of getting patted.

    We do play a label game with him – “find ” – but again, he’s very inclined to seek out a person anyway so it’s very hard to say if he associates the name of the person as a label for that person.

  16. says

    My young BC understands the word “get” and will wait for the name of what he is to go get. I tried to write down all the names of toys, objects, and people he knows — it’s a long list! He has down at least 30 object names. And since he destroys a lot of toys, the list is continuing to evolve. I’m not sure he understands “go”, as I use it in agility too often. He’s definitely more verbal than my Labs. The Labs tend to respond better to visual cues. Makes sense, in the field, you stop a Lab with a whistle, then give directional signals. BCs use the whistle as a verbal cue.

    Interesting idea with the flash cards, I should try that! A great rainy day game.

  17. Lacey H says

    I’ve noticed that several of my little foster dogs have learned the names of my dogs quickly, often before they’ve learned their own new names – new because either they were strays, names unknown, or because they showed a bad reaction to the name they came in with, such as cowering. They showed they knew my dogs’ names by turning to look at the dog I spoke to, without glancing at me first. I take glancing at me as the sign they’ve begun to learn their new names.

  18. Ed says

    My dog knows the names of a couple dozen things, except when she doesn’t. She’s either at high-90-something percent accuracy, or pure chance (or worse – sometimes she just keeps targeting or grabbing the same thing over and over or runs up and down not targeting or fetching anything!), depending on her mood. When I was graphing what she learned, it was very clear that sometimes she didn’t want to listen. (She knows several toys; some things I’m likely to drop in the kitchen; the DVD, TV, and Cable remotes, some of her body parts, etc.)

    However, she has never failed when I ask her to target a person by name. Not sure what that means.

    I know dogs aren’t supposed to generalize (and I’ve never quite understood what that means). One rainy day, when I was trying to find things to do, I moved her “place,” a cheap doormat, into various rooms and asked her to go to her place or put her back paws on her place or whatever. We eventually went into the den, which has a variety of throw rugs around, and I put her place behind a chair and called her in. Her first try was the closest rug – which by the way is at least 10 times the size of her place. Her next try was the next-closest rug. Obviously, although she knows the difference between her place and another dog’s place (they’re different colors and I assume smell different) she also knows that her “place” is carpet and not hardwood. That’s not an action, that’s the nature of thing.

    I’m with Willie – there are some things that seem to come easily to most that are completely beyond me!

    Merlin’s Mom – the need for regular refresher courses is interesting. Do you have any thoughts about why the flashcards won’t stick?

  19. Carrie says

    One of my BC’s got really frustrated, just like you described Willie, when I tried to teach him the names of toys and asking him to touch the so and so toy. After reading this blog posting I would definetly say that he thinks of the words he knows as actions. “Frisbee,” is not a round disc, it is the act of playing fetch with that particular shaped object. I can only think of one label he might know as a label. When I am finished playing and my husband is still up for it, I will tell him “I am all done, go play with daddy.” Even then, he might think of “daddy,” as the act of switching play mates.

    This is also my boy who does not enjoy doing anything unless it has a purpose. He will do agility, but gets very frustrated because he doesn’t see the point of it; “because I said so” just isn’t enough. His agility frustration is an almost idential reaction to his label frustration. Maybe he is frustrated because he does not see the point to labels? He has all the capability of learning them, he is incredibly smart, it may be that he doesn’t have the need for labels to communicate, so why learn them? I wonder how I could make labels “worth it,” in his mind to try this out.

  20. David says

    I’m wondering how the ability to identify an object in a photograph relates to being able to identify something being reflected in a mirror. We have an English Shepherd puppy who at four months figured out how to look at something in a mirror and react to it as a reflection. (Which was the equivalent of me waving a toy behind him as he looked at me in the mirror.)

  21. says

    Excellent observation for words. My dogs understand words as things, and words as actions. They can even retrieve a specific toy, but only because they like it.

  22. says

    Although I haven’t gone to really teach my two dogs a lot of different words, I know that they do understand many of them. For example, sometimes I’ll just say the word ‘ball’ without one in sight and my boy will actually look in the direction of one or go to get one. Same thing with ‘doggie’. When I say that word, they know that there’s either one outside or on TV.

  23. Susan says

    Hmmm, I’m struck by the dog retrieving an item when shown the picture, as I’m having difficulty teaching Arie “match to sample” with 3 dimensional objects. I’ve backed off and been pondering how to do this, as she has a large history of reinforcement for hot targeting, and often ends up choosing an item she had previously chosen before in a prior trial (correctly, though purely by chance).

    I’ve been considering using a stack of cards like what is used in the kids memory game where you try to find the matched pairs, there are usually simple graphics of toys, etc. But, I’m not sure if that would be too hard at this stage, was planning on doing that later. On the third hand, it might simplify things to not have a 3 dimensional object. And its a lot cheaper to print things up than to keep buying duplicates of various different objects, especially since at this point I’m trying to avoid re-using objects until she gets a better handle on what I’m looking for. On the fourth hand- maybe it is too visually difficult a task.

  24. Alexandra says

    Oh, that’s a very interesting difference between types of labels. I had never thought of it that way before, but it makes a lot of sense. Copper, my lab, knows a lot of words. Now that I think of it, most of the words he knows refer to actions and living things. He is very interested in whatever his people are doing and seems to listen very carefully when we speak, often staring intently or cocking his head. We joke that he uses telepathy as I can generally figure out what he wants from his expressions when he stares at me.

    I think he does know the words for “go get your… ball/stick/toy/tuggie/bone” with toy being whatever object he wants and the others being specific things. He will usually bring the correct object that you ask him for, although I didn’t make a concerted effort to teach him. I will play fetch or tug with any of those toys, so I don’t think he considers their names as referring to a specific activity. I think those are about the only object labels he knows. He definitely knows the names of living things like kitties, deer, bunny, my husband, other family members, his best dog friends, and places like upstairs, car, and outside. All the other many words I can think of refer to commands to action such as obedience and agility commands, or things he can do like “go lie down,” and “who wants dinner?”

    That is pretty amazing that the dog can find an object from a photo. I can’t wait to watch that BBC special.

  25. Roberta says

    I have observed that consistency matters. My husband calls my dog’s toys “toys.” I call her toys with eyes (no matter how many legs the toy has — or how many eyes — “dollies.” I tell her to go find a “dolly” so we can “fetch” and she will. If I tell her to go get the “other one,” she will. If I ask her to fetch and she comes back without her “dolly,” I tell her, “Where’s your dolly?” and she’ll go get it. She reacts much faster when I use the “sign language” of my hands opening up and out to show her that my hands are empty and I don’t have a dolly.

    At a time, I envisioned that I would teach her to distinguish between her toys, put her toys in her box, take a particular one out, and deliver a particular one to her papa, but she can’t see the use in it and would rather pester the cat, who doesn’t have to understand language because he reads minds (smiles).

    I’ve enjoyed reading the anecdotes here of what people (who are probably diligent in practice as well as in consistency) have been able to do with dogs. I particular enjoyed reading about Garrett. I had a “Garrett” (Bear Doggie) at a time in my life once, a part wolf. I wonder if Sam’s Garrett might have a bit of wolf in him, being a reservation dog and all. Sometimes I thought my Bear Doggie could just read my mind, like my cat does, but in reality I’m sure, like Garrett, he was just a gift.

  26. Kat says

    This is fascinating and since I read your blog this morning I’ve been experimenting with Ranger. I put two of his toys on his bed and asked him to get one by name. He grabbed the other one and played with it enthusiastically. I should mention that both were toys we hadn’t played with in months. I took the one he’d grabbed saying “give, tiger cat” and put it back on the bed. I put him in a down stay and put another toy, one that had been a great favorite months ago, on his bed with the tiger cat toy. I asked him to get booda and he didn’t move but looked puzzled which is when I remembered that we’d been in the habit of calling that particular toy booda booda. I asked him to find booda booda and he ran over and put his nose on it then grabbed the tiger cat and played with it tossing it up and down. I can’t say I didn’t indicate the booda toy with my eyes but he clearly understood I meant one toy over the other. I think that Ranger does recognize nouns as naming objects. He certainly recognizes nouns that name people. When he asks for a walk and someone else is going to take him I can say “Alexandra will take you” and he’ll go looking for my daughter or “Dad will take you” and he’ll switch his attention to my husband. I’m not sure I entirely grasp the distinction between noun as object and noun as action. If he can select the named object I’d argue that he knows that noun in some context. I’m not sure how to tell whether that understanding is comparable to mine–a symbolic sound representation/label for a specific object–or whether he is associating the sound label with an action that involves that object. I do know that his life is enriched by these discussions as they prompt me to experiment and try things to see what he can do, what he can learn and what he’s able to understand.

  27. Beth says

    Growing up, I had a very smart Chesapeake Bay Retriever, who would do anything to have you throw a tennis ball for him. I could ask him to get his ball (i.e. tennis) , another ball (i.e another tennis ball), a hard ball, a soft ball or a football. He knew the names of all of them. The softball and football were a bit big to fit in his mouth, but he would do his best to bring them to me. He keep many of his toys outside in a manzanita bush, where he kept all of his precious possessions, but sometimes they were left around the yard. I never really knew where they were, so I couldn’t be giving him clues about what to bring.

    He also would go and get the newspaper for us each morning. He would run to the top of our very steep driveway and bring it into the house. (If our paper was missing, he would take the neighbor’s!) I suspect that some dogs associate “Go Get XXX” with a what is typically gotten at that time of day or other specific situation. We usually didn’t play ball first thing in the morning. But, I believe he understood both nouns and actions. He also knew go for walk, go for ride or go for a swim.

    I also taught him hand signals. He would back up or move to the right or left and come forward (all the while facing me). So he could move in the shape of a square with out me saying anything. I was just 14 when we got him and really didn’t know much about training, but he was very food and ball motivated and easy to train.

    Now I have 2 cats and the only word I am really sure they understand is ‘birdies’. They are pretty smart (almost telepathic about many things). I have clicker trained them to respond to hand signals, but I am convinced they are unable to differentiate between words, even between their own names. Intonation of words is really all they get verbally. It’s really worse than the old Far Side Cartoon “Blah Blah Blah Ginger”.

    I think many of us are fascinated by what animals understand and feel. Understanding the differences in what and how they learn helps us create a better life for them.

  28. Debra says

    I had to laugh at your description of Willie when you tried to get him to bring toys by names. I had watched the same show and immediately tried to get my dog to bring me toys by name. Of course my expectations were way too high and he became confused and just tried to bring me toys randomly. I later noticed in the yard that he knows toys in a different context. There I have always asked him to interact with toys by name – a frisbee, a stick, and a ball. I was more disciplined about using a clicker when I was training outside and I wonder if that made a difference. (In the house I was more casual – in the yard I would have to walk farther if I couldn’t get him to bring the toy I wanted!).

    I love the pictures of Willie playing with his friend. Recently my younger aussie has found a new lab mix friend to play with in our corral. She funs as fast as him and I think it’s the same thing as you describe – she keeps up or passes him and he can’t do his usual herding turning or nipping trick. It was so funny to watch the first time he realized she was keeping up (or passing him). Total look of confusion as he had to readjust his thinking.

  29. Dobe mom says

    Interesting question. I’d say my doberman does know certain objects. He can find several of his toys by name. We are training in competitive obedience so for example in our last class, we were working on scent articles and the instructor put toys in the pile of unscented articles as a twist on the exercise. He seemed to understand I wanted the scented article, not the toy or the other articles. When we are training at home I can put him in the bedroom, hide his dumb bell in the living room and then send him to find it….he knows I don’t want the ball or the rope toy or his bone that are also lying around. I do use different words if I want him to retrieve his dumb bell vs. his scent article vs. his ball when we are playing.

    My former doberman surprised me one day….I had put his collar on the nightstand the night before and then forgot about it. When I was going to take him for a walk the next day, I was looking for it and said something to him about where I did I put his collar??? He ran in the bedroom and got it and brought it to me. I had no idea he knew what his collar was…..had never tried to teach him that word before.

  30. says

    My quadriplegic brother used to have a service dog, Marilyn, a blonde Great Dane, who could go get items whose names she knew and take them to specific people she knew. Once he asked her to go back to his van, get my purse and bring it to me. She knew “purse” was an unfamiliar object and selected it, and I was an unfamiliar person (not him or his wife) so she brought it to me–except she picked it up by the bottom and we had to go back to the van to get all the items that had fallen out on the way! Still, I was impressed.

    She was also trained to run around behind him at a curb cut and jump up, pushing him with all four paws on his back until he was up on the curb. They used to stop pedestrian traffic in the city as people watched this process.

    I have tried to teach my dogs to distinguish objects since then, but without the same success.

    Sadly, Marilyn didn’t live long–8 years–but we have great memories.

  31. says

    I have a dog I rescued from a trash container. I have him for 3 years now, and I taught him basic stuff like sitting, down and staying… except he doesn’t like staying much hehe… I like how simply calling his name he starts swirling his tail around.

  32. Frances says

    Very, very interesting discussion – I wonder if associating different actions or games with different toys might speed up learning the names? The noun versus verb thing could explain why I have struggled to get mine to differentiate – they know the difference between fetch it games, and find it games, but I have tended to use the same toys for both. I think I will experiment with different sets of toys for different activities, and see if they leafn more easily.

  33. Anna says

    I had not really thought about it until this blog came up but my dogs, even at their young ages, know quite a few words. How exactly the words work in their brain I don’t know. My14 month old Rudy knows “Where is ______?” Grammy, Mommy, Pattie (the cat), Penny etc. by looking at them if they are in sight or going to see where they are if not. However unthinking human that I am I sometimes ask “Where is Penny?” but at other times ask “Where is your sister?” referring to his corgi sister Penny and both statements invokes the same response toward Penny so two words represents the same thing to him. Then poor Penny if she is not in sight he goes, finds her, grabs her neck and drags her back to me… she is the only “thing” he ever fetches. Then again if he ever fetched his 84 year old Grammy he would be in BIG trouble so maybe he is smart enough to know that too.

  34. Ann W in PA says

    Interesting! I came up with a list of 55 cues Rowdy knows… while some are object names, each cue involves him interacting with the object in some specific way (jump through the tire, go collect dollars from people) and thinking it over with this new perspective, I’m sure he is perceiving those cues as me asking him to DO something. I generally always ask him to do the same thing with each object. He does seem to know some words that are sort of amplifiers, like ‘more’ and ‘farther,’ but those took longer for him to get, and I’m unsure of what they really mean to him, only that they elicit what it is I’m wanting at the time.

    Somewhat related… I’m teaching a clicker class for the first time, and it has been so interesting how differently each dog learns. One of the goals of class is to get the dogs to be more creative in trying to figure out what the person wants them to do, and just build the learning relationship, so we’ve been playing shaping games. Some of the “tricks” we’ve been shaping fall into different categories: interacting with an object (like 101 things to do with a box), interact with a person (targeting nose-to-hand), interacting with a place or area (like go to a mat or place marked on the floor), moving a body part (shake your head). I expected the dogs to just be more savvy or less savvy at learning, but it is so interesting that different dogs find some of the concepts way more difficult to understand than others, and have a hard time grasping what this new game is, when they are very savvy with the other ‘categories.’ My Rowdy had a hard time with body movements, and it was absolutely hilarious teaching him to move his head – I had to break it down to teeeeeny steps and have good timing since he was prone to choose some other behavior he was doing at the same time as the ‘thing I was looking for’ – thus far the result is a sort of Parkinsons-esque movement, but at least he’s getting that sometimes our games might involve him moving a part of his body around. One dog, who immediately grasped all the other games, at first just did not recognize the concept that the criteria might have something to do with being near a specific place on the floor. It seems like with shaping, you get to observe more closely just how the dog is sorting out the task on his own, since you’re giving less help than using a different method like luring, where all of the dogs have performed almost identically (like training the target stick, and then using it to teach them to spin.) But definitely, once the dog worked through a ‘category’ once, he was better able to understand a similar game later. I can’t tell if they actually are learning concepts, to pay attention to aspects of their environment they previously ignored, or if they are just adding to their mental list of ‘things that might be good guesses,’ but they definitely each came to class with a unique way of looking at a problem.

  35. says

    I loved the BBC program for lots of reasons. Facinating!

    About words
    I’ve tried (not persistently) to teach my dogs over time the difference between their toys with little success. I think it’s more a lack of my consistent persistence, than their inability to learn. My own lack of understanding about training, and impatience with the task, certainly played a part.

    About actions
    I’ve clicker trained my current guide dog to understand how to do certain things, like; show me elevator buttons, show me security card readers and automatic door openers, point to door handles, and find chairs. She’s a quick study and loves the clicker game.

    At the peak of our training exercises, I noticed we were inadvertently putting a chain of behaviors together, and I pondered how she thought about the words and actions she was learning. To me it became clear that as we entered my workplace, finding the elevator button, then finding the security card reader and pointing to the door handle were a short chain of behaviors, that though separate, behaviors, all lead to getting us to a certain destination.

    It sseems to me, that as you said Trisha, over time different dogs seem to have a different facility for learning words.

    The thing that was most facinating to me about the program, was the emphasis on the thousands, or possibly hundreds of thousands of years of bonding and domestication that have created a unique cooperative relationship between canine and human, and the impact that unique relationship has had on each species.

    I deeply value the learning I’ve experienced through clicker training, behaviorism in general, and ethology, yet I’ve felt there’s an aspect missing in each of those disiplins that doesn’t take into account the experience I’ve had with each of my guide dogs.

    There is a level of understanding that develops over time that comes from spending 24 hours a day, seven days a week together, in all kinds of situations.

    There is an interdependence that is deeply engrained in the relationship.

    Because blind handlers don’t have all the information they may need in a certain situation, and their dog often has more information than they do, there is a silent communication that develops over time between dog and handler.

    While guide dogs are taught to do many things, there are also certain behaviors, that because of their work, they are prohibbited from doing. That training in inhibiting some behaviors, spills over into their lives in other ways. An example is a friend’s guide was asked by her companion to pick up a frisby and carry it outside. Guides are taught not to take objects out of their homes, because the blind handler may not see that they have something in their mouths. The companion thought the dog was stupid, because another dog he knew could do this “simple,” behavior. He wasn’t taking into account the training of the dog, and why that training inhibited the dog from performing the behavior.

    A question that behaviorists have given me pause to think about, is how much of what we teach a dog inhibits them from doing things they might otherwise think to do on their own?

    With guides, we teach them to think and problem solve, but only under certain circumstances. A dog that is too “busy,” or thinks too much, may not be well suited for the work. Being still and quiet is an important part of the job as well.

    Thanks for turning us on to this program!

  36. says

    My little dog is quite intelligent, I think. She learns quickly, reasons well and is generally pretty quick on the uptake. That said, she is not terrific with verbal labels. Much better at following my “eyes pointing,” hand gestures etc. I haven’t really worked with her too much on labels … but she doesn’t seem to get “find Bruce” or “find Carolyn” although she certainly understands exactly “find it” which refers to a treat. I’ve been working with her on Left and Right recently, trying to give minimal visual cues and concentrate on verbal. She’s making progress but I wouldn’t say she’s quick on this one.

  37. DebC says

    This is all very fascinating – I love hearing about other’s experiences with their dogs. I think that Bo (border collie) does have some concept of words – at least I thought he did. Now I am not as positive because is it the words or the action? I still think he has some words….if I tell him to “go find (insert name)” he does actually go find that person. If I say “go get your chicken (or ball)” he does come back with that item. But if I say “time for bed” is he going to his crate because it’s bed time, or that ‘bed’ means go to your crate = get a biscuit? One big difference between Bo and Dozer (Lab mix): if I point at an item Bo will look in the direction I am pointing, Dozer will look at my finger/hand.

  38. Jane says

    I’ve recently been discovering that the capacity to learn a lot of words does not necessarily track with the dog’s overall intelligence, or at least intelligence in different areas besides language. My previous dog, a Lab/husky mix, was a poor problem solver in general and took a bit longer than average to learn new tricks, but had an immense vocabulary, many of the words picked up “accidently” (like when you suddenly realize one day that they react mightily when you say “toast” or “popcorn”, without ever having meant to teach them this). She truly was one of those dogs in whose presence you found yourself spelling things, rather than saying the full word. Yet this same dog could not in twelve years figure out how to get out of a room by pushing on the door that was merely ajar, and not closed.

    In contrast, my new dogs (both rescued cattle dogs) learn tricks lightning-fast, and have quickly solved every food puzzle toy I’ve ever put in front of them, but I’d be surprised if they’ve learned more than about fifteen or twenty words in the year I’ve had them. It still shocks me a bit. I really think they are incredibly attuned to subtle body movements and the overall tone of my voice, rather than the specific words, and that this is how they learn. I did try to teach one of them specific toy words, and I think you’re on to something with the “call to action” rather than “label” idea–every request to “get your XXX” was met with an enthusiastic leap toward whichever toy he preferred at that moment, which usually was not the one requested.

  39. says

    Interesting post and very provocative comments here. Makes me think a lot about how Puzzle and I interact at home and in the field and the sense she seems to make of words that might be different than the sense *I* think she makes of them.

    “Find” — that verb she loves, she understands is associated with human scent located all the way to its source. She also understand that, when presented with a scent article, “Find THAT” means ignore every other human scent and find only the one in the scent article. She does seem to understand that THAT can change from day-to-day, search to search, and that the whole sequence is about taking a good wuff of the scent article, memorizing its signature, and then locating that scent again in a host of others.

    At this point Puz can select some toys by name when I ask her to, but not all of them, and this specificity seems to have come to her with more difficulty than search commands. Certainly, we’ve worked much less at this! “Pink ball” she gets — but she may struggle to differentiate between the fuzzy octopus (Pooz-a-pus) and the fuzzy snake (Sneaky Snake), both of which have similar textures are long. Plush Leo she has learned — it’s her favorite toy of all, and she knows when I croon “Go get Leeeeee—ooohhhhhh” what to bring back, including all the bits that go inside.

    She knows “Where’s …?” means we’re looking for something. She has begun to find the other dogs or cats in the house by name — some more easily than others, which suggests they have some marked significance for her I don’t recognize? knows them better? likes them better? — but her finds of them aren’t with the consistency we see in the search field finding a human.

    Puz does know some sign language. All the basic commands — sit, stand, lie down, wait — but she also knows “ready to go?” and “good girl, pretty girl” and “I love you.” The first time she smiled and wagged when I signed good girl and pretty girl and I love you in the same way she smiles and wags when I say it, I was thrilled. That said, I know 50% or more of that message is carried in my body language, which is probably identical in both instances. Heck, maybe there’s just a twitch of endorphin when I say it or sign it, and she smells the I love you rather than sees it.

    I love the peeks and maybes such questions raise.

    A very interesting thread!

  40. Rebecca says

    My dog Maddie knows,”Go find Cleo.” as a cue to find one of the cats who needs daily medication, but would prefer to skip said medication. However, if Maddie is in a room that does not contain Cleo, she will show me where any other cat in the room is located. Which is sort of weird, since she differentiates if Cleo is in the room. Somehow, she has generalized find Cleo=action involving alerting on a cat, even though I never reinforce her for finding the other cats.

    Maybe annoying cats is self reinforcing for her, and hope springs eternal. Though about every 10th time, she will run out of the room and entice Cleo to chase her back to where I am (they are buddies). I haven’t figured out what the difference is in the two scenarios, except maybe whether or not Maddie knows exactly where Cleo happens to be at that specific time.

  41. says

    A busy mind keeps one out of trouble!

    Very interesting. I’ll have to do some testing at home. Off the top of my head.

    I find the males in my home are more action driven: Pug, Auzzie, Beagle where as my female beagle LOVES to learn specifics along with actions. She’s a brainiac.

    I wondered about that a while ago when watching THEM try to communicate with ME ;

    ie…they want me to pick something up like a ball, tug toy or some food that they can’t get to (I play hide popcorn), they look, sit, point with paw or bark once until I figure out what it is they want. If all the toys are in a pile, the boys will wait until I pick up the one he feels like playing with…

    Using their , body language for actions or states of mind/emotions…I can’t think of any way they use either to “label” something FOR me to discriminate.

    Maybe that’s why some of my dogs look to me like they don’t see the point of learning something specific, we already have a communication in place that works just fine :)

  42. Kim says

    My eldest dog has a heck of a vocabulary. I used to think that she was learning more of how things were said and less of what was said, but I did some proofing both with myself and others saying things in several different tones and her reactions were largly the same.

    So far she knows several toys by name (“get your” rope, ball, ring, frizbee and bear (all fuzzy things are “bears”)), bath (she hates them), hungry, thirsty, outside, potty, walk, bone (chewing time!), bye bye (car ride), bed, out (get out of the room), move (get out of my way), find it or find + toy name, not to mention all the trick, agility and obedience words (plus the “sign language”) she knows too.

    But funny enough, she doesn’t seem to attribute names to living things…with the exception of a very unusual cat I had for 9 years. She knew Foley alright, because he make sure she did ;) However, I’m the only living thing in this world that matters to her. The current cats and my other dog are definately not intentially on her radar!

    Kim

  43. says

    My BC Fenway is very vocal. The whines and barks are clearly his signals to me that he wants me to take action: throw the ball, I need to pee or I’m hungry. That’s simple and not at all remarkable.

    But there are times when he yodels at me. These are connected with happy play. I’ve begun to yodel back to him and we’ve been having “conversations”. He seems utterly delighted that I am talking “dog” to him. During these times he seems to stop wanting something and is just interested in sharing some talk. I’m thrilled at this development because it seems to me our bond is getting even tighter, if that’s possible!

    My former BC used to do some groaning (that’s the best way I can describe it) while he was getting cozy in bed. I would mimic his sounds and we would do a duet that could last for minutes, and get softer and softer until he drifted off to sleep. If I changed my pitch

  44. Kerry L. says

    Most interesting ideas. I would say that Walter associates my words with actions or the anticipation of actions. Sit, down, out, supper, ride, ready?, etc. I’ve deliberately avoided using words like treat and cookie just to avoid his over-exuberant reactions.

  45. lin says

    Old Pup seems to know actions much more than nouns, and body language and tone much more than words. She’s not interested in toys, and though she does seem to know my husband and me by name, when he tells her, “Find Lin,” if I’m walking with them, Old Pup doesn’t seem to know what to do. So she probably associates “Find Lin” with a hide and seek game.

    I wonder if, because scent is so important to dogs, that they could learn nouns better if they were associated by scent. If, say, each of their toys or objects were dabbed with a different perfume, and we gave them a scent strip and told them, “Find”

  46. Katie says

    When people ask me how many signs/cues my puppy knows I never consider half of them when I answer. I, too, finally took the time to sit and write them all down. One in particular caught my interest. Lily is only four months old and recently figured out how fun it is to grab the end of the toilet paper roll and run around the apartment until she reaches the couch. As cute as this looks in the movies, it’s definitely not cute after the first time. (Well, it kind of still is but I’m not about to tell her that.) Eventually I taught her “out,” which basically is the cue to walk out of the bathroom and not come back in. I never used it in any other context but started to wonder if I could. At four months and one week it definitely works! Just for fun I tried it from the kitchen and with no hand cues or body blocks she walked out and sat in the living room. Lily’s repeated this behavior in different rooms and different homes. I’m continually amazed by how smart she is!

  47. says

    This article really got me thinking about what my dogs understand. I would say mine primarily know things by the action and not the item. I think this is because it’s what I have taught them from pups or when they first became part of our family. Now I am going to have to observe mine and see what they react to more.

  48. Susanna says

    Great article, really anjoyed reading about how your BC play and herding gets mixed together, since my BC is exactly the same. Runs like mad and then ‘attacks’ the other dog and when they stop he tries to get them to run again. My terrier is so sick of him but is not ‘strong’ enough to tell him off perperly to stop him doing it

  49. Mihaela says

    I am thinking that dogs normally (i.e without the two-leggeds around) do not use vocalizations to identify objects. It may be that smell, taste, texture or a combination, are the equivalents of words in their minds for that purpose. So it’s very likely that they don’t really regard the words we say as labels for those objects. It might be more like “when the two-legged makes a group of sounds including ‘ball’, I have to DO something about ‘the ball’: run to it, touch it, fetch it etc”. On the other hand, even at this level, they may have the capacity to generalize (much like we do when we transform words into notions). Otherwise, how would they know, for instance, that we mean ‘ball’ for a small round evenly colored thing tasting of rubber (i.e small rubber ball), as much as for a larger round patterned thing with rough texture (i.e tennis ball)?
    Sorry, Trisha, this is what happens when you have a blog: you have to read through stuff like this and act interested…

  50. says

    A little late in response, but I think there is a lot of variation in dogs as to whether they understand words as relating to objects or actions. I believe the variation to be essentially innate, and not directly related to the intelligence of the individual dog.

    My oldest Staffordshire Bull Terrier, Elmo, came to me as an adult dog (almost 4) with very little training. I trained and competed with him in obedience and agility. He certainly understood words as relating to actions, but I sometimes ran into trouble training him because he had a different understanding of a particular word or phrase than I expected. For instance, when I went to teach him the directed retrieve for Utility, I could not get him to retrieve the glove. Which made no sense, he loved to retrieve, and had no hesitation about carrying around any object. Finally one day I tossed down the glove and said “take it!” and he went over to the piano and looked very deliberately towards the top, where his dumbbell was stored out of reach. I don’t know how much more clear he could have made it that “take it” to him meant the dumbbell! So I tossed the glove again, and said “glove!” (a new word to him), and he happily retrieved it. He would retrieve it on any command EXCEPT “take it”.

    Now I’m working with a bitch of the same breed, a few years younger, and I’ve had her and trained her from 13 weeks of age. I think she might know what “ball” means, though to her, that would only apply to a tennis ball, since that’s the only toy she has any interest in. Beyond that, I don’t think any words relate to objects in her mind, while she knows many commands, they all relate to actions. She’s a brilliant dog, one of the best canine problem solvers I’ve ever encountered, and she understands a lot of phrases I never actually tried to teach her, but they only relate to actions.

    So, 2 dogs, same breed, trained by the same person, but with very different understandings of words. And the one who relates words to objects the best is the dog that had little early mental stimulation. To me, that says it’s an innate variation.

  51. Michelle says

    I recall one evening when my shepherd-collie mix Sophie was about three years old. Unbeknownst to Sophie, I was planning to leave her at the kennel the next day and so had no intentions of giving her a bath, even though she was smelling a bit doggy. As I was petting her, I observed, “P.U. You stink!” Sophie looked at me, sighed, and obediently took herself into the shower for a bath. I had never intentionally taught her this, but I guess she’d heard “P.U. You stink!” before I got out the towels and dog shampoo enough to act on the verbal cue.

  52. Deanna in OR says

    Regarding dogs interpreting written or drawn symbols as having meaning or as cues–
    A couple of years ago, a newsletter from the Assistance Dog Institute (now Bergen University of Canine Studies) had an article about how they taught several of their service-dogs-in-training to respond to big “flash cards” with either symbols or simple written words as cues for behaviors They apparently could distinguish between several different cues, all on the same large white card stock. It was fascinating. One of these days I might try this with my collie and my BC.

    My collie, Willow, is really good with “Right” and “Left” in agility–sometimes when I get mixed up and say “jump right” while my body is screaming “jump left”, she’ll start to change leads to the right and then realize, no, she really meant left and save my run–while barking at me to Pay Attention, mom!

  53. Mary says

    I, too, have a border collie that seemed confused when I tried to teach him the names of toys, and I ALSO came to the conclusion that he thought the word was a command for action. After all, most of the words I say to him have an action associated with it. “Upstairs” means to go upstairs. “Outside” means to go outside. “Potty” means to urinate or defecate. And when we’re doing agility, “jump” means to go over a bar. “Tunnel” means to run through the tunnel. And when herding, “come bye” means to run clockwise, etc. I’m happy that someone else came to the same conclusion I did!

  54. Steve B says

    Well, I don’t have an answer, but my experience is my last dog could distinguish between 8 different toys. The game became “go get toy X,” which he would bring and then wait for me to tell him to get something else. But it started with him bringing me a toy and my saying “Good X!” as we played with each toy in turn.

    Watching my daughter’s language develop, I think it could be argued that people also start out interpreting words as action, but learn to associate them with an object. I’d like to believe that after a time, Talbot (who, by the way, was a Brittany/English Setter/(I think) Collie) made the transition to understanding “ball” or “tug” or “bone” or X did mean the object and not an action, but obviously I can’t say for sure.

  55. says

    Is it possible that since Willie’s life and training is oriented around action and cues as would be expected from a herder, that action-based brain development was dominant for him? I’ve had a Dane who seemed to easily distinguish over 200 nouns and verbs, and he was a typical Dane, not to0 action oriented.
    In fact most all of my danes seemed to have no trouble between a noun and a verb, unless I just didn’t really understand how I was communicating with them. And were awfully good self-initiating problem solvers.
    But Danes aren’t known for a high level of action oriented and human orchestrated activities. And these guys didn’t really have any “formal” training to learn words, they just seemed to on their own via hanging around me. Companions like the dog that learned so many words in the BBC show.
    I don’t know, you’ve just got me wondering if dog’s brains develop or don’t develop word/grammar recognitions according to their lifestyle and training purposes?

  56. trisha says

    I think that might be the exact issue: not only were all of Willie’s cues action oriented, but he is 100 % an action oriented dog. Not a ‘thinker’ at all… I have to work hard on stopping him from anticipating. One of my cues to him has become (almost inadvertently) “Take a breath.” Good observation from Great Dane Service Dog.

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=""> <strike> <strong>