Today’s question was inspired by the comments of sheepdog trainer Patrick Shannahan, who argued at a recent sheep dog clinic that I needed to ask Maggie to “take responsibility” for her behavior.
Eeeps! Isn’t that perspective one that trainers and behaviorists (me included) have been arguing against for decades? In most contexts, it is the last thing we should be doing. It calls to mind the ubiquitous comment “I KNOW he knows better!” when some poor dog pees on the rug because he’s never been house trained. Or the standard perception that dogs are jumping up even though they “know” they shouldn’t—and if we just stopped coddling them they’d straighten up and behave.
Perhaps the best extension of this misapplication of asking dogs to “take responsibility” is Will Ferrell’s Saturday Night Live skit on his new dog training method, called “Dissing Your Dog.” In it he advises owners to use sarcasm and humiliation to create an obedient dog. Check it out if you want to a good laugh (but don’t go there if you can’t handle his level of social parody… he begins by saying something like “I’ve tried all the standard dog training methods with no success, like screaming, beating, etc.).
So, yes, I have spent decades advising against people “putting responsibility” onto dogs for the dog’s behavior in most contexts, when all that did was befuddle or frighten an already confused dog.
And yet, sometimes basic rules are made to be broken. I’m reminded of my favorite line from the movie Babe, in which the barnyard animals discuss breaking into the farmhouse. Someone—the horse, the cow?—says “We barn animals can’t go into the house. That’s the rule.” The duck responds “That’s a good rule. I like that rule. But this is bigger than rules!”
Sometimes, rules are indeed made to be broken. In my experience, the basic rules like “don’t repeat a command” and “forget about whether your dog ‘knows’ what she should do” are essential to beginning dog trainers. The rules are there for a reason—to help people avoid common mistakes that at best lead to a confused dog, and at worst destroy the relationship between the animals at both ends of the leash. But perhaps, in some contexts, breaking a basic rule could be the best thing you could do IF a set of important conditions are met.
Those conditions are best understood with the example from the sheepdog clinic: Maggie and I were working on driving (the sheep away from me) at the clinic, and while doing so, Maggie kept creeping to the right and moving the sheep off the line I wanted her to drive the flock along. Each time she shifted right, I gave her a cue that put her back to the left where she should have been. After several repetitions of that, Patrick suggested that I stop ‘fixing’ it, because Maggie already knew what she should have been doing. “She knows what she should be doing,” he said, “she just rather not do it. Rather than telling her what to do, why don’t you just communicate to her it’s her job to ‘fix it,’ not yours?” And so, I said “Maggie!” My voice was quiet, but nonetheless expressed frustration. “What are you doing?!”
Maggie stopped and turned her head to look at me. She then put herself back in the correct position, and drove the sheep in a gorgeous, smooth line exactly where I wanted them to go. Ah. Interesting. I didn’t fix it, she did.
Of course, in one sense, I simply gave Maggie a verbal correction (essential in training sheepdogs, more on that in another post if you are interested) for her behavior, and she stopped shifting to the right, the side that she had preferred because it was more like following the sheep rather than taking charge of them. However, what’s most relevant to this discussion is that I was not telling Maggie what was “right,” but rather, telling her it was her job to figure it out. “She knows what she should do, Trisha,” Patrick said, “she just doesn’t want to do it.”
Patrick was right about Maggie. She knew exactly what needed to be done, and began doing it as soon as I called her on it. But… and here’s the kicker: How does one know when a dog “knows?” Look at all the trouble people get into thinking that their dog “knows” better? Visions of the viral video of the hapless Retriever, Denver, falsely accused of stealing the kitty treats come to mind. The perception that dogs “look guilty” over some perceived misdeed has been thoroughly negated by researchers like Alexandra Horowitz and Julie Hecht, whose research showed that a dog’s “guilty look” had nothing to do with his actions, and everything to do with appeasing an owner.
I’ve thought long and hard about this (and discussed it with some wise and experienced trainers, thank you all), and come to the conclusion that “breaking the rule,” and assuming that Maggie did indeed “know” what was right and what was wrong worked because: 1) She’d proven to me over and over again that she understood the concept of taking her own initiative to hold the sheep on a line that I established, 2) Taking charge of a flock of sheep is something she was bred to do, and dogs like her are actually much better than us at “reading the sheep,” and knowing what needs to be done, 3) Patrick has had decades of experience training working sheep dogs (he won the Nationals with Maggie’s father, Riggs) and if anyone would know when a dog “knew” what to do, it would be him, and 4) I tried it and it worked perfectly and instantly.
If similar criteria can’t be met, I would argue that one should never assume a dog “knows” what is “right.” But in this case, breaking the rule was the right thing to do.
The purpose of this rather long introduction, (apologies), is to ask you, dear reader, what basic dog training rules you think can be broken by more experienced trainers, and when and why. I’m all ears.
[Dece 4, 2015. I’ve added this addendum, based on the great discussion that has followed in the comment section: Oh how I love these discussions! Because of them I had an insight this morning while playing fetch with Maggie. I was continuing to think about my “What are you doing?” to Maggie while practicing driving, and had one of those great Ah ha! moments. First, it occurred to me that “asking Maggie to take responsibility” was similar to waiting for a dog to perform the right action from an operant perspective. Think clicker training, while you wait for a dog to move the correct way, rather than luring them. In both cases, we are asking the dog to figure it out what “right” was for itself. The primary difference between what I was doing with Maggie and waiting for the dog to do what you want in clicker training, is that rather than saying “wrong,” which many use to communicate to a dog who might need some direction, I said “What are you doing?” What is most relevant between those two cues is the tone. If I think a dog needs help while clicker training I’ll say “wrong” in as neutral a tone as I can. When I said “What are you doing?” to Maggie, it clearly had conveyed some frustration. (But keep in mind, it was ridiculously mild. She was working 100 yards away, and I didn’t yell it. I just changed the tone of my voice.) Because it wasn’t said in a neutral tone, I think it qualifies as a correction, which is where trick training and sheep herding do indeed diverge. Working sheep the way a dog wants to is the most reinforcing thing the dog can do, at least in the short term. You simply can’t let the dog do what it wants, like you can with trick training, and wait for the correct behavior. The dog is already being reinforced with something more powerful than anything in your repertoire. So, yes, sometimes mild corrections, like I gave to Maggie, are essential. But the motivation of my words was to communicate to Maggie that 1) what she was doing was wrong, and, 2) she needed to find ‘right’ herself, rather than me ‘luring’ her verbally into it. In both cases, the point is that it is better for the dog to figure it out for themselves, rather than you ‘fixing it’ (including using a lure). Make sense?]
MEANWHILE, back on the farm: Maggie and Willie had a blast yesterday playing with some new friends. Maggie had never met anything resembling a Bouvier, and initially lept off the edge of sanity into a full out panic when she saw her, but soon was playing “race horse” with Ready (Bouvier), Nellie (adorable Standard Poodle, not pictured) and Willie. Here they are about to begin another race around the pasture. Could Maggie look happier?
Willie hasn’t gotten to run like this for months, but his shoulder and paw both seemed healed up enough to let him run for awhile. He was so happy, and I was full of gratitude that he could run with other dogs again. He did go on leash after I thought he’d had enough. But oh, they had fun, and we had a blast watching them.
Maggie would like to thank the deer hunters who couldn’t find the buck they’d shot last weekend during hunting season. We found it beside our wood’s trail, long after the coyotes, raccoons and possums had eaten up a good deal of it. (Thank heaven for obedient dogs, who dropped the rotting strips of deer hide out of their mouths when asked.) We’ve avoided the trail for the last week, but Jim sawed off the antlers and Maggie is very, very grateful.
Regina R. Allen DVM says
Interesting post – great stuff for thought. Here’s my take with obedience dogs … If there have been enough (a relative term!) correct and rewarded (not lured) repetitions that the dog “knows” to sit straight in heel position at a halt, and the dog sits crooked during heeling practice, then I’ll tell the owner to stop fixing it and wait the dog out to see what he does. If he fixes it on his own, then great – praise, give a cookie the first time or two, and resume heeling. If you continue to give a cookie every time he fixes his crooked sit (even though he did it without prompting), you end up creating a behavior chain of heel-crooked sit-move butt in-cookie-heel. So after one or two cookies, stop rewarding, and try to set the dog up for success for a straight sit, such as heeling next to a ring gate. Praise that straight sit, reward, then heel on. Repeat enough times that the straight sit on the first try becomes muscle memory, and then work it away from the ring gating to see how you’ve progressed. The dog “fixing” the sit on his own shows me that he “knows” it, and just needs a little more work to get that muscle memory down to achieve straight sits 99% of the time. 🙂
Great post! I think there is a distinct difference between training a dog to do a “human motivated behavior” (like sitting in a particular position in relation to our legs) and what amounts to molding an instinctive behavior like herding, field work or even SAR/nosework. Their instincts and talents are often better than our thoughts. Ahhh… if I had a nickel for every time the herding trainer has said, “You need to just trust your dog. Shut up and let him work.” As you point out, one excellent clue that you have made the right decision is the dog’s immediate response.
Sooo glad you posted this. I am training my (rather difficult) border collie for herding, mostly on my own because I reject the usual methods used. Someone posed this thought to me in the context of herding: if you are sure the dog knows what to do, then why would they not want to do it? Esp border collies, who so love to be right. Not saying it never happens, I’m just struggling with the concept myself. (And yes I would love to hear more about your thoughts on corrections in sheepdog training.)
OK, in this example you said Maggie would rather follow the sheep than take charge of them. Maybe that’s easier? I don’t know. You switched from saying “come by” several times to “what are you doing?!” Is it possible Maggie really wasn’t sure what to do and it was just the last instruction/correction that convinced her to get back in the position you had already told her to get into (several times)? Of course with herding, there’s always the possibility the dog is sure the stupid human is wrong, and (darn!) often they are right. If you said the correction quietly, why was that different than the flank cue you gave her? We all (including the trained dog) know what “come by” means, but what does “what are you doing?!” mean? What does Maggie think it means and how did she learn that? Similarly, I always tell my students “No” is not dog training. It doesn’t give the dog any information about what they *should* do. I know mostly in herding “what are you doing?” means “stop that!” (just like “no” does), so I guess a dog could learn to stop and go back to whatever they were doing before. But I’m just not clear on how that translates to the dog “taking responsibility”. You are still somehow telling them what to do (or what not to do).
As to your question: I rarely break rules (although I only follow good rules!). Maybe this counts: I might take shortcuts sometimes to make a point or demonstrate something, e.g., if I am shaping a behavior I might push the dog a bit faster than I should so a client can see the process and/or the start of progress (I guess my criteria being that I explain to them I am only doing it for demonstration purposes).
When teaching basic obedience, I don’t encourage clients to use verbal corrections or no-reward-markers. Because getting the typical owner to focus on, and mark and reward when their dog is doing something “right” already requires a major shift in their thinking. So introducing the idea could set them on a slippery slope down a well worn path.
But I struggle with believing there is any real harm in using them if (1) you’re proficient at using positive reinforcement, (2) the behavior has been reliably proofed, and/or (3) the dog is operantly searching for the correct answer. Especially in the last case, it seems like withholding the NRM is like withholding part of the information they’re seeking.
Ohhh….I so loved this article!! I have attended several of Patrick’s clinics…when I was younger….and so enjoyed his clinics! I so miss having Border Collies…I miss being ‘young’ enough to have Border Collies and to ‘herd’….:) Patrick is a wonderful trainer/teacher/trialer ….you took me back to some of the very best of times for me 🙂
Thank YOU…and Maggie is beautiful….:) I remember Riggs 🙂
Just a question: Isn’t it possible that Maggie just thought that was how she should be running? I mean, instead of her thinking, “I’ll go right, and even tho I know I should go left, I’d rather go right”. She could be thinking, “I’ll go right, then Trisha will ask me to go left, and then I’ll go left” since that is the pattern she had learned. When you changed that pattern she then did what she thought she should do. Like you, I don’t like the idea of a rebellious reason for dog behavior. And honestly I would think it takes a lot of effort to think that way, especially in the heat of running a flock of sheep. The technique to fix it would be the same, but the reasons for it happening are different.
satarupa bhattacharya says
A few years ago I adopted a wire haired doxie who had been found as a pregnant stray around 9 months of age. She was (and to some extent still is) incredibly shy and skittish around most people but somehow took to me from the first day, and only me. She would have nothing to do with most other humans, including my husband, who most shy animals seem to gravitate to.
A wonderful behaviorist I consulted, told me to forget about correcting Kulfi when she jumped on me or begged for food or jumped on to furniture – many things I would correct w/ the other dogs in the household. I was to encourage her to do anything where she took initiative and/or approached me instead of running away. I think the idea was to gradually build her confidence. And it really worked to bring Kulfi out of her shell and for that I am grateful.
Your first paragraph said “Lassie.” I assumed you meant Maggie but just in case. I can’t wait to read the replies to this great question. I have wondered about some of the “rules” when it seems there are a few contexts in which a stern “knock it off” or a genuinely puzzled “how could you” seem really effective. I can’t wait to hear from the trainers.
Stacey Gehrman says
Regarding holding a dog accountable for his/her behavior I also will do this IF the dog has proven to me in many situations that he knows the asked for behavior and can apply it. A dog I show in Rally loves the work but occasionally will sit out of straight position. He and I have refined the “correction” to a sour look and sometimes a “where are you? “. My tracking and trailing dogs, now long gone, would stop if an animal crossed right in front of them. I stopped too, they would look at the animal, hold still a moment, then continue on. They had made a decision to continue with their work. When they were coming along in their training they would be tempted to deviate and when that happened I would do the “negative beep” noise and say their name. They would glance at me and continue the track. I ran Walker hounds, a bloodhound and GSD’S. I could give countless examples of asking the dog to make the right decision. The preliminary to that though is systematic and thorough training in man different venues. Lots of work.
Trish, I am glad to see that Willie was well enough to run with the other dogs, and I hope his test results are OK.
When is it acceptable to break the rules? When something else works better (keeping the dog’s well-being in mind, of course). Ultimately I rely most strongly on how the dog responds. An old dog training book had a phrase that stuck with me: the dog is the proof of the trainer.
An example: A few weeks ago Red Dog spotted a critter in the woods and went on an unauthorized adventure. Over the next few weeks we revisited by-the-book recall basics: simple on-leash recalls; on-leash recalls around critters; off-leash recalls in increasingly distracting environments; dragging the leash in the woods; and then eventually off-leash recalls in the woods (with particularly yummy treats).
Anyway, another setback caused me to deviate from “the rules”. After behaving beautifully on several hikes, Red Dog spotted a critter across the creek and initiated pursuit. When she did not stop on command I ran after her, screaming like a maniac. I am certain that my behavior would have upset small children, had any been present. Red Dog took notice and turned back towards me, giving me her “Huh, the human appears to be upset” look. I stopped, gave her a firm but friendly recall command, and she responded immediately.
I have never seen a dog training book recommend “run and scream like a maniac” for recall training, but in that situation I felt that it was necessary to accomplish two goals: 1) turn Red Dog’s attention from the fleeing critter to me; and 2) convince Red Dog that the “stop” command is not optional, even on the other side of a creek.
Later on in the hike Red Dog flushed a squirrel. She was close to me in a safe area so I let her bound after the squirrel for a while, and then called her. She recalled with only a momentary, wistful glance back at the squirrel. Did running and screaming earlier help? Hard to prove, but I think so. In this case, I think the training message was that the human is willing to jump the creek and surrender considerable dignity to enforce a command, coupled with a smidge of “Hmm, ignoring commands makes the normally calm human bat-feces crazy”.
For the record – these off-leash hikes take place deep in a 6,000-acre wooded park, far from any roads.
With regard to your criteria: 1) I have worked with Red Dog on recall frequently and successfully around critters (and at the dog park, the other big distraction); 2) Consequently, the recall command is well within Red Dog’s skill set; 3) Patrick Shannahan was not around, so I had to rely on my own decades of experience (no herding titles for me, though); and 4) It worked, without any evidence of trauma to Red Dog (who is a pretty robust dog, generally).
I would have taken a different approach with the less-robust Sammy. Not that the Sammy is inclined to run off – she rarely leaves my side, even off-leash or at the dog park.
I mentioned over Thanksgiving to my sister-in-law that my 5- and 7-year-old nephews (her kids) are incredibly responsive to her “cues,” regardless of the context – I believe what I said was that if she lifted her little finger they would obey. Her response was simply, “They’ve seen me melt down a few times.” I’ve never seen her punish them, and I’ve seen plenty of praise and thank yous. I can only surmise that the puzzled, scared or hysterical adult human can be a pretty impactful instructor/motivator.
Margaret McLaughlin says
What really gets confusing to me here is which set of “rules” I’m playing by: traditional dog training or operant learning. I started training dogs 20 years ago with j & p, not using food at all until I had graduated from Basic & moved into the competition class, & then not that much. Then lure/reward with a clicker in 2001, then the epiphany of shaping & capture in 2010. I’ve got a LOT of competing methods jostling around in my head.
I had screwed up Nina’s reset–lining up by sitting in heel position–with competing cues. I thought the cue was ‘sit’. Nina thought (determined by her behavior) that the cue was the slight upward leash pressure I didn’t even know I was using. So when she was off-leash she didn’t reset very well, & my verbal ‘SIT!’ was becoming a punisher–decreasing the frequency of the behavior. Facepalm time, but it took a great instructor to help me work it out. Exit one poisoned cue, reteach the behavior with a new cue. And a tennis ball. She now resets with a bounce & a tail wag.
But Nina “knew” how to sit. And an awful lot of the people I know would have said that she was blowing me off, & I should have popped her into position using a tab when she was off leash. And changed her flat collar for a choke collar or a pinch collar. Or a shock collar. But the responsibility was MINE. I had allowed her to think that the behavior that would earn reinforcement was sitting at the leash pressure.
If the principle here is (thank you eileenanddogs) that all behavior is something that I have either reinforced or allowed to be reinforced it comes right back on the trainer every time.
With herding (3 species chess) you have the additional problem that the actions of the sheep themselves may be aversive or reinforcing to the dog, & that if the trainer had full control of the sheep she wouldn’t need the dog in the first place….So it gets a lot more complicated. And it moves fast.
So, in a nutshell, go ahead & break the rules of traditional training. It’s a new world out there, & way more fun to achieve a traditional end with new means.
Jenny Haskins says
I think what this all boils down to is ‘micromanaging’ destroys self confidence.
It can also destroy willingness to work of co-operate.
At least I KNOW it can for humans — ‘Well why don’t you do it yourself?’ (and ruder comments/feelings).
I strongly suspect that his does for dogs to — generally they give up.
Jenny Haskins says
More thoughts re micro-managing.
It is VERY much like “nagging” 🙁
I don’t know if it is about breaking rules but I was thinking about mantrailing lessons I take with Spot. When he is trailing, I can’t ‘see’ what he smells. I walk him, the trainer is with me an she knows where the trail is, I don’t. So all I have to do, is read Spot to see whether he is still working or of doing something else (did I mention there are rabbits where we train??) . Anyway, even rading him is diffcult enough and requieres alot of concentration. Now at first, when I saw him starting to do something else, I would feel irritated. But my trainer mostly sees it as evidence that the trail might be too difficult. Gradually, we’re learning. He is learning not to go too fast (because then he ‘overshoots’ on crossings) and I am learning to put him back on the trail. Now the only way to do that is to take him back to the last place where I was sure he had the scent and let him pick it up again. What I find so amazing is to see how PROUD he can be when he solves a difficult puzzle. He really loves doing it. Off course there is a reward at the end, but in the meantime it is up to him to figure it out and up to me to guide him, help him along but I can’t ‘train’ him to use his nose properly since he is already so much better at it than I am ? .
O gosh…. sorry for all the typing errors…
What I forgot to say: when he does go astray (like last time when he flushed a rabbit out of the bushes) the only thing I can do is correct him, tell him to stop sniffing and by taking him back on track ‘hoping’ he’ll start to track again.
Now he is off course motivated by the target (we train on finding the nice man/woman with something very nice to eat, sausage or cookies or…) but he also likes it for being able to do it. And perhaps there is something in that if a dog truly enjoys something, the want to learn how to do it and there is nothing wrong with occasionally telling them that they are on the wrong track.
Great comments so far. To jewel, who was asking why a BC wouldn’t do the right thing if they know what it is (because they BC’s “love to be right.”) Great question. In this case, Maggie does indeed love to be right, but she also doesn’t like putting a lot of pressure on the sheep. Long story, not sure I can do it justice here, but “pressure” is when the dog moves forward into the ‘personal space’ or ‘bubble’ of the sheep. You know that feeling when someone stands too close to you so you back up to relieve what feels like pressure? That’s the ‘bubble,’ and it is exactly the place in which a BC needs to push against to move the sheep forward. Maggie is somewhat uncomfortable in that zone; she’d rather work far from the sheep and, rather than pushing them, play defense and stop them from going where they shouldn’t. In other words, a dog can either take control of the sheep by pushing into the edge of that personal space and moving them exactly in the correct direction (preferred, especially on Midwestern sheep who aren’t very flighty) or let the sheep move forward, but block any direction they don’t want the sheep to go. Maggie prefers the latter, I need her to get better at the former. Making sense so far? So my most important job with Maggie is teaching her to become more comfortable putting pressure on the sheep. Her fading to the right was yet another example of her trying to work the sheep while avoiding the area with the most pressure. We’ve been working on it for months, and we’d also been working on “once I establish the line, it’s your job to keep the sheep on it yourself.” That’s why I am confident that once I asked her what she was doing to immediately starting doing her job correctly, even if it was a bit harder for her.
To gayla: I agree 100% that sometimes dogs are desperately searching for the behavior that you want, and it’s actually punishing to them to withhold information. Such a fine line, isn’t it, between helping a dog too much and yet not torturing the poor thing when he or she is clearly struggling? I have seen dogs, trained with “100% positive reinforcement” who begin stress whining or become frantic because the owner is waiting for them to do something that they can click. I suspect we all have a tendency to do one or the other a bit too much (I’m too much of a helper, and work on it on a daily basis). But that’s what makes good trainers, right? Knowing our own behavioral predispositions and how that links up with a dog’s…
To Bruce: I love your story. Yup, I too would never recommend “screaming like a maniac” after a dog, and like you, agree it wouldn’t work anyway with most dogs. But I love that you had worked so hard to work Red Dog up to distracted recalls, and especially that you later let Red Dog 1) run after a squirrel for awhile and 2) gave her the moon for her subsequent recall. I’ve often added my own riff to that: Let them run after something, ask them to stop, then stay OKAY! and let them continue the chase.
To Margaret: I LOVE that you thoughtfully deconstructed what was going on with your obedience dog and her “Sit.” Well done! A lesson for us all to think about what cues are truly relevant to our dogs–so very often not what we think.
HFR: Good question about a possible behavior chain as a reason for Maggie’s behavior, but see my answer to jewels. Those behavior chains are so easy to fall into, and I’m glad they’ve come up in the comments–I should’ve mentioned them actually in the post itself, so thanks to the readers who mentioned them. However, as I explained above, Maggie simply tries, in all contexts, to avoid putting pressure on the sheep, and that’s the best and simplest explanation I’ve got for her behavior. Training working sheep dogs is SO complicated, and includes so many variables, that one’s head can start spinning sometimes, but the more I understand about the interaction between the sheep and the dog, the clearer it becomes.
One of Jack’s many self-assigned jobs in the household is to tell me when one of the other pets breaks house rules (he is a perfect example of the fact that smart working dogs without a job invent their own). So for instance, when I forgot to close the treat cupboard and Maddie dragged out a bag, he barked for me rather than joining fee in trying to open the bag.
One of the rules is that dogs don’t touch foid on the coffee table. Jack is 8 and always follows the rule.
So when I caught him nosing at a plate I simply said “Jack, what are you doing?” in a normal voice and he backed up.
He did NOT put his ears back, slink away, or give any other appeasement signs. He just looked at me like “What AM I doing???” and stopped
I never assume Maddie knows that what she’s doing is wrong though. Different dogs, different personalities, different relationship.
I break the “don’t repeat a command” rule all the time in agility. Basic obedience commands when things are fairly quiet and I know I have my dog’s full attention are a different story. If we’re in the kitchen and my dogs are underfoot and I ask them to sit, they really should (and do!) without fuss. But on an agility course when their energy is really up, and there are tonnes of distracting noises and smells and obstacles to divide my dogs attention, repeating commands is incredibly helpful. Especially with a verbal cue to turn tightly, I start the verbal as early as possible so my dog has time to adjust their stride to the task. I’m also backing that verbal up with several physical cues at the same time, just to make sure my dog knows their job, so really I’m telling them the same thing in all sorts of manners. Repeating a verbal is fairly consistent with that. My dogs actually get weirded out when I don’t talk to them on course, as they seem to find it reassuring and helps with our connection.
One personal rule what I have broken for some of my own dogs is: bribery. I taught the dogs that “magic leash”, a short leash would appear and we would go for a ride without all the regular leashes, car harness, yada. I would hold it up and the dogs would run to me and to the door as they loved to go for a ride even if it were only around the block. If a dog darted out the door, I would walk out with “magic”, the dog would come to me and we would go for a ride.
I have since worked on teaching touch my hand near the door as an alternative to magic when I observed another trainer yelling this as her dog darted out (dog came back). I also now have a dog who does not like car rides.
I want to add a few points. Jack is NOT trained to look at me whenever I say his name. I have never used “What are you doing?” as a correction. I do use “What…?” and “Are you…” questions in many different contexts when we communicate with each other. What do you want, what do you hear, what do you have there, are you bored, etc.
I find myself “breaking” the rules I tell others to follow all the time. I only do so however, after I have seen them doing the correct thing plenty of times. I will only insist that a dog lay down when I have already seen that a) he understands and already lays down other times and places quickly when asked, and b) that he has laid down of his own accord already in this place, so I understand he is not uncomfortable. I have watched plenty of dogs who learn that it is too easy to push their owners around and not do what is asked (aka small toy dogs mostly) simply by playing dumb. As soon as I step in and ask, without letting them get away with not doing it (by putting them on a leash) they quickly comply. Then you go “a-ha”.
Ashley Hill says
So interesting to think about these kinds of questions. Dog training murky waters!
I think this is just further independent verification you were right when you determined the answer to every dog training question is “it depends.” Thanks for the first cut on four criteria when these rules are to be broken, though. Very helpful.
This reminds me of Dr. Dunbar’s instructive reprimands which, of course, you used like a good textbook trainer as I think that logic is intuitive to many excellent dog trainers. Certainly, for a “green” dog and a typical owner who does not specialize in behavior, this is a great way to go. But I have also seen him look at his wife’s dog and say, “let’s not go there…” or something similar preceding an undesirable behavior. In some situations, I see this as ideal, because we do not want to live in a world (to the extent this ideal is attainable) where we re-direct the dog constantly upon pursuit of an undesirable behavior and have to police its behavior all the time (in this case, we policed the intention — which hopefully we inferred correctly from the dogs observable behavior and history of behavior — not the action itself). I think it was smart of your trainer to cut short Maggie from getting reinforcement for that behavior before it became even more likely to happen, but did so in a way where the dynamic won’t ultimately be you having to constantly watch her to police that specific desire.
It reminds me also of Susan Garrett who uses no reward markers on occasion even though the crux of her training seems to be, at least in part to a novice like myself, such a healthy and extensive reinforcement history on a desirable behavior that it is practiced enthusiastically to the exclusion of less desirable options. If one of her border collies, for example, is doing weaves wrong (usually in early stages of training), she will drop the level of difficulty/review but she will also stop the dog with a gentle verbal correction, especially since border collies can get a lot of reinforcement for a bad behavior if it involves running around in a vaguely circular fashion, haha!
To put it simply, I guess we want to, as people have observed, guard against a relationship dynamic that is solely “I try to do things I want and then you tell me when it’s wrong.”
It seems to me when we enter murky waters with a “well-trained” dog, the range of options to which they will go if we give them a non-specific correction is less likely to be some wild behavior we never wanted them to practice in the first place — which isn’t always the case with a “green” dog who, if jumping is inhibited with a verbal correction, might very well just try pulling on the visitor-at-the-door’s sleeve to get reinforcement! (Although the purpose of the instructive reprimand is to communicate to them what they should do, of course, instead of just putting them in a scary world of thousands of inhibited behaviors). I think one goal of not relying on verbal corrections and NRMs is to prevented development of an inhibited dog who is afraid to behave in ways the owner has not specifically told them is correct — that is a stressful life! But I think if we complement our training with some shaping and rewarding of creativity-within-criterion, we can help balance that against some gentle guidance when it is important to stop a behavior, particularly when it is self-reinforcing, even in a “well-trained” dog.
I learned early on (from Karen Pryor’s book, I think?) that when training a new behavior it is good to “end on a high note.” I love this rule, and I follow it, and I implement it with my students. It’s great because it’s short and catchy and easy to remember. However, “end on a high note” isn’t actually the rule, it’s only shorthand for “when you are working on a new behavior and you get a great response, resist the urge to keep going, it’s better to stop right there and end on a high note before the behavior deteriorates and you end up having to withhold the reinforcers or reward lesser behaviors and also it seems to make an impression on the dog . . .” Some people misunderstand the rule and when things are going poorly will keep going and struggling, trying to get a good repetition so they can stop. So, it’s not really breaking the rule in my view, but when things are going poorly I will end the training session on whatever note we are on, and regroup.
pat mommaerts says
Interesting blog, especially in light of a feature article in the Nov 29 sunday N.Y. Times about Werner Ehrhard, founder of EST and the Forum whose central premise is “taking responsibility for your own life”. It does presume that one knows (be it either end of the leash)what it is they “should” be doing.
Very good blog… I’ve emailed you before about training a sheepdog as I find it perplexing shall we say! I would love to read your other blog on sheepdog training… It could be because I’m reading it on a phone that I can’t find the link! From a behaviourist point of view I find it hard with so many factors to think about!
This doesn’t sound like Maggie “knew what to do,” it sounds like you gave her a verbal correction or interrupted her behavior and prompted her, and she did what she had done several times before just moments before. She was going one way and you cued her to go another. After a few repetitions of this, you could have said “Maggie! What’s the capital of Alabama?!” and she would have done the thing you had cued her to do the previous times you interrupted her. She didn’t know what you wanted and deliberately crossed you until you reminded her who was boss, she was just doing what she thought was best until you cued her several times and then when you prompted her.
This sounds dangerously similar to the justifications made by people who say that the warning tone on a shock collar or the “eh!” sound is “just a reminder” to make sure the dog isn’t “disobedient.” The fact that you tried it and it worked does not mean that what you thought happened is actually what happened.
No one, no matter what fancy dogs they breed or how many fancy ribbons they win, knows what’s going on inside another being’s head. For that reason alone, it’s wise never to assume that someone else is being defiant or “stubborn.” It opens up too many avenues for mistreatment.
Unless I’m missing something I don’t think this is “breaking a rule” at all. I see it as part of a logical and necessary transition to having a true working partnership with a dog. There’s a great deal of difference between house manners (say) and advanced training where a dog must be trusted to do the right thing (as opposed to NOT doing the wrong thing, like peeing on the rug or barking out the window). I’ve been blessed with several wonderful dogs that I was able to take to advanced levels in several different activities – including herding -and have always seen the process of “handing off responsibility” as integral to advanced work. It is astonishingly wonderful thing when you are able to absolutely rely on a dog to do as s/he has been taught.
Wow, such great comments. I loved Margret’s comment about wondering about the many re-enforcements going on in sheep herding. Also, great article. With my dogs, the responsibility for behavior is built right in, or trained right in I should say. I work in a building which holds events in our big lobby almost daily. Lots of times, there are tables and all sorts of stuff being set up when I leave or enter the building. Seamus has to figure out how to get us around all of that. Usually, he does it without me even being aware of a clearance, the term we use when he’s successfully guided me around something. Sometimes though, he’ll just stop. Several things could be going on here. He could be distracted by something, there could be something in his way, or he could be confused as to how to proceed and those last two usually go hand-in-hand. Seamus doesn’t especially like tight clearences and he becomes hesitant to guide me through a tight spot. If I can tell it’s a tight clearance and I also know that this is the only way through, I cue him with the Forward or Hop-up command and he sucks it up and we squeeze through. sometimes though, I’m at a loss as to what we should do. A few weeks ago, we were coming out of the hall and into the lobby and there was tons of stuff around, so much I had no idea what was there or how we’d get around it all. Seamus stopped and we stood still for a moment. What should I do? There wasn’t another human around to ask for assistance and I didn’t know why my dog was stopped. Here’s where his responsibility comes in. I asked him to move forward, adding in a, “Figure it out,” as encouragement. I couldn’t tell him what to do and I couldn’t help him either. All I could rely on was his training and his steady nerves under pressure. That may sound a bit overdramatic but it isn’t. I can’t see what’s out there and I trust this dog not to lead me into a situation which might get both of us killed. He moved forward and took me in a wide arc around what ever was in our way and then picked up his line of travel again and we headed for the door. My point in saying all of this is, sometimes there comes a point where you’ve done all you can to help your dog and the dog may just have to take matters into his or her own paws.
This article reminded me of a great article I just read a few days ago written by a guide dog instructor. She wrote about “Soundness” in dogs raised for guide work. She wrote about a sweet dog she was training who did everything she asked of him, but when he had to make decisions by himself, he buckled under the pressure. There wasn’t anything wrong with the dog, he just couldn’t take the pressure of that kind of decision making when he had to. Maybe he’ll become a loving pet, maybe another type of working dog, but guiding a human isn’t for him. Tricia, your Maggy sounds like a great dog in this area, a bit uncomfortable, but when you asked her to move out of that comfort zone, she did and fixed the problem herself. Could she be uncomfortable because she’s young? I actually forget how old she is, but isn’t she about two? I found, with my first two dogs, who were older when I received them, that they reacted differently to situations than Seamus, who was younger when I got him. this was especially true with groups of people on a sidewalk. My other two would part a crowd, no problem, but Seamus needed some initial encouragement to get passed his causion. He’s fine now, or, appropriately careful I should say, but it was because he was a young dog who wasn’t fully confident yet.
for me a sound dog is now one of the key things I’ll look for in the future. I had a dog who hated the pressure of guide-work and though he never put me in danger, I could tell then, and can see now, that the stress was making him miserable. Remember those tight clearences? Torpedo was even more careful than Seamus is. It got to the point, at the end, where he wouldn’t guide me through them and I was heeling him. Marlin on the other hand, my first dog, blew through clearences like they were no problem. In fact, we had to re-work a few because he wasn’t careful enough. Having three dogs has taught me so much about how dogs handle certain kinds of pressure, especially when they have to make decisions for themselves and I know they’ll continue to teach me as they come into my life.
I think it’s OK to break the rules when you’ve fully mastered the rules. The first example that came to my mind is with writing. An author needs to understand how to do it right/according to the rules before they can violate those rules. I’m sure that’s what came to mind because I’d just finished Patrick Rothfuss’ novella *The Slow Regard of Silent Things* which was one of the most lovely enjoyable things I’ve read in awhile and yet it broke every single rule about writing short fiction.
I’ve noticed my dogs are lazy in the sense that they will do jobs only as well as they need to. I wonder if Maggie is similar. It’s harder for her to pressure the sheep than to head them off so she opts for the easier for her choice despite knowing it isn’t what she should be doing.
For myself, I’m sure I break training rules all the time not because I chose to but simply because I’m enough of a novice to not have mastered the rules. I suppose too, that the kind of relationship you want with your dog and the kind of life you want them to have makes a big difference in how you train. I often suspect that Finna would be best pleased if I micromanaged every bit of her life and she never had to make good choices for herself. And if I were a different kind of person and wanted a dog that was totally dependent I could do that. But I don’t want that for her or the responsibility of making all her choices for myself. I want her to learn to make good choices on her own. It’s harder for her to make good choices than bad when confronted with something she finds alarming. It requires thought to make a good choice but only reaction to make a bad one. She spent her formative year never being asked to think. She gets better day by day but it is a long long road.
We’ve had a lot of changes lately, including a move out of our suburban neighborhood out into the country. The move is temporary, and we’re staying with a relative who has a large farm property. She also had four dogs of her own, which initially made me very nervous with dog-reactive Remus.
Remus and I have worked a lot on his reactions to strange dogs, and he has gone from uncontrollably fear reactive to giving me beautiful auto-watches when he sees a new dog. He loves the work we do so much that he looks for opportunities to practice every time we’re out- to the point of waiting outside houses where he has seen dogs in the past, hoping they will come out to “play.”
The property where we are staying now is fenced, with a large dog run near the middle of the yard for three of the resident dogs. When we moved in, I started training near the run, and Remus was maintaining his calm and watching me when near the other dogs. I was taking extreme care not to allow him off leash near the run, even though I know the resident dogs are dog-friendly and not barrier aggressive.
The other day, my husband had the dogs loose while he cleaned up some fallen trees from a storm we’d had. I came out of our house to see Remus, 70 yards away and headed straight for the dog run. He was well within his reaction distance, and one of the other dogs was just inside the fence.
I wanted to call him. His recall is good, he would have come. But, in the seconds I had to watch him before he got to the fence, I saw that his body was loose, tongue lolling, and his fluffy fur was down and sleek against his body. I let him keep running.
He stopped by the fence, briefly greeted the dog on the other side, and then trotted on to find the (human) kids who were playing somewhere beyond the dog run. I heard the kids call to him. I could see him through the trees as he circled the kids. He spotted me and made a beeline back to me, passing the dog run again.
The same scenario has played out over again multiple times since. He will trot happily to the dog run, greet the dogs he finds, and then run back to me without being called. The other dogs, all much older, seem perfectly happy to sniff through the fence and let the bouncy youngster continue on his way. It’s not a by-the-book way of doing things, and it isn’t how I would have planned it out, but giving Remus a little more freedom to practice his learning seems to be paying off. I’m certainly thrilled to see him looking briefly at his new “friends” at close range, then bounding back to me or the kids with only a big dopey smile on his face.
Ok, not sure if I have assessed this correctly but it is a fascinating post so keen to learn.
It seems like you are asking Maggie to do something that she is not 100% comfortable with? So, how do you ‘correct’ that? You can’t correct it because she isn’t doing anything wrong (operant)as far as I can understand. What she is doing ‘wrong’ is not doing exactly what you ask her to do. The options for that are because she doesn’t understand the cues, hasn’t heard you or doesn’t want to do it, because she is not comfortable with it. By a process of elimination you came to the latter conclusion but as we will never know what a dog is thinking and you have to work as a team how else would you progress past a sticky learning point in your working relationship? What are the alternative options?
When you are working as a team, sometimes you may have to nudge a colleague out of her comfort zone to get the job done. However, if you have built a relationship with a solid, safe and trusting foundation, providing she is fairly resilient and you have assessed her behavioural needs, then she can probably handle it. She can learn. What is interesting and important is understanding how far to push. Because pushing into the brown zone can backfire. For example, if you had popped her on sheep and done this the second week you had her then it may have backfired badly?
Micromanaging rarely works out for dogs and humans who thrive on working independently in my experience. Choices are empowering but it seems to me that all dogs (pets, working dogs etc.) need plenty of opportunities to learn what choices WE would prefer them to make as opposed to correcting them when they make the wrong choice. What seems fascinating is that you are learning so much about Maggie’s nature by working her on sheep and continuing to learn new stuff about yourself too – both as a trainer, dog lover and as a sheep dog handler.
Would love a post on correcting sheep dogs please as I know diddly squat about that.
One thing I have noticed with my Aussies is that it is often very effective if I ask them to stop & think. Sometimes they get so “sure” of what they think is right that they fail to understand what I am really asking. If I do pretty much what you did and ask them to think, they often figure it out. One of my favorite Kiva stories (my first Aussie who passed away last summer) is from Nosework. He was trying to figure out a hide that was placed under the lip of a table which was set out in the middle of the room. He went under the table and lost the scent, so backed up and put his paws up on the table…and lost the scent. Back under…back up…back under…back up. He was obviously perplexed. After repetition 4 or 5 of this, I said, “Kiva! Think about it! If it’s not on top of the table and it’s not on the other side, where could it be?” He looked at me for a second, then put his nose under the lip of the table and nailed the hide. I swear – there were witnesses!!
Similar things happen all the time with my dogs, though not usually that dramatic. That is how I would define letting them take responsibility; it’s not always telling them what the right answer is, but letting them know they need to work it out. And then letting them know they got it right!
When is it ok? I tend to focus on your #3 – When you’re a expert. When you’ve done your 10,000 hours and become a (Malcom Gladwell/Anders Ericsson defined) expert… From my 25 years of bench science I’ve discovered when I can and can’t bend the experimental rules, but boy it’s taken a while and some times serious mistakes were made. And paradigms shift and rules change with the new paradigms, whether it’s classical aversive based training to clicker training or our understanding of classical Darwinian evolution vs the more recent understanding of Lamarkian like (gasp) epigenetic modifications of individuals within their own lifespan.
In some cases it may be a matter of semantics and/or careful definition that’s skipped for the sake of brevity. Take luring for example. The rule: don’t lure (more than twice). I had to have my instructor define luring for me a few years into training. I early on took luring to mean any form of enticing an animal whether it was with food, my body or a toy and I was a miserable trainer. It turns out a good operant conditioning trainer will set up the subject for success during shaping exercises with all sorts of non-food based lures and obstacles. If you define a rule too broadly and hold stock still and do nothing because you’re paralyzed by a rule it’s not that beneficial. I love the back and forth of the further clarification in the comments section of whether Maggie “knew” what she was supposed to do. That needs to happen to understand the rule. I also like you 4 guidelines for when to break that particular rule; can’t think of anything to add to that.
Liz Shaw says
Ha. I have been thinking about teaching dogs “autonomy”, for lack of a better word, for about 10 years now. I think one of the most wonderful things about working with a herding dog is that they have their job that they can do far better than we can and we have our job. Ours is to come up with the grand scheme and the basic outline for how work is to proceed (we are going out to the pasture and put the sheep in the electrinet by the woods today) and theirs is to manage the details (silly sheep wants to eat the grass on the side of the lane. Not going to happen). My dog hates to put pressure on the sheep, but he knows how to drive a straight line. This fall has been a lesson (for me) in letting him handle the details and when I put the reponsibility on HIM to hold the line his drive is straighter and truer. It has occurred to me that my back seat driving (pun intended) might actually be *distracting* him from the details (“well maybe that isn’t really the line she wanted” ). It then saps him of his initiative. When someone else is always telling you what to do, what is the point of taking responsibility? And perhaps in other dog training scenarios you don’t want a lot of initiative but in herding I sure do. When I let him take responsibility he does do better. But I really think this is true in other scenarios. Once you have taught the dog various behaviors and they understand them, maybe a dog taking responsibility for her own behavior is a good thing. I think of reactive dogs. Very often reactive dogs don’t have the idea that they could “walk away” in a situation. We can teach them that it is an option. But we often teach them to turn to us for “advice”. If we teach them to turn to us exclusively, if we aren’t around they don’t have the tools to help themselves. I have seen some really powerful things happen for a dog, once they know they have the choice, when they choose to walk away. Their choice – not me leading them with a cookie – but them saying “I don’t have to get into this”. It empowers them and ends up being incredibly reinforcing. I guess taking responsibility for your behavior once you know how to do something is fundamental to living responsibly along side other mammals. Absolutely true that we set our dogs up for failure all the time because we want them to take responsibility for their behavior when they don’t have the tools to do so. But I see the reverse as well – people micromanaging a dog, sapping them of their initiative, when the dog has the skills set and just needs support, not direction, to make the right choice. I will make a bet Maggie learned something really important about driving just from you saying “hey you need to do this yourself, I can’t fix it for you”
I am reminded of a short-haired young BC I met at a pound earlier in this year. He comes out flip-flopping on his leash like a fish on a line, lunging at other dogs, barking, zig-zagging, he’s all over the place. I got him moving and started giving him lots of leash signals with pressure and release, and started getting his attention and rewarding him for it. In a few minutes, he had quit most of the silly out-of-control business and was surprisingly responsive. Acting on nothing but sheer gut feeling, I firmly made my expectations of that dog’s behaviour known. You will walk nicely, respond to my signals, and you will not lunge or bark at other dogs. Completely not what I would normally do with a green dog I had met 5 minutes ago, but there was something about this dog. His quick change in behaviour and responsiveness made me think he was acting up because why not? It made him feel good. Me suggesting I did not want him to was all the “why not” he needed, though. I fell in love. Unfortunately, I could not take him home and he very nearly got put to sleep, but a rescue group got him a last minute save. Occasionally one of my own dogs is a bit like this. He will start to act up and bark and lunge at something I have seen him handle fine in the same conditions time and time again. It’s a dangerous call to make IMO. If I’m wrong, he gets scolded and told to pull himself together when he actually can’t. If I’m right but I don’t let him know I expect him to do better, he gets himself into a state, kind of enjoys the cathartic bark and carry on, perhaps, or undoes some of the conditioning I’ve been doing with him. If he genuinely is that upset, I should get us distance and help him calm down. It’s probably not a disaster if I do that when he could have pulled himself together after all, but I do think it makes working on triggers go slower and sets a precedent I don’t necessarily want.
Other rules I break regularly:
1. Repeating cues – Sometimes I just want my dog to perform a behaviour the instant their brain is back in thinking land. I don’t know when that is, so I just repeat the cue until they do it. I trust that they will do it when they can.
2. Negative reinforcement – There are times when I want the dog to learn to navigate around things they don’t much like. Some think it’s evil to use this quadrant in training, but I think it’s perfectly humane if the goal is to help a dog find coping strategies that are effective and not so anti-social and we are very, very careful about using this in moderation.
3. Bribing and luring – who cares? I don’t. There is a time and place for such things, like when you are asking for something big. I bribed one of my dogs to hold still for eye ointment application. He will do it for pate. Fair enough, I say. Corneal ulcers hurt. Don’t you worry, the pate is waiting for you right here.
4. Automatically creating distance if a dog goes over threshold – I think you have to be careful what the dog is learning. They can actually learn when over threshold, so let’s not forget that and inadvertently teach them how to create distance.
Btw, an interesting off-shoot of this discussion that might make for as good blog topic would be “Is great training just 10,000 hours or are there innate physiological talents that some trainers are just gifted with?” I.e. professional baseball players and 20/13 vision, Lance Armstrong and large lung capacity, etc. Are there innate sensory or cognitive capabilities that some trainers are gifted with? It’s not a topic to marginalize those that are differently equipped but to better understand trainers and training.
I think the simplest explanation for what happened with Maggie is this:
Initially, Trisha kept repeating “Don’t do that, do this.” Maggie, being capable of doing the right thing but finding it mentally uncomfortable, was happy enough to keep drifting back to her preferred behavior because there was no consequence.
In the second phase, Trisha’s new approach was “Your choice has disappointed me.” For a dog, as for a person, a social consequence is very important. As a child the worst thing my parents could say was that I disappointed them.
Faced with this consequence, Maggie chose to stop taking the path of least resistance and do the right thing, to avoid social disapproval.
This only works, as already stated, when the dog (or person) knows already what the preferred behavior is.
The reason it can be dangerous for novice handlers is they may lack the experience to judge if the dog understands or not.
I think it also depends on if we see our dog as another adult, capable of realizing that choices have consequences; or if we see our dog as someone always in need of guidance.
Some of that depends on us but some of it is the dog. Not all dogs ever mature out onto solid decision-makers and are happiest when most of their decisions are made for them. Others are practically begging to have more responsibility.
Alas, I have lied to my dogs! Call it bluffing, fibbing or lying, I have used words guaranteed to get their attention in high pressure situations. I have repeated “good…treat,” to my child-phobic dog. Like when I knew I was fresh out of any rewards and ten or so kids proceeded to run out of a school bus and swarm our car at a stop sign. (I considered throwing the car in reverse once I saw the direction the kids were headed, but uh, not safe.) Also, when out of rewards and/or energy, I have falsely used the names of animals to distract my off leash, high prey drive dog away from the real animals. Like if there’s a rabbit she hasn’t spotted yet off in the field, I will gesture to the ground in the opposite direction and say, “mouse,” then lead her on a mousing mission rather than risk a failed sit/stay, emergency stop, or recall. I’d like to think it hasn’t shaken their trust in me, but I do always feel guilty whenever I say something to them I can’t back up.
I thought I’d only lied to my own dogs, and that our history made this situation possible, but I actually lied a stray dog as well! When two strays showed up together on our property along a busy road, I was able to get one quickly with minimal effort (food and sweet talk). The smaller of the two strays, however, was wholly committed to marking over all of my dogs’ urine spots. I remembered a tip from Trisha, perhaps, and asked the small one if he wanted to go for a car ride. Opened up my door, he ran over and jumped right in! Both had collars and tags and waited in my car until the owner could come pick them up. Turns out the man was dog sitting for his son, and the small stray wasn’t familiar with the neighborhood or road-savvy at all. So in this case, a lie led to a happy ending and I was much more relieved than guilty . Technically, they did get to go for ride, too, just not not with me.
While you did break a rule, you did not apply an aversive, well at least not of the type we have used in the past to make a dog do what (we think) he knows. Maybe it was more aversive to be “nagging” her to change, as one comment called it? I also agree with the comment that Maggie might have thought the behavior included your direction request (“more to the left” or whatever). In every case however, your blog and the others’ comments show 2 differences from the “blame and correct” old days (of which I was a part of all those years ago): 1) we now understand so much more about learning, training, our role, the dogs role (thanks Patricia for your role in that!) and the ethology of dogs that we can make way better decisions and training plans and 2) We are actually asking ourselves why the dog is doing what it’s doing. Such a paradigm shift from totally blaming a fresh dog or a not dominant enough handler! So essentially when a “positive ” (still struggling to define) trainer says “take responsibility” to their dog it is a different question with a different outcome.
FWIW, I have been asking some top trainers over the years and yup, they all use a verbal reprimand with their own dogs, when living with them, not necessarily training (for competition). I am not a dog sport competitor but a pet dog trainer and I can tell you that pet owners break the “no aversive or corrections” rule all the time with generally good results. I’m not saying we should teach that way, but I am saying that, when asked from a new paradigm, breaking that rule and initiating some evil-eye parenting and maybe a little physical or emotional stress, in the context of an otherwise stable, trusting relationship, is a viable way to gain some good decision making on the part of the dog (or kid).
I totally think that ultimately it still comes down to learning theory, it’s just a question of what we are reinforcing. Are we reinforcing the feeling of being independent when that’s what we want (or when we don’t want it!) when we think we are reinforcing a behavior (go more left)? A lot to be learned in that area, I think, and amazing that dogs and people seem to be very similar in terms of things like picking the easy way out, rebelling or forgetting to think for ourselves due to micro-management, or wanting to be right.
I could never have thought this comment let alone write it if it weren’t for your post and asking the question in the first place, but now my mind is reeling!
Andrew N. says
I’ve always felt that good dog training does just that – makes a dog aware and responsible for their own behavior. After all, “operant,” to some degree, refers to the learner intentionally operating on the environment.
Empowering the dog doesn’t mean I just leave them out to dry, of course, but it does mean that I help them achieve the ability to be autonomous if at all possible.
I love this discussion, but I am posting to comment on the picture of Maggie chewing the antler. My BC is a Redtop dog(Riggs and Jen), and Maggie looks so very much like him that I cannot stop smiling at the picture. Thanks for sharing all of the pictures of your dogs!
Cynthia Mills says
The thing about herding dogs is that it is a collaboration between handler and dog, moreso than most other training tasks (we used to say that we “educated” our retrievers for field trials, we didn’t train them–instead we provided experiences that helped them learn how to put visual memory together with the change of terrain and wind and smell.) Herding dogs have the tools to herd but we handler/trainers just have to show them what works best, like getting out far enough to be seen in the sheep’s outside eye, and foregoing chase for flanking, and keeping enough distance between them and the sheep to keep them all in sight so the dog only needs to lean out to be seen by the errand sheep heading in the wrong direction.
I think most training that involves “guilt” is mistaken in a basic way: that the thing we think dogs should feel guilty about are not natural for the dog to feel guilty about. Stealing food is what a dog does, naturally, chewing shoes and breaking house training are only “bad” because we say they are bad.
Conversely, a dog working in a hunting situation (herding) with other companions (handler) must act in concert with the others, and must step up or the game is lost. Feeling shame and guilt in that situation would be an evolutionary advantage.
Here’s a rule I break sometimes: always follow a click with a treat. I break this rule with my own dogs once I’ve been clicker training them for at least a year if it’s a really critically bad click.
Just a simple example.
Lindsay Pevny says
I just have pet dogs, who have few expectations of them, nothing as complicated or intensive as herding, but even in basic training, I find myself wondering what my dog is thinking, and if she understands what I expect of her.
For example, when we’ve gone a long time without practicing a trick, she will seem to not remember what to do – until I have something really valuable like hot dog. Then, she remembers very well! Either way, I blame myself for not practicing.
Donald McCaig says
Glad you got a chance to work with Patrick. Sheepdog clinicians have different strengths: some are dancers (dance as I do), some are looking for the great dog they could win with (win with me); Though I don’t think he’d agree with my choice of words, Patrick is a theoretician (see the dog with me). He’s a deeply gentle soul.
There isn’t a “rule” for the sheepdog’s taking responsibility though that’s where you want to end up because often you cannot see what the dog is doing.
Mary Gessert says
Hi Trish, I am so delighted to come across your blog site. This is a very interesting discussion, and I can’t help but add my two cents. I wonder if it might be our human penchant for competition that creates these training dilemmas. A stockdog doesn’t need to move livestock in a straight line or make tight turns to bring a herd from the pasture, through a gate, and into a pen, and they will do it happily with few commands and a “good dog” for reward. Likewise, a dog does not need to sit straight to be a lovely companion and good canine citizen. One that sits calmly to be petted by visitors will gain much praise for its owner, and will be rewarded by the attention. I think that sometimes we may set our standards so high that we and our four legged companions fail, even when doing amazing things.
I knew you all would come through with some insightful comments, and I was not disappointed. So many great points made, here are few additions of my own: First, thanks for all the ideas for new articles. Sasha especially got me thinking about the path to being a good trainer–how much practice and experience, how much innate qualities that someone is born with? Food for thought there for sure.
To Mary Gessert (sheepdog colleague from way back!): Great to hear from you, and thanks for your important comment about what we are asking of our dogs. That leads to another potentially interesting topic–sometimes we push our dogs beyond their initial comfort level, and their expertise in a certain task improves. Is that good for them, or perhaps a better way to ask that is when would that be good for them (if ever) and when are we causing harm. My belief (belief only) is that lots of sheepdogs love learning that they have the power to control the sheep and take charge of them. In Maggie’s case (as suggested by an early comment), a lot of what is going on with her is related to her age. She’s a young 2 and a half in my opinion, and is growing and maturing in front of my eyes. Part of that maturity is learning “Wait! I CAN do this! Oh Wow! I think everyone involved in dog sports (from agility to herding to scent work) has seen a dog leave the field with her head high as if floating on air. Are they proud? Is there such as thing as pride in a dog? Eeeps, another article awaits….
I shouldn’t neglect to thank LisaW for pointing out that I wrote Lassie when I meant Maggie. Lassie was a BC who I still miss on a daily basis, I guess that is obvious from my mistake. My brain probably stores them in a similar place, because both Lassie and Maggie make me smile as much as any dog ever has.
To Laura: “The fact that you tried it and it worked does not mean that what you thought happened is actually what happened.” Absolutely true, I agree completely. However, I am not a fan of the “we can never know what is in the mind of another” perspective. Yes, of course, at one level that is true, both for other animals, for other people and perhaps most importantly, based on the new research on the role of the unconscious, our own mind. But, and this is a significant qualifier, that doesn’t mean we can’t or shouldn’t try. We simply couldn’t function as a member of a social group if we could speculate what is going on in the mind of another. Are they angry about what we said? Are they in agreement with our suggestion? I think that is equally true of our relationship with dogs. We simply couldn’t be decent owners if we didn’t have some idea of their mental state. With Maggie, I agree completely that I could have said
“Maggie, go to China!” and she would have put herself in the right position. But the difference between saying her name and then something she hadn’t been taught (and letting her make a choice about what to do) and telling her what to do next was important. Here’s my question: How much did the way I said it influence her response? It was a very different tone than I usually use–did she have any clue that I was frustrated? Ah, another topic for a blog article!
Also, Laura equated my saying “Maggie! What are you doing?” with giving a dog a shock from an electric collar before the dog began to work to remind the dog of the power you have over them. I can’t see the analogy myself, and I smiled when thinking of how actually powerless one can feel when working a high powered furry machine 100-500 yards away with nothing but your voice to influence their behavior.
And to Patrick–Gotta love the connection you made between Ehrhard, EST and dog training!
I totally understand the explanation of Maggie’s behavior now, thanks, Trisha.
I guess what worries me is that I’ve experienced first hand what happens in a competitive atmosphere when it comes to breaking training rules. There is a lot of justification that goes on, especially when it comes to well-respected, highly experienced trainers and their all-too-willing followers and possibly frustrated students. To use an overworked phrase, it’s a slippery slope. It’s not that far from “she knows what to do” to “why won’t she do what I want her to do?” to “she’s really making me angry now”.
I guess my fear is that trying to get the result you desire can so easily overshadow the journey to get the result. Let’s face it, these are hobby sports we are doing with our dogs. You may actually be using Maggie to herd sheep, but I’m assuming it’s how you make your living. So even tho you may want her to burst the bubble around sheep, does it really matter all that much in the larger picture? Most people I know who do herding have never touched a sheep let alone owned one and the intensity of their training goes over the line a lot. The same thing happened in agility, which is why I left that sport completely. When is breaking the rules just permission to go the easier route and when is it situation specific? I don’t know the answer to that, but it does concern me that all too often it’s not even asked.
BTW, I meant “I assume it’s NOT how you make your living”…kind of crucial word to leave out. 🙂
Cynthia Mills says
Regarding the statement that border collies love to be right–sure, probably most dogs do also. But being right when driving sheep means so many things: it means moving away from the handler which means losing confidence, after all, now the dog is the only one taking charge of the sheep and they just might get away, and I think all of us can imagine how scary it is to be the only one responsible. Plus the pressure thing is not just holding pressure on sheep, but also maintaining concentration, maintaining the message you, the dog, are giving the sheep (as in “go this way, not that way”). If you are worried (as the dog as a beginner may be) that you don’t have it “right”, it’s hard to stay confident and “on message.” Plus there is a bit of laziness involved–it’s easier to zone out and just follow. Sometimes just saying their name gets them back on track.
Perhaps I’m way off base here but “What are you doing?!” was the corrective cue Maggie learned, which may be the reason she continued to stray with the use of hand cues and why she responded immediately to a specific verbal cue. Also, giving that cue or any cue to correct would not be allowing the dog to correct itself, the dog was responding to the learned cue and corrected because she was told to.
There is no doubt in my mind that Maggie knows the routine, but for whatever reason, she is occasionally non-compliant and motivated to stray off line. Perhaps that motivator should be more closely examined rather than just assume she’s exercising free will to flagrantly disregard the rules rather than satiate an innate need, like walking behind the flock instead of leading, to feel more comfortable. Bred for the herding life or not, Maggie is still a sentient being and an individual creature that may just not be entirely cut out for that specific task.
This topic certainly has given me some great food for thought. My dogs are pets, so what I need from them, behavior-wise, is much less complex than what is required of a working dog, but allowing them to build a sense of autonomy and confidence is something that I feel very strongly about, and I feel that too much direction absolutely can work against this goal.
Self-reliance comes very naturally to Otis, who has a loooooong attention span and a true gift for cooperative motion- I make vague noises and gestures, and he intuits what I want a very high percentage of the time.
Sandy needs to be shown more clearly, and I can practically see the gears whizzing in her brain. The difference is visible in their eye contact. Otis makes less direct eye contact, and holds it for less time. He’s more likely to pause for a moment and consider not just me but the entire scene before deciding what to do. He’ll adjust smoothly enough if I give him additional feedback, but I almost never need to. On the down side, if he decides to do something that I emphatically do not want, I have to be VERY insistent to stop and redirect him, and he makes no secret of his contrary opinion.
Sandy is more likely to lock her gaze on my face. Her eye motions are smaller and quicker, compared to Otis’, and she is much less inclined to take in the big picture before responding to my cues. If she doesn’t KNOW a cue through repetition and explicit teaching, she will guess by running through her familiar repertoire, rapidly and with increasingly frantic energy. On the upside- she is very quick to respond and much more trusting if I interrupt her chosen path.
When Sandy first came to us, she went from a home where she had almost no autonomy at all, to one where she was given daily doses of tremendous freedom within broad but firm limits. She was beautifully obedient, but in some circumstances she really struggled to “make the call” for herself. For instance, one of my broad rules is that dogs need to stay in sight on off-leash walks. Obviously there’s a lot of logistical variation at play when hiking in mixed terrain. In the woods, dogs need to stay very close, and out in the open fields, they can run much further before I call them back. After very few walks, Otis developed a sense of how far is “too far”, and I can walk for an hour or more without needing to give him any cues at all.
Sandy had a SENSE of when she was too far, she would be listening for the cue and ready to turn on a dime (impresses the heck out of people when a gently called, “Far Enough!”sends my chipmunk hunter screeching to a halt mid-bound 150yds up the hedgerow- little do they know that she wouldn’t respond half so dramatically if she were right beside me), but she wouldn’t ACTUALLY stop doing what she was enjoying until she heard me tell her that she HAD to. In this situation, I can say anything to her, because she knows what I want and what to do, so I don’t have to be as careful to ask for a specific desired behavior as I would otherwise be.
Anyhow, as time has passed, she’s come a LONG way. She’ll always be a bit more cue-dependent than Otis, but her confidence in her own decisions has truly bloomed, and she seems so much happier now that she can “take care of herself” without a constant stream of instruction and commentary from the humans. No matter who is doing it or how, I wonder if the process of learning can ever be truly “pain” free?
Whether the unpleasantness comes from anxiety, confusion, social pressure, (godforbid) actual pain, or just the discomfort of dealing with novelty, isn’t all learning – manners, tricks, work skills- ultimately a case of the ends justifying the means?
Veronica Sanchez says
Thank you for a wonderful article – and the link to the Sat. night live skit, hilarious! I think I break dog training “rules” more often than I care to admit. One example, I might cue a “sit” and reward a different but still pretty desirable behavior in a distracted or reactive dog. Usually I want “sit” to really mean “sit.” But if circumstances are hard, and the dog is just not going to do that, and happens to offer a different behavior like a down or eye contact, I would go ahead and reward that. The end goal would be of course, to set the dog up for success and not to cue a sit in a situation that the dog is not ready for, but the real world being what it is, sometimes I’ll take what I get.
Just to be clear to Joanna: Maggie’s cues were all verbal, and she’d never heard the words “What are you doing?” in her life. I wish I could magically explain the complexity of sheepdog work to those who have never done it; it is so different than training standard ‘obedience’ or tricks! Three sentient species in the picture, all with their own predispositions, backgrounds and desires. The more I do it, the more complex it becomes. That’s part of why I love it so much!
Trisha, your comment on tone really hit home with me, and yes please, a blog article would be great.
Tone or inflection is very, very important in my opinion. If I asked Olive in a voice wrapped with confusion and tinged with a dash of hurt, “What is the capital of Alabama?” She would run over to me very worried, give a bark, and try to figure out how to Google U.S. capitals.
If I asked my other dog the same question using the same inflection, she would say, “What’s Alabama? Can I eat it or lick it?” And then move on to something else.
I guess that is a rule I’ve broken. I have used the “oh my, how could you” tone of voice on Olive a few times when it is not necessarily 100% warranted. It gets her attention and then I can react more appropriately. Hmm, I use it to buy a little time so I can figure out the next best step for me to take. It’s like a pause button 🙂
Herding is an amazing sport and to watch a dog do what it was truly bred to do is an awe inspiring sight. And, believe me, the last person in the world who needs to worry about whether breaking the rules is dangerous, it’s you! You WROTE most of the rules (or the good ones anyway), so there is no worry that you would break a rule and have it lead down a road that shouldn’t be taken.
I’m just speaking from my experience (and, yes, I’ve taken multiple herding lessons, but never took to it or it’s training methods) of watching what happens to people when the element of competition is brought into what used to be a fun time with their dog. Sometimes that 39 cent acetate ribbon becomes something much bigger to a person. It becomes a validation of what kind of person they are and how good their dog is (see parents at a soccer game for something very similar). I know one person who went from just having fun competing with their dog to “collecting” multiple BCs and giving them away when they weren’t up to snuff. Granted, not everyone is that extreme. But it’s not that uncommon nowadays.
Perhaps I’m just taking a simple question about rules and extrapolating that to mean potential extremism, but a lot of time that’s how it starts.
Again, certainly not with someone like you or most people on this blog. But with people who are susceptible to it (and I’ve been tempted myself), it can be a dangerous game to play.
A simple example of breaking the rule for me was when my dog was in basic obedience for the first semester and so was I. When we were teaching HALT…I taught my dog to sit at attention when I said Halt. My teacher told me she did not have to sit. I chose to disregard this for 2 reasons….1) she had already learned it this way and did it perfectly and I did not want to confuse her, and 2) I have a large pit mix rescue dog who always went to greet people when they passed us. By having her sit at attention, I had much more control over her and could maintain a short leash length, and people afraid or uncertain of dogs especially this type could see I had some control over my dog. I have followed every other suggestion they have given me prior to this and since.
Ellen Jefferies says
Oh my! So many responses from those who think dogs understand the words. Well, maybe border collies do, but I doubt it. My experience is it is all just noise and is the least of the things they respond to. Not that I think dogs don’t know what is wanted, some better than others. Of course, almost all my experience is with the untrainables. Right now, we’re working with our beloved English Bulldogs, a golden retriever, and Irish Wolfhoung and a great dane. The English Bulldogs are the best at establishing a team with their human(s) and knowing what is wanted. They’re a joy to train – show them once and they know – attentive, focused BUT they are the worlds best at offering behaviors and they are predominantly motivated by…being entertaining and competeing to see who can out wit whom, so expect a lot of NQ’s. I once watched a Bulldog score a perfect zero in obedience trial open. That means he did every single part of every single exercise wrong. How well did he have to know the work to accomplish that?! He had a very big crowd, the judge and the stewards included dissolved in laughter as he sauntered out of the ring with a big grin, leaving his owner still standing by the broad jump. Then the IW. I have 3 seconds to “get it right” before he leaves; getting it right is my somehow eliciting a behavior I am willing to reward. Aaaargh! He loves me dearly, but his standards for trainer performance are high and he won’t compromise. I’m improving daily. The GR is Mr. Make Me. I do; he is perfect. With my husband, a much softer, kinder, gentler soul, not so much so. Lots of crooked sits for him.
My humble opinion is that “the rules” and a lot of what the last 20 yrs of research on animal behavior has come up with is nonsense and has had (based on what I see on the other end of the leash) has had truly unintended consequences which people seem to be very happy with. Who would have guessed people wanted to be drug around by their out of control (rescued!?) dogs? Training is the fine art of getting another being to do what one wants and honestly, I think the trainee makes the rules and the trainer’s job is to figure them out and follow them. Or maybe it would be more accurate to say the trainer is the dog and we humble servants are the trainee. On the other hand, I don’t have herding breeds, so what do I know?
To HFR: I agree completely that we humans can become obsessed with competing. I too have seen people do things they would never do in any other context (because they’d be ashamed….). It is a fascinating question to me about our species: What is it about ‘winning,’ when that, as HFR reminds us, might be result in a thin piece of ribbon, that is so seductive? Of course, having competing a bit my self, I know that much of it is driven by a feeling of accomplishment, which I think is a great motivation, as long as we remember that yes, this is a game we are playing (at least when we are competing, not so when loading lambs onto a trailer) and it is our responsibility to not get carried away and put too much pressure or expectations on our dogs.
Check out Amanda Owen’s Twitter feed, aka as the Yorkshire Shepherdess. I am in awe of this woman. 8 kids, numerous dogs, hundreds of Swaledale sheep, horses, pigs and one devoted farmer and husband. I love that she refers to her Border Collies as colleagues. I was watching her round up sheep with her dogs on those unforgiving fells in with a baby strapped to her back – ultimate teamwork. A hardy Yorkshire lass indeed! And she most definitely expected her dogs to work hard until the job is done. Difficult to assess her handling as I am not a sheep dog handler but she certainly has a deep respect and love for the animals she is responsible for. Sometimes it seems as if the dogs know that they are really working (as opposed to pretending, if that makes sense) as they seem so serious in their intent. She has written a book too about her life on Ravenseat Farm and it’s on my pile of books to read that keep growing because I come to this website and find yet another great recommendation.
@HFR so right about the competitive element. I choose not to do races with my Siberians but just to train & have fun together for just that reason. I know I am competetive and I just didn’t want that to happen.
Well, anyway, put too much pressure on a Siberian and they teach you humility anyway ;-). I once was grumbling to much “don’t stop for another pee, don’t drink from that water, don’t… and they just quit on me. Lay down on the path and refused to move. That makes you feel right silly I can tell you…
@ trisha love reading about the sheepdog work. For me scootering, or even better, skijoring with my dogs feels a bit similar. Less difficult (no other animals involved) but a true cooperation. And yes, the get corrections occasionally. I was scootering this evening in the dark and suddenly I saw Spot’s ears go in full alert modus and he was getting ready veer of the path to the left and then I only have a split second left. So I barked “UH< SPOT, ON BY" (which basically means "ignore what you want to get at/chase and keep to the path" and he kept moving in the right direction. Now I also use 'on by' in a steady voice as a command, but when he is so high strung I do not get trough to him with a normal tone of voice. This worked great. (and really, 26 kilo siberian at full speed in the dark will have me bouncing of trees if he does decide to go after the deer) . With my sibes, I sometimes have to match my tone of voice with their level of excitement.
I would certainly fall into the novice/non-existent category in regards to training herding dogs. For some reason, in my head, I pictured a hand signal to redirect prior to the finale of asking a dog a question. Thank you for the correction!
Pun intended :p
It still appears as though she was responding to your inflection rather than the actual words. To me, it seems as though she didn’t correct herself as much as simply respond to your body language and tone, which would still mean she was corrected by you, just with a different ‘cue.’
I suppose what I’m grappling with is that, she knows the routine and what is expected of her, and had she caught herself straying and re-aligned without human intervention, I would definitely say that she “knew better”. To me, it would be like my dogs knocking the trash over. They have been corrected for it, they know I will be angry if they do it, but I have yet to see them hurriedly scooping the trash back in the can when they hear the door open. If I had trained them to clean up the mess on cue, they would do it but the action would not be a self-prescribed decision to do so. They wouldn’t be fixing the problem themselves, they would be fixing it because my tone, body language or verbal cue prompted them to do so. Even if they were trained to clean up trash they knocked over, I’m hard pressed to believe they would do this voluntarily (because they know better) without being told in some way to do it.
I didn’t see in the article as to whether these herding dogs self-correct without human presence. If so, that would definitely add some more food to the table of thought.
Dr. Katharina Graunke says
In fact, I let dogs take responsibility quite soon. BUT only after training, proofing, generalising etc. I sometimes even have to get my own, very agitated dog to focus by giving him a “what did I just say”. If it doesn’t work the second time, I know he didn’t know (in that special occasion/environment/distraction etc.) and I go back in training one or two steps. I have very good experience with all my dogs with it. I think, having them think for themselves lets them learn even more, better and the behaviour/rule gets more engraved.
I teach my more advanced clients to realise when their dog “knows”, because all too often it happens that the dogs “train” the owners to give them more treats by not sitting pretty, lying down half way or coming all the way except the last meter. Especially the border collie-type-will-to-please-breeds are super fast with that and mostly learn so much faster than their human (and get bored), even more so when the human is a dog-newbie 🙂 Always with the rule “if it doesn’t work the second time…”. It’s a “line-dance” (as we say) and it’s certainly something I would not tell every owner with just every dog.
As to the publications on the “guilty dog”. I have trouble with their conclusions that there is no guilty dog, just one that reacts to appease their owners. I think they tested whether dogs know the right face expression to appease their owners. Period. And that, dogs can do independently of whether they did something wrong or not (who would have guessed). They did not test whether dogs know they are wrong/don’t obey the rules or have a sense for rules or can feel guilt. At least that’s my take on the methods of these publications. As a former researcher I am now telling you a very unscientific “one case study” 😉 :
Maybe I am biased because of Charlie. It happened maybe 2-3 times a year that he came into the living room in the evening (where everybody of the family was), behaving “suspicious”. Often we didn’t take too much notice, because his overall body language was relatively submissive in and of itself. Only to find out later that a knife in the kitchen had fallen to the ground or a sausage or cheese on the counter was missing (when nobody was there). It went so far that my mother thought my father had eaten up everything and only accidentally found out he didn’t. And usually then we “knew” 😉 Charlie had acted “suspicious” for a reason (it must have been him, otherwise it was the house elf 😉 ). That is proof enough for me, that dogs very well know rules and that some obviously have such strong feelings about obeying those rules, that they show signs of guilt even when no one of the family can know about the disobedience.
Did Charlie know the rule to not steel food even when no one was in the room? – 100 %
Did Charlie care about obeying this rule? – 100 %
Did Charlie obey this rule more than say 95-98 % of the time? – Absolutely!
Did Charlie act as if he was feeling guilt when he didn’t obey this rule? – Yes.
Did he feel guilt? – I cannot say for sure. But why not? Had he acted normal, we’d never known. Of course, we tought him this rule, proofed it, generalised it, corrected him etc. But we weren’t in the room or on the way to go in to see his disobedience. He wasn’t behaving guilty in the room where his disobedience happened but elsewhere. Had he experience with negative consequences after something like that? – No.
Did Nico (another one of our dogs) ever behave guilty? – No, never. He still had an idea about rules. He just didn’t have obedience as priority #1 in his life and felt strong about obeying the rules (in contrast to Charlie).
For the record: Charlie was trained solely with positive reinforcement and relationship-based training. Nico likewise.
Patricia, I encourage you to try giving your dogs responsibility even more with your certainly very well trained dogs. You can easily see them think and go like “oh right, that’s what X (insert asked behaviour) means”. You can see it sink in, I assure you 🙂 And in case it doesn’t you do as always.
Oh how I love these discussions! Because of them I had an insight this morning while playing fetch with Maggie. I was continuing to think about my “What are you doing?” to Maggie while practicing driving, and had one of those great Ah ha! moments. First, it occurred to me that “asking Maggie to take responsibility” was similar to waiting for a dog to perform the right action from an operant perspective. Think clicker training, while you wait for a dog to move the correct way, rather than luring them. In both cases, we are asking the dog to figure it out what “right” was for itself. The primary difference between what I was doing with Maggie and waiting for the dog to do what you want in clicker training, is that rather than saying “wrong,” which many use to communicate to a dog who might need some direction, I said “What are you doing?” What is most relevant between those two cues is the tone. If I think a dog needs help while clicker training I’ll say “wrong” in as neutral a tone as I can. When I said “What are you doing?” to Maggie, it clearly conveyed some frustration. (But keep in mind, it was ridiculously mild. She was working 100 yards away, and I didn’t yell it. I just changed the tone of my voice.) Because it wasn’t said in a neutral tone, I think it qualifies as a correction, which is where trick training and sheep herding do indeed diverge. Working sheep the way a dog wants to is the most reinforcing thing the dog can do, at least in the short term. You simply can’t let the dog do what it wants, like you can with trick training, and wait for the correct behavior. The dog is already being reinforced with something more powerful than anything in your repertoire. So, yes, sometimes mild corrections, like I gave to Maggie, are essential. But the motivation of my words was to communicate to Maggie that 1) what she was doing was wrong, and, 2) she needed to find ‘right’ herself, rather than me ‘luring’ her verbally into it. In both cases, the point is that it is better for the dog to figure it out for themselves, rather than you ‘fixing it’ (including using a lure). Make sense? (Although it’s late in the game, I’m going to add this to the post itself. I’d love to see more discussion on it.)
Re: whether or not dogs understand words: Tone IS important, and words might not mean to the dog what they mean to us, but many dogs have a very large vocabulary. Jack, as I’ve mentioned, knows all his toys and learns a new one now with 2 repetitions (since he has the concept down). Just the other day, I asked him “Where’s ball-in-ball? Get ball-in-ball.” He went on a hunt, and I said (without moving my eyes or changing my tone) “I think it’s in the basket.” He immediately stopped his search, ran to his toy basket, and pulled it out. I never trained him “basket.” I don’t refer to the basket on anything like a regular basis. But he picked it up because it’s something that is meaningful to him.
Re: Guilty dogs. I absolutely agree that a huge portion of what we call “guilty” is appeasement behavior and it does not mean the dog did something wrong. BUT dogs are capable of experiencing something like horror if certain rules important to them are broken. Jack, as a puppy, defecated in the house exactly once and was not corrected nor was I displeased, since it was his first day home (peeing was another matter all together).
However, if he is ill (on rare occasion) and has messed in the house, when I come home, he is frantic. He is waiting for me at the door and he will run to me and leap up at me repeatedly. He is, as best I can tell, mortified. Since he comes to me with a “help me, help me!” look and not an “I’m so sorry” look, I doubt he is expecting punishment. Indeed, not only am I genuinely not angry, I feel bad for him.
If Maddie was the one who could not hold it, he is not frantic but he does look a bit disgusted and will carefully avoid the area til it’s thoroughly cleaned.
So clearly, “feces in the house” is an obvious source of disgust, but “I could not hold it and I had to go on the floor!!!” is a source of extreme distress.
Re-reading my earlier comment (made, I confess, when I was crunched for time), I realize that I was not as clear as I could have been re: Sandy and her recall. What I said only included half of what I meant: When Sandy waits for my cue, despite knowing that it is time to turn back, she is not doing anything ‘wrong’, just pushing the envelope to see how far she can go before the party is over. Since she’s just a pet, I have no problem just remembering that she needs to be cued EVERY time when she’s in that situation and calling out a redirection, not a correction.
If, instead, I changed my tone and instead uttered a shocked and disappointed, “Sandy!”, I have no doubt that she would do exactly the same action: whip back around and come dashing back. As I say, she knows what to do in that situation. But I also have no doubt that she would view that as a mild correction. I do NOT think, for a second, that it would do her any harm, and I have had rare occasion to use just such a correction when she is headed for trouble.
The difference is that I don’t really mind having to cue her at the hedgerow. There’s no real danger, and no real reason to introduce even a mild aversive. IF, however, Sandy’s envelope-pushing were somehow dangerous, rude, or counterproductive, I WOULD have a good reason to try to stop her from engaging in that behavior in the future. Since habitually cuing and rewarding the desired behavior is not enough, I would need to choose a different path.
In Sandy’s case, I could probably escalate the value of the reward I’m offering, since there’s no problem if she runs to me in that spot rather than exploring the hedgerow (I don’t really want this, because I enjoy our free rambles and think it adds satisfaction and variety to her life- I don’t really want her glued to my heel instead of enjoying her walk), but in a herding dog, it may be wildly impractical (if not impossible) to offer an enticing enough reward, mid-drive.
OR, I could offer a correction, to encourage her not to make the same choice next time. Obviously, I’d have to weigh the cost vs. benefit: how much will the correction distress her? How unpleasant is the task itself? How valuable to her health, safety, and future happiness is it that she learn to perform the behavior I want and avoid the one I don’t without cues from me? What possible consequences could there be if she takes this too hard? (again, the specter of the dog glued to heel, not enjoying the hedgerow that brings her so much pleasure-unlikely in this case, since I know how resilient she is in similar circumstances, but a different dog is always a different story).
As I did manage to say above, over time, Sandy has gone from needing LOTS of cues and feedback, to requiring fewer and fewer. Most of this was accomplished with positive reinforcement, some of this with mild (verbal) corrections, but in both cases the effect: a dog who can be trusted to make at least 95% of her own choices when we’re out on a hike has been absolutely worth what small and temporary stress I may have caused in teaching her the skills she needs. The gains in confidence, emotional comfort, and uninterrupted pleasure have more than paid for the (minimal) pain of the learning process. Still, I try hard to weigh each circumstance on its own merits- where I can spare her discomfort or distress without compromising her health, safety, or general happiness, I do. When I can’t, I try to get the result I need from my dogs with the least fear, pain, danger, or aggravation to them possible.
I aim for compassionate training, but I don’t really know if a companion animal can EVER be taught with “all positive” stimuli- meaning that the animal never feels any distress in the course of training- and sometimes I find myself making an instinctive call based on what I know of my dogs, about whether a single firm verbal reprimand will be better or worse than weeks of redirection. Anyhow, I’m so glad to have had the chance to really think about some of these issues!
Woooop! Congratulations Trisha.
I can not wait and it’s going straight to the top of my reading pile.
Re dogs knowing the words: My experience is that dogs’ vocabularies vary considerably. Red Dog pays very little attention to words; she knows the basic obedience commands but is much more attuned to tone of voice, body language, and patterns of behavior.
In contrast, we had a Jack Russell mix who had an incredible vocabulary. Typical “conversation”: Human: “Go get your horse.” Dog: quizzical expression. Human: “It’s in the kitchen.” Dog: Runs to kitchen to retrieve stuffed horse.
The Jack Russell mix also figured out that spelling out “W-A-L-K” was the same as “walk”. We had to consult the thesaurus regularly to confound the dog: “Honey, would you like to perambulate with the canines?”
@Bruce, Dogs always learn the words that are relevant to them. Currently at my house we take excursions to the exterior of the domicile I estimate we have about a week before we need to change how we refer to walks again. Ranger has a huge vocabulary of spoken words Finna has an equally huge vocabulary of human body language. She can tell just by how I stand up out of my chair whether I’m going to take her outside or take Ranger to work. I have no idea how she does this since the initial steps are identical right up until the moment I say to Ranger “ready to go to work?” (ready, go, and work are words he knows) Finna, however, is not dancing with impatience by the door the way she is when she’s going to get to go out. When I’ll be taking Ranger to work (he’s a therapy dog) she’ll have gone to what we call her sad place. She sits on the couch at the end nearest the window where she can see the car come home and looks so abandoned and forlorn.
@kat do they? If a say to to my hubby “shall we go for a walk” the guys just keep on napping. If I ask them “shall we go for a walk” in a different intonation, they run to the backdoor. I tried teaching them to find toys by name, and they do not seem to grasp the subject.
When we close our laptop/iPad in the evening, they jump up “time for the last walk” to sometime be disappointed… because I am just going to make tea 😉
Because they ALSO come when we do go for a walk, it seems they know what we are about but there are many misses.
My former Siberian did have one amazing behaviour: he followed me whenever I went to the kitchen to get the cheese out of the fridge. Not when I went to get something to drink or veggies or.. only when I got the cheese even before I had my hand on it. But then again: he also sang in the car for the last part of our journey (whahaheaawawhawhaha) if we went somewhere nice. One day we were driving to the breeder – het loved visiting there – but the road was blocked. He was singing “whahehawawhwah” but stopped abruptly with an amazed WHHAAAA?? when we took an unexpected turn because of the roadblock. Hilarious! Still miss him and his antics 😉 he trained us more than we him…
@Kat – Yes, exactly. The language-attuned dogs knew that when I said, “So . . .” in a particular tone of voice, something good usually followed (“. . . would you guys like to go for a walk?” [. . . eat dinner, ride in the car, etc.]). People laughed to see how quickly the dogs spooled up for action after hearing just the word “So . . . .”
As you pointed out, having a language-attuned dog has its pros and cons. I do miss being able to speak to the JRT mix in ordinary sentences, and to have him understand what I meant.
I wonder why some dogs develop such a strong understanding of spoken language, and other dogs do not – probably some interesting nature-vs-nurture research to be done.
Re dogs and human language,
I love the comments about this and I’m right there with people who’ve had dogs who understood spoken words and dogs who either did, and couldn’t have cared less about what I was saying, or didn’t understand at all. Seamus has a huge vocabulary. It comes in handy when we’re out working because I can teach him things like how to find the elevator or a bench or a trash can. He knows both “elevator” and “button” for the same thing, though usually I use elevator and I’ve taught him to find the panel with the buttons on it rather than just finding the doors like they teach at his school. What’s amazing to me though, is the vocabulary he’s picked up without me deliberately teaching it to him. When I’m discussing going to sleep for the night with my husband, sometimes he’s up later than I am, the dog will get up and walk into our room and into his crate. I haven’t even commanded him or anything, just up and off to bed. If I say, go bug “Other Human,” my husband’s preferred nickname regarding the dog. I’m Mama-human, but as he says in his southern drall, “Ain’t no kid of mine,” so he’s other-human, Seamus runs to him. It’s the same with my sister, who is Auntie. He also knows toy, a general term for all of the toys in his basket. I’ve named them all, but haven’t made an effort to teach him those names. this is all fun to watch, but what amazed me was what happened this morning. I was up, getting ready for work and my ride was late. I told them to call when they were at my apartment and when Seamus heard the ringing of my phone he jumped off the couch and walked to the front closet where I keep his harness. What’s awesome about that is that if he hears my phone ring at other times, he’ll get up and go into his crate, assuming he isn’t going with my husband and me. It’s so context rich and what he pays attention to is just amazing. I’ve never had a dog who paid as much attention to what I was saying as he does. The only other time I had a dog pay attention to anything I said outside of working commands was my first, Marlin. I was sitting at the kitchen table, reading an email my friend had sent me about her dog. The dog had gotten into a lovely chocolate cake she had made her husband for his birthday. The dog felt sick, and was a bit sick, but in no danger. Still, not a good doggy thing to do. As I read I gasped out “nauty!” The only way I can describe the tone was dismay/surprise. Marlin came running over and put his head in my lap. I hadn’t even known he was in the kitchen with me and here he was, letting me know what ever had happened, he hadn’t had a thing to do with it. I laughed at his reaction and he got an ear-scratch for being so good. did he respond to my tone, the word I’d used? I don’t know, but something made him come over to investigate. I still miss that wonderful dog.
When reading your post I thought of the horse I owned once – a very smart Tennessee Walker. I was leading her through the pasture to take her back to the barn for a ride when I walked to the left of a small sapling and she tried to go to the right. I just stopped, turned to look at her (and I am sure I probably raised an eyebrow at her) and she backed-up and went to the left side with me. Sometimes we humans over rate words.
I personally love any comments you make about sheepherding. I did agility with my BC, and then started herding, and herding is by far the most challenging. I think the problem novice handlers (like me) have is that in many cases the dog DOES know more than the handler. So there is a steep learning curve for us newbies. I know a veterinarian sheepherder whose professional practice is primarily behavior. She said that she has lain awake nights pondering the unique challenges of training herding. By the way, she does use clickers for training the “in here” command when starting shedding. And as you know, clickers are not a familiar word in the sheepdog training world. It has taken me a while to even be able to give a correction; fortunately my dog is very biddable, but as we know, that can change around sheep!
Re: words vs tone: With a language-attuned dog (I love that phrase), words have meaning, regardless of tone (though just like with people, tone matters— watch the comedy sketch about how many different things “dude” can mean to see how much tone matters to us as well).
My favorite example of this is that when Jack was younger, he had a set of three interlocking hard rubber rings (Maddie the Destroyer made such a toy no longer ownable). We called them, logically enough, “rings.”
We were out on our deck with a small telescope, looking at planets. Jack was with us. Hubby and I were chatting about what we were seeing, and he said “Oh, look, you can really see the rings of Saturn.”
Jack went to the door and barked. I thought maybe he had to go potty (our deck is gated and he can’t get off it). I opened the door and followed him in the house. But instead of running to the front door to go out, he ran right into the living room and came out with his rings, eyes shining, ready for a game.
We were not interacting with the dog at all. We did not give a “pay attention” word like “Do you want to….” or “Where are….”.
We just said the word “rings” in an ordinary tone in the middle of a conversation in which we were totally ignoring Jack, and off he went to get his rings.
I often think that humans talking sounds a lot like the Charlie Brown grownups to dogs, and then you say a word they know.