How do I stop my dog reacting - for real, not just on the surface?

The case where the distinction is the quality bar for everything, stated on day one before a single command is taught. The claim is diagnostic rather than moral: it is very easy to make a nervous dog look opened up when what you have actually done is repress her, so compliance cannot be the measure and the absence of repression has to be. He adds that with a dog like her you cannot do even a hint of it, and mid-programme he grades her as borderline suppressed already and says they are working to keep her open - which is a handler reporting a near-miss against his own standard. The housemate's session supplies the test that makes the standard checkable: a suppressed dog reverts to its old behaviour the moment it goes home, which means the family paid two and a half to three and a half thousand dollars for nothing. And the same file names what actually does the damage, which is not pressure but pressure the dog cannot read - non-contingent pressure, with no understandable cause and no way out.

The dog on whom the argument is made twice, in two different currencies, to the same family. At the first owner lesson it is a decay curve with a timescale: a two-week board-and-train can hand back a dog who stays when told, because he has been repressed into staying, and a week or two after go-home the repression fades and the training goes with it. That is the reason the programme deliberately takes longer than anybody would like - he would rather spend the time than traumatise the dog into impulse control. At the handover it arrives again as a moral line, drawn by degree rather than by kind, and volunteered unprompted: he is very morally against repressing a dog, and in very light doses it is fine. A smidgen of repression of a behaviour is acceptable; repression of the personality is not. Then he cashes it out in the smallest possible way - he would not want to make her less goofy - and carries it with the job analogy, which is the most owner-usable version of the idea in the corpus: a person who learns to focus during working hours is the same person, and a person who has lost themselves is not. The same session shows the cost of the position: he refuses to be a hardcore enforcer about the shape of her down and settles for the position most of the time.

The dog whose file contains the model rather than the slogan. The two tracks are the body and the mind, and the target is the two of them in sync - correct position and a positive mental state at the same time, not one bought with the other. What makes the passage unusual is what the trainer says next: you can go too far in either direction, and he has done both. Too far towards the picture, and the dog's emotional state goes into the toilet; too far towards the feeling, and the dog is happy and not doing the things. He also states the trade honestly - whatever speed you gain on the front end by forcing the picture you pay for later, and it looks bad as well - and declines to add everything at once with the e-collar on the explicit grounds that he could get the result and would repress the dog getting it. In the same file the corpus's clearest statement of the method's own name arrives: they are not force-free, they run a balanced method, and they say so on camera. And the way progress is reported for this dog is entirely in the second track: more interested in sniffing than he has ever been, acting like a dog who is not in constant fear, curious about people, and - stated as the goal for the family - a life with adventures in it rather than a list of commands.

The case that puts a price on the distinction by offering to give up the obedience. The trainer says outright that obedience is the easy part and not the point of this programme, and then makes the trade explicit: he would rather the dog went home not knowing a lick of heel or down - recall excepted - if he could hear dogs barking without panicking, because that is the family's actual functional problem. The session's argument is built on one ugly repetition and what caused it: the rep fell apart not because the dog forgot the heel but because dogs barked at him and he went hypervigilant, and then he had pressure coming from two directions at once. The lesson worth having, in the trainer's account, is that solving the puzzle makes all of it vanish at once and something good arrives immediately - and that teaches something far beyond obedience, which is that stress is survivable and recovery can be fast. In the following session the same logic is instrumental rather than ethical: confidence is what will make integrating the two household dogs possible at all.

The dog on whom the principle is stated in its shortest form and its sharpest form, at opposite ends of the programme. On a field trip to a store: emotional state change is the key to progress, not being able to make her sit and make her down. At the go-home lesson: stacking a down on top of her fear is a band-aid, because you have addressed the physical state of her body and made her stationary, and what is happening inside may not have changed at all. Between those two sits the mechanism she is the corpus's best evidence for - management kept deliberately loose so that her own predictions can fail, because restricting her options too tightly stops learning and produces a dog who panics when the restriction is removed. And her file supplies the reason skills matter at all in a case like hers, which is not obedience but transfer: how a dog responds to pressure inside a command bleeds over into how she responds to pressure everywhere.

The constructive half, and the only place in the corpus where what replaces the behaviour is specified. The house approach is stated as phases rather than as a punishment: they do not go in and punish the behaviour, they change the emotional state around the thing and make it cooperative with authority and obedience in it. Then the specifics. The two patterns the dog has are named. The crate is ruled out as a punishment in the strongest terms in the whole corpus - no timeout for dogs in these cases, and he says he wants to emphasise that. The old pattern is violated by not leaving; obedience is asked for and reinforced in the same place; and then the dog comes out to play and walk, so a complete different pattern is installed over the top of the other one - "and when we just layer over that, then it it goes away." Ten months later the outcome is reported in a store in the currency this concept asks for: not that he complies near strangers, but that he wants hugs from them, and that at intake he came out of his own front door bristling at everybody he saw.

The mechanism claim, and the only one in this set that is falsifiable. The trainer says flatly that he does not repress dogs, and then explains why he does not need to on this case: fix the over-excitement and there is nothing left to repress. The arousal is allowed to exist and she is taught not to act on it - when you feel a high impulse of energy you can have fun, just do not let go of your impulse - and because the bites came out of over-excitement in the first place, removing the over-excitement is expected to take the bites with it. He generalises it to a breed group and then, unusually, applies the generalisation to her own specific events, and the owner confirms it from her side. He is also careful about a confound the rest of the corpus mostly ignores: how strict the handler is, is how tight the dog behaves, so what the family are watching includes him.

The dog whose file contains the bluntest version of the anti-suppression line and the most specific instance of how suppression gets built by accident. The blunt version arrives at the first owner lesson, immediately after telling the family not to expect a perfect dog: trainers turn their dogs into robots, they repress them for two weeks and the dog loses its soul, and I am not about that life. The specific version is at the handover, and it is an e-collar argument made on emotional grounds. Do not hit the collar every time the dog sees a dog - because a dog with nervous tendencies will build an aversion and conclude that all dogs are bad and dangerous, since every time she sees one she feels the collar. The alternative given is to cue something else every single time instead, so that what an onlooker eventually sees is a dog who does not care. Her file also contains the corpus's frankest note on owner-facing language: the trainer keeps 'conditioning' for what is happening and 'muscle memory' for explaining it, and says so.

The second independent statement of the moral line, in a different handler's mouth and about a very different dog, which is what makes it a company position rather than one person's phrase. The formulation is the shortest in the corpus - keep the personality, just make them obedient - and he names what he is refusing in the dog's own second person: his goal is not to repress this dog into something frozen that sits there all day. Submission is not the objective and personality dictates the approach. Around it he gives the mechanism for how suppression turns into a bite, which nothing else in the corpus supplies: heel grinding, keeping a dog at heel constantly, produces frustration in nervous dogs, and about a year later they start biting people out of nowhere. And he gives the corresponding failure on the other side, which is a word with nothing behind it: use the marker and never the button, and it looks good for a couple of weeks and then goes downhill, because there is no backup.

The physiological evidence, and the best single piece of it in the corpus, because hackles are not a decision. At the handover the resident dogs both have their hair standing up along the ridge of the back and the trainer points it out to the household as a thing to read. Then he grades the dog he trained against them: his hackles did not go up, he looked and did nothing - and, he adds, that is not what he was doing for the first three quarters of the programme, when he always put his hackles up and the same situation would turn into a humongous barking fest. The rest of his case sits behind that: the arrival behaviour re-read as insecurity, the weeks of presence work in which no obedience was asked at all, and the trainer's own statement of what those weeks were teaching - that a person can be in his space and nothing happens, so he does not have to protect himself or his space.

The dog whose file argues the same position from the uncomfortable end, and is the honest counterweight to everything else on this page. The trainer states the method without softening - negative reinforcement and obligation, and the dog is uncomfortable a few times, you shock them, you are a little mean to them sometimes - and then justifies it by the life it buys: teach them things and by doing so you open up their life. His measure is not compliance either; it is freedom, and he quantifies it against the same dog three weeks earlier, when he could not take his eyes off him for one second. He also puts a condition on the pressure that matters more than the rest of the sentence: it has to clarify, in a teaching way, and it sits inside a great deal of positive reinforcement. The evidence he offers in the same session for a real change rather than a suppressed one is the kind this page asks for - squirrels he never noticed while he was sniffing trees, and a down held with a strange dog a foot away.

The inverse case, and the reason this concept is not only about reactive dogs. Dolce is a protection dog being started in defence work, so here the worry runs the other way: not that the training has flattened her, but that it has made her internally defensive in situations that do not call for it. On her first public outing the stated purpose is to check exactly that - she is going into defence work, and they want to make sure she is not becoming overly defensive in random scenarios - and the observable that says she passed is about her state rather than her behaviour: she is checking things out properly, not fixating and not being defensive or reactive just because there are people about. The same session shows the pre-condition being met before anything is asked of her, which is the whole page in one sentence of method: let her open up, sniff and relax before adding any public-area work.

What the existing concept page says, against what the corpus now shows. The page at concept/internal-change.html keeps its URL; its text is superseded in four ways.

Not a disagreement between trainers, and the old page is not wrong - it is one sentence where the corpus has a body of argument. Recorded rather than resolved because what the live page says is the owner's decision. Four things would change if it were rewritten: the question would widen beyond dog reactivity, since the strongest statements of the idea in the corpus are about a crate case, a nervous rescue, a puppy's personality and a protection dog's defensiveness; the decay curve and its price would appear, because that is the version that works on a family comparing two-week quotes; the diagnostic - that you cannot tell the two apart at handover, and that the test is delayed - would appear, because it is what makes the concept actionable rather than reassuring; and the limit would be stated, because 'we never repress anything' is not what the corpus says and a page that implies it will not survive contact with the footage.

Is any amount of repression acceptable?

Unresolved, and the two may be talking about different things rather than disagreeing: one draws the line between behaviour and personality, the other makes the dog's nervousness the reason for a stricter standard ('a dog like her'). But they are stated as general positions, to clients, in the same weeks, and one of them uses the word 'ever'. A staff handbook cannot carry both sentences as written.

Is obedience the vehicle for the emotional change, or a distraction from it?

Both sides are the same business in the same season, and the dogs differ in exactly the way that would explain it - an over-aroused dog versus two frightened ones. Nobody says that, though, and the two positions imply different deliverables and different invoices. It is also the conflict most likely to reach a client, because 'we might send him home without a heel' and 'we teach the obedience and the rest follows' are not the same sale.

How the same method is described to the viewer.

Not a contradiction - both are the same balanced method, and the corpus names it as balanced on camera - but they are very different pieces of publishable material, and a page that carries only the second one is not showing what the first one admits. Both are worth keeping for exactly that reason.

My dog is quiet on walks now. Isn't that what I wanted?

Quiet behavior is one observation; it does not by itself tell you why the dog is quiet. Lola’s trainer distinguishes an appearance of openness from the change he is trying to produce (FZ4UONAzt8s, 06:44). Stormie’s discussion raises the question of what continues after the training setting changes (TN1flHN40Vw, 07:46). Those explanations support looking beyond stillness to the broader response and later settings. The mentioned week-or-two timing is the speaker’s example, not a follow-up result established for every dog in this archive.

Will an e-collar break my dog's personality?

The trainers answer this without ducking it, and they answer it in two halves. The first half is that the risk is real and they name it as the thing they define themselves against: trainers who repress a dog for two weeks and hand back something that has lost its soul. One says he is very morally against repressing a dog, and then draws the line by degree rather than by kind - a small amount of repression of a behaviour is acceptable, repression of the personality is not - and cashes it out by saying he would not want to make a goofy puppy less goofy. The second half is the cost, stated by a different trainer in the plainest available terms: the dog is uncomfortable a few times, you shock them, you are a little mean to them sometimes. He puts a condition on it in the next breath - it has to clarify rather than just hurt, and it sits inside a lot of positive reinforcement - and then says what it is for, which is a dog who can be let off the lead at all.

Why can't you just correct him when he reacts to another dog?

Because with a nervous dog you would be teaching the wrong lesson. The argument is made explicitly at a handover: do not hit the collar every time she sees a dog, because what she learns is that dogs are what hurt her - every time I see a dog I feel this, so all dogs are bad and dangerous. That builds an aversion rather than removing one. What is done instead is to cue something else at every single instance of the trigger until the cue is what she expects when a dog appears, at which point an onlooker sees a dog who simply does not care. The same trainer adds the rule that keeps you out of the other ditch: never punish a position you did not ask for, because a dog told off for getting up from a sit she was never put in cannot learn anything from it.

Why is the programme taking so long?

Because the fast version is the one that expires. The trainer says it to the family directly: sometimes it takes a bit longer in the programme than anybody would like, and he would rather the dog learned the impulse control and became comfortable with it than be traumatised or repressed into it. The comparison he is making is to a two-week board-and-train, which can produce a dog who stays when told by repressing him into staying - and which costs the dog its personality and, a fortnight after go-home, the training as well. He also refuses to inflate the progress report in either direction, telling the same family that their dog is on schedule and that he has told other people their dog was behind.

What should I actually be watching for, instead of obedience?

The things the dog does when nobody has asked him for anything. In this corpus the progress reports that count are: hackles that no longer rise when another dog appears; a dog who sniffs and explores instead of scanning, and who is described as acting like a dog who is not in constant fear; curiosity about strangers, tracked separately from tolerating them; squirrels going unnoticed because he is busy with a tree; and - the plainest one - a dog who used to bristle at everybody wanting hugs from people in a shop. Note what these have in common. None of them is a command, and none of them can be produced by pressure.

Is it different if my dog isn't reactive at all - if he is just badly behaved, or if he does protection work?

The same test is applied, and the corpus shows it running in both directions. On a soft, over-friendly puppy the concern is the personality: he would not want to make her less goofy, and obedience is explained as learning to focus during working hours rather than becoming a different animal. On a protection dog it runs the other way - she is being started in defence work, and the thing being checked at her first public outing is whether she is becoming internally defensive in ordinary situations she has no reason to be defensive in. The observable is the same either way and it is about her state rather than her compliance: is she investigating things properly, or fixating on them?

Buying the fast version. A two-week board-and-train can hand back a dog who stays when told because he has been repressed into staying, and the compliance expires with the repression.

Correcting the trigger instead of the behaviour. Tapping the collar every time a dog appears teaches a nervous dog that dogs are what hurt her.

Punishing a position the dog was never put in. A dog who sits down because there is nothing to do, and is then told off for getting up, has no way to learn anything from it.

Holding the dog in a constant restraint, which makes you the restrainer rather than the source of freedom - and is more work than training, for ever.

Keeping the dog at heel permanently as a management strategy. Named as heel grinding, and given a consequence with a timescale: frustration, and a bite out of nowhere about a year later.

Using the crate, or any place the dog has to live in, as a punishment. Ruled out in the strongest terms anywhere in the corpus, with the trainer saying he wants to emphasise it.

Expecting a finished dog to come home. Stated to an owner before the training was done: dogs are not robots, and a dog who comes back perfect is neither the goal nor achievable.

Reading a quiet dog as a trained dog. The whole point of the concept is that the two look the same, and the difference only shows up a week or two after he gets home.

He does what he is told near the thing, but he never stops staring at it - the body is in position and nothing else has changed.

He came back from training brilliant, and a couple of weeks later he was back to his old self.

He is obedient and he is not himself - flatter, quieter, less silly than he used to be.

On a walk your whole job is holding him back, and if you let go you have no idea what happens.

He never sniffs anything on a walk - he scans. And when a dog barks somewhere, the rest of the outing is ruined.

His hair goes up along his back when he sees another dog, and you have stopped noticing because it always does.

His commands are fast and clean and he still cannot be let off a leash, because the problem was never the commands.