My dog follows me everywhere - is that a good sign?

The dog the diagnosis is made on, and the only case where the whole arc of it is on film. It arrives as an improvement: she started out refusing to be touched, and by the end of the first long session she is holding onto a leg and circling it - which the trainer names as her having switched to using physical contact to manage anxiety. His prescription is two things at once that sound contradictory and are not: bond hard with a nervous dog, which he calls one of the most important things a handler does, and be very detached about supplying the contact. The consequence he cares about is practical rather than emotional: she is tied to him because she is nervous, and he cannot honestly test her recall, because there is nothing to call her away from. The drill is a two-hander. Let her Velcro to handler B, recall her off him, mark the refusal, repeat, and vary the setup so she cannot pattern it - and he catches her patterning it anyway three repetitions in, which is the detail that makes the drill credible. He also states the reason with the family in mind: this is the failure he predicts for them, a dog who tries to hide behind her people when a command arrives. His own Tier 1 note records that she had already done it with two handlers. And her file contains the argument for why the alternative is worse: the Velcro dog is made with an e-collar and enough reinforcement to frighten a dog off leaving your side, and what you have then is a dog following you around for life with, in his words, no soul anymore.

The case where the problem is nothing but a training obstacle, stated in one honest line. A session opens with the plan announced - teaching come, so that if she goes far and he says come she can come - and then the session cannot happen, because she will not go far. What he says instead is the cleanest statement of the concept's practical half anywhere in the corpus: how can I train her to come if she will not leave my side. He tries walking away, retrieving the leash, sending her out, and cueing come anyway; none of it produces distance, and the long line that would have is not there. It is worth recording for two reasons. It is the mirror image of the flight-risk work the same program does elsewhere, where the whole problem is a dog who will leave. And it says something about prerequisites: before you can train a recall you need a way to produce distance, and distance is equipment as much as it is training.

Not a nervous dog at all - a ten-month-old who does not know how strong he is - and that is why his case matters here. The concept appears on him as prevention rather than treatment, and as a deliberate action taken at the least intuitive moment. Two minutes into marker work, at the exact point where control has just become easy, the handler lengthens the line: now that he understands the concept, loosen up a little, because a Velcro dog is the kind that sticks by you all day and cannot go anywhere. The lengthening is explicitly not a reward for good behaviour; it is a step in the plan. At his handover the same instinct shows up as a statement of what the program is for - keep the personality, just make them obedient - and as a refusal said in the dog's own second person, that the goal is not to repress him into something frozen sitting there all day. The same handover carries the mechanism for why the Velcro version is dangerous rather than merely sad: nervous dogs kept permanently at heel get frustrated, and a year or so later they bite somebody.

The dog where the reading is made on a walk, out loud, with the behaviour on screen - and it is the reading an owner is most likely to get backwards. She closes distance on the handler and he stops to point it out, because somebody watching might not know: she comes close to him, and that is an anxious behaviour that cannot be allowed. No correction is narrated; the behaviour is simply named and worked away from. Her file is the shortest evidence in this concept and the most useful for an owner, because the same twenty seconds contains the thing the dog does and the interpretation the trainer puts on it, and every owner who has been pleased by a dog that shadows them will recognise the picture.

The case that supplies the moral line rather than the diagnosis. Her handover contains no Velcro dog and no recall problem; what it has is the positive statement of the same value, made to a client who is about to take her dog home. He would not want to make her less goofy - he trained her, and she is still going to be a goofy dog, because he still wants them to be them. And then he draws the boundary explicitly: he is very morally against repressing a dog, a smidgen of repression is acceptable because what is being repressed is a behaviour, and it becomes unacceptable at the point where the personality is what is being repressed. Read against Lola's soulless-Velcro-dog line, this is the same position stated from the other end, by a different handler about a different dog - which is what makes it a company position rather than one man's turn of phrase.

Whether a nervous dog coming to you is progress or a symptom.

Both are said about Lola, by handlers whose transcripts do not name them, weeks apart. They are not obviously reconcilable from the footage: the same behaviour - the dog closing the distance and making contact - is the symptom in one file and the result in the other. The distinction the concept needs is presumably about the dog's state when she does it rather than the act itself, and nobody on film says so.

How a nervous dog's refusal inside the two-handler drill should be read.

The same handler, ninety seconds apart, gives two different readings of the same failure - and the second does not retract the first. It matters because the two point to different fixes: one says keep running the drill, the other says the drill has become a pattern and must be varied. He says vary it; the corpus does not say how you tell which is happening on any given repetition.

Whether the shadowing is a welfare problem or a training problem.

Not a disagreement so much as two concepts sharing a name. The welfare version is about a state a trainer can manufacture; the training version is about a dog who simply will not produce distance, in a case where the dog is not described as nervous at all. They have the same cure - send her away - and different stakes, and the corpus never distinguishes them.

My dog follows me everywhere. Isn't that what a bonded dog does?

Sometimes, and on a nervous dog it is usually something else. One trainer describes the change happening in front of him: the dog had refused his hand for days, and then started holding onto his leg and circling it - and what he names that as is her using physical contact to manage her anxiety. His answer is not coldness. He says in the same session that bonding very well with a nervous dog is one of the most important things a handler does, and also that in this situation you have to be very detached - available, and not supplying contact on demand. The test is simple enough to run at home. In a big safe space on a long line, walk away from her. If she will not leave you at all, what you have is not a bond you can rely on; it is a dog who cannot cope without you standing there.

Her recall is perfect - she always comes. Why do you say it is untested?

Because a recall is a command for the case where the dog is far away, and yours has never been that. One handler puts it as a question while a session falls apart around him: how can I train her to come if she will not leave my side? He tried walking off, he tried sending her out, he cued come anyway, and nothing produced distance - and the long line that would have was not with him. Another says the same thing about a nervous dog in the plainest terms: she is tied to me because she is nervous, and he has a hard time testing her recall at all. So the first job is not the recall. It is producing distance, which is partly training and partly a fifty-foot line.

What do I actually do about it?

Send her away, and if there are two of you at home, call her off the other one. The drill is named on film and it is unfussy: let her stick to the second person, then call her to you, mark a refusal with an uh-uh and repeat, and vary the setup so she cannot work out what is coming. The stated aim is not obedience - it is to get her out of the mental habit of Velcroing to anybody. It also rehearses in advance the exact failure the trainers predict for a family at home, which is a dog who goes behind her people when a command arrives. And it is worth knowing that the handler running it caught the dog anticipating the pattern after three repetitions, which is why the varying matters rather than being a nicety.

Isn't a dog who sticks to you exactly what obedience training is supposed to produce?

It is what some obedience training produces, and the trainers in this corpus are against it in unusually strong terms. There is a name for the result - a Velcro dog - and one of them explains how it is made: take a nervous dog, or one that does not listen, use the e-collar and enough reinforcement that the dog is frightened to leave your side, and you have a dog that follows the owner around for the rest of its life, repressed so hard, in his words, that it has no soul anymore. A second handler puts the same position positively about a different dog: keep the personality, just make them obedient, and no intention of repressing the dog into something frozen sitting there all day. A third draws the line for a client at the handover - a smidgen of repression is fine because what is being repressed is a behaviour, and it stops being fine at the point where the personality is what is going.

My dog is not nervous, he is a happy puppy who wants to be near me. Does this apply?

It applies as prevention, and there is a case in this corpus where that is exactly how it is used. A ten-month-old with no nerves at all is having his first marker work, and two minutes in - at the moment the handler has him nicely under control - the line gets lengthened on purpose. The stated reason is not that the dog earned it; it is that the handler does not want to build a Velcro dog, the kind that sticks by you all day and cannot go anywhere. That is the useful transferable habit: give the distance back at the point where holding him close has just got easy, because that is the point at which it is tempting not to. The neighbouring warning in the same dog's file is what makes it worth bothering with - nervous dogs kept permanently at heel get frustrated, and about a year later they bite somebody.

So should I stop cuddling my nervous dog?

Nobody in this corpus says that, and one of them says nearly the opposite - bonding very well with a nervous dog is named as one of the most important things a handler does. The instruction is about who initiates and what the contact is doing, not about whether it happens: be physically available and emotionally detached, and do not supply contact on demand. It is also worth being honest that the corpus does not settle this cleanly. The same dog whose leg-holding is diagnosed as anxiety management is, in another session, described as far more open than when they met, and the evidence given is that she will come up and snuggle. Those are the same behaviour read two ways, weeks apart, and nothing on film tells you how to tell them apart in the moment.

Reading the shadowing as devotion, and reinforcing it - feeding contact every time the dog asks for it.

Reading the dog closing the distance on a walk as her checking in with you.

Letting the dog hide behind you - or behind whoever else is in the room - when a command arrives, so the refusal is never answered.

Holding the dog close because it is working, and never giving the distance back. The corpus names the endpoint of that habit and it is a bite about a year later.

Buying a quiet, glued-to-you dog as a trained one. The corpus says how that dog is made and what is left of him.

Assuming the recall is fine because the dog always comes. If she is never more than a few feet away, the recall has never been tested.

She will not go more than a few feet from you outdoors, even off leash in a safe space - and you have been pleased about it.

You cannot practise a recall, because she is never far enough away for the word to mean anything.

She leans on your leg or circles it when something worries her, and it started after she began to trust you rather than before.

When you ask her for something she goes behind you, or behind whoever else is in the room, instead of answering.

She closes the distance on you on a walk whenever anything happens, and you have read that as her checking in with you.

Her commands are genuinely good and she is still no better in herself - fast, clean responses, and the same dog underneath.