How do I teach my dog to make better choices?

Bullet's park session is where the distinction is drawn, and it is drawn against something the viewer can see. Other people's dogs go past on leads, pulling hard, and the handler says those owners do technically have control over their dogs, just not obedience. In the same session he states the ambition that follows from it - knowing commands is good, but ideally you do better than that, so rather than teaching commands and having them followed you teach the dog to make better choices - and then demonstrates it with a dog who declines a close greeting and ignores squirrels. What makes his account credible is that he does not claim the urge has gone. He says the dog thought about it, because the other dog came very close, and that he responded to being pre-empted. On a walk a week earlier, a non-chased squirrel gets four words that put the credit in the right place: it was a good decision.

Roxy's second in-program lesson contains the passage that ought to be the top of this page, because it is the one that answers the complaint families actually arrive with. The trainer walks the owners through it from the dog's point of view: suppose she follows every command every time, and then goes straight back to what she was doing, because she never learned that the thing itself was the problem. When you get frustrated, will she know why? She followed every command, was released to freedom, and did what she wanted - and by the end of it you are saying stop and she is thinking stop what, I did everything you asked. The alternative is stated as a procedure a few minutes earlier, in four sentences: let them make a choice; wrong choice, e-collar; right choice, good job; then back to freedom. And it is stated as a preference with a reason rather than as doctrine - it makes a more responsible dog instead of a more micromanaged one, and the best thing you could possibly have is not having to ask them to do many things because they do not do stupid things. Her file also carries the ordering rule, which is the most easily lost detail in the whole concept: mark the wrong choice first and recall second, because you want the dog to think that running over to that dog is bad rather than that a recall was needed. And her home-context session shows the finished picture, which is a two-item list: she may walk around, she may be in calm liberty, she must settle herself - and jumping on the handler or on the couch is not among her options.

King's lessons supply the word. In the second in-program lesson the trainer says that the route he always takes first is the one that becomes hands off and automatic, because the dog starts learning what is lawful and what is not lawful, and learns to make good choices instead of being managed with obedience. He then applies the word with unusual precision at the following week's walk, to the difference between two cues: if you say let's go, the dog may come to you and circle right back to what he was doing, and that is lawful; if you say come, he is supposed to come to you and stay there. The reason attached is not tidiness but safety - the reason you said come is usually that you want him away from whatever he is going towards, so a recall you have let slide is a recall you cannot use to protect him.

Bo's clip is the shortest and most complete demonstration in the corpus, and it is a clip in which nothing happens. A dog is walked near people with the leash loose and the handler deliberately silent. The dog sniffs the air, looks at what they are doing, and then does nothing. The narration is the point: he never told him anything, did not communicate with him at all, the dog made his own choices and made good ones, and no obedience was necessary. Then the argument is generalised, and it is the argument an over-managing owner needs. You can only get there if you practise it; a dog who is not regularly in that window will not make the right choices, because he has never had to try, because you have always pre-emptively done something to force the choice. And the same clip names the mechanism by which management creates the problem it is managing - the resistance you build by holding a dog back, so that now he wants to go because he is being held back, where a dog simply given the choice may turn out not to care.

Zeus's file shows the mechanics rather than the philosophy, and it shows them twice. In the first in-program lesson the choice point is equipment: with a flexi there is no tension and nothing for the dog to feel, so you do not lock it, you let it go, and that is what gives him the choice - and if he chooses to act stupid you tap the collar, lock the line and give him something else to do. At the go-home lesson the judgement is handed to the owner in the same breath as the word: if the dog looks, and then you say let's go and he walks on and leaves it alone, that is totally lawful, that is acceptable, he did not do anything. What makes his sessions the most rigorous on this page is a remark the handler makes to nobody in particular after a correction - that it counted because he was giving the dog the choice and was not locking the leash and holding it tight. He then names the hidden help most handlers never remove: he needs to be able to tell the dog to down with no leash attached that the dog feels he might use.

Lola's first in-program lesson is where the term is introduced by name - I call them choice points, you have got to get choice points, find them, allow the choice, then either reward or penalise - and it is said to a family rather than to a camera. Her second lesson is where the whole concept gets compressed into a handover. Asked how not to undo the work, the trainer gives two rules and no more: the big one is allowing space and allowing choice, and not pulling her in; and the other big one is punishing when necessary, because you have to penalise the behaviours that made you come to see us in the first place. That pairing matters, because the first rule on its own is the one owners hear. Elsewhere in her file the release contract is extended from the recall to every position, which is what makes choice a method rather than an attitude: the instant she complies she gets the okay and gets to run around.

In the shared Rocko/Kodak owner lesson, this discussion occurs during Rocko's demonstration. The trainer contrasts managing every moment with commands with developing the dog's choices, then explains how he wants these owners to maintain Rocko's work. A separate Rocko recording describes an approaching dog as a choice point. These passages document the trainer's reasoning in these cases; they do not establish the order of the recordings or provide a general instruction to allow unsafe encounters.

Luna's sessions carry the argument for choice and the limit on it, and the limit is the more interesting half. The argument is an analogy, made in ordinary language and immediately demonstrated - you have to give them the chance to make the right decision, like children who are helicopter-parented their whole lives and then go off like a spring the second they move out, because they have never had the chance to make a right decision on their own. It is said seconds after the dog overran a position and corrected herself with no help. The limit is stated in a different session, and it is the answer to the question owners actually have, which is when to do nothing: he would not punish the sniffing, and the reason he would not is that there was no clear command afterwards - if he had said let's go and she had not picked up, that is direct disobedience to a command and is a different story. A third passage explains why he enforces something he has just said he does not care about, which is the rule a family is likeliest to break in their first week: once you have told the dog he has to do it, the principle holds even though you no longer mind.

Finn is where the concept is aimed at the handler rather than the dog, and it is the warmest material on this page. In a home session, one of his owners has just walked him herself for the first time, on a loose leash, past something she was frightened of - and the trainer's response is to clap for her rather than for the dog, and then to say the whole method back to her as praise: you let him make a choice, and what did he do? Nothing. He did nothing. Earlier in his programme the same idea is given as the goal of the case, and the goal is not obedience. Watching a fearful dog behave like an ordinary dog, the trainer says that he can be happy, that he has control over his life, and that this is what you want to see.

Sage is the nervous-dog version, and her file supplies the mechanism the rest of the page leaves implicit. The claim is that tight management is not a mild form of treatment but the absence of it: restrict her options too tightly and she will panic when she is not restricted, and that kind of tight management does not allow for any kind of learning at all. What the freedom is for is named, and it is the most technical sentence in this concept - he has to allow enough freedom for prediction errors to occur, for management to be out of the way, so that she can start feeling differently about what is going on. That is choice as a route to an emotional change rather than to compliance, which is the reason a field trip is not a walk.

Dutchess gives the cleanest possible piece of evidence for the whole idea, and it is evidence of the handler doing nothing. Off leash at a park with a squirrel left in play, she backs off it by herself, and what the trainer says is that he did not have to do anything - she automatically kind of backed off on her own. That is the form of proof this concept needs and the hardest to fake, because the claim is about the dog's judgement and the only way to show it is to be visibly idle.

Deuce's late-programme session shows what the trainers are actually looking for by the end, and it is small. The dog drops what he was chewing, looks up and starts doing the right thing - and the handler's note is that he did not hit anything, the dog did it of his own volition. That is the same success the rest of the page describes, recorded in passing rather than argued for.

Louie's session adds the part that complicates the concept, and it is said to a new owner. Allowing the dog to refuse is deliberate, because it is important to have those little moments where you are able to have some kind of authority over the dog - and then the trainer says the thing the rest of the corpus mostly leaves out: the authority does not transfer. The new owner is going to have to have a couple of those moments himself, when the dog decides he does not want to be picked up or does not want somebody near him. That places the choice point in a relationship rather than in a curriculum, which means the family has to run one too.

instead of saying come first is because I want them to think that running over to that dog is bad. Mhm. Not just that they need to recall.

Watch at 22:11 ↗ · Current transcript match · Audio confirmation pending

only get there with your dog if you practice that if you don't practice that uh them being in that window on a regular basis they won't make the right choices because they've never had to try to make the choices before you've never given them the opportunity you've always preemptively done something to them um to force their choice right

Watch at 01:46 ↗ · Current transcript match · Audio confirmation pending

So I have to allow enough freedom for prediction errors to occur for management to be out of the way for her to start feeling differently about what's going on.

Watch at 19:10 ↗ · Current transcript match · Audio confirmation pending

Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.

with that solution that's always option A for me because then it becomes hands off and automatic because they start learning what's lawful and what's not lawful. They learn to make good choices instead of you having to manage them with obedience.

Watch at 45:57 ↗ · Current transcript match · Audio confirmation pending

Supersedes the existing page

The existing concept page at concept/lawful-choice.html is called 'Lawful choice points' and defines the concept as arranging a situation where the right choice is available and letting the dog choose, with the headline quote 'Without choice points, you're just stuck managing. You're your dog's micromanager slash prison guard.' That is not wrong, but it is half of what the deep read found, and its evidence comes from outside the dogs read here.

Recorded as superseded rather than corrected. The old definition stands as far as it goes; what it lacks is the ordering rule, the control-versus-obedience distinction, the second half of the handover instruction, and the fact that the trainers use 'lawful' as a working term about particular behaviours rather than as a name for the concept. The decision about what the published page says is the owner's.

Unresolved

Allow the choice, against arrange it so the behaviour cannot be completed. Both are taught in this corpus, sometimes about the same kind of moment, and no session says which governs when.

Left open here and cross-referenced rather than argued, because the conflict belongs to the sibling concept where both sides' evidence sits together. What can be said from this page's material is that every handler who allows the choice also insists the leash be genuinely loose at that moment, which is the opposite of holding the situation - so the two instructions are not being blurred by accident.

Unresolved

Whether the credit for a good choice belongs to the dog or to the handler's preparation. The corpus says both, warmly, and never reconciles them.

Not a contradiction in practice so much as an unresolved question about what the page should claim. It matters because the concept's whole evidential form is the handler's inaction, and in at least one of the cited successes the handler says he did in fact intervene, early and quietly.

How do I teach my dog to make better choices?

By arranging a moment where the wrong thing is genuinely available, saying nothing, and then answering what he picks. The trainers call these choice points, and one of them gives the procedure in four sentences to a family: let them make a choice; if they make the wrong choice, the e-collar; if they make the right choice, tell them so; then back to freedom. The part that owners find hard is the silence in the middle. It only works if the leash is actually loose and you actually do not intervene, which is why one handler makes a point of saying, after a correction, that it counted because he was not locking the leash and holding it tight. And it has to be practised rather than waited for - a dog who has always been pre-empted has never had to try, and on this account will not get good at it.

My dog knows all his commands and still does it. Why?

Because from where he is standing he has not done anything wrong, and this is the passage in the corpus worth reading twice. Suppose he follows every command you give, every time - and then goes back to whatever you gave the command about. He complied, he was released, and then he did what he wanted. When you finally lose patience he does not know what the problem is, because he did everything you asked. Obedience was installed, but nothing was learned about the behaviour. The fix is to put the consequence on the choice rather than on the command, and there is an ordering rule: mark the wrong choice first, then recall, so that going over to the other dog becomes the bad thing rather than something a recall interrupted. The claim is that a few of those and the walks get simple, because you stop having to tell him things.

Isn't it enough that I can hold him on the leash?

The trainers say that is control, and that control is not what they are selling. The line is drawn at a park, out loud, with two dogs pulling hard past the camera: those owners do technically have control over their dogs, just not obedience. The distinction is worth having because it tells you what to test. If the only reason your dog is not doing the thing is that you are physically preventing it, you do not yet know what he would choose - and you will find out at the worst possible moment, which is the one where the leash is not in your hand.

What am I allowed to let him do? What counts as a bad choice?

The trainers use the word lawful, and they use it about particular behaviours rather than as a principle. Looking at another dog is lawful - if he looks, and then you say let's go and he walks on and leaves it, that is acceptable, he did not do anything. Circling back to what he was doing after a let's go is lawful; doing the same thing after a come is not, because come means come to you and stay there. And sniffing when you have said nothing is not punishable at all. One handler draws that line in a single breath: he would not punish the sniffing because there was no clear command afterwards, and if he had said let's go and she had not picked up, that is direct disobedience to a command and is a different story. The practical version: correct disobedience to something you actually said, and correct the behaviours that brought you to a trainer. Do not correct a dog for being a dog.

Doesn't letting him get it wrong just teach him he can?

The trainers' answer is that a choice point is not permission - it is a choice with a consequence attached, and the consequence is the half owners forget. Asked directly by a family how to avoid undoing the work, one trainer gives exactly two rules: allowing space and allowing choice, and not pulling her in; and punishing when necessary, because you have to penalise the behaviours that made you come to see us in the first place. Both, or neither. The version with only the first rule is not the method, and the version with only the second is what the trainers call managing - or, in their harsher phrasing, being your dog's micromanager.

My dog is nervous. Surely I should manage him more, not less?

On a nervous dog the trainers say the opposite, and they give a reason rather than a reassurance. Restrict her options too tightly and she will panic when she is not restricted - and that kind of tight management does not allow for any kind of learning at all. What the freedom is for is stated precisely in the same session: he has to allow enough freedom for prediction errors to occur, for management to be out of the way, so that she can start feeling differently about what is going on. The change being aimed at is emotional rather than behavioural, and it cannot happen if the situation never gets close enough for her expectations to be wrong. That does not mean no structure. It means boundaries on her choices rather than removal of them.

What does success actually look like?

Quieter than most people expect. In the clearest example in the corpus a dog walks near a group of people, sniffs the air, checks what they are doing, and does nothing - and the handler's narration is that he never told him anything, did not communicate with him at all, and no obedience was necessary. In another, a dog off leash in a park backs off a squirrel by herself and the trainer says he did not have to do anything. And on a fearful case the goal is stated in words that are not about obedience at all: he can be happy, he has control over his life, and that is what you want to see. If your dog's good behaviour still requires you to be doing something, the work is not finished.

Pre-empting. Shortening the leash, turning away or cueing something else before the dog has had the chance to decide - which means he never gets practice at deciding and, on this account, never will.

Holding the dog back from the thing and then wondering why he wants it more. The corpus names this as management creating the problem it is managing - the resistance is built by the restraint.

Recalling first and correcting afterwards, or recalling instead of correcting. The dog then learns that a recall happened, not that the thing he went for was the problem.

Answering every problem with another command, and then getting angry. From the dog's side he obeyed each time and was released each time, so your frustration has no referent.

Taking the first half of the handover instruction and leaving the second. Allowing space and allowing choice is one of two rules; the other is that the behaviour which brought you to the trainer has to be penalised when it happens.

Correcting a dog for something that was never asked of him. Sniffing with no command given afterwards is not punishable; a let's go that was ignored is a different matter, and the trainers draw the line in one breath.

Managing a nervous dog tightly and calling it care. Restrict the options too much and there is no room for the learning, and the dog panics the first time the management is not there.

Letting a command go the first few times because you do not mind on that occasion. Once it has been said, the principle is the thing being trained - the trainers enforce cues they have just admitted they do not care about.

He does everything you say and then goes straight back to the thing you said it about, and you have started to find that maddening.

You are talking to your dog constantly - the whole walk, or the whole evening - and it is exhausting rather than improving.

You can hold him, and that is what you rely on. If the leash were loose you have no idea what he would do, because it never is.

He seems to want things more when you hold him back from them - the lunging got worse after you started shortening the leash.

Your nervous dog copes as long as you are managing every detail, and falls apart the moment you are not.

You cannot remember the last time he did something right that you had not asked for.