What does progress look like for a fearful dog?

The dog the phrase was invented for, and the only case in the corpus where the finish line is described as an image rather than a behaviour. The deer speech is said to him, not about him: deer freeze and bolt when they see a person, and he is asked to stop looking like that. What makes the case readable is that his handler grades sessions by what did not happen - no okay and no running to hide - and by carriage rather than compliance. His milestone is specifically a voluntary approach to a stranger, graded against his own previous best of walking past neutrally, and against a baseline the same handler states plainly: at their first meeting his eyes went like saucers and he offered a fear bite out of terror rather than aggression. The way the approaches are produced is a kind of anti-technique. He is taken where willing strangers are and nothing is staged: "I'm just here, like, hoping that we'll want to pet him." The stranger is warned it is a severe fear case, the dog is invited, and the handler's own surprise when it works is left in the recording.

The dog whose recovery is measured in both directions, and where the in-house phrase gets defined. The company's word for the outcome is dogging - "they're being a dog and we want to see them be a dog" - and the failure state has the same name: a nervous dog who will do nothing at all is, in the trainer's words, not a dog. Her file also supplies the part of a recovery that catches families out. She ignored squirrels and rabbits for a month and would not go near water; four days before going home she swam after a duck. The trainer names both halves at once - great to see her be a dog, and also stop that - and then explains why: behaviours surface once there is less apprehension. So the new problem is read as evidence of the recovery rather than against it, and is then trained like any other problem. One more thing about her belongs here: opening up around her trainer is treated as the prerequisite for opening up around anybody, which makes the handler's own relationship with her a staged objective. The success signal at the other end is its mirror image - at the handover she opens up more around her family than around the staff who trained her.

The dog with three instruments, none of them obedience. Her nerve progress is reported through the crate - she now walks in as she is supposed to instead of falling apart beforehand - which is a behaviour she performs unwatched and which a family could read just as easily as a trainer. Indoors on a cot, drool is her readout, used the way a tail is used outdoors. Outdoors the tail itself is the gauge, and a drop is acted on immediately rather than noted, because the thing at risk is the outing itself. And the fourth reading is the one that inverts an owner's instinct: when she closes distance on the handler, that is logged as an anxious behaviour to be removed, not as engagement to be rewarded. Taken together her file makes the general point rather than the specific one - each dog has a legible sign and it is worth naming which one, because the cheap per-dog instrument is what lets an owner keep score between sessions.

In the shared owner lesson, the trainer contrasts Kodak's vigilance during walks with his engagement when a ball is available. He explains his belief that putting learning inside a game helps Kodak cope with pressure. A separate down-teaching recording contains the trainer's estimate of how quickly that exercise developed; the estimate is a report about that session, not an independently established starting baseline or a guarantee of later transfer. In another walk, he names coping with barking dogs as a priority over polished obedience, while retaining recall as essential. Read these passages as the trainer's observations, interpretation and program aims. To evaluate progress, compare Kodak's visible responses across settings rather than treating enthusiasm for a ball as proof that every concern has resolved.

Whether the new behaviour that surfaces as the fear lifts should be celebrated or stopped.

The same handler holds both positions in the same conversation and does not treat them as in tension - the celebration is about what the behaviour signifies and the punishment is about what it costs. A page that carries only the first reads as sentimental; a page that carries only the second reads as punishing a dog for recovering. The corpus does not say how a family is meant to hold both at once.

Whether a fearful dog's closeness to the handler is progress or a symptom.

Two dogs, two handlers, and no acknowledgement of each other. The distinction the corpus implies but never states is between contact the dog offers when she is comfortable and contact she uses to manage fear; the concept sticking-close-is-not-attachment carries the other half of this argument. Nothing on film tells a handler how to tell the two apart in the moment.

Whether the milestone is the dog's choice or the handler's production.

Not a disagreement between people so much as an ambiguity in how the result should be reported. It matters for the honesty of the claim: the approach was not trained and was not accidental either, and the corpus records both facts without reconciling them.

How will I know my frightened dog is getting better?

By the picture rather than the commands. The trainers describe the target as a dog who sniffs, wanders, wags and walks up to people - one of them literally says he saw an image of it in his head, a normal dog walking along looking happy with his tail going. The deer version is the same idea inverted and it is said to the dog: a deer freezes and bolts when it sees a person, and the ask is to stop looking like that. Day to day, though, you will be reading something much smaller: how she goes into the crate when nobody is watching, whether the tail is a little low, whether she drools indoors, whether she sniffs in curves. Pick the sign your own dog has and check it, because that is what the trainers do.

Nothing good happened on that walk. Was it wasted?

Probably not, and this is how the trainers keep score on a case where the good thing is still weeks away. Progress in a fear session is graded on what did not happen. The clearest example in the corpus is a session whose whole result was that the dog did not run and hide when he was released, and stood next to a person instead, and let himself be handled. That is written up as something rather than as nothing. If your dog came home from a walk having not bolted, not hidden, and not shut down in a place where she previously would have, that walk did work - and it is worth writing down, because the change in a fear case is slow enough that you will not remember the baseline otherwise.

My dog will never let a stranger touch her. Is tolerating them enough?

Tolerance is a real step and it is where this method's most-filmed fear case sat for weeks - the best his handler had managed was the dog walking past people neutrally. But the milestone they hold out for is different: the dog choosing to walk up. When it finally happened the handler graded it out loud against that previous best, and against the day they met, when the dog was so frightened his eyes went like saucers and he offered a fear bite. The way it is produced is less a technique than a decision about where to spend an afternoon: take the dog somewhere with a reliable supply of friendly strangers, warn them what they are dealing with, invite the dog, and let the people do the work. Declining is allowed. If he does not want the treat, he does not want it.

She has started chasing things she never used to. Is the training coming undone?

The trainers read this the other way round. The account they give is that fear had been suppressing the behaviour, and it appears now because there is less apprehension in the way - a dog who ignored squirrels and rabbits for a month went into the water after a duck four days before she went home. Both halves get said in the same breath: it is a good sign, and it also has to stop. So it is trained like any other problem rather than treated as a failure, and on this dog the suggestion was to penalise the sighting itself. What it is not is evidence that you have a different dog back. It is evidence that the dog underneath the fear has opinions.

She is glued to me. Isn't that the bond working?

Be careful with this one, because the corpus reads it two ways and the difference matters. On one dog, coming in close to the handler is logged out loud as an anxious behaviour to be removed, not as engagement to be rewarded - which is the opposite of how most owners would read the same picture. On another, a dog coming up to snuggle is quoted as the proof that she has opened up, and bonding hard with a nervous dog is called one of the most important things a handler does. The reading the corpus implies, without ever stating it, is about which direction the fear is running: contact offered when she is comfortable is the bond, contact used to manage fear is the symptom. Nobody on film says how to tell them apart in the moment.

He is only ever confident when there is a ball. Does that count?

It counts, and on one dog in this corpus it was the whole route in. He is described as hypervigilant for most of an early session and then, in the same sentence, as the most confident thing in the room the moment a ball appears. The trainer's reason for caring is mechanical rather than sentimental: a dog who does not normally tolerate pressure, and panics under it, can learn from pressure while the game is running - and what he learns in that context can then be moved into the others. The evidence is checkable on the footage. A dog who could not be played with at all in the earliest recordings picked up a verbal-only down inside a quarter of an hour of fetch, and days later held it while the ball flew past him.

Scoring the dog on commands. A fear case can have fast, clean obedience and be no better at all - one dog's commands were called genuinely good while she still could not be trusted off leash because she was so frightened.

Accepting a held position that looks right while the dog is still braced - a body arranged correctly around an unchanged state.

Reading a dog who does nothing as a well-behaved dog. The in-house phrase runs in both directions: a nervous dog who will not do anything is not being a dog.

Treating a new problem that appears late in a recovery as the program coming undone, when it is the fear letting go of something it had been sitting on.

Reading the dog crowding your legs as devotion, and rewarding it.

Forcing the greeting once a stranger has been found. The whole point of the outing is that the dog chooses it, and declining is allowed.

He carries himself like something that expects to be hunted - head down, tight, ready to leave - even on an ordinary walk.

She sticks close to you when she is worried, and you have been reading that as her loving you.

She falls apart before the crate or the kennel rather than in it - the trouble is in the approach.

She drools heavily indoors, in no heat, while being asked for nothing much.

He holds a sit or a down but there is nothing relaxed about it - he is braced rather than settled.

She will do nothing at all - no play, no exploring, no interest in anything except getting out of the situation.

There is one place, or one object, she has never gone near - a lake, water, a particular corner - and you have quietly stopped taking her there.

He is a different animal the moment there is a ball involved, and unreachable the rest of the time.