The scene is always the same. Another dog appears at the end of the block, and your dog detonates. Barking, lunging, spinning at the end of the leash while you apologize to a stranger's back. At home, an hour later, he's asleep on your feet like nothing happened.
The gap between those two dogs confuses everyone. It also holds the answer.
What's actually happening
Leash reactivity is not one thing, but it usually runs on one engine: arousal past the point of thinking. Trainers call it threshold - the moment the brain switches from processing to reacting. Under threshold, your dog can see the other dog, notice you, take food, make choices. Over it, the thinking part goes quiet and the ancient part drives. Barking and lunging are what driving looks like.
Why the leash makes it worse: it removes the option every social animal wants in a tense moment, which is distance. A off-leash dog can curve away, sniff the ground, approach in an arc. A leashed dog has one line and you're holding it. Trapped between an approaching trigger and a short rope, many dogs learn that making noise and rushing forward is what creates space - because from their side of the leash, it works. The other dog leaves. Every time.
That's the loop: trigger appears, explosion happens, trigger leaves, explosion gets reinforced. You're not dealing with a bad dog. You're dealing with a rehearsed one.
What it's not
It's rarely "dominance." Your dog is not auditioning for pack leadership at the end of a six-foot leash.
It's not always aggression, either. A large share of reactive displays are fear or frustration in a loud costume - the dog who desperately wants to greet and the dog who desperately wants distance can look identical at full volume.
And it is not a life sentence. Which brings us to the useful part.
What actually fixes it
Three moves, in order.
First, management. Stop the rehearsals. Every explosion practices the explosion, so the first job is environmental: walk at quieter hours, take the route without the fence-fight house, master the cheerful U-turn. This feels like avoidance. It's actually anesthesia - you can't do surgery on a patient who's running around the room.
Second, threshold work. Find the distance where your dog can see another dog and still think - still take food, still hear his name. That distance is the volume knob, and it's where training lives. Too close and you're just rehearsing the explosion again. At the right distance, you can start changing what the trigger means.
Third, counter-conditioning. Other dog appears, chicken falls from the sky. Other dog leaves, chicken stops. Repetition by repetition, the prediction flips: other dog stops meaning brace-for-impact and starts meaning where's my chicken. You're not rewarding the barking - you're rewiring the association underneath it. The barking fades because its fuel is gone.
The part nobody likes hearing
This takes months, not days, and the curve isn't smooth. There will be a great week, then a dog who explodes at a poodle he used to ignore. Progress in reactivity looks like shorter explosions, faster recovery, longer stretches of ordinary walking - not a straight line to a dog park.
But the arc is real, and it's reliable when the work is done in the right order: stop the rehearsal, find the distance, change the meaning.
The dog asleep on your feet and the dog who detonated at the end of the block are the same dog. The difference between them is information - what he expects the world to do next. Changing that expectation is trainable. It's just quiet, repetitive, distance-by-distance work. The kind that finishes.
(CALM ON THE LEASH lays out the full protocol, week by week.)
Explore the books: find your starting point.