Defensive sabotage looks like friendly fire
I.
Suppose a major strategic bridge collapses in your country during a war. You are the general in charge of logistics, and you demand to know why the bridge fell. There are three very different causal models for what just happened:
- Enemy Action: The invading army sent bombers to blow up the bridge.
- Friendly Fire: Our own artillery was shelling an enemy battalion near the river, missed, and accidentally took out the bridge.
- Defensive Sabotage: The enemy was advancing rapidly toward the capital. Our local commander realized they were about to cross the river, so he intentionally rigged the bridge with dynamite and blew it up to stop them.
Why does it matter which model is true? Because your job is to figure out how to prevent bridges from falling in the future. And your choice of intervention depends entirely on the causal model.
If the bridge fell due to Enemy Action, the correct intervention is bridge hardening. You build the next bridge out of reinforced concrete, you put anti-aircraft guns on the towers, and you make it indestructible.
If the bridge fell due to Friendly Fire, bridge hardening still sort of works, but the better intervention is improving targeting. You train your artillery crews better, or maybe you tell them to stop shooting so close to your own infrastructure.
But if the bridge fell due to Defensive Sabotage, bridge hardening is absolutely catastrophic.
Imagine you spend billions of dollars researching an unbreakable, ultra-hardened titanium bridge. You install it over the river. The enemy advances, the local commander panics and pushes the detonator—and nothing happens. The dynamite goes off, but the titanium bridge stands strong. The enemy rolls their tanks across your beautiful, indestructible bridge and burns your capital to the ground.
So you’re pinned. The same crossing that carries your supplies carries their tanks. Every bridge you can use, they can use.
II.
Consider the lizard.
If you grab a lizard by the tail, the tail will snap off in your hand. The lizard will scurry away into the underbrush, leaving you holding a twitching piece of meat.
“Wow, biological materials are so flimsy,” you might think. “Pure wear and tear. The mechanical stress of my hand was too much for the poor lizard’s tensile strength.”
Or you reach for the cleverer diagnosis: friendly fire. “The lizard’s muscles spasmed so hard in a panic that it accidentally tore its own spine apart. Evolution is so sloppy.”
But if you actually look at the anatomy of a lizard’s tail, you will find something incredible. Lizards have specialized fracture planes built into their vertebrae. They have complex sphincter muscles inside the tail designed to immediately clamp down and prevent blood loss the second the tail detaches. The tail breaking isn’t an accident. It is autotomy. It is Defensive Sabotage.
The easy breaking is the point.
Now imagine a well-meaning bio-hacker decides to “cure” the lizard’s fragile tail. He gene-edits the lizard so that its tail vertebrae are reinforced with carbon nanotubes. The tail is now indestructible. The lizard goes out into the wild, a hawk swoops down and grabs it by the tail, the lizard tries to drop the tail… and fails. The hawk carries the lizard away and eats it.
The intervention was counterproductive because the bio-hacker mistook defensive sabotage for friendly fire.