Where signals become an assessment.

Doctrine is written in peacetime. The bill arrives later, and it is never sent to the people who wrote it.

In the autumn of 1452 a Hungarian engineer named Urban walked into Constantinople and offered to sell the emperor a cannon.

He was specific about what it would do. It would be the largest gun ever cast, it would throw a stone ball weighing half a tonne, and it would bring down any wall then standing in the world. He named his price. It was, by the standards of a state, not large.

Constantine XI could not pay it. This is the part that is usually skipped. He was not a foolish man and he did not disbelieve the engineer. He was the emperor of a city with a treasury that had been empty for a generation, whose revenues had been signed away to Genoese and Venetian merchants, and whose walls had held every assault for eleven hundred years. He weighed a speculative weapon against a proven fortification and an empty purse, and he decided the walls would hold. Every one of those facts was true when he decided.

Urban crossed the water and made the same offer to Mehmed II, who paid him four times what he had asked and gave him a foundry.

The gun took fourteen months to cast. It was so heavy that sixty oxen were needed to move it and it could be fired perhaps seven times a day. On 6 April 1453 it began to fire on the Theodosian Walls. On 29 May the city fell, and the Roman Empire, which had existed in one form or another for fifteen centuries, ended on a Tuesday morning.

Constantinople was not lost to a better army. It was lost to a procurement decision — a line item declined by a man who had run out of money and out of time, in favour of an asset that had never yet failed him. The walls were not wrong. They were obsolete, and obsolete is a word that only exists in hindsight.

Every defence budget on earth is a version of that afternoon in 1452. Something is being declined right now because it is speculative and the existing thing has always worked. Somebody else is being offered the same technology at four times the price and is about to say yes.

Force is the oldest instrument of policy and the least honestly priced. Every capability has a cost. Every cost has a source. Every source has a weakness an adversary can find. This section follows that chain — from the line in a budget, through the factory and the supply route, to the border where the decision is finally tested.