Doctrine lag

Why does a victorious army so often enter the next war with exactly the wrong doctrine?

"Generals always fight the last war" is the kind of line that feels explanatory while explaining nothing. It implies stupidity, and armies are not obviously staffed by stupider people than banks or hospitals. So set the insult aside and ask a harder question: what would an intelligent, well-run organisation do with the experience of a war it had just won — and would the right answer look any different from what actually happens?

Where can doctrine come from? Only two places: from experience, or from prediction. Experience is evidence about a war that occurred; prediction is a claim about a war that has not. One is checkable and the other is not. Which would a serious professional body prefer to bet its officers' lives on?

So doctrine is written from experience, and that is not a failure of imagination — it is the only epistemically respectable source available. But notice what victory does to the evidence. If your methods were used and you won, the methods are credited with the win. The counterfactual — that you won despite them, or because the enemy blundered, or because you simply had more coal — is never observed. Victory is the weakest possible test of a doctrine, because it confirms whatever was in use.

Now follow the human machinery, because this is where the lag really lives. The officers who succeeded were the ones who executed the doctrine well; they are the ones promoted; a decade later they run the staff colleges, write the manuals, and sit on the promotion boards. What is an officer arguing that the doctrine is obsolete really asking of them? Not a technical revision. He is asking the men selected for mastering it to declare the basis of their selection void.

Then it hardens in things that are not people at all. Training exercises are designed to rehearse the doctrine, and because both sides in an exercise obey it, the exercise confirms it. And procurement locks it into metal: a ship or an aircraft takes the better part of a decade to field and then serves for thirty years, so the doctrine that specified it still shapes the force long after the argument that produced it is forgotten.

But here is the point I want to press hardest, because the popular telling never reaches it. Suppose an army did rewrite its doctrine from a prediction. Would that be safer? Between the wars, air power theorists predicted that fleets of bombers would break civilian morale and decide a war almost on their own. That was a bold bet against the last war's experience, and it was substantially wrong — expensively wrong, in aircraft and crews and cities. Conservatism is not merely inertia. It is a refusal to spend a nation's defence on an unverified hypothesis, and the known-imperfect doctrine has at least been tested once.

Which brings us to the case everyone reaches for, and which is more complicated than its reputation. France in 1940 is remembered as an army that ignored the tank and hid behind a wall. That is not what happened. France fielded roughly as many tanks as Germany, and some — the Somua S35, the Char B1 — were better armoured and better gunned than anything the Germans had. The Maginot Line did its job, which was to make the fortified frontier not worth attacking. Nor was the Wehrmacht uniformly modern: most of it moved on foot and by horse, with a minority of armoured and motorised divisions.

The difference lay in how the same equipment was organised and commanded. French doctrine, the bataille conduite or methodical battle, assumed a tempo set by centrally planned artillery, with tanks distributed among infantry formations to support them; fewer radios and a slower command cycle followed from that assumption rather than causing it. Germany had concentrated its armour and pushed decisions downward. So this is not a story about failing to notice a machine, but about an organisation whose validated command model could not run at the speed the new machines permitted.

Two corrections to the picture as usually told.

First, "armies always prepare for the last war" is a tendency, not a law, and the exceptions are informative. Defeat prompts revision far more reliably than victory, because the losing side has both the motive and the institutional permission to rewrite — Germany after 1918 put large numbers of veteran officers to work dissecting what had happened, an exercise the victors had less reason to run. But victors innovate too: the United States Navy built carrier aviation and modern amphibious doctrine between the wars, and Britain assembled an integrated radar-and-control air defence system that had no precedent to copy.

Second, what distinguishes the innovators is less brilliance than plumbing. The historian Stephen Rosen argues that successful peacetime innovation depends on a career path for officers who master the new way of fighting — someone senior protecting a promotion route, so the new specialism can produce its own generals. An idea without that pathway stays an idea. That reframes doctrine lag as an organisational problem rather than an intellectual one, which fits the evidence better, though it is one account among several and the field still argues about weighting.

• Why defeat is a more reliable teacher than victory, and what that implies for an army that has not lost in living memory.
• How exercises can be designed so that they can actually falsify the doctrine they are testing.

Every term the collection defines is gathered in the glossary.