Verdun: what was contested was not ground but the exchange rate

By the beginning of 1916 the Western Front had been immobile for more than a year. The trenches ran unbroken from the North Sea to the Swiss border, so there were no flanks. You could not go around, only through — and every attempt to go through killed more attackers than defenders.

What faced Erich von Falkenhayn, the German Chief of the General Staff, was not a tactical problem but an arithmetic one. If you cannot break through, the only move left is to put winning somewhere else.

Where he put it was the exchange rate. Not taking ground but consuming the other side — and not indiscriminately. Choose a place the enemy cannot give up, and site the machine there. If they cannot abandon it they will keep sending men, and the guns will deal with whatever arrives.

The place chosen was Verdun: a fortress town on the Meuse that had been treated since the nineteenth century as somewhere France does not lose. The reason for choosing it was symbolic value, not military value. It did not have to be taken. It only had to look takeable.

On 21 February 1916, 808 guns put roughly 1,000,000 shells in 10 hours into a box some 30 kilometres wide and 5 deep¹⁴. It was the heaviest bombardment of the war to that date. And 302 days later the front line was back at roughly its starting position.

Read Verdun as a tragedy in which two armies bludgeoned each other out of stubbornness and the structure disappears. There is a design here. And the design ate its designer.

What was contested was not ground. It was the exchange rate. Falkenhayn assumed France would lose heavily and Germany lightly: the law of the Western Front, that the attacker dies more than the defender, reversed by concentrating artillery. The infantry did not go forward to take ground; they went forward as bait to draw the French counter-attack, and the guns dealt with what came out.

The result was 377,231 French against about 337,000 German¹⁴. A ratio of roughly one to nine tenths. They consumed, and were consumed almost equally. The stated purpose of the design was not achieved.

Here is the thirteenth pattern. A design of attrition — a design whose object is the consuming itself. And it comes with a defect none of the other twelve cases carries.

There is no stopping condition. How many is enough? What do you do if the ratio comes out worse than assumed? Those lines are not in the plan for Verdun. So the side that built the machine went into it too.

Verdun was a fortified zone. The Meuse runs north through it and forts stood on the hills of both banks: Douaumont, Vaux, Souville. During 1915, however, much of their heavy artillery had been stripped out and sent to other fronts. The place France would not lose was, by this point, substantially empty.

The German attack was directed only at the right bank. Leaving out the left bank was what allowed the guns to be concentrated. It was also the first thing to go wrong. French artillery left on the left bank fired into the flank of German infantry advancing on the right, and from March the Germans had to reach across for the left bank too — Côte 304 and Le Mort Homme. The frontage of the machine widened, and it widened to suit its owner, not its target.

On 25 February, on day 5, Fort Douaumont fell. There was almost no fighting for it. The garrison was thin, a small party of German soldiers found their way inside, and that was that — about 100 men, by the usual account¹⁴.

That single incident decided everything afterwards. Douaumont became the name of a thing France could not be without. Getting it back took 8 months, and tens of thousands of men were sent at it in the meantime. This was exactly the state Falkenhayn wanted — and the identical state grew on the German side as well.

On 25 February, the day Douaumont fell, Pétain was given the Verdun front. What he did was logistics, not tactics.

The railway into Verdun lay under German guns. What was usable was a single road of 75 kilometres to Bar-le-Duc. Pétain turned that one road into a machine. Units were assigned to the road itself; a broken-down lorry was not repaired but pushed onto the verge; gangs stayed permanently on it spreading crushed stone. From March a vehicle passed every 14 seconds on average, carrying about 90,000 men and 50,000 tonnes a week¹⁴. It became known as the Voie Sacrée.

The second device was the noria, the word for a chain of buckets that lifts water. Divisions were not left standing in the line; they were rotated through quickly and continuously. In the end 70 of the French army's 95 divisions passed through¹⁴.

This design is not a way of winning a contest of consumption. It is a way of being consumed more slowly. And it carries a price. Because losses were spread thinly and widely, no formation was destroyed and the same memory entered the whole French army. What happened in 1917 was laid down here.

This site already has Stalingrad, whose pattern it calls a design of fixation. There is a reason Verdun was not put in the same box.

A design of fixation pins the enemy at one point in order to decide the matter somewhere else. Pinning is the means; the winning happens elsewhere. So when the pinning stops working, there is something to fall back to.

A design of attrition has no elsewhere. The consuming is itself the route to winning, so the moment you step out of the contest, everything already spent becomes waste. You have built, yourself, a structure you cannot step out of.

To see what that does in practice, look at 29 August. The designer was removed and the battle ran on for another 111 days. Hindenburg and Ludendorff moved to wind the Verdun offensive down, and even so it did not stop at once. When a plan has no stopping condition written into it, stopping becomes a fresh decision, and a fresh decision needs grounds. The numbers that could serve as grounds do not exist inside the design.

The designs at Cannae and Austerlitz say when winning is confirmed: when the encirclement closes, when the heights are taken. Verdun has no such line.

The line came back to roughly where it began. France held Verdun and the German offensive did not achieve its purpose. But writing "France won" makes the case unreadable.

Pétain's noria prevented the destruction of formations, and in exchange it distributed the experience of Verdun across almost the whole French army — 70 of 95 divisions. The damage was thin, and it reached everybody.

Then in the spring of 1917, Nivelle, who had retaken Douaumont, scaled the same method up and attacked on the Chemin des Dames. He had promised a breakthrough in forty-eight hours. It failed, and mutinies spread across much of the French army immediately afterwards.

This is the heaviest part of the case. A design of attrition transfers not only to the side that built it but to the side that endured it. Having survived makes "one more push and it opens" an easy reading. The organisation that lives through a contest of consumption goes on to start one of its own.

The German consequence has the same shape. The manpower spent at Verdun and on the Somme in 1916 did not come back, and the German army of the following years became an army that could fight one decisive battle and no more. Attrition sends the same invoice to the winner.

Everything above rests on a design — bleeding France white — and that design raises a problem aimed at this site's own method.

The evidence for it has always been a memorandum Falkenhayn is said to have submitted to the Kaiser at Christmas 1915. That document has never been found. The only source for it is a summary in the memoirs he published in 1920²³.

The Prussian army archive burned in 1945, so the original cannot be checked and will not be. Work by Afflerbach and Foley questions whether the memorandum was ever a plan made in advance, or a rationalisation written afterwards for an offensive that had failed²³. The actual operational orders do not state attrition as the object; what they preserve looks more like an attempt at a breakthrough.

This is inconvenient for us. Reading events as structure carries a permanent risk of mistaking a blueprint drawn afterwards for one drawn before.

The case is still treated here as a design of attrition, because whether the memorandum is genuine and whether Verdun actually operated as a machine of attrition are two different questions. 302 days, about 714,231 people, net movement close to zero. Whoever designed it, the machine certainly ran.

And if the memorandum was written afterwards, the story gets worse rather than better. It would mean a machine of attrition assembled itself unintentionally and acquired its rationale only once it was running. The stopping condition may be missing not because someone forgot to write it, but because there was no design at the start to write it into.

Stop at "Verdun was appalling" and the usable part is left behind. What is here is not madness but a design with the stopping condition missing. And designs with the stopping condition missing are far closer to hand than madness is.

That makes thirteen. Austerlitz — space; Cannae — momentum; Midway — time; Trafalgar — command; Stalingrad — fixation; Agincourt — frontage; Waterloo — junction; Vicksburg — supply; Sekigahara — defection; Normandy — deception; Tet — the verdict; Dunkirk — what to lose. And the thirteenth is attrition.

What separates it from the other twelve is that this is the only pattern that eats its designer first. Designs of space, deception or supply can fail and leave the designer free to think of the next one. A design of attrition puts the designer inside the machine at the moment it starts, because stepping out is a total loss.

So there is only one line worth writing about this pattern. If you are going to start consuming each other, write down first how it stops. Without that line it is not a strategy. It is only a machine.