Combined arms

Why does mixing tanks, infantry and guns beat a force built from the single best arm?

Suppose you were given a fixed budget and told to build the most effective ground force you could. A natural instinct is to find the best single instrument — say the tank — and buy as many as the money allows. Specialisation usually pays: a factory that makes one thing makes it well.

Yet no competent army does this, and the ones that have tried it have been punished quickly. Notice that this cannot simply be because tanks are overrated, since armies keep buying tanks. Something about mixing unlike things must be producing an effect that no quantity of the best single thing reproduces. What could that be?

Let us take the pure-armour force out for a walk and see what happens to it.

Where is a tank strong? In the open, moving fast, protected against small arms and splinters, delivering accurate fire. Where is it weak? In close terrain — woods, towns, broken ground — where it cannot see, cannot manoeuvre, and where a small team with an anti-tank weapon can approach within a few dozen metres unobserved. Ask yourself what an opponent who knows this will do. He will not meet you in the open. He will fight you where your strength is inoperative, and he will do it cheaply.

That is the first thing to notice, and it is more general than tanks: specialisation creates a matching vulnerability. To be very good at one kind of fight is to be legible — an opponent can read what you are optimised for, and choose the other thing. The more purely you specialise, the more precisely he can aim.

So what does infantry do for the tank? It goes into the close ground where the tank cannot see, finds the anti-tank teams, and kills them. And what does the tank do for the infantry? It carries protected, accurate fire forward at a speed infantry cannot manage on its own. Each covers the other's weakness. That is mutual support — and it is already better than either alone. But it is not yet the deepest part of the answer.

Bring in the third arm and watch what happens. Artillery does not need to destroy anything to be useful; it needs only to make people take cover, because a soldier who is flat in a trench is not aiming an anti-tank weapon. That is suppression, and it buys time rather than casualties.

Now put yourself in the defender's position and see what you have been handed. Shells are falling. You can stay down in your trench, in which case you survive the artillery — but you are not shooting at the tanks, and in a minute or two they are on top of your position. Or you can come up to engage the tanks, and be caught in the open by the artillery. Which do you choose?

Sit with that for a moment, because this is the mechanism. Every action that answers one arm exposes you to another. Not "we have more capability", but "there is no correct response". This is what tacticians mean by the combined arms dilemma: the mix does not simply add strengths together, it converts the enemy's remaining choices into a set with no good member. A single-arm force, however large, always leaves its opponent one correct answer — and a competent opponent will find it.

Is the mixing free? No, and this is where the idea is often oversold. Combining arms means synchronisation in time and space: artillery must lift as the assault arrives, not before and not after; infantry and armour must be able to talk to each other. That is difficult, it is the first thing to fail under stress, and a poorly coordinated combined force can perform worse than a simple one, because half-timed suppression warns the defender without pinning him. The historical record is full of armies that owned all the arms and could not combine them — possession is not integration.

Two corrections worth making.

First, combined arms is not a fixed recipe of tanks-plus-infantry-plus-guns. It is a principle about complementary vulnerabilities, so the correct mix changes with the terrain and the threat: add engineers where there are obstacles, air defence where the sky is contested, electronic warfare where the enemy depends on links. Cheap aerial drones are only the newest arm to be folded into the same logic.

Second, "best arm" is not a coherent idea in the first place. An arm's effectiveness is conditional on what is next to it — the same tank is decisive with an infantry screen and a liability without one — so ranking arms independently and buying the winner mistakes a relational property for an intrinsic one.

• How the same dilemma logic scales up from a platoon assault to an operational campaign.
• Suppression as a distinct effect from destruction, and how it is measured.
• What new arms — drones, electronic warfare — change about the mix, and what they do not.

Every term the collection defines is gathered in the glossary.