Fog of War | Kaelan

In a strategy game with fog of war — StarCraft, Age of Empires, most of the genre — you can only see the ground you occupy. The rest of the map is dark. Your opponent is out there building something, and you don’t know what. You have to commit to plans against a position you can only guess at.

The phrase comes, by a long and crooked route, from Carl von Clausewitz, the Prussian officer whose On War was published by his widow in 1832, the year after he died. He never actually wrote “the fog of war” — that compound got attached to him later. What he wrote was sharper: “War is the realm of uncertainty; three quarters of the factors on which action in war is based are wrapped in a fog of greater or lesser uncertainty.” The not-knowing, he argued, isn’t an accident of war. It’s the medium.

Fog is what makes the game a game. Strip it away and you get chess — perfect information, everything on the board, no surprises — which collapses into pure calculation. The dark is the thing you’re really playing against.

It turns scouting into a real cost. You spend a unit, some time, some risk, to lift a corner of the fog and learn one fact: where their army is, what they’re building toward, whether that base is real or a decoy. Every scout is a small purchase of information, paid for in the resources you didn’t spend building. The skill is knowing what’s worth scouting for — which unknowns, once resolved, would actually change your move, and which you can act under without ever clearing. This is the same arithmetic as diagnosis and navigating by estimate: you rarely get certainty, so you buy a bounded amount of it on the questions that matter and commit on the rest.

Hidden information also makes deception possible, and this is the deeper move. Because your opponent can only see what he scouts, you can shape what he scouts — show an army on one flank while the real attack masses on the other, fake a strategy to bait a response, hide your intent in the dark. The fog isn’t only a problem to solve; it’s a medium to manipulate. Half of high-level play is managing what the other side believes about the parts he can’t see.

This is exactly why bluffing exists in poker and not in chess. Game theorists draw the line as perfect versus imperfect information: chess hides nothing, so there’s nothing to misrepresent. Poker hides your cards, and the instant information goes private, belief and reality can come apart — and feints, threats, and lies become not just possible but central. When Facebook’s poker AI Pluribus beat professionals in 2019, it had to learn to bluff, because you cannot play the game well without it.

The general condition is simply life. You almost always act on a partial map — in markets, relationships, careers, war. The competent response isn’t to demand more certainty before moving, because the fog never fully lifts. It’s to scout cheaply on the questions that would change your decision, commit on the rest, and stay alert to the possibility that what you can see was arranged for you to see.

Treat certainty as something you buy in small amounts, never something you wait for. And remember the map you’ve been shown might be a map someone wanted you to have.

Go Deeper

Books

On War by Carl von Clausewitz — The source of “war is the realm of uncertainty.” The Howard and Paret translation is the standard.
The Theory of Poker by David Sklansky — On playing well under hidden information, and why bluffing is mathematics, not bravado.