Frictional and Structural Unknowns

The Category Error

When we face uncertainty, the natural impulse is to reduce it - to gather more information, analyze more carefully, wait for things to become clearer. This impulse is often correct.

But it is sometimes a category error. Some uncertainty yields to more effort. Some does not. Treating structural uncertainty as if it were frictional uncertainty leads to wasted effort and, worse, to a deferred action that could have been taken without waiting.

The distinction between frictional and structural unknowns is one of the most useful distinctions in decision-making.

Frictional Unknowns

Frictional unknowns are uncertainty that resolves with more information or more time. You do not know what price a vendor will offer because you have not asked. You do not know how a customer will respond because you have not run the test. You do not know how long the project will take because you have not broken it down into tasks.

These unknowns are unknown because of friction - the gap between the information that exists and the information you have. Reducing that friction (asking, testing, analyzing) produces the information.

The appropriate response to frictional unknowns is the one that resolves them most efficiently: gather the information. The decision should wait until the information is available if it can, or proceed on a range of estimates if it cannot.

Structural Unknowns

Structural unknowns are uncertainty that does not resolve with more information or more time. You do not know whether this product will find market fit because no amount of analysis can substitute for actually attempting it. You do not know whether this relationship will work because the answer depends on choices neither party has yet made. You do not know how this market will evolve because the future does not exist to be gathered.

These unknowns are unknown because of the structure of the situation - not because of any friction that better analysis could remove. The uncertainty is real and irreducible.

The appropriate response to structural unknowns is not to gather more information but to structure your action to be robust to the range of possible outcomes, or to take the smallest action that will produce real information, or to accept the uncertainty and commit anyway.

Gathering more information in response to structural unknowns is the category error. It delays action, consumes resources, and does not reduce the uncertainty. It just moves the decision to a future moment at which the structural unknown is still present.

Misclassification

The challenge is that frictional and structural unknowns are often hard to distinguish from the inside. Both feel like "I don't know enough to decide."

The diagnostic question is: what would change if I had more information? If the answer is "I would know the answer to this specific question" - if the uncertainty is about a fact that exists and could in principle be gathered - the unknown is probably frictional.

If the answer is "I still wouldn't know, because the thing I'm uncertain about doesn't yet exist / hasn't happened / depends on choices that haven't been made" - the unknown is probably structural.

A related diagnostic: how long have you been gathering information about this question? If substantial information-gathering has not reduced the uncertainty, that is evidence the unknown is structural. More of the same approach is unlikely to produce different results.

Acting Under Structural Uncertainty

Structural unknowns do not resolve before the decision - they resolve through the decision. Action in conditions of structural uncertainty is not reckless. It is often the only way to produce the information the situation requires.

The key is not to avoid action but to take actions that are robust: actions that produce information regardless of how the uncertainty resolves, and that preserve future options if the uncertainty resolves unfavorably.