Rittel and Webber on wicked problems
Some problems cannot be solved the way tame ones can, and mistaking one kind for the other is expensive.
Every wicked problem can be considered a symptom of another problem.
Why this matters
Levels of Design says an isolated artifact cannot solve a systemic problem. That is an assertion until you can say why, and the clearest why in the literature is Rittel and Webber's account of what they called wicked problems. Their work names the reason a brochure cannot fix a problem that lives in the system that produced the brochure.
The problem they were trying to solve
Writing in 1973, Rittel and Webber were reckoning with the failure of the confident, engineering-minded planning of the postwar decades. The methods that worked for tame problems, problems that can be defined, solved and checked, were failing on social problems, and failing in a way that looked like a category error rather than a lack of effort.
Core contributions
They separated tame problems from wicked ones. A tame problem can be stated clearly, has a stopping rule that tells you when it is solved, and admits solutions that are right or wrong. A wicked problem has none of these. It cannot be definitively formulated, because how you frame it already presumes a solution. It has no stopping rule, so work ends when time or money runs out, not when the problem is solved. Its solutions are not true or false but better or worse. And, most useful for design, every wicked problem can be considered a symptom of another problem, so the level at which you intervene decides whether you are treating a cause or a symptom.
The central model
The tame and wicked distinction is the spine, and the symptom-of-another-problem property is the piece that speaks directly to design. If a problem is a symptom, then an intervention aimed at the symptom, a new artifact, will leave the cause untouched, however well the artifact is made.
What is commonly misunderstood
Wicked does not mean evil. It means resistant, tangled, resistant to the clean methods that dispatch tame problems. The word is also thrown loosely at anything merely hard, which drains it. A problem is wicked in the specific sense of lacking a definitive formulation and a stopping rule, not just difficult.
Where the work remains strong, and where it is contested
The framework is still the reference for why systemic and social problems resist tidy solutions, across planning, policy and design. The contested edges are whether the ten properties they listed are all genuinely distinct, and whether the tame and wicked split is a clean binary or a spectrum, which most now treat it as.
How Jason builds on it
Levels of Design uses the symptom insight as the engine of its mismatch diagnostic. A systemic, wicked problem handed an artifact stays unsolved, because the artifact was aimed at a symptom. And the executive test the article offers, asking at what level the problem actually exists before authorizing the work, is a way of asking whether you are facing a tame problem or a wicked one, and matching the response to the answer.
- Horst Rittel and Melvin Webber. Dilemmas in a General Theory of Planning. Policy Sciences, 1973.The tame versus wicked distinction and the ten properties, including the symptom-of-another-problem claim.
- Read the 1973 paper; it is short and remarkably readable.
- Pair it with the Alexander field guide on fit and misfit for the design side of the same idea.
Where this thinking meets the principles on this site.
- Provides evidence forLevels of DesignA wicked problem is a symptom of another problem, which is why an artifact cannot resolve it.
- Parallel toReality WinsA wicked problem has no flattering formulation that quietly makes it tame.