Wicked Problems and Complexity

Some problems don't yield to more analysis. Homelessness, climate policy, a failing school system, a chronically dysfunctional team — push on one part and another bulges out; agree on a fix and you find you never agreed on the problem. These are not tame problems waiting for a cleverer solver.

Treating a wicked problem as a tame one is a common and expensive mistake. You define it too narrowly, declare victory when the visible symptom fades, and discover the problem has simply moved. Recognising wickedness early changes what "progress" even means — from solving to coping, steering, and improving.

Tame versus wicked

A tame problem can be clearly stated, has a definable stopping point, and its solution can be tested as right or wrong. Hard mathematics, a broken engine, a chess position — these are tame, however difficult. They demand skill, not a new way of thinking.

A wicked problem resists clean definition. Horst Rittel and Melvin Webber, both at Berkeley, named the category in a 1973 paper and listed properties that still define it: there is no definitive formulation, no stopping rule that tells you when you are finished, and solutions are not true-or-false but better-or-worse, judged by stakeholders who disagree.

Why they resist solution

Rittel and Webber noted that every wicked problem is essentially unique, so past templates transfer poorly; every solution is a "one-shot operation," because you cannot run a controlled trial on a city and then undo it; and every wicked problem can be seen as a symptom of another problem, so there is no natural floor to the analysis. Crucially, how you explain the problem determines the solution you will reach — and the explanation you pick is itself a value-laden choice.

One clarification worth making early: "wicked" means resistant and tricky, not morally evil. Rittel chose the word to capture how these problems fight back against resolution.

From solving to coping

If you cannot solve a wicked problem, what can you do? Reframe the goal from a final solution to sustained improvement. Involve the stakeholders who hold the conflicting definitions, because their agreement is part of any workable answer. Make small, reversible moves and watch what happens rather than betting everything on one grand plan. And expect to keep steering; there is no moment at which you are simply done.

Consider a city trying to "fix traffic congestion." Tamed naively, the answer looks obvious: widen the roads. Do it, and within a few years traffic is as bad as before — the extra capacity invited more driving, a pattern transport researchers call induced demand. The problem was never "not enough road." It was tangled with land use, housing cost, transit quality, and where people can afford to live.

A wicked-aware approach treats congestion as a symptom, engages residents and businesses whose definitions of the problem differ, tries reversible experiments — a bus lane, a trial of congestion pricing, a pilot cycle route — and measures the effect. Nobody declares congestion "solved." They manage it, adjust, and keep going.

Not every hard problem is wicked, and calling a tame one wicked is its own failure mode — a way to excuse vagueness and dodge commitment. Building a bridge across a river is genuinely difficult, but it is tame: the requirements can be specified, the solution tested, and the project finished. If a problem has an agreed definition and a testable endpoint, treat it as tame and solve it. Reserve "wicked" for problems that actually resist definition, not merely ones that are large or politically awkward.

The concept has a precise origin. Horst Rittel, a design and planning theorist, and Melvin Webber, an urban planner, both at the University of California, Berkeley, introduced "wicked problems" in a 1973 paper titled "Dilemmas in a General Theory of Planning," published in the journal Policy Sciences. They were reacting against the optimism of postwar planning, which assumed social problems could be solved with the same rational methods as engineering ones.

Their argument was that planning problems in an open society are fundamentally different. They set out ten properties of wicked problems — among them no definitive formulation, no stopping rule, solutions that are good-or-bad rather than true-or-false, and the observation that the planner "has no right to be wrong" because real consequences fall on real people. The paper reshaped a generation of thinking in public policy and design, and the term is now standard in those fields. Rittel used "wicked" to mean resistant and tricky, explicitly not evil.

• "Wicked means evil." It means resistant to definition and resolution, nothing moral.
• "A wicked problem just needs more analysis." More analysis can shift the definition, not settle it.
• "There is a right solution we haven't found." Solutions are better-or-worse and contested, not true-or-false.
• "Every big, hard problem is wicked." Many large problems are merely tame and difficult.

You will be given a set of problems — some engineering, some social, some organisational — and asked to classify each as tame or wicked using Rittel and Webber's tests, then justify the harder calls where reasonable people might disagree.

Think Like a Maester: Before choosing a solution, ask whether everyone even agrees on the problem; if they don't, you are coping with a wicked problem, not solving a tame one.

1. Name three properties that distinguish a wicked problem from a tame one.
2. Why is "no stopping rule" such a practical difficulty for planners?
3. What does it mean that a wicked solution is a "one-shot operation"?
4. Why is calling a tame problem "wicked" also a mistake?
5. What does progress look like when a problem cannot be solved outright?

Wicked problems, named by Rittel and Webber in 1973, resist clean definition, have no natural stopping point, and admit no true-or-false solution. They are not tame problems that need more effort. The response is to reframe from solving to coping: engage the people who disagree, make small reversible moves, measure, and keep steering.

• Systems Thinking
• Design Thinking