Wicked problems

Working Definition

A wicked problem is a social or policy problem that has no definitive formulation, no objective stopping rule, and solutions that are not true-or-false but good-or-bad in the eyes of plural groups whose values and interests irreconcilably diverge. Every attempt at a solution is a “one-shot operation” that irreversibly alters the system and spawns new downstream problems, leaving no room for costless trial-and-error experimentation. While colloquial usage flattens the phrase into a synonym for “any very hard problem,” Rittel and Webber’s technical concept contrasts wicked problems sharply with tame problems—such as solving a chess puzzle, analyzing an unknown chemical compound, or landing a module on the Moon—where the objective is agreed upon in advance and any competent specialist can verify whether the problem has been solved. Because wicked problems lack a value-neutral test of resolution, there can be no purely technical expert qualified to settle them on behalf of the public.

Genesis and Global Lineage

The term originated in the mid-1960s seminars of Horst W. J. Rittel, a German mathematician, physicist, and design theorist who taught at the Ulm School of Design (Hochschule für Gestaltung Ulm, 1958–1963) before moving to the University of California, Berkeley. Its earliest appearance in print was a December 1967 guest editorial by the philosopher of systems science C. West Churchman in Management Science, recounting Rittel’s challenge to operations researchers. Rittel and the Berkeley urban theorist Melvin M. Webber published their canonical statement, Dilemmas in a General Theory of Planning, in Policy Sciences in June 1973, enumerating ten defining properties of wicked problems. Their essay was a frontal reckoning with the technocratic optimism of 1960s America—specifically the belief that NASA- and Pentagon-style systems engineering, operations research, and cost-benefit analysis could “solve” urban poverty, crime, highway routing, and curriculum design just as they had solved missile guidance. Rittel and Webber argued that in an increasingly pluralistic society there is no single, undisputed public welfare function, so social planning can never be an applied natural science. In 1992 Richard Buchanan imported the concept into design theory (Wicked Problems in Design Thinking), making it a foundational text of the modern design thinking movement; in public administration, it became so ubiquitous after 2000 that scholars like B. Guy Peters (2017) warned against its conceptual stretching, while Kelly Levin and colleagues (2012) introduced super wicked problems to capture global climate change.

Central European Lineage

The sole documented link to Central Europe in Rittel and Webber’s canonical 1973 paper runs through Vienna and Karl Popper. To formulate their tenth and final property—“The planner has no right to be wrong”—Rittel and Webber explicitly invoke Popper’s falsificationist philosophy of science from Logik der Forschung (The Logic of Scientific Discovery): in science, solutions to problems are bold hypotheses offered for refutation, and the scientific community does not blame or punish a physicist whose conjecture is refuted by experiment, so long as the work was conducted honestly. Rittel and Webber use Popper’s scientific republic as a sharp contrast class: in the world of urban and social planning, no such immunity is tolerated because the aim is not to discover abstract truth, but to alter the concrete world in which people live, and planners are held directly liable for the irreversible consequences their interventions inflict on citizens. Educated at the University of Vienna (where he earned his doctorate in 1928 and published Logik der Forschung with Julius Springer in autumn 1934, dated 1935), Popper provided Rittel and Webber—who cite the 1961 New York printing of his 1959 English translation—with the exact epistemic boundary marker separating laboratory science from public policy. Rittel himself, born in Berlin and active in post-war West Germany (Ulm) before emigrating to California, falls outside our strict regional definition.

Why Now

Between 2018 and 2026, wicked problems ceased to be a retrospective critique of 1960s urban freeways and became the daily reality of crisis governance. First, the COVID-19 pandemic (2020–2022) displayed all ten of Rittel and Webber’s properties at once: every lockdown or school closure was a one-shot operation that could not be tested in a control group first, and the very definition of the problem (minimizing viral mortality, preserving small businesses, preventing learning loss, or protecting civil liberties) depended on who was asking. Second, post-2018 climate and energy transition politics—from French Gilets Jaunes fuel-tax protests to European battles over agricultural emissions and onshore wind-turbine setback rules (such as Poland’s decade-long conflict over the 2016 “10H” distance rule)—demonstrated that decarbonization is a super wicked problem in which time is running out and every technical fix redistributes visible costs across polarized communities. Third, the governance of generative AI (from the 2024 EU AI Act to disputes over copyright and safety) reproduced Rittel’s core dilemma: because stakeholders disagree on what the problem is, no regulatory threshold can count as a definitive solution.

Explanatory Power

The concept allows analysts to distinguish failures of execution from category mistakes: many policy disasters fail not because civil servants were incompetent, but because politicians misclassified a wicked conflict as a tame engineering task. First, it explains why “policy pilots” and randomized trials have narrower limits in macro-social policy than their champions hope: piloting a benefit rule in one district alters expectations across the country, and scaling up a local trial shifts general-equilibrium prices and coalitions in ways the pilot could never test (as seen in Poland’s 2022 Polski Ład tax overhaul, rewritten multiple times within a year). Second, it explains the life cycle of bureaucratic indicators: lacking a natural stopping rule, agencies introduce proxy targets that quickly become ends in themselves. Third, it shows why defining a problem is already choosing the solution: if urban traffic congestion is defined as “insufficient road capacity,” the solution is widening highways; if it is defined as “excessive car commuting,” the solution is congestion pricing; if it is defined as “spatial mismatch between housing and jobs,” the solution is zoning reform. No traffic model can settle which definition is “correct,” because each expresses a competing normative vision of the city.

Axes of Debate

Underappreciated Angle

The least read section of Rittel and Webber’s 1973 paper is its fourth part on social context, which rejects 1950s theories of homogenizing “mass society” and links wickedness to the growing differentiation of values in pluralistic societies. That diagnosis inverts conventional wisdom: public problems do not become wicked primarily because technical systems grow more complex, but because society loses a shared vocabulary of evaluation. From this follows an uncomfortable corollary for both technocrats and AI enthusiasts: neither bigger datasets nor better predictive models will reduce wickedness, because its source does not lie in a shortage of data. What can reduce it is only the hardest political work of re-negotiating what counts as success—which led Rittel to develop “second-generation design methods” treating planning as structured argumentation among stakeholders rather than optimization.