The Committee of Public Safety

Horst Rittel, in inventing the “wicked problem“, described it as having ten characteristics:

1. There is no definitive formulation of a wicked problem.
2. Wicked problems have no stopping rule.
3. Solutions to wicked problems are not true-or-false, but better or worse.
4. There is no immediate and no ultimate test of a solution to a wicked problem.
5. Every solution to a wicked problem is a “one-shot operation”; because there is no opportunity to learn by trial-and-error, every attempt counts significantly.
6. Wicked problems do not have an enumerable (or an exhaustively describable) set of potential solutions, nor is there a well-described set of permissible operations that may be incorporated into the plan.
7. Every wicked problem is essentially unique.
8. Every wicked problem can be considered to be a symptom of another problem.
9. The existence of a discrepancy representing a wicked problem can be explained in numerous ways. The choice of explanation determines the nature of the problem’s resolution.
10. The planner has no right to be wrong (planners are liable for the consequences of the actions they generate).

The explanation offered by the TRADOC pamphlet Commander’s Appreciation and Campaign Design is in the spirit of Herr Rittel but passes it through the cold, clammy grip of the color sucking vampires of DoD terminology (e.g. the rather colorful and memorable phrase “wicked problem” becomes the rather limp “ill-structured problem”):

1. There is no definitive way to formulate an ill-structured problem.
2. We cannot understand an ill-structured problem without proposing a solution.
3. Every ill-structured problem is essentially unique and novel.
4. Ill-structured problems have no fixed set of potential solutions.
5. Solutions to ill-structured problems are better-or-worse, not right-or-wrong.
6. Ill-structured problems are interactively complex.
7. Every solution to an ill-structured problem is a ‘one-shot operation.’
8. There is no immediate and no ultimate test of a solution to an ill-structured problem.
9. Ill-structured problems have no ‘stopping rule’.
10. Every ill-structured problem is a symptom of another problem.
11. The problem-solver has no right to be wrong.

Jeff Conklin narrowed this down to six characteristics:

1. The problem is not understood until after the formulation of a solution.
2. Wicked problems have no stopping rule.
3. Solutions to wicked problems are not right or wrong.
4. Every wicked problem is essentially novel and unique.
5. Every solution to a wicked problem is a ‘one shot operation’
6. Wicked problems have no given alternative solutions.

CACD provides this helpful breakdown on how dealing with the wicked problem differs from more righteous problems:

Well-Structured “Puzzle”

• Problem Structuring: The problem is self- evident. Structuring is trivial.
• Solution Development: There is only one right solution. It may be difficult to find.
• Execution of Solution: Success requires learning to perfect technique.
• Adaptive Iteration: No adaptive iteration required.

Medium-Structured “Structurally Complex Problem”

• Problem Structuring: Professionals easily agree on its structure.
• Solution Development: There may be more than one “right” answer. Professionals may disagree on the best solution. Desired end state can be agreed.
• Execution of Solution: Success requires learning to perfect technique and adjust solution.
• Adaptive Iteration: Adaptive iteration is required to find the best solution.

Ill-Structured “Wicked Problem”

• Problem Structuring: Professionals will have difficulty agreeing on problem structure and will have to agree on a shared starting hypothesis.
• Solution Development: Professionals will disagree on:
  • How the problem can be solved.
  • The most desirable end state.
  • Whether it can be attained.
• Execution of Solution: Success requires learning to perfect technique, adjust solution, and refine problem framing.
• Adaptive Iteration: Adaptive iteration is required both to refine problem structure and to find the best solution.

Conventional human problem solving breaks down when confronted by the wicked problem. Humans usually throw two problem solving approaches at problems. One uses the Automatic System and the other uses the Reflective System. Thaler and Sunstein comment in Nudge:

The Automatic System is rapid and is or feels instinctive, and it does not involve what we usually associate with the word thinking. When you duck because a ball is thrown at you unexpectedly, or get nervous when your airplane hits turbulence, or smile when you see a cute puppy, you are using your Automatic System. Brain scientists are able to say that the activities of the Automatic System are associated with the oldest parts of the brain, the parts we share with lizards (as well as puppies).

The Reflective System is more deliberate and self-conscious. We use the Reflective System when we are asked, “How much is 411 times 37?” Most people are also likely to use the Reflective System when deciding which route to take for a trip and whether to go to law school or business school. When we are writing this book we are (mostly) using our Reflective Systems, but sometimes ideas pop into our heads when we are in the shower or taking a walk and not thinking at all about the book, and these probably are coming from our Automatic Systems.

The Automatic System is what we in software engineering call legacy software. It’s the results of solutions to obvious problems that were so obvious that evolution hardwired them into our brains over eons. Nassim Nicholas Taleb’s illustration of a “well-structured” problem is what happens when you see a leopard. Your Automatic System has arrived at the one clear solution: run away as fast as possible (a good option if, as this guy postulates, early humans could run as fast as 37 mph). The Automatic System is fast while the Reflective System is somewhat slower. The Reflective System is geared to the CACD’s Medium Structured “Structurally Complex Problem”. It’s good at problems where the desired endstate is clear but there’s more than one way to get there. Wicked problems, however, are too complex for Automatic responses and strain Reflective responses. A good working definition for a wicked problem is any problem that falls into the gap between evolved and Automatic responses that Reflective calculations haven’t filled.

It seems that wicked problems will only succumb to distributed problem solving where many heads are knocked together and thrown at the problem. This creates multiple lines of approach that crush the wicked problem between converging columns of adaption. Some would identify this happenstance with the phenomenon known as crowdsourcing but that’s a misreading. Most of the great paradigm shifts come from a few aggregating minds but as much of 90% of the upfront processing will be done by lesser minds with varying degrees of ability and effort. At the end, a few super aggregators will step in and finish the final formulation. At that the multitudes can look back and see that they had been whittling away at a giant problem that they didn’t even known was there. The solution to wicked problems is culture, a form of Lamarckian natural selection where attributes acquired in life can be passed on to others without transfer through direct biological interface (though that’s a matter of choice).

The question that we face is how best to conduct the concentric cultural attacks on the looming wicked problems we confront on personal, group, national, and world levels. Any such detached question is complicated by the problem that any solution to a wicked problem has political implications, meaning that it will shift power from one party to others. The airy detachment of pure intellectual debate will inevitably be befouled by appeals to the Automatic System since triggering Automatic responses is a cheap, powerful, and time-tested method for achieving power. However, political infighting, whether expressed as logrolling, politicking, or outright war, may only tangentially contribute towards solving the wicked problem in the most tangential way: by creating an even worse wicked problem than the existing wicked problem. Techniques of dialogue have changed little since the dawn of time. If rhetoric and precision guided munitions have more conversational impact than war cries and spears, it’s more of a change in quantity than quality. Violence and sophistry are part of the wicked problem’s definition and its eventual solution. However, their underbrush must be cleared to get at the wicked problem, especially if, as I’d argue, solving the wicked problem is primarily a distributed effort. As a communication problem, noise imposed by sender, recipient, and medium must be minimized as much as possible to enable clarity.

Rittel’s own solution to defeating the wicked problem was IBIS, the Issue-Based Information System. IBIS involves at minimum four elements:

• Questions
• Ideas
• Pros
• Cons

An IBIS map starts with a root question:

Root Question

A question can only be responded to with another question or with an idea. An idea is best seen as first a potential answer to the question and secondly a chance to evolve into further questions:

Questions and Answers

Pros and cons can only respond to ideas. Further questions can also respond to ideas:

Questions, Answers, Pros, and Cons

Following those few principles, Rittel argued, even wicked problems could be mapped. A shared map would be capable of establishing a shared understanding, enabling distributed problem solving to begin. Actually doing the IBIS mapping requires skill; as the old Othello commercial said, it takes a minute to learn and a lifetime to master. There’s several approaches to utilizing IBIS for creating shared understanding. Rittel’s original version used pencil and paper, relics of the 1970s. Nowadays you can use Compendium, a free open source (LGPL) IBIS mapping tool in conjunction with techniques like dialogue mapping or argument mapping. Less structured approaches can be taken with techniques like mindmapping using free tools like Freemind or XMind or concept mapping using tools like Cmap or VUE. More structured approaches also exist but too much representational granularity leads inevitably to uses of words like ontology. IBIS is a nice balance between too little structure and too much.

How do you extend such an approach over a sufficient distribution space to generate solutions for wicked problems? That’s a question for another time.