log entry ID: a.r.i.... 2025-06-08 19:13:35 EESTSo, I had to read an article called Wicked Problems by Rittel and Webber as part of a course in the basics of design. It is kind of junky and pretentious, but I think I understand the basic idea. Please don’t make me read something like this again, though.
“”“ Still another group has arrived at the conclusion that liberty and equity are luxuries which cannot be afforded by a modern society, and that they should be substituted by "cybernetically feasible” values. “”“ —Rittel and Weber
I like it when my values are cybernetically feasible.
“”“ The professionalised cognitive and occupational styles that were refined in the first half of this century, based in Newtonian mechanistic physics, are not readily adapted to contemporary conceptions of interacting open systems and to contemporary concerns with equity. ”“"—R. & W.
R. & W. know exactly what they are talking about.
Anyway. This is my attempt at unfucking this pulpy LSD-saturated whad of nonsense.
Rittel and Weber incorrectly diagnose the problem. They give up on the idea of an indisputable public good.
“”“ … in a pluralistic society, there is nothing like the undisputable public good … ”“”
I say, nonsense. The indisputable public good is simply what the most people agree upon. After a solution is implemented and the people have experienced it, they can vote to keep it. The more votes a particular solution gets, by definition the better it is. A 99.8% solution is better than a 95% solution is better than a 68% solution.
Indeed, measuring the performance of a solution is the easy part. Finding good solutions is the hard part.
Everybody can approve of a bridge until the day it collapses. That is, voting only works in assessing past performance and not future performance of a solution. This is a huge practical drawback, but it doesn’t mean performance cannot be measured. It only means that performance cannot be completely known until after the solution’s lifetime has completely passed. After the day that bridge collapses, everybody will vote that the bridge is no good, and never was. The votes before then are irrelevant.
There are problems that are too complicated to be solved by one person. Within this class, there are problems that divide cleanly into sub-problems, each of which can be solved by minimally communicating people. These are the tame problems. One might also call them local problems.
Also within this class, there are problems that do not subdivide at all. These are the wicked problems, alternatively non-local problems.
Between the two, there are problems that can be subdivided, but the people working on the problem must be in constant communication. The time spent communicating can be comparable to the time spent in independent work. Indeed, one can define the semi-wicked threshold as the point where time communicating exceeds time in independent work.
There will come a point where so much communication is required that the solution attempt fails. It can fail because problem parameters will have changed by the time a solution is arrived upon, rendering the solution obselete before it is implemented. It can fail because of a failure in communication from the past to the future. The progress on a problem can be forgotten. Such a problem effectively cannot be subdivided. This is the wicked threshold.
In the absence of timely exact reasoning, the people trying to solve a wicked problem are forced to rely on aesthetics, intuition, and history. It is a very human form of problem solving. It is like religion, and that is no accident. It is plain that religion’s primary purpose is to overcome the wicked threshold.
Existing religions today seem to have laws that work fairly well. Opponents of religious reasoning will argue that its success is an illusion. The problem is that over the millenia, there have been thousands of competing religions, each with their own guess at a good set of rules. When all of the dead religions are counted, it is plain that religious reasoning is no better than random chance.
By addressing the actual problems determining the wicked threshold. Every time our ability to communicate is improved, the threshold lowers and new problems can be solved. Particularly good progress has been made on our ability to communicate between generations of people. When the printing press was invented, scientific knowledge could finally accumulate and spread.
This can’t go on forever, though. Fundamental barriers to communication have now been reached. It turns out that a remote meeting is no better than a meeting in person, which can only accomplish so much. It turns out that one can’t do much better than Wikipedia when it comes to summarising and disseminating accurate information about the universe. Biology is a great mass of tightly integrated systems, whose comprehension is beyond the abilities of any group of people to understand, even collectively. A heart connects to its lungs better than a heart specialist could ever connect to a lung specialist.
It’s all the religions do anyway, and it at least served some of its adherents well. By adhering to rules that are known to work, a society may survive a world that it does not understand. The religious aspect helps it adhere to these rules in the face of doubt and ignorance. Five surviving societies out of a hundred is better than zero, which is what could have happened if everybody had abandoned religion.
I suggest that we approach the guess-and-check method more honestly. The transformation of Western cities in the ‘50s and '60s in response to the invention of cars was a guess. It was wrong. Those cities in Europe that accepted the failure honestly changed course. They brought back the trams and pedestrian neighbourhoods that were bulldozed for highways and parking lots. The cities of North America did not. Some do not admit that the great car experiment was just a guess. Others are failing at the “check” step, paralysed by political systems unable to respond to the will of the people they were made to serve.
If a problem is so difficult to solve, it’s worth asking if a solution is necessary. The answer is almost always no. Almost. Most people agree that humanity is worth perpetuating. It is only the barriers to this that truly fundamentally need to be broken. Extinction level events like climate change and meteor impacts. Everything else is optional.
Just keep living, keep guessing, and keep checking.
There is one more way to solve the problem of collective reasoning: remove the “collective” in collective reasoning. This goes back to the beginning. A problem too complicated for one human to solve is solvable by a superhuman intelligence. But only a fool would fail to recognise the problems with giving power to a creature whose reasoning they could not possibly understand. How do they know the AI is helping them, and not merely pursuing its own hidden, incomprehensible goals?
We must not resort to such a method unless humanity is truly at risk of extinction. The only thing worse than a wicked problem is a wicked solution.