Wicked problems and their connection to aleatoric uncertainty

In the most recent lecture, Jörn Behrens explained his conception of aleatoric and epistemic uncertainty and their connection to the social scientific concepts of wicked and tame problems. He related wicked problems to aleatoric uncertainty by stating that wicked problems are ‘inherent in the social system’ and in opposition to tame problems cannot be solved by gaining more knowledge. Aleatoric uncertainty describes the statistical uncertainty inherent in a system due to unknown behavior of system variables. As such the connection to wicked problems seems obvious. Wicked problems are assumed to be unsolvable due to being inherent in the social system. Thus, superficially their wickedness can be assumed to stem from some basic aleatoric uncertainties such as future developments or the undeniable randomness of human behavior. In this short text, I will argue that this connection has significant shortfalls and a definitive attribution of aleatoric or epistemic uncertainty to one of the two problem kinds should be avoided.

But first, the terms wicked and tame should be explained once again in order clarify at least my personal understanding of the two as a basis for my later line of thought. As social scientific concepts, they refer to the division of social and political problems depending on society’s ability to solve them. In that regard wicked problems are defined as not being solvable over time ‘in the sense of definitive and objective answers’ (Rittel and Webber, 1973). Instead, they have to be constantly tackled, as they indeed are inherent in the social system. As the concept has become more frequently used, this definition has been expanded to include ten overall characteristics, such as the inability of policy responses to be right or wrong, with them instead being categorized on a scale from good to bad (Peters, 2017)1. Current research remains inconclusive regarding whether wicked problems have to fulfill all ten preconditions to be categorized as such or if a loose resemblance with the core might be enough. In this context, I will deploy a rather loose definition of wicked problems with the central condition being the lack of a clear solution and a temporal stopping point. In their development of the two concepts in 1973, Rittel and Webber defined wicked problems in oppositions to problems that have a clear and objective path to a solution marked by scientific consensus (Rittel and Webber, 1973).
Exploring a possible explicit connection between wicked problems and aleatoric uncertainties requires the clarification of whether aleatoric uncertainties are even part of wicked problems and if so if they are part of the essence of their wickedness2. Starting at the central condition of wicked problems, their lack of a clear solution and a temporal stopping point (at least in the foreseeable future), it quickly becomes obvious that epistemic as well as aleatoric uncertainties are involved in a number of wicked problems. Furthermore, it seems like especially the longevity of wicked problems has no necessary connection to the involved uncertainties whatsoever. As an example, one could examine the treatment of refugees in the EU in general and Germany specifically. Without a doubt, policy decisions regarding the acceptance of refugees in Germany as well as the treatment of refugees at the European borders do not have a stopping point in the foreseeable future nor do they have a clear solution. Furthermore, they even fulfill broader characteristics, for example that refugee policy decisions cannot be categorized as scientifically right or wrong but rather as morally good or bad. Nonetheless, it is not the involved uncertainties and certainly not purely the aleatoric ones that cause these features. Instead, it is rather the division in the political landscape and irreconcilable divisions in the ethical orientation causing the wickedness of the problem. Similar features can be seen in numerous other example, such as the American middle eastern policy. On the other end, tame problems can posses a lot of aleatoric uncertainty, which does not prevent them from being solved quickly. In the classic example of a tame problem, the human induced loss of ozone (Grundmann, 2016), some aleatoric uncertainty concerning the origins of the ozone loss in the Arctic regions remained into the 90s, especially concerning possible feedbacks and implications for future developments (Solomon, 1990). Nonetheless, the ban on CFCs was already decided by the international community of states in 1987, which prevented possible stronger damages.

With these examples, I have shown that aleatoric is neither a sufficient nor a necessary part of wicked problems. Therefore, I personally would not attribute any of the kinds of uncertainty to either wicked or tame problems. Instead, I would suggest that the political discourse surrounding a topic is the foremost factor in its classification as wicked or tame. I would view aleatoric and epistemic uncertainty as part of any natural scientific process independent of its political and societal implications. Please let me know what you think, I am looking forward to your takes on this issue and hope to have given you a little insight and possibly a better understanding into the concepts and their relationship.

1 For the complete list of preconditions, take a look at Peters’ paper, which gives a good overview of the characteristics of wicked problems.

2 Exploring the philosophic concept of essence and possibly the essence of uncertainty would be another interesting topic for a blog post but sadly goes beyond the scope of this text. Thus, the essence in this context is utilized in the simplified meaning of necessarily and suffiently being part of something.

 

References

Grundmann, R. (2016). Climate change as a wicked social problem. Nature Geoscience, 9(8), 562-563.

Peters, B. G. (2017). What is so wicked about wicked problems? A conceptual analysis and a research program. Policy and Society, 36(3), 385-396.

Rittel, H. W., & Webber, M. M. (1973). Dilemmas in a general theory of planning. Policy sciences, 4(2), 155-169.

Solomon, S. (1990). Progress towards a quantitative understanding of Antarctic ozone depletion. Nature, 347(6291), 347-354.

1 Comment

  1. Thanks for your post! I really appreciated your effort to provide a conceptual clarification of this analogy. I also totally agree with you that such connection is problematic and should be avoided, and the approach you chose especially when trying to investigate examples of wicked problems looks to me really valid. While reading your post I found two connected aspects that seem to me worth further discussion in a wider setting. The first one is what does “inherent to social system”really mean, i.e. the idea that uncertainties are part of the system variables looks to me potentially really slippery, since pretty much everything could wind up being a social system variable. Parallel to that, your discussion on wicked problems as involving moral justice rather than scientific correctness definitely hits the point and involves, in my opinion another important concept, which is that of conflict. This latter aspect can clarify in my opinion the first issue, since at this point the “system variables” that incorporate uncertainty are the very moral and (aesth)etical principles that, at a given time, in a given point in space, may well come to conflict. This dynamics is also normally incorporated by juridical systems via the concept of right balance, I would say the pragmatical way to deal with ordinary, but potentially wicked problems. Besides that, I want to thank you again, for your discussion provides a clear and granular description of the specific problem, which I personally find really stimulating.

Leave a Reply