Posts Tagged climate change

Is Clean Energy a Wicked Problem? – Part 3


WitchIn two previous posts I looked at the established definition of wicked problem and tested whether a rough statement of the clean energy problem met the 10 (adjusted to 11 by me) points of that definition. I found that clean energy met about half the requirements to qualify as wicked. Next I want to look at whether characterizing the problem of clean energy as wicked is productive.

Outside the usual hyperbole of climate journalism, there are a number of serious, credible authors who use the term. The Hartwell Paper (London School of Economics, 2010), referenced in yesterday’s post, features it rather centrally. Its authors sought a means of putting climate policy on track after failure of the Copenhagen climate conference. They made some excellent points and recommendations, noting that climate policy and energy policy are not the same thing. They suggested that reframing the climate issue around matters of human dignity will likely be more effective than framing it around human sin and atonement. They also asserted that the UNFCCC/Kyoto model was doomed to failure from the start because it approached climate change as a tame problem when in fact it is a wicked one. I believe The Hartwell Paper errs considerably in concluding that mischaracterization of a wicked problem as a tame one was the main reason for failure of Kyoto. Doing so implies much too sharp a distinction between tame and wicked and overstates the value of that distinction in determining how to attack a problem. Kyoto’s failure can be understood by simple economics; some parties saw insufficient benefit for the cost.

The Hartwell Paper says that presence of open, complex and/or nonlinear systems make a problem wicked. Hartwell does not address nonlinearity by name, though one of its authors, Gwyn Prins, does in related discussions. Though I agree with most of the conclusions reached by Hartwell and separately by Prins, I think Prins’ work might benefit from a better understanding of systems engineering and design and less reliance on the notion of wickedness. To clarify, my only quibble with Prins is terminology, not intent or conclusion. The terminology wouldn’t matter except that it becomes fuel for trumpery and creates an air of unsolvability.

For example, Prins contrasts the wicked problems of climate and energy with the tame problem of aircraft carrier design (The Wicked Problem of Climate Change on YouTube). He offers that in the case of an aircraft carrier, after a certain amount of study into metallurgy and propulsion systems, you can know that it’s time to quit studying and start building, but the lack of definitive formulation of the climate problem prevents us from identifying a similar point in the problem solving sequence for climate.

But this comparison – fix climate change versus build aircraft carrier – is inaccurate. The goal in the case of an aircraft carrier is not an armored boat with 40 fighter jets on it. The carrier is a system, itself a component within a larger weapons system having the objective of national defense. National defense might further be elaborated something like the capacity to defend the US and allies against various military threats, to operate efficiently with minimum risk to its occupants while being reliable, maintainable and fuel-efficient.

In other words, a better comparison would be national defense versus climate change. These problems probably have similar wickedness. If national defense were a tame problem, we could, with a finite amount of analysis and calculation, derive the horsepower requirements of an aircraft carrier’s nuclear-driven turbines and the BTU requirements for its cooling system, through some complex but finite analytical process, from the requirement for national security. But translating peace-keeping and defense-readiness into horsepower first requires making a bunch of subjective and qualitative decisions using an arbitrarily large number of very human judgments. These judgments have no stopping rule; the design has an infinite number of potential solutions, and is close to a one-shot solution that is prone to unintended consequences (case in point, the French carrier Charles de Gaulle). Once implemented, products like the aircraft carrier have no ultimate test of efficacy. Weapons system design – and almost all engineering design problems – are wicked problems using Rittel’s criteria. So how useful is the characterization of wickedness?

One potential value of calling a problem wicked is to convince management and government that study is needed before quantitative requirements can be set, but I think that point is now firmly established. Many engineers would see this as the usual need for requirements analysis, which has always been a subjective and social process involving operations analysis, identification of stakeholders, ethnography, focus groups, scenario and persona modeling, interviews with subject matter experts, consensus tools, fall-back methods, and possibly a dictator or tie breaker.

Steve Rayner of Oxford is another fan of wicked problems. He’s done great work in bringing rationality and pragmatism to climate policy, but his application of wickedness (e.g., Wicked Problems: Clumsy Solutions) can easily be read (erroneously) as an admission of insolvability. If the category wicked once had value, it now seems a liability – an immobilizing one at that. We have work to do; roll up your sleeves.

Rittel and Webber concluded their paper with no advice on how to deal with wickedness; but they imply early on a need for the social professions to advance beyond the view that “instruments of perfectability can be perfected.” I take that to mean they see limits to the utility of science and flaws in viewing organizations, governments and societies as mechanisms. I agree; the mid 20th century was rife with such flawed thinking. However, governments, managers and product design teams have always had to deal with deciding what to tell the engineers to build. If this is the reason climate and energy writers find their topic to be wicked, the term is useless.

A related problem revealed by press covering climate and energy wickedness is that many journalists confuse the difficulty of reaching consensus with the difficulty of making calculations. An open system in physics is merely a means of modeling a physical process; we model problems as open or closed as a convenience for analysis. Social scientists use open system to discuss adaptive agents, co-evolution and social or political interactions. They’re both good definitions in the their contexts, but confusing them leads to the bad conclusion that physical open systems are unanalyzable by the tools of science. The same applies for the term, nonlinearity. In engineering, it means a second- or higher-order system – standard engineering stuff. In new age literature, it sometimes (at its worst) implies a style of thinking that refutes logic and rationality. We can’t blame equivocation of the terms open system and nonlinearity on the use of the term wicked problem, but we can recognize that choice of language has a dramatic effect on popular uptake of science (see post Toward a New Misunderstanding of Science).

Assigning wickedness to the problems of climate/energy or national defense adds little value toward dealing with them. Nor does calling them super-wicked as do Levin et al in “Playing it Forward: Path Dependency, Progressive Incrementalism, and the ‘Super Wicked’ Problem of Global Climate Change,” which does, thankfully, take pains to avoid a lost-cause position. But wicked and super-wicked do have the power to bewilder and demoralize because of our inability to divorce wicked from its more traditional context. Characterizing the problem as wicked is a self-fulfilling prophecy; it convinces that if some of the questions are unanswerable then no action can be taken. We don’t have to know how the global climate works in order to know how to avoid interfering with it any more than we currently do. We know that China is booming  and will accept no external constraints that hamper its economic growth. But we also know that China’s air pollution kills half a million people a year, that the US is good at inventing things, and that China is good at manufacturing them. We also know how to calculate the extent to which solar and wind can contribute to US and global clean energy. We know that governments can stimulate demand as well as supply. That’s something to work with, despite the lack of consensus or transcendent authority.

Further, we can know that solar-powered cell phone chargers, biodegradable phones, eco-beer, and gloves heated with USB-power are truly wicked, in the old-fashioned sense of the word. They’re wicked because of the point made by Rittel, Webber and Churchman in their original papers on wicked problems. Taming a small part of a wicked problem is morally wrong, as is outright faking it – surely the case with much of the greenwash. But even where there’s no fraud, minor taming with major fanfare is still reprehensible. It creates an illusion of progress and distracts us from the task at hand.

Next I want to look at whether our major clean energy efforts – wind and solar power, biomass, hybrid cars and the like  –  are wicked and morally wrong for these same reasons.


The price of metaphor is eternal vigilance – Arturo Rosenblueth and Norbert Wiener

, , ,

1 Comment

Is Clean Energy a Wicked Problem? – Part 2

William Storage           19 Sep 2012
Visiting Scholar, UC Berkeley Science, Technology & Society Center

Nowhere to Run AnymoreIn the last post I looked at Rittel and Webber’s  definition of wicked problem toward determining whether clean energy met that definition. Answering that involves figuring out what we mean by clean energy.

The clean energy problem is closely linked to the issue of climate change, though they are not equal. The climate change problem is usually taken to mean that, given that anthropogenic warming has occurred and will continue unless greenhouse gas emissions are substantially reduced (note this is a premise I don’t care to argue about here), either geoengineering or dramatic changes to energy production techniques are urgently needed. Clean energy assumes that dramatic changes to energy production techniques are urgently needed to correct man-made climate change along with other constraints and provisions.

The  energy problem also includes the need for a continuous supply of energy for the lifetime of the human race, along with getting that energy to developing nations. I.e., even if coal could be made clean, through carbon sequestration or similar, the energy problem would not be solved by burning coal, since it is in finite supply. We may disagree about size of that supply, but not about its finitude. Security of supply must be included too. If oil were clean and in near-infinite supply, but only sourced by hostile governments, design of an energy production system should accommodate that constraint. Terms like green, sustainable, renewable, and alternative are off the table for this discussion. They are too nebulous, ideological, or overloaded. Clean does not necessarily imply renewable. If coal were infinite and clean, it would suffice, as would fusion if it existed. Further, many energy sources today called renewable, my not be sufficiently clean for indefinite use since their energy production densities are too low to supply a significant portion of global demand without major modifications to the earth. More on that in a later post.

Others have put far more thought into defining long term energy requirements than I, so I’ll draw from some experts in the field. Combining David MacKay’s three motivations (Sustainable Energy – without the hot air with, p. 5) and The Hartwell Paper’s three overarching objectives yields something along these lines:

  • The energy supply cannot be finite (in practical terms).
  • It must be secure.
  • It cannot change the climate.
  • It must ensure energy access for all.

I’m specifically not including adaptation and I’m aware that we can quibble over whether universal energy access is a principle, a constraint or a goal. Still, I think this is decent working set. The beginning of an attempt to convert these goals into a requirement might look something like this:

A means of providing sufficient energy for the human race to flourish for 10,000 years without significantly altering the surface and atmosphere of the planet in the acquisition of energy (population growth may require extensive modification of the planet, but that’s out of scope here).

You might then attempt to quantify “flourish” and “significantly alter” by coming up with an energy quantity per person, a percentage of earth’s surface devoted to energy production, and an allowable carbon production per unit of energy.

I’m not saying getting agreement on the numbers will be easy or even possible; I’m merely outlining the process toward the goal of deciding how wicked the energy problem is.

With this in mind let’s have a look at Rittel’s properties of wicked problems against the energy problem as summarized above to see which of them apply (Yes or No, below). Refer to yesterday’s post for more detail on each of the 10 properties.

1. No definitive formulation – solving the problem is identical to understanding its nature: No
Understanding the nature of clean energy and even anthropogenic climate change is mostly independent from solving it. The social components of climate change, energy demand and energy production are not mysterious or unpredictable. Economists and scientists have had great success in that area. The vagaries of climate prediction and extent to which climate change is manmade are rather independent of the solutions that might be put in place based on any such predictions and analyses. This one clearly does not apply; clean energy is not wicked based on this criterion of wickedness.

2. No stopping rule: No
Since atmospheric carbon, temperature, population, sea level, disease, starvation, and energy production and consumption are reasonably measurable, there clearly is a stopping rule in place for clean energy.

3. No formal decision rules – better/worse, not true/false: Yes
One might argue that if a set of metrics could be agreed-upon, clean energy actual does become true/false, but I don’t think that is fair to Rittel’s intent for this rule.

4a. No ultimate test of solution: No
For the same reasons stated in rule 1, clean energy solutions are reasonably testable.

4b. Unintended consequences: Yes
Leaving geoengineering out of the picture, we’d still need to watch for surprises, especially from low density production schemes that would involve large transformations, e.g., massive solar or wind farms, tide and ocean wave modification, geothermal plants, and carbon sequestration schemes.

5. One-shot operation – no second chance: No
Some concern over the ramifications of expending all a government leader’s political capital on short-term measures with trivial contribution toward a solution is warranted; but overall, energy initiatives are very tolerant of experimentation and learning by trial. This is especially on a global scale, even with disasters like Chernobyl and red herrings like fuel cells in the 1990s.

6. No enumerable or exhaustively describable set of potential solutions: No
Nature, physics and economics combine to yield a finite set of policy and technology components to a solution. Yes, there are infinite permutations of the components, but this is always true. In any case, the potential solutions and their elements are enumerable.

7. Unique problem: Yes
Aren’t they all?

8. The problem is a symptom of another problem: Yes
Human breeding habits, materialism, inequitable distribution of wealth, sexy car ads, inefficiency, indifference toward nature, bad science education, the Roman Empire and the Han Dynasty are all problems of which the need for clean energy is symptomatic.

9. Numerous explanations: Yes
Yes, for the same reasons listed in number 9 above. The numerous explanations are in fact relevant, because they could materially affect the solution. For example, realizing that waste and inefficiency is significant can lead to product requirements that result in a lower figure for per-capita energy requirements. Japan has had remarkable success at this.

10. Planner has no right to be wrong: Yes
In the case of clean energy, answering Yes for item 10 seems to be in conflict with answering No for 4a. and 5. Repeated readings of Rittel and Webber have not allowed me to see a real difference between this and number 5 above. The difference between them may be more apparent in problems whose scope is urban planning, the original context of Rittel and Webber. Nevertheless, for sake of charity in argument, I’ll answer Yes here to represent the voice that, in the long haul, we have to get this right or civilization may fail.

So for Rittel’s ten properties, here presented as eleven, we have five No and six Yes responses. On that basis, clean energy can be said to be a half wicked problem. Systems engineers, product managers and designers might say that all engineering and design problems are partly – perhaps equally – wicked. This and other considerations make me wonder whether characterizing a problem as wicked has any practical use.

That will be the topic of my next post. I vow to make it more controversial.




Photo: “Nowhere to Run Anymore” by Thomas Hawk on Flickr

, , ,

Leave a comment

Is Clean Energy a Wicked Problem?

Deciding whether clean energy is a wicked problem involves two tasks. One is to define wicked problem and the other is a formulation of the clean energy objective.

Advocates of Design Thinking and Systems Thinking, among others, are fond of the term, wicked problem. Popular examples include climate change/clean energy, drug trafficking, homeland security, nuclear energy, natural hazards and healthcare. In the next few posts, I’ll argue that the characterization of clean energy as a wicked problem is, at best, not very useful and, at worst, detrimental to the stated goals of those who use it. I think the clean energy challenge is partly wicked – but only partly – and not for most of the reasons one might guess. In upcoming posts I’ll also argue that to some degree the clean energy problem is made wicked by characterizing it as wicked.  There is a Keyser Söze effect (seemingly omnipotent criminal whose omnipotence derives from his scaremongering) at work here. It demoralizes us and misdirects thinking that could be put to better use solving problems. My previous post, on philosopher Richard Rorty, ends wth Rorty’s appeal that if a solution to the problem of climate and energy exists, it is a matter for the engineers. Indeed. Let’s get to work.

The term wicked problem was first used around 1967 in lectures by Horst Rittel of UC Berkeley according to systems guru West Churchman, who first used it in print, in reference to Rittel’s lecture. The context of Rittel’s use of the term was social policy and urban planning. Six years later, Rittel and Melvin Webber defined wicked problems in detail in “Dilemmas in a General Theory of Planning,” published in the journal of the Society for Policy Sciences.

Rittel and Webber list ten distinguishing properties of the planning-type problems they classify as wicked. They note that wicked does not mean that anything in the problem space is ethically deplorable or that malicious intent exists, but that such problems are tricky, malignant, vicious and aggressive.

Both Rittel & Webber and Churchman do, however, go to some length to describe an ethical issue related to wicked problems. This important point is lost in most modern use of the term. The authors indicate that it is usually morally objectionable for a planner to treat a wicked problem as though it were a tame one, or to tame only part of a wicked problem. Churchman says that taming part of a wicked problem, but not the whole, is morally wrong, because doing so can create the illusion of safety where danger exists. He then calls for a new level of maturity and morality in operations research and management science. Churchman urges that his profession not only avoid telling management what it wants to hear, but that operations researchers should not tame parts of wicked problems even if they warn management that only part of a problem was solved. It takes more than a verbal caveat, said Churchman, to convince the management that a solution is incomplete. For the energy/climate problem, it seems to me this aspect of Rittel, Webber, and Churchman’s work may be considerably more important than examining the wickedness of the energy/climate problem. More on that in a later post.

Rittel’s ten distinguishing properties of wicked problems are listed below. These descriptions are excerpted directly from Rittel’s wording with very minor additions and clarifications. I’ve split Rittel’s item number 4 into two parts because I think he inadvertently connects two related but distinct characteristics – solution testability and likelihood of unexpected consequences. I differentiate these because non-function and malfunction (and the likelihood of each) are fundamentally different engineering concerns.

1. There is no definitive formulation of a wicked problem. In order to describe a wicked-problem in sufficient detail, one has to develop an exhaustive inventory of all conceivable solutions ahead of time. The process of solving the problem is identical with the process of understanding its nature.

2. Wicked problems have no stopping rule. You never know whether you’re finished.

3. Solutions to wicked problems are not true-or-false, but better-or-worse. Parties may be equally interested or entitled to judge the solutions, but none has the power to set formal decision rules to determine correctness.

4a. There is no immediate and no ultimate test of a solution to a wicked problem.

4b. Wicked problems are prone to unintended consequences.

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. Every implemented solution is consequential, leaving “traces” that cannot be undone.

6. Wicked problems do not have an enumerable (or an exhaustively describable) set of potential solutions.

7. Every wicked problem is essentially unique. Despite long lists of similarities between a current problem and a previous one, there always might be an additional distinguishing property that is of overriding importance. The conditions in a city constructing a subway may look similar to the conditions in San Francisco, say; but planners would be ill-advised to transfer the San Francisco solutions directly. Differences in commuter habits or residential patterns may far outweigh similarities in subway layout, downtown layout and the rest.

8. Every wicked problem can be considered to be a symptom of another problem. The process of resolving the problem starts with the search for causal explanation of the discrepancy. Removal of that cause poses another problem of which the original problem is a “symptom.”

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. Crime in the streets can be explained by not enough police, by too many criminals, by inadequate laws, too many police, cultural deprivation, deficient opportunity, too many guns, etc.

10. The planner has no right to be wrong. As Karl Popper argues in The Logic of Scientific Discovery, it is a principle of science that solutions to problems are only hypotheses offered for refutation. In the world of planning and wicked problems no such immunity is tolerated.

The definition of wicked problem has remained consistent through its usage. It appears in Design Thinking and climate-change circles often, with substantially the same meaning, usually referencing Rittel and Webber. Given that consistency of usage, we can next take a crack at what we mean when we say we want clean energy. With a useful definition of wicked and a fair formulation of a clean energy objective, we can then look at whether clean energy is a wicked problem and how that characterization might impact planning and design of solutions.

More on that tomorrow.

, ,