The Enemy of the Good

Authors


Foundations of Ecological Resilience . Gunderson, L. H., C R. Allen, and C. S. Holling , editors . 2010 . Island Press , Washington , D.C. 466 pp. $35 (paperback). ISBN 978-1-59726-511-9 .

Tackling Wicked Problems through the Transdisciplinary Imagination . Brown, V. A., J. A. Harris, and J. Y. Russell , editors. 2010. Earthscan, London, U.K. 332 pp. $57.05 (paperback). ISBN 978-1-84407-925-4.

Far better an approximate answer to the right question … than the exact answer to the wrong question.

               John Tukey (1962)

I wouldn't give a fig for the simplicity on this side of complexity; I would give my right arm for the simplicity on the far side of complexity.

       Attributed to Oliver Wendell Holmes Sr.

My wife, who was a math major in college, has been volunteering in our daughters’ primary school math classes. She has found that perhaps the most difficult skill to teach the kids is estimating the answers to their math problems. One problem, for example, might require the students to roughly estimate how many hours are in a year. She was able to teach them how to do the exact but relatively difficult calculation of 24 h/d × 365 d/year = 8760 h. But she found it surprisingly challenging to convince the kids to multiply 25 × 400 to obtain an estimate of 10,000 h that could serve as an approximate answer to the problem. There is a real feeling among the kids that only the exact answer will do. The perfect is the enemy of the good.

Having spent much of the past two decades working with conservation practitioners to develop performance metrics to monitor the effectiveness of their conservation actions, I am not so surprised by the human tendency to prefer the precise over the approximate, even when the precise is not practical. A reasonably approximate answer to the correct management question is sufficient in many conservation situations. For instance, qualitative evidence that a reintroduced mussel or bird population is breeding at the reintroduction site is a strong confirmation that restoration efforts are beginning to work. Spending orders of magnitude more money to learn precisely how many mussels or birds are reproducing may not be a wise investment of scarce project resources. And yet, in my experience, practitioners and their scientific advisors are often so worried about getting the exact answer that they either design overly complex and ultimately unimplementable monitoring systems or do no monitoring at all.

Sophisticated monitoring systems certainly have their place in conservation. But I like to think about investing in monitoring as analogous to parental investment as represented in behavioral ecology. Some organizations that fund or implement many small, simple, short-term projects (akin to an oyster generating millions of eggs per year) are best served by an r monitoring strategy in which they invest very little in assessing the success of any one project, but assume that at least some of these projects will succeed. Other organizations that take on fewer, but large, complex, or long-term projects (like a gorilla that has only one offspring every few years) require a K monitoring strategy in which they invest large amounts in scientifically rigorous monitoring designs that may even include replicated controls. As a general rule, an organization will be well served to follow a K strategy for projects in which the stakes are particularly high (e.g., managing the last remaining population of an endangered species), that require large investments, that have high levels of uncertainty, that pose reputational or other forms of risk to the organization, and that have a high potential for learning. But there is also a role for r-level investment in monitoring the vast majority of conservation projects that do not meet these criteria. Best-practice standards demand that you match your investment in monitoring to your position along the r/K gradient and then clearly state the confidence levels associated with the analyses.

The 12 articles reprinted in Foundations of Ecological Resilience exemplify the K end of this gradient. The volume is essentially a festschrift celebrating four decades of publications and ideas of C. S. “Buzz” Holling who, with his colleagues, developed and formalized the ideas of ecological resilience and adaptive management. As in many edited volumes, the first section of papers covers the basic theory, the second provides examples from several systems, and the third examines how these ideas might be applied to policy.

The core idea of resilience theory is that ecosystems are dynamic nonlinear systems in which small perturbations that drive ecological responses across certain thresholds can quickly and sometimes surprisingly lead to extensive changes in ecosystem state. As a result, we cannot manage these systems on the basis of traditional steady-state assumptions, but instead need to take an adaptive, experimental approach. These concepts seem self-evident today, but reading the canonical papers made me remember how revolutionary this thinking was given the equilibrial theories that prevailed in the 1970s and 1980s when the papers were written. It also reinforced my already strong faith that adaptive management is the only way to solve the complex conservation problems humans are collectively facing.

At the same time, however, in reading through the examples illustrating sudden losses of resilience and accompanying state changes in coral reefs, kelp forests, spruce forests, and other systems, it became clear that the active adaptive management Holling et al. propose requires massively K-levels of monitoring investment to be useful. As the editors themselves write, the documentation of “the phase shift that occurred rapidly in Jamaican reefs once their resilience was exceeded…was only possible because of intensive monitoring and research over decades.” In effect, the case studies in this book describe potential management applications of basic scientific research, not targeted research to answer specific management questions.

Furthermore, it is hard to imagine most managers or policy makers reading any of these case studies and being able to transfer them to real-world management decisions regarding the ecosystems of interest. For example, Clark et al.'s 1979 paper, included in the book's section on “Lessons for Policy Design,” explicitly advocates “compressions and simplifications” as “essential to encapsulate understanding and help intuition play its central role in the analysis [and] to facilitate communication in the transfer process.” And yet, the remainder of Clark et al.'s paper is full of complex mathematical equations and multidimensional graphs. We have not quite found the simplicity on the far side of all that complexity.

At this point, I turned with some hope to Tackling Wicked Problems, which promises a “transdisciplinary” approach to solving the multifaceted societal issues that bedevil conservationists and other do-gooders. A wicked problem is “a complex issue that defies complete definition, for which there can be no final solution, because any resolution generates further issues, and where solutions are not true or false or good or bad, but the best that can be done at the time” (p. 4). Unfortunately, although the obligatory theory, example, and application sections of this edited volume may be of great interest to philosophers, they do not offer much for conservation practitioners. The core of this book is a series of 15 essays from different disciplinary perspectives that are not about solving wicked problems, but instead are meta-level philosophical discourses about the modalities of inquiry for solving wicked problems. Mostly, the essays consist of definitions and explanations of all kinds of postmodern jargon and concepts: epistemological trade-offs; ontological divides; ontological commitments; underdetermination; designerly ways of knowing; the partiality, plurality, and provisionality of knowing; individual-focused inquiry; organizational inquiry; collective inquiry; transdisciplinary inquiry; open transdisciplinary inquiry; imaginative transdisciplinary inquiry; and on and on. And yet, despite the promise of its title, this volume has almost no practical guidance or even helpful examples as to how one might frame or assess, let alone actually tackle, real-world wicked problems.

Without a doubt, we need to take holistic, transdisciplinary approaches to solving the wicked problems involved in ecosystem management and to be conscious of the biases inherent in the scientific process. But I do not think conservation professionals need more edited volumes stating these basic propositions in abstract language or promoting only K-level monitoring rigor. What we most need, and what neither of these volumes provide, is guidance on how to practice adaptive management in the over 90% of the situations for which we can only afford r-level monitoring investments. Our community needs examples of how a “barefoot clergy” of professional researchers can work directly with practitioners to frame the right management questions and then estimate answers to the questions. And then our community needs advice on how to distill these approximate answers into the simplicity-on-the-far-side-of-complexity that can be used by decision makers.

Ecosystems are complex, and, as such, there is certainly a need for quantitative scientists and mathematical modelers to ply their trade at the K end of the monitoring gradient. But just as there are ecological niches for both oysters and gorillas, so too there is a need for a balance of r and K strategies in the ecology of conservation-project monitoring and adaptive management. And unlike in the natural world, in the conservation-project system, practitioners of different monitoring and adaptive management strategies need not compete, but can actually be symbionts. Imagine a world in which approximate lessons from thousands of replicated interventions can be pooled and systematically analyzed to learn under what conditions different interventions are successful. Or a world in which the results of detailed ecological studies in one system can be easily extrapolated to r-level conservation efforts in related systems. All that is necessary is to extend the theory behind these ideas, develop some examples, and offer suggestions as to how to put the ideas into practice. Now that would be a good edited volume.

Ancillary