Uncertainty and Anxiety

Published in Scenario 02: Performance
Spring 2012

The Call for Data

When it comes to reading about sustainable landscape design, it’s practically a given that you’ll encounter a call for more research. This thread runs strong through the academic literature on green infrastructure, in the recently released spate of sustainable design guidelines, and in day-to-day professional practice. [1, 2, 3]  Regardless of the medium, the message is surprisingly consistent: we – the researchers, the architects, the landscape architects, the planners, the policymakers – want to do the right thing but until science tells us what the right thing is, we’re doomed to do the wrong thing. We want more information. Why? Because we are uncertain and uncertainty begets anxiety.

There’s nothing wrong with wanting more information but we should not forget that more information is not necessarily the cure for uncertainty.

For decades, designers have sought to fold ecological knowledge into the design and construction of the built environment. Leaving aside the question of success, there has been considerable and consistent tension in this effort. The reasons for this are varied, including many of the challenges typical to interdisciplinary work. The tension between the supply, by scientists, and the demand, by designers, for more information has significant impact on efforts at ecologically informed design.

Grappling With Uncertainty

There’s nothing wrong with wanting more information but we should not forget that more information is not necessarily the cure for uncertainty. In some cases, data can even lead to increased uncertainty for a number of reasons, including the nature of data, the methodology of science, and the biases inherent to research.

In order to understand this situation, it’s important to distinguish between different types of data and to acknowledge that information can be mapped on to a gradient of certitude. There’s a reason why there’s a term, “physics envy”, but no such thing as “ecology envy.” Mathematical precision and fundamental laws are unique to physics but we nonetheless expect precision from all scientific research; in this, we will always be disappointed. For example, we have a precise understanding of electricity but only partial understanding of forest ecosystems. In both cases, however, one of great certainty and one of great uncertainty we have made decisions of tremendous import. We have built electricity infrastructure across the globe and managed forests for resources and recreation, both with varying degrees of success. The data that exist for forest ecology increase every year, but we are nowhere near and may never attain the precision and irrefutably of physics.

Uncertainty is, in fact, the foundation of much scientific research. Scientists understand that “science depends not on the inductive accumulation of proofs but on the methodological principal of doubt [whereby any] scientific tenet… is open to revision… in the light of new ideas or findings.”[4] Ecology, in particular, is notable for its lack of a generalizable theory. There are three reasons for this: one, the field of ecology, and in particular urban and landscape ecology, is young; two, ecological research is synthetic, focusing on systems as opposed to single elements; and third, every ecosystem is unique.

In seeking data for our decisions, reasonable expectations are essential. As the statistician George E.P. Box said, “all models are wrong, some are useful.”[5] Not only can data be useful when incomplete, an expectation of uncertainty is crucial. We must not conflate actionable data with irrefutable data and we should expect that much data will, sooner or later, be called into question. In addition, if ecology can, at present, only provide case specific answers that are locally relevant but cannot provide answers about ecosystems that have not been directly studied, we would do well to accept that there can only be so much research and only so much data available. Given the enormous number of ecosystems that exist at any given moment, the task of amassing knowledge about each of them is elusive. Every bit of ecological research may get us closer to the broad principles we are seeking in making land use decisions, but not in the time frame that interests us. As designers, we must move from anxiety to acceptance of uncertainty.

The urgency with which we seek information is understandable: the plans we draft today have long-term impacts. Our burden of responsibility is heavy. Still, we would be remiss if we did not ask: is information what really stands between us and a good decision? The answer to this question is context dependent, but it is often true that facts ultimately take a back seat to beliefs, values, biases, politics, and power. It’s entirely possible that we would get closer to our goals by calling for better environmental education or environmental policy than by consistently focusing on information as the most effective leverage point. We are justifiably anxious to make the best decision possible but, in focusing on a call for more information, we may be reducing our ability to do so.

Coping with Anxiety

Since humans built the first sedentary settlements, we have designed and built in the absence of complete knowledge; we know much more than we did thousands of years ago, but we still toil in ignorance. The ongoing native vs. non-native plant debate is a good example. When choosing plants we must balance calls for native plants, both from scientists and the public, with the constraints of site conditions—soil, hydrology, micro climate, etc.—that bear little resemblance to those that preceded human disturbance. In some cases, we can find data that support the decision to use native plants. For example, a recent study on the impacts of suburban non-native plants on bird and butterfly diversity reveals that native plants better support local insect populations that in turn support bird species of local regional concern. The authors are careful in their abstract to state that their conclusions apply only in their study area.[6] So how should we balance these conclusions about the performance of native plants with the knowledge that keeping non-native plant species out of the suburbs requires tremendous resources including financial, chemical, energy, and human?

A project rarely has a single goal or measure by which all decisions can be evaluated. The conclusions of the aforementioned study may support the choice of native plants but we must still grapple with the knowledge that the very concept of “native plants” is a human construct we bring to bear on our ecosystem manipulations. In direct conflict with the conclusions of that study, a recent review paper co-authored by eighteen scientists takes a strong position against the dominant practice of favoring native plants over non-natives stating that, “it is time for scientists, land man­agers and policy-makers to ditch this preoc­cupation with the native–alien dichotomy and embrace more dynamic and pragmatic approaches to the conservation and manage­ment of species — approaches better suited to our fast-changing planet.” [7] Both papers are from highly respected, peer-reviewed journals, demonstrating that there is no scientific consensus on the issue. The native vs. non-native debate is not unique; scientific consensus is consistent in that it is always elusive. To navigate through these complexities, we have to turn from quantitative information to qualitative information, acknowledging that it is our human values that determine many of our decisions. Decisions are made based on a wide range of factors including experience, values, power, politics, time, budgets, public opinion, and yes, the so-called facts. No amount of data will clear the path to clear-cut decisions.

Even though our knowledge may be incomplete, or subject to revision, and may not be generalizable, we can and should make decisions based on the best available data.

There is another field of applied science that is equally, if not more so, riddled with uncertainty and ambiguity: medicine. In a recent article in the British Medical Journal, Dr. Des Spence wrote,

“We are scientists… The only certainty of science is uncertainty. Medicine is often little more than an opinion, a faith system: We believe that what we do is right. This is despite history telling us that what we do now is almost certainly wrong.” [8]

Although Dr. Spence may be unique in publicly discussing the topic, the dilemma of uncertainty is incorporated from the start into the medical school curriculum when new students are told that everything they are about to learn may soon be obsolete. [9] What is remarkable about the medical profession is that, in spite of this acknowledged uncertainty, doctors not only take action, they do so with authority. Although there are fundamental differences between treating patients and designing the built environment, both entail long-term care that is largely beyond the scope of work. The two professions differ, however, in one crucial way: follow-up. In many cases, doctors follow up with patients to assess the course of treatment and make any necessary adjustments. Designers, however, are often disconnected from both the construction and maintenance of their work. This exacerbates the impact of uncertainty on decision-making since designers feel they only have one shot at making the right call. This robs designers of both the opportunity to take corrective action and to learn from their projects. The solution to this condition lies not in chasing a certitude that cannot be found but in folding uncertainty into the process. Monitoring and adjustment need to become standard items in design contracts. Just as doctors cannot wait for knowledge that will never be complete, designers must do the same in applying ecological data to the design of our built environment. We must act with both uncertainty and authority.

Although uncertainty forms the baseline of scientific understanding, this doesn’t mean we cannot and should not make use of it. We can bolster the public perception of our decisions because even if all data are subject to revision, human psychology tilts us towards information-based decisions. Clients and the public feel more secure knowing that decisions are made using the best available data. We can also treat information as protean, making decisions based on it even as it shifts. Even though our knowledge may be incomplete, or subject to revision, and may not be generalizable, we can and should make decisions based on the best available data. This is, in effect, what we’re already doing but we do so with tormented minds, yearning and calling for more information.

Will we ever be comfortable with uncertainty? As designers, data can seem the perfect counterpoint to a field that many see as trafficking in the subjective. Although aesthetics are but a small part of what we do, they are the most visible and therefore the most dominant in public perception of design. The prospect of legitimizing our decisions with indisputable data holds tremendous appeal.

Fortunately, we do not need to look only to other professions for a way out of this quandary. Within the broader community of landscape design and management, there already exists a mechanism for folding uncertainty into the decision-making process: adaptive management, an iterative framework that works from the premise that data, and therefore best management practices, are forever evolving. Adaptive management institutionalizes both monitoring and modification, integrating adaptation into a design/construction/maintenance process that has traditionally been discrete and monolithic, thus creating a loop in place of linearity. Adaptive management also works from an understanding that generalized approaches need to be field-tested, with each harvest of data informing the next round of changes.

By taking uncertainty as a given, adaptive management harnesses it as a tool. Instead of working to eliminate uncertainty, adaptive management places it at the center of the process, thereby robbing of its bite. Some designers have embraced this mentality and begun to write contracts that allow them to monitor and then change elements of a design post-construction. Although there are financial and legal aspects to such contracts, as designers we have to take responsibility for our unwillingness to admit to uncertainty. The call for more research should not be muted but matched with an equally powerful call to act with authority and an acknowledgement of uncertainty.

Renee Kaufman is a landscape architect and urban ecologist who seeks to wield a diverse array of tools in the management of urban natural resources. In other words, she wrestles with the design, ecology, policy, and public perceptions of urban space to achieve the best possible outcomes for competing needs. At the moment, she is trying to understand how academic concepts such as ecosystem services and biodiversity can best be applied to decision making in the built environment. Renee is currently working with the Yale Office of Sustainability to develop on ecosystem services plan for the university.

Renee will soon complete her Master’s in Environmental Management program at the Yale School of Forestry and Environmental Studies.  Prior to Yale, she earned a Master’s in Landscape Architecture from Harvard’s Graduate School of Design. She then practiced for 4 years at the NYC office of EDAW/AECOM, working on a wide range of public and institutional work, including the PlaNYC Reforestation Initiative. She also served on Manhattan’s Community Board 2 and was an active advocate for bicycle transportation.

[1] Diane Pataki et al., “Coupling biogeochemical cycles in urban environments: ecosystem services, green solutions, and misconceptions,” Front Ecol Environ 9(1) (2011): 27–36.
[2] ASLA, Sustainable Sites Initiative Performance Guidelines and Performance Benchmarks (Report, 2009) accessable at http://www.sustainablesites.org/report/.
[3] Steven Windhager, Frederick Steiner, Mark T. Simmons and David Heymann, “Toward Ecosystem Services as a Basis for Design,” Landscape Journal 29 (2010): 2-10.
[4] Giddens, Anthony. Modernity and Self-Identity: Self and Society in the Late Modern Age (Stanford: Stanford University Press, 1991)
[5] George E. P. Box, and Norman R. Draper, Empirical Model-Building and Response Surfaces (Hoboken: Wiley Press, 1987): 424
[6] K. Burghardt et al., “Impact of Native Plants on Bird and Butterfly Biodiversity in Suburban Landscapes,” Conservation Biology, Vol.23, No. 1(2009): 219–224.
[7] Mark A. Davis et al. “Don’t Judge Species on Their Origins,” Nature, Vol.474 (2011): 153-154.
[8] D. Spence, “ Bad medicine: cardiology” BMJ (2011): 342