Surprise And Global Environmental Change
Report Of The 1994 Aspen Global Change Institute Summer Session On Surprise And Global Environmental Change
Stephen H. Schneider- Stanford University
Billie L. Turner, - Clark UniversityAbstract
Surprise and uncertainty are often confused in the literature and in public discourse; various meanings are used within different communities and cultures. Risk is classically defined as the condition in which the event, process, or outcome and the probability that each will occur is known. In reality, of course, complete knowledge of the probabilities and range of potential outcomes or consequences is not usually known and is sometimes unknowable. Uncertainty is defined as the condition in which the event, process, or outcome is known (factually or hypothetically) but the probabilities that it will occur are not known. When probabilities are assigned, they are subjective, and ways to establish reliability for different subjective probability estimates are debatable. Surprise is defined as the condition in which the event, process, or outcome is not known or expected. The research issue is how can we anticipate the unknown, improve the chances of anticipating, and, therefore, improve the chances of reducing societal vulnerability?
Use of a strict definition of surprise logically entails that we cannot anticipate the event, process, or outcome, because the very act of anticipation implies some level of knowledge. Assessments designated as "surprises," however, indicate that the events, processes, and outcomes so registered were, in fact, knowable in one manner or another. This second type of "surprise"- - a broad use of the term- - is one from which the globalchange community may learn much. After Holling, we adopt the following working definition of this second type: surprise is a condition in which perceived reality departs qualitatively from expectations. Using this working definition does not deny the existence of surprise of the first type (narrow definition). There are many possible typologies of surprise and uncertainty. Focusing on surprise of our second type and its subcategories, we will present a typology that may be particularly useful in distinguishing the sources of surprise and the difficulty in identifying and anticipating some types of surprise. This typology recognizes risk, uncertainty and ignorance. Here, risks are possible (usually undesirable) outcomes whose probability and existence are known. Uncertainty characterizes outcomes that are known to be possible but whose probabilities are not known. Ignorance, the main subject of the typology, is the most intractable: we are ignorant when we cannot or do not know a possible outcome. Reducible and hard to reduce ignorance categories will be distinguished. We will also present a long list of potential "gobal change surprises" cutting across physical, biological and social science disciplines, compiled by participants in an Aspen Global Change Institute workshop in which this paper's authors were the co- chairs.
1. Surprise and uncertainty are often confused in the literature and in public discourse; various meanings are used within different communities and cultures.
- risk: The condition in which the event, process, or outcomes and the probability that each will occur is known.
Issue: In reality, complete knowledge of probabilities and range of potential outcomes or consequences is not usually known and is sometimes unknowable.
- uncertainty: The condition in which the event, process, or outcome is known (factually or hypothetically) but the probabilities that it will occur are not known.
Issue: The probabilities assigned, if any, are subjective, and ways to establish reliability for different subjective probability estimates are debatable.
- surprise: The condition in which the event, process, or outcome is not known or expected.
Issue: How can we anticipate the unknown, improve the chances of anticipating, and, therefore, improve the chances of reducing societal vulnerability?
3. Working definition
Use of a strict definition of surprise logically entails that we cannot anticipate the event, process, or outcome, because the very act of anticipation implies some level of knowledge. Assessments designated as "surprises," however, indicate that the events, processes, and outcomes so registered were, in fact, knowable in one manner or another. This second type of "surprise"-- broad use of the term--is that from which the global- change community may learn much.
Following Holling (1986: 294), the AGCI group adopts the following working definition of this second type--
- Surprise is a condition in which perceived reality departs qualitatively from expectations.
Use of this working definition does not deny the existence of surprise of the first type (narrow definition). Unless otherwise designated; however. the remainder of this report deals with surprise of the second type.
LOGIC OF ANTICIPATING SURPRISE
1. Given the second meaning, it is possible to anticipate a subset of surprises.
2. For example, complex systems, chaos, and other such theories provide a conceptual and analytical basis for understanding that surprises will occur, and a variety of methods (e.g., simulations, backcasting) and assessments facilitate seeking and finding surprises.
3. Coupled with experience, this understanding permits the identification of potential arenas wherein surprise may to take place.
4. This identification may (should) inform the public and policy makers of the issues, and thus potentially allow reduced vulnerability and enhanced environmental and societal resilience to surprise.
5. The probabilities that suspected "surprises" will take place within a specified arena are generated on a subjective basis (or by objective methods or models that rest on subjective assumptions), and vary significantly by individual, community, and culture.
WHO IS SURPRISED AND WHY?
1. Surprise is dependent on expectations. and thus we must analyze how expectations are formed by individuals and groups.
2. This view implies that the degree of surprise depends on the extent to which reality departs from expectations and on the salience of the problem (e.g., hazards).
3. Expectations reside not only in the individual, but with groups, communities, or cultures, such as experts, policy makers, managers, and educators, who can share common ranges of expectations that are generated by group dynamics, leaders, and signal processors.
4. In many cases, surprise lies in the policy/managerial mindset and response to an unexpected or improbable (lowly anticipated: e.g., Three Mile Island) event.
5. A variety of factors contribute to this subcategory of surprise (#4), including: differences of opinions among the expert community; fit with broader policy agendas; and vested interest of an agency or group to maintain a particular view.
6. Factors that may contribute to surprise (of our second type) among the science and policy communities are those involved under conditions of systems complexity and connectedness. Integrated systems modeling, for example, informs that (i) one surprise may lead to another because of sub- system coupling and other such issues and (ii) cascading surprises may emerge.
TYPOLOGICAL MAP/(Figure I)
There are many possible typologies of surprise (and uncertainty) (e.g., Brooks 1986; Timmerman 1986). Focusing on surprise of our second type and its subcategories (e.g. #4 above), we have been informed by one that may be particularly useful in distinguishing the sources of surprise and the difficulty in identifying and anticipating some types of surprise. Adapting from Faber, Manstetten, and Proops (1992), Figure 1 shows a topology that recognizes risk, uncertainty, and ignorance. Here, risks are possible (usually undesirable) outcomes whose probability and existence are known. Uncertainty characterizes outcomes that are known to be possible but whose probabilities are not known. Ignorance, the main subject of the typology, is the most intractable: we are ignorant when we cannot or do not know a possible outcome. Following this typology and definition, ignorance may be where the most significant surprises lie. (It should be noted, however, that some do not make such strong distinctions among these three sources of surprise but see each as a variant on the same basic insight that outcomes are indeterminant.)
Ignorance comes in two varieties. Closed ignorance is the unwillingness or inability to consider or recognize that some outcomes are not known but are perhaps possible. Open ignorance is the opposite and much more complicated. The willingness to acknowledge ignorance is a start to the identification of possible outcomes and anticipating surprises, but some forms of ignorance are easier to reduce than others. Ignorance that is relatively easy to reduce comes in two forms, depending on whether an individual or the group is ignorant. Personal or individual ignorance can be reduced by education, after which "surprises" may become "risks" on some typologies. On the other hand, communal ignorance requires creation of new knowledge through research, broadly within existing scientific concepts, ideas, and disciplines (what some call "normal" science--science within an existing paradigm but not necessarily science that causes a revolution to a new paradigm).
The other type of open ignorance is more complex and less tractable. Ultimately all ignorance might be reducible, but much of it is very hard to overcome. Part of this hard- to- reduce ignorance stems from epistemology--the rules that we think govern how the world works and the language and symbols we use to describe what we think and observe. Some people use the term "paradigm" to describe those rules, relationships, symbols, and language. (Some point out that "epistemological ignorance" can be a form of "closed ignorance" because epistemological blinders lead to an unwillingness or unwitting inability to consider alternatives.) The other part of this "hard"- to- reduce ignorance is intrinsic to the phenomenon at hand. Some phenomena may simply be unpredictable, at least from the technologies and analytical perspective now in existence. Notably, systems characterized by chaos are currently thought to be unpredictable in detail--for example, detailed weather forecasts six months in advance are not possible, no matter how accurate the initial state of the weather condition is known because of chaotic dynamics of the atmosphere. And yet, the general character of some chaotic- like systems can be better understood, permitting models of them and, hence, forecasts of their impacts (e.g., E:l Nino or ENSO events). A further example of phenomenological ignorance is a change in the underlying forces of a system, producing markedly different observed outcomes.
This typology is helpful because:
- it makes a distinction among risk, uncertainty, and surprise;
- it also makes clear that phenomenological surprise is only one category of ignorance; and,
- it suggests that many surprises are easily reducible,
- whereas others are blocked by epistemological blinders that create expectations that exclude some categories of outcomes and, hence, surprise.
FITTING THE MAP FROM THE BOTTOM UP
Table 1 and Table 2 present a series of surprises pertinent to global environmental change presented at the AGCI summer session on "Anticipating Global Change Surprises." To each candidate surprise (and in some cases highly uncertain outcomes that were perceived by many as surprises) in Table I are attached the sources attributed to them as understood by our group. Without reviewing each table entry here, it is possible to fit these sources within the typology presented above. A few cases of phenomenological ignorance were presented, particularly those in which the technology of data retrieval outpaced the analysis of data (e.g., misreading remotely sensed imagery, led to exaggerated estimates about the spatial scale of land- cover changes; erroneous assumptions about outlier values of stratospheric ozone delayed detection of the Antarctic ozone hole). Most of the cases, however, suggested sources of surprise in global environmental change may be closely aligned with the following: l
- narrowness of "paradigm" (epistemological ignorance)
- organizational goals and structure of organizational decision making not consistent with the problem (closed ignorance; epistemological ignorance)
- organizational goals in conflict with the outcome (closed ignorance)
- purposeful obfuscation and blocking (closed ignorance)
- rigid common frameworks (epistemological ignorance--frameworks/mindsets that impede effective use of normal science and learning)
SCIENTIFIC VERSUS SOCIETAL SURPRISE
Outcomes are frequently a surprise to some individual, group, institution, community,and culture, or to society as a whole. Many of the surprises noted in the literature on the subject are scientific surprises--surprises to the community of experts of a phenomenon or area. In contradistinction to these are societal surprises--surprises involving events, new discoveries, or assessments that are processed by social institutions and agents in ways that focus social attention on the surprise and place it on society's agenda for debate and possible action.
Figure 2 illustrates this process. At any time, a number of new events or surprises vie for the attention of society as a whole. They enter a process that Kasperson and colleagues (1988) describe as social amplification and attenuation, whereby the processing of the event or discovery by information and response systems either strengthens or weakens the signal value to managers, policy makers, and publics. Thus, some genuine scientific surprises fail to be taken up by the mass media, watchdog groups, or policy makers and fail to make it onto the societal agenda. Other surprises, perhaps less salient to scientists, undergo substantial amplification in signal value due to intense coverage in the mass media, lobbying by critics or environmental groups, connection to social movements, or concern on the part of policy makers or regulators. Thus, it is important to distinguish between scientific and social surprise and to evaluate how events interact with societal processes to amplify or attenuate the perceived significance of the surprise to managers, social institutions, and publics.
IMPROVING THE ANTICIPATION OF SCIENTIFIC SURPRISE
The sources of global- change surprise noted above point to several ways of improving the anticipation of the arenas or domains of surprise.
l. Encourage and integrate the role of synthesis and synthesizers--appreciating "putting the puzzle together" and searching for connections across problem domains, disciplines, and perspectives.
2. Focus a larger fraction of the research effort on "outlier" outcomes (e.g., applying methods to sample the opinions of a broad range of knowledgeable experts as to the likelihood of a wide range of imaginable outcomes).
3. Support work at the edges (and across edges) of conceptual and problem areas.
4. Promote process- as well as product- oriented research and encourage multiple disciplines and communities to communicate and integrate their knowledge about global- change problems.
5. Insure the following attributes of research discourse and funding that have been insufficiently appreciated to date:
- skeptical welcoming of advocacy science/scientists and of the airing and professional evaluation of unconventional views; and
- multiplicity and constructive duplication of research domains among approaches and institutions.
6. Work backwards from posited future states to identify events or processes that might happen along the way: backcasting scenarios or reconstruct past scenarios in alternative ways to examine what might have happened (e.g., Brooks l986).
7. Encourage the "strategic paradigm" as well as the "efficiency paradigm" to build resilience into social and environmental systems.
PREPARING FOR SURPRISE: BEYOND THE SCIENCE
Many potential surprises can be anticipated as noted above. It is clear, however, that many hazard or problem arenas are intrinsically subject to surprise due to system complexity, lack of experience, or poor theoretical understanding. The scientific and managerial community and society as a whole should expect and prepare for the reality that, whatever anticipatory measures are undertaken, some surprises will inevitably occur. Put somewhat differently, the hubris that science and social science can predict the future sufficiently to anticipate the full range of both positive and negative surprises should be constrained. (For example, the recent Kobe earthquake has put to rest the notion that Japanese cities are adequately prepared to withstand major earthquakes.)
It is, of course, the negative and potentially catastrophic surprises that are of particular concern. Managers and social institutions are not helpless to these surprises simply because specific events and outcomes cannot be predicted reliably or even (perhaps) anticipated. What can be done is to increase the resilience and adaptability of receptors (human and ecological) that are at risk, thereby decreasing the sensitivity to the impacts of the unexpected or uncertain perturbations. Actions aimed at increasing the resilience and adaptability of potentially affected systems are noted below. They do not represent recommendations of AGCI but are provided as examples of the broader ranging amplifications of surprise and global change.
l. Diversifying economic productive systems: the tendency towards increased economic specialization carries the risk of vulnerability to controls (e.g., markets or absentee landlords) well beyond the local area which can have both positive and negative impacts on local resilience to environmental perturbations.
2. Avoidance of technological monocultures: reliance on a single technology, such as nuclear power, may be vulnerable to environmental or other perturbations with negative impacts on the economy.
3. Strengthening the broader entitlement structures: providing robust safety nets to respond to unforeseen events is a critical part of resilience.
4. Adaptive management systems: organizational theory suggests that different management systems have different capacities for dealing with surprise; those doing better are characterized by openness, participation of all parties, and flexibility, while those faring less well are characterized by command- and- control systems.
5. Disaster coping systems: improving designs of early- warning, monitoring, and alerting systems, and strengthening the capability of private and public sectors to respond rapidly to potential disasters should be encouraged.
6. Organizational memory and social learning: measures that improve memory and the ability to learn from surprises improve overall resilience to vulnerability to surprise.
Writing over a decade ago, Kates (l985:50) noted that "one of the distinguishing features of the past l5 years is that surprise persists and, paradoxically, grows." Looking to the next l5 years, he concludes: "Finally, there will be surprises--surprises that in turn will generate new concerns and activities. There will also be other concerns and surprises unrelated to technological hazards, international tensions, social change, and resource needs" (p. 57). The professional community recognized global environmental change as new source of surprise and concern more than a decade ago. The international community, including the public and policy makers, now have the same recognition.1 There are, of course, many ways to typologize surprise. Our method focuses on the nature of source of surprise within individuals, communities, or cultures which largely involves different sources of ignorance. In contrast, Casti ( 1994: 263) provides a typology based on source of ignorance within the expert community. These sources are: paradoxical conclusions, discontinuity from smoothness, deterministic randomness, output transcends rules, behavior cannot be decomposed into parts, and self- organized patterns.
Brooks, Harvey. 1986. The Typoloay of Surprises in Technology, Institutions, and Development. In Sustainable Development Of The Biosphere, W. C. Clark and R. E. Munn, eds., pp. 325- 348. Cambridge: Cambridge University Press for the International Institute for Applied Systems Analysis.
Casti, John L. 1994. Complexification: Explaining a Paradoxical World through the Science of Surprise. New York: Harper Collins.
Clark, William C. 1986. Sustainable Development of the Biosphere: These for a Research Program. In Sustainable Development of One Biosphere. W. C. Clark and R. E. Munn, eds. pp. 5- 48. Cambridge: Cambridge University Press for the International Institute for Applied Systems Analysis.
Clark, William C., and R. E. Munn, eds. 1986. Sustainable Development of One Biosphere. Cambridge: Cambridge University Press for the International Institute for Applied Systems Analysis.
Cohen, Jack and Ian Stewart. 1994. The Collpase of Chaos: Discovering Simplicity in a Complex World. New York: Viking.
Faber, M., R. Manstetten, and J. L. R. Proops. 1992. Humankind and the Environment: An Anatomy of Surprise and Ignorance. Environmental Values 1 (3): 217- 241.
Heaton, Thomas H., John F. Hall, David J. Wald, and Marvin W. Halting. 1995. Response of HighRise and Base-lsolated Mw 7.0 Blind Thrust earthquake. Science 267: 206- 211.
Holling, C. S. 1986. The Resilience of Terrestrial Ecosystems: Local Surprise and Global Change. In Sustainable Development of One Biosphere. W. C. Clark and R. E. Munn, eds. pp. 292- 317. Cambridge: Cambridge University Press for the International Institute for Applied Systems Analysis.
Kasperson, Roger E., Ortwin Renn, Paul Slovic, Halina Brown, Jacque Emel, Robert Goble, Jeanne X. Kasperson, and Samuel J. Ratick. The Social Amplification of Risk. A Conceptual Framework. Risk Analysis 8: 177- 187.
Kates, Robert W. 1985. Success, Strain, and Surprise. Issues in Science and Technology 2, No. 1 (Fall): 46- 58.
Svedin, Uno and Britt Aniansson, eds. 1987. Surprising Futures: Notes from an International Workshop on Long-Term Development, Friiberg Manor, Sweden, January 1986. Stockholm: Swedish Council for Planning and Coordination of Research.
Timmerman, Peter. 1986. Mythology and Surprise in the Sustainable Development of the Biosphere. In Sustainable Development of One Biosphere. W. C. Clark and R. E. Munn, eds. pp. 436- 453. Cambridge: Cambridge University Press for the International Institute for Applied Systems Analysis.
Toth, Ferenc L., Eva Hizsnyik, and William C. Clark, eds. 1989. Scenarios of Socioeconomic Development for Studies of Global Environmental Change: A Critical Review. RR 89- 4. Laxenberg, Austria: International Institute for Applied Systems Analysis.
CANDIDATES FOR GLOBAL- CHANGE "SURPRISE"
by Participants at the AGCI Summer Session on Anticipating Global Change Surprises
[entries not intended to be comprehensive or independent]
A. Anthropogenic Forcing Functions
- South remains proportionately behind the North in economic development.
- Tansfer of wealth from South to North accelerates, widening the economic disparities between the two.
- The nation state demises, leading to conflict and collapse of economic growth.
- An underclass of nations is maintained owing to the diminished process of globalization..
- World mortality patterns are transformed by the emergence of a new, highly contagious virus.
- Medical technology increases life expectancy substantially.
- Human population growth rate does not decrease; the demographic transition does not happen globally.
- The smooth population trajectory foreseen in all standard projections of world population becomes woefully inaccurate in the face of sharp departures from them.
- Funding stops for technology development that would facilitate a low carbon future.
- Change takes place in the political consciousness of the value of nature and the will to act accordingly.
- The global market does not dominate (control) natural resource allocation locally, especially for land and water use; rather non- market institutions (e.g., control economies, quasi- market economies, local institutions) remain important.
- India matches China in CO2 emissions.
- Siberia incurs large- scale resource depletion/degradation and deforestation.
- Several catastrophic nuclear plant accidents lead to ban on nuclear power before inexpensive non- carbon backstop technology is available.
- A very inexpensive, noncarbon backstop technology is developed.
- China burns its coal without significant improvement in technology efficiency.
- China shifts to lowcarbon alternative energy source (e.g., finds ample supply of natural gas or develops viable biomass industry).
- Energy use reverts to a parallel track with economic growth because (i) the cost of energy conservation proves too expensive and/or (ii) a switch from an industrial to a service economy proceeds slowly.
- CO.2 emissions from developing countries do not increase.
- Land-cover change stabilizes in South America and South East Asia; deforestation slows dramatically.
- Synergism of habitat fragmentation, chemical assault, introduction of exotic spp., and anthropogenic climate change affect biodiversity in unforseen ways.
- Spatially varying (regional scale) competing forces create unforseen regional climate anomalies (e.g., land- use changes, aerosols, or tropospheric ozone).
- Chemical pollution causes significant genetic change in humans and other species, possibly affecting fertility.
B. Nonanthropogenic Forcing Functions
- A gradual reduction in "conveyor belt" oceanic overturning leading to cooling at high latitudes occurs, despite general (but slower) global warming.
- Heat stored in the ocean at intermediate depths is released to the atmosphere, leading to rapid warming.
- Stratospheric cooling causes increased Polar Stratospheric Clouds and loss of ozone.
- Antarctic volcanoes lubricate ice-stream flow causing glacial surge and rapid sea level rise.
- The Greenland ice sheet surges.
- Changes in volcanism is induced by change in climate.
- High latitude forests are not a CO2 sink.
- Dimethyl sulfide emissions decline with reduced sea ice.
- Dimethyl sulfide emissions change with sea- surface temperature change.
- Positive or negative biogeochemical feedbacks become climate forcing.
C. Environmental Consequence
- Regional climate anomalies lead to economic and political dislocations.
- Regional environmental degradation has global impacts on economic and political systems.
- Differential movement of species ranges in response to global environmental change causes irreversible or very long- term ecological damage (extinctions or cascading effects).
- Warmer climate could be more stable/less variable.
- Enhanced hydrological cycle leads to unanticipated extreme floods or droughts.
- Cloud liquid water increases causing increased cloud albedo and negative feedbacks on warming.
- Increased snow accumulation compensates faster outflow in West Antarctica when the Ross Ice Shelf disintegrates.
- Land- cover stabilizes in South America.
- Hurricane intensity changes with warming.
D. Human Response to the Advent or Prospect of Global Change
- Geoengineering is adopted.
- The climate convention increases funding for low- cost noncarbon backstop technologies.
- The creation of wildlife reserves and migration corridors lowers impact on biodiversity.
- Improved climate- change scenarios and better understanding of climate impacts identifies specific winners and losers and thereby destroys consensus in the international community for emissions reductions.
- CO.2 build- up stalls for five years, derailing the current convention process.
- Society of 2100 chooses to be relatively carbon free and resilient to climate change.
SELECTED CANDIDATES FOR GLOBAL CHANGE "SURPRISE" ARRANGED ACCORDING TO "SURPRISE" ARENAS
Greenhouse gasses are less than 2 X CO2.
--The world emission rates peak and decline in the near future.--
Strong international agreements are implemented.
Rapid decarbonization of energy system takes place because
- low cost biomass alternatives are developed.
- artificial photosynthesis is mastered.
- inherently safe, inexpensive nuclear power is developed.
- large natural gas discoveries are made in China.
- China and Brazil develops a large biomass- energy industry.
Energy/GNP ratio declines sharply because
- Low- energy technology is improved and adopted globally.
- development increases per capita GNP sharply
World economic growth rates decline sharply because
- of the demise of the nation state leading to conflict and collapse.
- of the emergence of a new, quick acting and highly contagious virus reducing populations globally.
Minimal deforestation takes place because
- land- use cover stabilizes in South America and elsewhere in the tropics.
Greenhouse gasses are far more than 2 X CO2.
--More than 50% of incremental CO2 remains in the atmosphere, sinks become saturated, and world emisson rates grow sharply.--
No significant policies are adopted because
- improved understanding of climate impacts identifies winners and losers, thereby destroying consensus in the international community for emission reductions.
- people place low value on environmental impacts.
- of the demise of the nation state leading to conflict and collapse.
Decarbonization of the energy system stops because
- R&D on low- carbon sources halts.
- nuclear accidents cause a shutdown of all nuclear plants.
- China continues its commitment to coal use.
- India increases coal- based energy significantly.
Energy/GNP ratio stops declining because
- the Demographic Transition does not take place in developing world.
- energy prices remain low.
- cost of energy conservation proves too expensive too implement.
- a switch to a service economy in the Western or developed world proceeds slowly.
Increased deforestation takes place because