The Problem of the Future: Sustainability Science and Scenario Analysis

Abstract

Unsustainable tendencies in the co-evolution of human and natural systems have stimulated a search for new approaches to understanding complex problems of environment and development. Recently, attention has been drawn to the emergence of a new “sustainability science,” and core questions and research strategies have been proposed. A key challenge of sustainability is to examine the range of plausible future pathways of combined social and environmental systems under conditions of uncertainty, surprise, human choice, and complexity. This requires charting new scientific territory and expanding the current global change research agenda. Scenario analysis—including new participatory and problem-oriented approaches—provides a powerful tool for integrating knowledge, scanning the future in an organized way, and internalizing human choice into sustainability science.

Full text

Global Environmental Change 14 (2004) 137–146
The problem of the future: sustainability science and scenario analysis
R.J. Swart
a,
*, P. Raskin
b
, J. Robinson
c
a
National Institute of Public Health and the Environment (RIVM), P.O. Box 1, 3720 BA Bilthoven, Netherlands
b
Boston Center of the Stockholm Environment Institute, Tellus Institute, 11 Arlington Street, Boston, MA 02116-3411, USA
c
Sustainable Development Research Initiative (SDRI) of the University of British Columbia, 1924 West Mall, UBC Vancouver, BC, Canada V6T 1Z2
Abstract
Unsustainable tendencies in the co-evolution of human and natural systems have stimulated a search for new approaches to
understanding complex problems of environment and development. Recently, attention has been drawn to the emergence of a new
‘‘sustainability science’’, and core questions and research strategies have been proposed. A key challenge of sustainability is to
examine the range of plausible future pathways of combined social and environmental systems under conditions of uncertainty,
surprise, human choice and complexity. This requires charting new scientific territory and expanding the current global change
research agenda. Scenario analysis—including new participatory and problem-oriented approaches—provides a powerful tool for
integrating knowledge, scanning the future in an organized way and internalizing human choice into sustainability science.
r2003 Elsevier Ltd. All rights reserved.
1. Core questions
There is wide consensus in both science and policy
that world development continues to move in an
unsustainable direction. A recent comprehensive review
of the state of the environment (UNEP, 2002) finds that,
over the last 30 years, human and environmental
circumstances have changed considerably and inequi-
tably across the world. Social and environmental
conditions have deteriorated in many places, and the
integrity of life support systems has come under
increasing threat.
Since the seminal report of the Brundtland Commis-
sion (WCED, 1987), the policy response has centered on
a call for sustainable forms of development. While
definitions of sustainable development vary widely
(Robinson, 2002), most call attention to the need to
maintain resilience in environmental and social systems
by meeting a complex array of interacting environ-
mental, social and economic conditions. One version of
such an approach suggests three ‘‘imperatives’’ for
sustainable development: the ecological (staying within
biophysical carrying capacity), the social (providing
systems of governance that propagate the values that
people want to live by), and the economic (providing an
adequate material standard of living for all) (Robinson
and Tinker, 1998).
These concerns stimulated a scientific response, as
new research initiatives addressed various biophysical
aspects of global environmental change. In the 1980s,
international global environmental change research was
coordinated within the frameworks of the World
Climate Research Programme (WCRP), the Interna-
tional Geosphere-Biosphere Programme (IGBP) and
DIVERSITAS. As the importance of social and
economic aspects increasingly became recognized, the
International Human Dimensions Programme (IHDP)
for the social sciences and humanities was formed in the
1990s. Global environmental change research matured
into a broader agenda under the rubric of global change
research. The interdependence of natural and social
systems increasingly led to calls for interdisciplinary
research efforts (e.g. Lubchenko, 1998;NAS, 1999;
IGBP/IHDP/WCRP/Diversitas, 2001). In the context of
IGBP, the global analysis, integration and modeling
(GAIM) task force is advancing a framework for
integrated research (Schellnhuber and Sahagian, 2002).
For the 2002 World Summit on Sustainable Develop-
ment, the world’s scientific research programmes com-
mitted themselves to integrate societal problems into
their endeavors (ICSU (International Council for
Science), 2002).
ARTICLE IN PRESS
*Corresponding author. Tel.: +31-30-274-3026; fax: +31-30-274-
4464.
E-mail addresses: rob.swart@rivm.nl (R.J. Swart),
praskin@tellus.org (P. Raskin), johnr@sdri.ubc.ca (J. Robinson).
0959-3780/$ - see front matter r2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.gloenvcha.2003.10.002

While the link between science-driven research activities
and sustainability policy development remains weak
(Cohen et al., 1998), the call to strengthen the contribution
of science to a sustainability transition
1
has grown louder
(Raven, 2002). Recently, Kates et al. (2001) identified a
framework for an emerging ‘‘sustainability science’’ for
generating useful knowledge to support a transition to
sustainable development. Such a sustainability science
would seek to illuminate the interactions between nature
and society at different geographic scales from global to
local. It would address the behavior of complex self-
organizing systems and responses of the combined nature-
society system to multiple and interacting stresses,
involving different social actors. It would develop tools
for monitoring key environmental and social conditions,
and guidance on effective management systems.
The seven core questions proposed for sustainability
science by Kates et al. (2001) are collected in Box 1. The
emphasis is on understanding the systems complexities
associated with sustainability, including the provision of
information to help social actors to develop transition
strategies. These core questions are a particular way of
looking at the problem of sustainability, which reflects
the scientific perspective it is intended to represent.
Sustainability science is seen as fundamentally a
problem of representing the interactions, behaviors
and emergent properties of combined natural and social
systems, and providing decision makers with better
advice about the effects of various forms of behavior or
intervention. These are indeed critical issues. Our goal
here is to suggest ways in which scenario analysis can
contribute to addressing them, and in so doing to help
broaden the focus to encompass a richer set of
considerations. These are derived in part from a ‘‘human
science’’ perspective that emphasizes the need to develop
approaches for evaluating future options, recognizing
diverse epistemologies and problem definitions, and
encompassing the deeply normative nature of the
sustainability problem
2
.
Where the questions in Box 1 touch upon the future,
the proposed core questions focus on trends and
discontinuities in those trends, and how these trends
may be changed ‘‘in ways relevant to sustainability’’.
However, they do not underscore the critical role of
envisioning alternative futures, exploring plausible path-
ways, and identifying the factors conditioning long-term
outcomes. To highlight some of these issues we have
added an 8th question in Box 1: ‘‘How can the future be
scanned in a creative, rigorous and policy-relevant
manner that reflects the normative character of sustain-
ability and incorporates different perspectives?’’
The core questions outlined in Box 1 raise a number
of research challenges for sustainability, which we
discuss in the following section. We go on to summarize
parallel developments in the use of scenario analysis to
illuminate sustainability problems. We then argue that
ARTICLE IN PRESS
Box 1
Core questions for sustainability science
1. How can the dynamic interactions between nature and society—including lags and inertia—be better incorporated in emerging models and
conceptualizations that integrate the Earth system, human development, and sustainability?
2. How are long-term trends in environment and development, including consumption and population, reshaping nature-society interactions in
ways relevant to sustainability?
3. What determines the vulnerability or resilience of the nature-society system in particular kinds of places and for particular types of ecosystems
and human livelihoods?
4. Can scientifically meaningful ‘‘limits’’ or ‘‘boundaries’’ be defined that would provide effective warning of conditions beyond which the nature-
society systems incur a significantly increased risk of serious degradation?
5. What systems of incentive structures—including markets, rules, norms and scientific information—can most effectively improve social capacity
to guide interactions between nature and society toward more sustainable trajectories?
6. How can today’s operational systems for monitoring and reporting on environmental and social conditions be integrated or extended to provide
more useful guidance for efforts to navigate a transition toward sustainability?
7. How can today’s relatively independent activities of research planning, observation, assessment, and decision support be better integrated into
systems for adaptive management and societal learning?
Source: Kates et al. (2001).
8. How can the future be scanned in a creative, rigorous and policy-relevant manner that reflects the normative character of sustainability and
incorporates different perspectives?’’
Source: this paper.
1
Detailed analysis of the integrated requirements for a sustainability
transition was introduced by Raskin et al. (1998), and further
developed by the National Research Council of the National Academy
of Sciences of the United States (NAS, 1999), which spells out
international targets for meeting human needs and preserving life
support systems, as a basis for priorities both for research and action.
Subsequently, on the policy side, the General Assembly of the United
Nations adopted the Millennium Declaration (UNGA (United
Nations General Assembly), 2000), which includes targets for a
number of social and environmental sustainability problems.
2
The scientific perspective described by the proposed core questions
corresponds to the ‘‘descriptive’’ approach to the social sciences and
humanities (Rayner and Malone, 1998), in which norms and values are
not addressed explicitly. Our approach in this paper attempts to
incorporate ‘‘normative’’ (explicitly accounting for the value-ladenness
of the issues at stake and associated human goals) and ‘‘interpretive’’
(characterized by a focus on the meaning created by human agents in
the conduct of social life) approaches.
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146138

scenario analysis is a natural and powerful tool for
advancing important aspects of sustainability science,
and close with some observations on future directions.
2. Research strategies
Kates et al. (2001) conclude that the structure,
methods, and content of the scientific enterprise would
have to change in order to pursue sustainability science
adequately. From the core questions they derive four
research strategies:
(i) spanning multiple spatial scales from local to global
processes;
(ii) accounting for temporal inertia and urgency of
problems that are both long-lived and perilous;
(iii) reflecting functional complexity and multiple stresses
in human and environmental systems; and
(iv) recognizing the wide range of outlooks in order to
generate knowledge usable for people with differ-
ent perspectives.
These are important strategic challenges, indeed,
which science has only begun to address. But others
should be highlighted or, where implicit in the four
generic strategies above, made explicit, in order to draw
attention to the critical problem of the future (Core
question 8). These are:
(v) integrating across policy themes and issues to
capture the intricately linked ecological, social,
economic, ethical and institutional dimensions of
sustainability problems—adequately addressing
any specific problems requires a framework that
reflects the breadth and depth of interconnections
(e.g., poverty, deforestation, climate change, land
tenure systems, income distribution, trade regimes,
etc., are co-determining);
(vi) reflecting uncertainty: incorporating large uncer-
tainties of very different kinds, surprise, critical
thresholds and abrupt change that are immanent in
non-linear natural and social systems—the poten-
tial for and implications of novel events and rapid
structural change beyond critical thresholds are
fundamental to assesse the resilience, vulnerability
and suite of possible future states; understanding
and reflecting large uncertainty in complex socio-
ecological systems should be an explicit part of the
sustainability science strategies;
(vii) accounting for human volition as a key conditioning
factor of combined human and environmental
systems—the constitution, reproduction and re-
formulation of human needs, wants and values is
key to illuminating consumption, social goals,
institutional innovation, social learning and the
prospects for alternative futures;
(viii) combining qualitative and quantitative analysis to
elevate non-quantifiable cultural, institutional and
value aspects of the integrated system is required
to avoid limiting the analysis to quantifiable
aspects which are not necessarily the most crucial;
3
and
(ix) making sustainability science more relevant to
policy development and action through stakeholder
participation: the incorporation of the relevant
values, perceptions and preferences of societal
actors about the future into the research process is
needed to encompass forms of knowledge nor-
mally not considered by analysts, and normative
values.
3. Scenario analysis: history and current frontiers
Scanning these various research strategies for a new
sustainability science, a recurrent theme is the challenge
of integration. The systemic character of sustainability
problems demands a holistic perspective that unifies
across sectors, problems, methods, disciplines, spatial
scales and time. Furthermore, the strict distinction
between the realm of the normative and the objective,
the ‘‘ought’’ and the ‘‘is’’, is not useful when the system
under scrutiny entrains human values and choices as
irreducible and critically important system constituents
and drivers of change.
What are scenarios? In the context of sustainability
science, integrated scenarios may be thought of as
coherent and plausible stories, told in words and
numbers, about the possible co-evolutionary pathways
of combined human and environmental systems. They
generally include a definition of problem boundaries, a
characterization of current conditions and processes
driving change, an identification of critical uncertainties
and assumptions on how they are resolved, and images
of the future. The characterization of the nature of
human and environmental response under contrasting
future conditions is key in scenario formulation.
Reflecting respect for the uncertainty inherent in such
systems, scenarios are neither predictions nor forecasts.
Scenario analysis is an evolving concept. The term has
been applied to diverse efforts ranging from literary
descriptions to model-based projections, from visionary
thinking to minor adjustments to ‘‘business-as-usual’’
projections. Although scenario development as a sys-
tematic way of thinking about the future has a long
ARTICLE IN PRESS
3
Global change science continues to be dominated by quantitative
approaches, whereas non-quantifiable factors, such as trust, power,
emotional attachment and many others, which profoundly affect
human behavior and the prospects for sustainability are left implicit or
ignored.
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146 139

history it has not been codified into a common set of
definitions and procedures. Such methodological ambi-
guity is in many ways a source of strength for this
evolving field of inquiry. The range of aims and the
sheer complexity of the problem demand flexibility and
creative exploration.
The broad use of the term ‘‘scenario’’ for characteriz-
ing the systematic framing of uncertain possibilities can
be traced to post-World War II strategic studies. In
particular, Kahn and Wiener (1967) explored possible
consequences of nuclear proliferation, defining scenarios
as ‘‘hypothetical sequences of events constructed with
the purpose of focusing attention on causal processes
and decision points’’. In the private sector, Shell has
played a leading role since the 1970s developing
scenarios to highlight world development possibilities
that are relevant to the company’s future, and to prepare
company managers for responding to an uncertain
future (Wack, 1985a, b), a process that has been used
more widely in the business community since (Schwartz,
1991;Shell, 2002).
Systems modeling is another antecedent to contem-
porary scenario analysis. Mathematical simulations
were used to forecast the behavior of the economy, its
pressures on the environment, and resource constraints.
A controversial early effort was Limits to Growth
(Meadows et al., 1972) for the Club of Rome, which
was followed with more complex modeling exercises
(Mesarovic and Pestel, 1974). These studies found that
trends in economy, demography and technology would
overshoot the planet’s carrying capacity in the 21st
century, and explored the adjustments to growth
necessary to avoid such a crisis. While calling attention
to critical problems, the rigidity and uncertainty of the
model specifications, and underestimation of society’s
adaptive capacity and of human ingenuity in the face of
emerging problems has been used to discredit the work.
In order to emphasize social and political aspects of
development, particularly in developing countries,
Herrera et al. (1976) developed the Latin American
World Model to explore the requirements for a more
egalitarian future. H.
afele (1981) focused on long-term
energy options in a finite world.
Another stream of scenario work focused on envi-
sioning desirable futures, particularly in the energy field,
in order to stimulate discussions on how to get there.
Such ‘‘backcasting’’ studies (Robinson, 1982), mostly at
the regional or national level, were conducted in dozens
of countries, inspired by the early work of Lovins (1976,
1977) in developing scenarios of ‘‘soft energy paths’’.
More recently, this backcasting approach has been
applied in the context of sustainable futures, at both the
regional and global scales (e.g. Robinson et al., 1996;
Raskin et al., 1998, 2002).
After the Brundtland Report (WCED, 1987) and the
1992 Rio World Conference on Environment and
Development, a second ‘‘wave’’ of global scenarios
was launched in the context of the sustainability
challenge. Some were model-based, and focusing on
one issue such as climate change (Rotmans, 1990,
Rotmans et al., 1994), but also broader efforts were
undertaken, such as the updated work of Meadows et al.
(1992) and new integrated studies on such themes as
climate change, water scarcity, public health, and land-
use (Rotmans and de Vries, 1997). The IPCC series of
greenhouse gas emissions scenarios studies became
successively more sophisticated (IPCC (Intergovern-
mental Panel on Climate Change), 1990;Leggett et al,
1992;Nakicenovic and Swart, 2000). Meanwhile, a
number of studies adopted the narrative scenario
tradition (e.g., Svedin and Aniansson, 1987).
4
The distinction between quantitative (modeling) and
qualitative (narrative) traditions of scenario analysis
should be underscored. Quantitative analysis often relies
on formal models, using mathematical algorithms and
relationships to represent key features of human and
environmental systems. Quantitative modeling is often
used for predictive analysis, which is appropriate for
simulating well-understood systems over sufficiently
short times. But as complexity increases and the time
horizon of interest lengthens, the power of prediction
diminishes. Quantitative forecasting is legitimate to the
degree the state of the system under consideration can
be specified, the dynamics governing change are under-
stood and known to be persistent, and mathematical
algorithms can be created that map these relationships
with sufficient accuracy for simulation. These conditions
are violated when the task is to assess the long-range
future of socio-ecological systems—state descriptions
are uncertain, causal interactions are poorly understood
and non-quantifiable factors are significant. In such
case, even probabilistic forecasting of a given future
state, or a spectrum of possible states, is not feasible.
Systems can branch into multiple future pathways, each
consistent with current conditions, trends and drivers,
and some entailing discontinuous and novel behavior.
This suggests the desirability of non-predictive forms of
quantitative scenario analysis.
The limitations of quantitative analysis mean that it
should be complemented by qualitative scenario ex-
ploration, which can probably better capture other
factors influencing the future such as system shifts and
surprises, or non-quantifiable issues such as values,
cultural shifts and institutional features. The scenario
narrative gives voice to the important qualitative factors
shaping development such as values, behaviors and
institutions, providing a broader perspective than is
ARTICLE IN PRESS
4
For a recent review of both modeling and narrative global
scenarios analyses, mainly related to climate change, see Morita et al.
(2001) and for an overview of applications of scenarios for climate
change impacts, see Carter et al. (2001).
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146140

possible from mathematical modeling alone. Recent
combinations of long-term narratives with scenarios
quantification are attempting to combine the advantages
of both approaches (Raskin et al., 1998;Nakicenovic
and Swart, 2000,Tansey et al., 2002). Narrative offers
texture, richness and insight, while quantitative analysis
offers structure, discipline and rigor. In this sense,
scenario analysis, with its rich history of alternative
traditions and approaches, offers the potential of
fostering the integration of descriptive and normative
or interpretive traditions.The art is in the balance
(Raskin et al., 1996).
Another way of distinguishing between types of
scenarios is between primarily descriptive scenarios,
i.e., scenarios describing possible developments starting
from what we know about current conditions and
trends, and primarily normative scenarios, i.e., scenarios
which are constructed to lead to a future that is afforded
a specific subjective value by the scenario authors. We
add the word ‘‘primarily’’ because in practice scenarios
have elements of both types, but with great differences
in emphasis. Neither of these types is value-free, since
both embody extra-scientific judgments about how the
problem is to be framed, and what are reasonable or
feasible assumptions. However, they differ in terms of
overall purpose. That is, the choice between descriptive
or normative scenarios is dependent on the objectives of
the scenario development exercise. Normative scenarios
represent organized attempts at evaluating the feasibility
and consequences of trying to achieve certain desired
outcomes or avoid the risks of undesirable ones.
Descriptive scenario analysis, on the other hand, tries
to articulate different plausible future societal develop-
ments, and explore their consequences.
From a methodological point of view, scenario
authors can attempt to discern the likely outcome of a
range of ‘‘expected’’ trends,
5
outline the implications of
different assumptions not chosen on the basis of
likelihood (what-if analysis) or examine the feasibility
and implications of desirable futures—or risks of
undesirable ones (backcasting). For sustainability pro-
blems, a combination of backcasting from an array of
possible end-states and forward-looking analysis from
initial conditions and drivers of change is appropriate.
The latter helps to identify long-term risks and to specify
sustainability conditions, while the former identifies the
bandwidth of initial trajectories and available actions to
‘‘bend the curve’’ (Raskin et al., 1998) toward long-term
sustainability goals.
4. Integrating scenario analysis into the sustainability
science toolkit
With this background, the potential for scenario
analysis as a tool for addressing the core questions and
methodological challenges of sustainability science
comes into focus. Sustainability science must consider
the interplay and dynamic evolution of social, economic
and natural systems—it requires an integrated and long-
term perspective. It must address the sustainability
process as tentative, open and iterative, involving
scientific, policy and public participation. It must
capture the possibility of structural discontinuity and
surprise in socio-ecological systems. And it must
recognize the critical importance of alternative, and
sometimes competing, stories, beliefs, institutional con-
texts and social structures.
Modern scenario methods are well-suited to these
tasks. They can help to organize scientific insight into an
integrated framework, gauge emerging risks, and
challenge the imagination. They can provide a means
for integration of descriptive and narrative elements,
and qualitative and quantitative information. They ease
communication with non-scientific audiences, and can
engage diverse stakeholders as actors in scenario design
and refinement. Though their subject is the future,
scenarios can catalyze and guide appropriate action
today for a sustainability transition.
Table 1 summarizes how scenario analysis can
contribute to the various strategies introduced in the
previous section. Several examples of recent projects
demonstrate that various of the strategic challenges
discussed above can be successfully met by scenario
development and analysis. The work of the Global
Scenario Group (Gallopin et al., 2001;Raskin et al.,
1998, 2002;seehttp://www.gsg.org) not only couples
regional to global scales, it also demonstrates how
narratives can be successfully combined with model-
based quantification of socio-economic developments
and the environmental changes they cause. Their
regional elaboration by UNEPs Global Environmental
Outlook (UNEP, 2002) added a participatory regional
consultative process and a direct link with policy
processes. In the Netherlands, a series of projects have
adopted a participatory scenario approach to analyze
climate change response options. Initially, a project was
implemented in which an integrated assessment model
was used as the basis for a dialogue with negotiators of
the UN Framework Convention on Climate Change
(Klabbers et al., 1996;van Daalen et al., 1999), from
which various interactive tools addressing specific
questions developed (e.g. Swart et al., 1998;Berk and
Janssen, 1997). The Climate OptiOns for the Long-term
(COOL) project followed-up this work linking three
spatial levels: national, European and global, and
involving a variety of stakeholders such as national
ARTICLE IN PRESS
5
This does not imply that scenarios can be (‘‘single’’) forecasts, or
represent ‘‘most likely’’ or ‘‘business-as-usual’’. Such terms mislead-
ingly suggest that the probability of particular futures is objectively
known.
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146 141

ARTICLE IN PRESS
Table 1
How can scenario analysis contribute to the methodological challenges of sustainability science
Research challenges Key aspects Contribution of scenario analysis
1. Spanning spatial scales Socio-ecological change must be studied at various levels of
spatial resolution. Planetary, regional and local change are co-
determining, but the linkages are not well understood.
Absent full scientific understanding of cross-scale linkages,
‘‘what-if’’ analysis can explore possible linkages and link local,
regional and global perspectives in a common and consistent
framework. Examples of regional scenarios associated with
global scenarios are the GBFP, GSG and GEO projects (see
main text).
2. Accounting for temporal inertia
and urgency
Processes play out at disparate time scales. System inertia
characterizes both natural and socio-economic systems, while
system change is increasingly indeterminate as the time horizon
of analysis lengthens. Yet, societal decisions in response to
long-term changes must be made in the short term.
Through backcasting, scenario analysis can link long-term
goals (e.g., associated with a sustainability transition) to the
shorter time horizon of today’s decision makers. Examples are
the tolerable windows/safe landing approaches to climate
change, the GSGs ‘‘bending the curve’’ scenarios and the
GBFP scenarios.
3. Recognizing the wide range of
outlooks
Sustainability science cannot be divorced from the values that
shape people’s perspectives and preferences with respect to
sustainable futures and on ways and means to get there.
Traditional scientific research is poorly equipped to deal with
such normative issues.
Participatory scenario development, involving key
stakeholders, is one of the ways to address different views on
how the world works, how a sustainability transition could be
envisaged and how it could be achieved. The GBFP and IPCC
work are good examples.
4. Reflecting functional complexity
and multiple stresses
Multiple stresses affect the interdependent functioning social
and environmental systems at all scales affecting their
resilience. Scientific explanation of these complex interactions
is likely to remain limited.
Qualitative expert knowledge applied in ‘‘what-if’’ scenario
analysis can help charting possible complex linkages and
multiple stresses, minimizing inconsistencies.
5. Integrating across themes and
issues
Component aspects and features of combined human and
environmental systems are highly interdependent while policy
interventions ripple and cascade through the combined system.
The scientific research and policy communities are both highly
segregated into individual disciplines, subject areas, or themes.
Integrated scenario analysis provides opportunities to broaden
perspectives and reveal links. The GBFP, GSG and GEO
projects are examples (see main text).
6. Reflecting uncertainties,
incorporating surprise, critical
thresholds and abrupt change
Conventional methods are not well-equipped to deal with
discontinuous changes in complex, non-linear systems.
Scientific uncertainties in the complex socio-natural system
relevant to sustainability issues are deep and may not be
resolved. It is crucial to understand their importance and to
make them explicit in research output.
Since surprises, critical thresholds and abrupt changes can have
dramatic consequences for nature and society, creative, ‘‘what-
if’’ scenarios offers the possibility of exploring the possibilities
and analyzing the consequences. Svedin and Aniansson (1987)
analyzed some options.
7. Accounting for volition The dynamics of change is influenced by individual and
collective human decisions at all scales. Human behaviors and
choice will determine both the goals and pathways of
development and, hence, the prospects for a sustainability
transition.
Scenario analysis is a powerful method to explore normatively
distinct future images and alternative pathways for getting to
each. It also helps researchers and users of scenarios alike to
reflect on their own worldviews, biases and values, and thus
enrich the science and practice of sustainability. The
backcasting analyses of the COOL and the GBFP projects are
examples.
8. Combining qualitative and
quantitative analysis
Cultural, institutional and value aspects of sustainability,
although difficult to quantify, must be considered in a unified
framework with those bio-physical, economic and social
features in which quantitative analysis adds insight.
Narrative scenarios capture qualitative features, which can be
used in conjunction with quantitative descriptions. The work
from the Global Scenario Group and the IPCC are examples of
this integration. Models can also be designed to capture
qualitative aspects of scenario analysis, as in the QUEST model
used in the GBFP.
9. Engaging stakeholders Human agents are an important internal feature of the system
that sustainability science must address. Stakeholder
engagement allows for taking into account the normative
dimensions of sustainability, widens the knowledge base, and
enhances mutual learning.
Scenarios promote communication between researchers and
stakeholders, provide a structured framework for
incorporating feedback into iterative analysis and offer a
laboratory for testing (and influencing) human perceptions and
goals.
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146142

climate negotiators; sectoral policy-makers; representa-
tives of the private sector; representatives of environ-
mental NGOs, politicians and government policy
makers; and scientists (Berk et al., 2002). The main
objective was to develop a process in which the
implications of long-term sustainability goals related
to climate change and equity were tied to short-term
decision-making, using backcasting techniques. In the
Georgia Basin Futures Project (GBFP) sustainable
development scenarios in the Georgia Basin (on the
west coast of the Canadian province of British
Colombia) are explored in a ‘‘second order’’ backcasting
framework (Robinson, 2003) by emphasizing commu-
nity engagement through participatory integrated as-
sessment methods (Tansey et al., 2002; see also
www.basinfutures.net). The project addresses the chal-
lenges of linking between temporal and spatial scales,
combining qualitative and quantitative scenario analy-
sis, recognizing diverse outlooks and the influence of
human choice, and linking with policy choices through
stakeholder involvement. The Intergovernmental Panel
on Climate Change (IPCC) has developed and used
emissions scenarios for all three of its major assessments
to date (IPCC, 1990;Leggett et al., 1992;Nakicenovic
and Swart, 2000). The evolution of these scenario sets
illustrates some of the challenges discussed in this paper.
One business-as-usual scenario with three policy scenar-
ios was conceived by a very limited number of modelers
in relative isolation in the first assessment report (IPCC,
1990). In the second assessment a broader set of future
scenarios were explored by the same small group,
simply reflecting low, medium or high levels of devel-
opment (Leggett et al., 1992). For its 3rd assessment,
IPCC engaged a much wider set of experts and
stakeholders to develop a much richer array of possible
future developments, quantifying these with the help of
a multitude of different modeling techniques and placing
greenhouse gas emissions in the much broader context
of socio-economic futures (Nakicenovic and Swart,
2000).
While scenario analysis cannot provide, of course, all
the answers to the questions posed by sustainability
science, it has an important role to play in synthesis,
thinking about the future, and linking to policy and
stakeholder communities. This role is complementary to
the daunting amount of non-scenario work required for
a robust sustainability science. Each type of scenario
analysis can address different elements of the questions,
and different challenges posed by the proposed research
strategies. For example, backcasting approaches are
useful when it comes to analyzing today’s implications
of long-term risks or long-term sustainability objectives,
taking into account the inertia of natural and social
systems, and for exploring different strategies. ‘‘What-
if’’ analysis can be useful to evaluate under which
conditions and when particular types of surprises could
occur or thresholds be passed, and then to explore how
to take these into account in today’s decision-making
processes. Forward-looking analysis is the appropriate
methodology if we want to explore how different
plausible socio-economic trends would work out in the
short-term future and how these might interact with
changes in natural systems, taking into account all
relevant scientific uncertainties. Narrative scenario
analysis facilitates a debate about normative aspects of
sustainability, quantitative analysis can contribute to an
adequate knowledge base and structural consistency.
Two aspects of scenario analysis in the context of a
sustainability transition deserve special attention. First,
scenario analysis for sustainability science as we see it
goes beyond the traditional view on the relationship
between science and policy, which assumes that science
provides—sometimes on request, sometimes not—in-
formation to policy-makers in order to improve the
quality of the decision-making process. Because of the
apparent objective nature of scenarios developed from
this perspective, they can be used by policy makers as a
means of legitimizing rather than informing policy
decisions.
Second, scenario analysis in the context of sustain-
ability science has a potentially important role to play
with regard to the increasing demand for more public
and stakeholder involvement in the scientific activities,
driven by a complex mix of factors, including increased
public distrust of expert-driven decision making, grow-
ing awareness of a diversity of opinions in the scientific
community, and increased sophistication of NGO,
private sector and public involvement in regulatory
and other decision-making fora. These evolving dimen-
sions of the policy–science interface suggest that
participatory forms of scenario analysis could be
particularly effective in addressing the strategic and
normative elements of the sustainability questions by
incorporating values and preferences into the scenario
analysis process itself. The objectives and design of a
participatory scenario development exercise would be
different for involvement of key stakeholders with
advanced levels of scientific and technical expertise, as
compared to engagement of the general public. But in all
cases, scenario analysis for sustainability science should
encompass mutual learning about the knowledge,
positions and preferences of those involved, hopefully
leading to better informed decision-making.
6
ARTICLE IN PRESS
6
When combined in the exploration of desired futures, these two
factors can give rise to a ‘second order’ form of backcasting, in which
the desired future is an emergent property of the (participatory)
process of creating and evaluating scenarios (Robinson, 2003). Such
processes can combine the descriptive and normative approaches
described earlier.
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146 143

5. Conclusions and future directions
In summary, the emergence of a globalized phase of
development is bringing both new opportunities and
new perils. In many regions technological advances
thrive, incomes increase, and health conditions improve.
At the same time, poverty and hunger continue to
plague hundreds of millions of people, conflicts abound,
and ecological resources are under continuous pressure
around the globe. With the possibility of the world’s
population doubling in this century and economic
output increasing significantly, these problems can be
expected to become even more pronounced and urgent.
There is broad international agreement that a transition
to sustainability is needed. But incomplete understand-
ing of the problems, their causes and possible
solutions makes society poorly prepared for such a
transition. Moreover, the prospects for effective
responses are complicated by contradictory philosophi-
cal views underlying policy and scientific discussions,
views that are rarely made explicit, on how natural and
human systems operate and interact, or how they should
be managed.
Science can contribute to the sustainability transition
by providing knowledge and guidance for navigating the
journey from unsustainable contemporary patterns to a
sustainable future. We have argued that scenario
analysis offers one promising approach for operationa-
lizing and enriching the required ‘‘sustainability
science’’. The process and product of scenario analysis
are equally important. In many contexts, scenario
exercises will benefit by having both science and affected
actors shape the analysis. Scientists bring knowledge of
relevant processes and their linkages to the discourse
and stakeholders enrich scenarios by bringing the
perspectives of the human participants in the story of
the future.
Scenario analysis can play a major role in addressing
the challenges of sustainability science, especially the
core question of how to scan the future in a structured,
integrated and policy-relevant manner. Based on our
scenario experiences over the past 20 years, we offer the
following principles of good practice for those who
would like to pursue scenario analysis as a means of
addressing the vital questions of sustainability science.
While each scenario exercise should be tailored to the
specific problems and context, the following broad
aspects should be considered:
*A sufficiently large and diverse group of participants:
The typical size of the core group developing the
scenarios in the examples cited earlier was between 10
and 40, usually involving experts from different
disciplinary backgrounds and stakeholders with
different interests. Interestingly, scientists, represen-
tatives from the private sector, governments and
NGOs can find common ground through a scenario
development process, since all these groups have
usually had at least some exposure to scenarios, many
even actively. Key stakeholders can be integrated
directly into the problem definition, research design
and scenario generation components of the research.
For global scenario work, balanced representation
from all regions is important and, ideally, a wider
community of experts and stakeholders should be
consulted, e.g., through consultations or review
processes. The process of scenario development
should be a process of mutual learning, and co-
production of knowledge by those involved.
*Adequate time for problem definition, knowledge base
development, iterative scenario analysis, review and
outreach: It is vital to devote sufficient time and effort
to build trust among team members and a shared
appreciation of critical questions, the research
strategy, and the audience. Over a period of
months, storylines, possibly supported by quantifica-
tion of key scenario elements, are best developed
through several iterations, with penultimate results
reviewed by peers beyond the core team. In many
cases, the processes of interacting with stakeholders
in the course of generating scenarios are as important
as the scenario analysis itself. Finally, effective
communication with the external audience needs
careful attention, often requiring specialized
skills and creative approaches beyond conventional
scientific reports.
*Full account of available scientific knowledge and rigor
of methods: All relevant scientific knowledge about
what is known, and what is unknown about the
problem being considered, should be incorporated
into the scenario process. Existing scenario analyses,
modeling exercises and databases can be useful
sources of information and, in some cases, pitfalls
to be avoided. This includes predetermined elements
and critical uncertainties. Care should be taken when
applying tools developed for one type of question for
addressing other questions.
*An explicit discussion about normative scenario
elements: Scenario analysis is a means to address
the inherently normative dimensions of sustainability
that takes sustainability science beyond the bound-
aries of the traditional scientific enterprise. The
normative aspect enters the scenario in two ways.
First, the storylines make assumptions about future
behaviors and worldviews of scenario actors, invol-
ving assumptions on religion, spirituality, norms and
values, as well as socio-political and institutional
design questions. Second, the worldviews of the
people creating the scenarios shape the way the
story is told and what policy lessons are drawn.
Transparency and rigor require that each of these
normative dimensions be made explicit in developing
ARTICLE IN PRESS
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146144

and analyzing the scenarios. Through open internal
discussions and well-designed external communica-
tion, mental maps of both participants and audience
can be revealed, challenged and enriched.
*The development of coherent, engaging stories about
the future: While quantitative analysis can add insight
and consistency, the power of scenarios lies in telling
compelling stories that capture the imagination,
understanding, and beliefs, hopes and dreams of
participants. The stories should have a consistent
logic and take into account evolving positions and
power balance of stakeholders.
The narratives help reveal and address critical questions
that might otherwise be neglected and offer a powerful
vehicle for effective communication with target audi-
ences.
*Explore the possibility of surprise events and address
possible seeds of change: Many scenarios are quite
narrow, restricting their analysis to dominant trends
and incremental variations. The future however, can
be influenced by surprise events and novel phenom-
ena. Explicitly considering such possibilities will
enrich the scenarios discussion and ultimate narra-
tives. Similarly, ‘‘seeds of change’’ should be probed,
societal or natural developments with the potential to
significantly change society, and that are presently
dormant or in their early stages of evolution (van
Notten, 2002).
*Place the focal problem in a broader context: Often,
scenario exercises focus on a single dimension of
sustainability, such as climate change, biological
diversity, poverty, International security, demo-
graphics, water, agriculture and energy. These issues,
which become manifest at different spatial resolu-
tions and time scales, influence each other. A systemic
and integrated perspective will help real key linkages
that influence the focal problem.
The development of an effective science of sustain-
ability is an urgent endeavor that requires further
clarification of the contours of its key questions and
research agenda. The character of the sustainability
problems compels a systemic exploration of the future,
that is, sensitive to normative issues grounded in
cultural, spiritual and philosophical attitudes of people,
while incorporating methodological rigor. To succeed,
sustainability science will need to be a dynamic, evolving
field of inquiry and application, which seeks integration
across disparate natural and social science perspectives.
Scenario analysis offers a powerful platform for evol-
ving an integrative, conceptually rich and inclusive
process of relevant knowledge generation for a sustain-
able future.
References
Berk, M., Janssen, M., 1997. The interactive scenario scanner (ISS): a
tool to support the dialogue between science and policy on scenario
development. National Institute for Public Health and the
Environment, Bilthoven, pp. 31.
Berk, M.M., van Minnen, J.G., Metz, B., Moomaw, W., den Elzen,
M.G.J., van Vuuren, D.P., Gupta, J., 2002. Climate OptiOns for
the Long term (COOL)—Global Dialogue Synthesis Report RIVM
Report No. 490200003, National Institute for Public Health and
the Environment, Bilthoven, pp. 75.
Carter, T.R., Larovere, E.L., Jones, R.N., Leemans, R., Mearns, L.O.,
Nakicenovic, N., Pittock, A.B., Semenov, S.M., Skea, J., 2001.
Developing and applying scenarios. In: Mccarthy, J.J., Canziani,
O.F., Leary, N.A., Dokken, D.J., White, K.S. (Eds.), Climate
Change 2001:Impacts, Adaptation, and Vulnerability. Cambridge
University Press, Cambridge, UK.
Cohen, S., Demeritt, J., Robinson, J., Rothman, D., 1998. Climate
change and sustainable development: towards dialogue. Global
Environmental Change 8 (4), 341–371.
van Daalen, E.C., Thissen, W.A.H., Berk, M.M., 1999. The Delft
process: experiences with a dialogue between policy makers and
global modellers. In: Alcamo, J., Leemans, R., Kreileman, E.
(Eds.), Global Change Scenarios of the 21st Century: Results from
the IMAGE 2.1 Model. Elseviers Science Publishers, London,
pp. 267–285 (ISBN 0-08-043447-9).
Gallopin, G., Funtowicz, S., O’Connor, M., Ravetz, J., 2001. Science
for the 21st century from social contract to the scientific core.
International Journal of Social Science 168, 219–229.
H.
afele, W., 1981. Energy in a Finite World: A Global
Systems Analysis. Ballinger, Cambridge, MA, USA, 834pp (ISBN
0-88410-642-X).
Herrera, A., Scolnik, H., Chichilnisky, G., Gallopin, G., Hardoy, J.,
Mosovich, D., Oteiza, E., de Romero Brest, G., Suarez, C.,
Talavera, L., 1976. Catastrophe or New Society? A Latin American
World Model. International Development Research Centre,
Ottawa, Canada.
ICSU (International Council for Science), 2002. Report of the
Scientific and Technological Community to the World Summit
on Sustainable Development (WSSD). Series on Science for
Sustainable Development, No. 1, ICSU, pp. 20.
IGBP/IHDP/WCRP/Diversitas (International Geosphere-Biosphere
Programme/International Human Dimensions Programme/World
Climate Research Programme/Biodiversitas international biodiver-
sity research programme), 2001. The Amsterdam Declaration.
IPCC (Intergovernmental Panel on Climate Change), 1990. The IPCC
Response Strategies. Island Press, United States.
Kahn, H., Wiener, A.J., 1967. The Year 2000: a Framework for
Speculation on the Next Thirty-three Years. Macmillan,
New York.
Kates, R.W., Clark, W.C., Corell, R., Hall, J.M., Jaeger, C., Lowe, I.,
McCarthy, J.J., Schellnhuber, H.-J., Bolin, B., Dickson, N.M.,
Facheux, S., Gallopin, G.C., Gruebler, A., Huntley, B., J .
ager, J.,
Jodha, N.S., Kasperson, R.E., Mabogunje, A., Matson, P.,
Mooney, H., Moore III, B., O’Riordan, T., Svedin, U., 2001.
Sustainability Science. Science 27(292), 641–642 (extended version
published by the Belfer Center for Science and International
Affairs, John F. Kennedy School of Government, Harvard
University, USA).
Klabbers, J.H.G., Swart, R.J., van Ulden, A.P., Vellinga, P., 1996.
Climate science and climate policy: improving the science/policy
interface. Mitigation and Adaptation Strategies for Global Change
1, 73–93.
Leggett, J., Pepper, W., Swart, R., 1992. Emissions Scenarios for
IPCC: an update. In: Houghton, J., Callander, B., Varney, S.
(Eds.), Climate Change. The Supplementary Report to the IPCC
ARTICLE IN PRESS
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146 145

Scientific Assessment. Cambridge University Press, Cambridge,
pp. 1–25.
Lovins, A., 1976. Energy Strategy: the road not taken? Foreign
Affairs, October 1976, 55 (1), pp. 186–217.
Lovins, A., 1977. Soft Energy Paths. FOE/Ballinger, New York.
Lubchenko, J., 1998. Entering the century of the environment: a new
social contract for science. Science 279, 491–497.
Meadows, D.H., Meadows, D.L., Randers, J., Behrens III, W.W.,
1972. Limits to Growth. Universe Books, New York.
Meadows, D.H., Meadows, D.L., Randers, J., 1992. Beyond the
Limits: Confronting Global Collapse, Envisioning a Sustainable
Future. Chelsea Green Publishing Company, Post Mills, VT.
Mesarovic, M., Pestel, E., 1974. Mankind at the Turning Point: the
Second Report to the Club of Rome. Dutton, New York.
Morita, T., Robinson, J., Adegbulugbe, A., Alcamo, J., Herbert, D.,
Lebre La Rovere, E., Nakicenovic, N., Pitcher, H., Raskin, P.,
Riahi, K., Sankovski, A., Sokolov, V., de Vries, B., Zhou, D., 2001.
Greenhouse gas emission mitigation scenarios and implications. In:
Metz, B., Davidson, O., Swart, R., Pan, J. (Eds.), Climate Change
2001: Mitigation. Cambridge University Press, Cambridge, UK.
Nakicenovic, N., Swart, R., (Eds.), 2000. IPCC Special Report on
Emissions Scenarios. Cambridge University Press, UK.
NAS (US National Academy of Sciences), 1999. Our Common
Journey: a Transition toward Sustainability. A report of the Board
on Sustainable Development of the National Research Council.
National Academy Press, United States.
van Notten, P., 2002. Foresight in the face of scenario diversity. Paper
presented at International Conference on Probing the Future:
Developing Organizational Foresight in the Knowledge Economy.
Glasgow, United Kingdom.
Raskin, P., Chadwick, M., Jackson, T., Leach, G., 1996. The
Sustainability Transition: Beyond Conventional Development.
Stockholm, Stockholm Environment Institute.
Raskin, P., Gallop!
ın, G., Gutman, P., Hammond, A., Swart, R., 1998.
Bending the Curve: Toward Global Sustainability. PoleStar Series
Report No. 8, Stockholm Environment Institute, Stockholm,
Sweden.
Raskin, P., Banuri, T., Gallopin, G., Gutman, P., Hammon, A., Kates,
R., Swart, R., 2002. Great Transition: the Promise and Lure of the
Times Ahead. Stockholm Environment Institute, Boston, USA.
Raven, P., 2002. Science, sustainability, and the human prospect.
Science 297, 954.
Rayner, S., Malone, E., 1998. The challenge of climate change to the
social sciences. Human Choice and Climate Change. Battelle Press,
Columbus, OH.
Robinson, J., 1982. Energy backcasting: a proposed method of policy
analysis. Energy Policy 10 (4), 337–344.
Robinson, J., 2002. Squaring the Circle: On the Very Idea of
Sustainable Development. In: Chesworth, W., Moss, M., Thomas,
V., (Eds.), Sustainable Development: Mandate or Mantra? The
Kenneth Hammond Lectures on Environment, Energy and
Resources—2001 Series, Faculty of Environmental Studies,
University of Guelph.
Robinson, J., 2003. Future subjunctive: backcasting as social learning.
Futures 35, 839–856.
Robinson, J., Biggs, D., Francis, G., Legge, R., Lerner, S., Slocombe,
S., Van Bers, C., 1996. Life in 2030: Exploring a Sustainable Future
in Canada. UBC Press, Vancouver.
Robinson, J., Tinker, J., 1998. Reconciling ecological, economic and
social imperatives. In: Schnurr, J., Holtz, S. (Eds.), The Corner-
stone of Development: Integrating Environmental, Social and
Economic Policies, International Development Research Centre,
Ottawa.
Rotmans, J., 1990. IMAGE: an Integrated Model to Assess the
Greenhouse Effect. Kluwer, Dordrecht, The Netherlands.
Rotmans, J., Hulme, M., Downing, T.E., 1994. Climate change
implications for Europe: an application of the escape model.
Global Environmental Change 4, 97–124.
Rotmans, J., de Vries, H.J.M., (Eds.), 1997. Towards Integrated
Assessment of Global Change. Perspectives on Global Change: The
TARGETS approach. Cambridge University Press, Cambridge,
UK.
Schellnhuber, J., Sahagian, D., 2002. The twenty-three GAIM
Questions. IGBP Newsletter 49, 20–21.
Schwartz, P., 1991. The Art of the Long View: Paths to Strategic
Insight for Yourself and Your Company.Doubleday.
Shell, 2002. People and Connections; Global Scenarios to 2020;
Public Summary. Global Business Environment, Shell
International.
Svedin, Uno, Aniansson, B., 1987. Surprising Futures, Swedish
Council for Planning and Coordination of Research, Sweden.
Swart, R., Berk, M.M., Janssen, M., Kreileman, E., Leemans, R.,
1998. The safe landing analysis: risks and trade-offs in climate
change. In: Alcamo, J., Leemans, R., Kreileman, E. (Eds.), Global
Change Scenarios of the 21st Century. Results from the IMAGE
2.1 Model. Elsevier, London, pp. 193–218.
Tansey, J., Carmichael, J., VanWijnsberghe, R., Robinson, J., 2002.
The future is not what it used to be: participatory integrated
assessment in the Georgia Basin. Global Environmental Change 12
(2), 97–104.
UNEP, 2002. Global Environment Outlook 3. Earthscan, London,
United Kingdom.
UNGA (United Nations General Assembly), 2000. The Millennium
Declaration. United Nations, New York, United States.
Wack, P., 1985a. Scenarios: uncharted waters ahead. Harvard Business
Review 5 (Sept./Oct. 1985), 72–89.
Wack, P., 1985b. Scenarios: shooting the rapids. Harvard Business
Review 6 (Nov./Dec. 1985), 139–150.
WCED (World Commission on Environment and Development),
1987. Our Common Future. Oxford University Press, Oxford, UK.
ARTICLE IN PRESS
R.J. Swart et al. / Global Environmental Change 14 (2004) 137– 146146