Reflexive Government For Sustainable Davelopment
Reflexive Government For Sustainable Davelopment
Reflexive Government For Sustainable Davelopment
Development
Reflexive Governance
for Sustainable
Development
Edited by
Jan-Peter Vo
Dierk Bauknecht
ko-Institut Institute for Applied Ecology, Germany
and
Ren Kemp
MERIT, Maastricht University, The Netherlands
Edward Elgar
Cheltenham, UK Northampton, MA, USA
Jan-Peter Vo, Dierk Bauknecht and Ren Kemp 2006
Published by
Edward Elgar Publishing Limited
Glensanda House
Montpellier Parade
Cheltenham
Glos GL50 1UA
UK
Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall
Contents
List of figures vii
List of tables ix
List of contributors x
Preface xiv
INTRODUCTION
1. Sustainability and reexive governance: introduction 3
Jan-Peter Vo and Ren Kemp
v
vi Contents
CONCLUSIONS
16. Reexive governance: a view on an emerging path 419
Jan-Peter Vo, Ren Kemp and Dierk Bauknecht
Index 439
Figures
5.1 Four phases of transition 106
5.2 A transition is the result of system innovations and other
innovations and changes 107
5.3 Current policy versus transition management 110
5.4 Transition visions: images and pathways 114
5.5 Activity clusters in transition management 121
5.6 The Dutch vision for biomass 124
6.1 The Tisza river basin with tributaries in catchments in the
Carpathian mountain range across portions of ve
dierent national territories (Romania, the Ukraine,
Slovakia,the Federation of Serbia and Montenegro, and
Hungary) 133
6.2 An artists interpretation of dierent landscape mosaic
patterns in the Tisza river oodplain both before (a) and after
(b) the execution of the original Vsrhelyi river engineering
plan in 1871 138
6.3 The adaptive management process as a structured learning
cycle that iteratively links four phases: assessment,
formulation, implementation, and monitoring 141
6.4 Adaptive management framework revised according to system
dynamics modelling approaches of Sterman (2000) to
incorporate performance indicators explicitly 151
6.5 Conceptual model (mental model map) of key variables and
causal loops that interact to aect Tisza river oodplain
resilience to climate related hydro-dynamic variability 154
9.1 Working denitions for unreectiveness, reection and
reexivity for the present chapter 228
9.2 Risk, uncertainty, ambiguity and ignorance as degrees of
incertitude (with examples) 239
9.3 Contrasting methodological responses appropriate under
dierent degrees of incertitude 241
9.4 Implications of reection and reexivity for the role of
science in sustainable governance 247
13.1 Variables characterising sanitation technologies 342
vii
viii Figures
ix
Contributors
Peter Balogh is a geographer and head of research of the Floodplain
Management Pilot Project in Nagykru, introducing the principals and
practice of sustainable water and land use in the Tisza-region.
Dierk Bauknecht is a research fellow and project leader with the Energy
and Climate Group, ko-Institut, Institute for Applied Ecology, Germany.
He works on the transformation of energy infrastructures, innovation
studies and transdisciplinarity.
Ulrich Beck is professor of sociology at the University of Munich and at
the London School of Economics. He is editor of the Second Modernity
Series published by Suhrkamp Verlag. Research foci are theories of moder-
nity and the sociology of risk.
John Grin is a professor of policy science, in particular system inno-
vation, at the Department of Political Science of the University of
Amsterdam. His current research focuses on system innovations, that is,
reexive modernisation of socio-technological practices, especially in the
areas of agriculture, water management and health care.
Ren Kemp is a senior research fellow at MERIT (Maastricht Economic
Research Institute on Innovation and Technology). He has been working
since 1988 on innovation and environmental policy issues, as an economist
and policy analyst. He has a broad range of expertise, covering various sectors
(manufacturing, transport, energy, waste and to a lesser extent services).
Kornelia Konrad is with the research group CIRUS (Centre for
Innovation Research in the Utility Sector) at EAWAG (Swiss Federal
Institute for Aquatic Science and Technology) in Switzerland. The CIRUS
research focus is on sustainable transformation of infrastructure sectors
from a social science perspective.
Marie Cline Loibl is with the Austrian Institute for Applied Ecology. Her
research interests include: methods for interdisciplinary and transdisciplin-
ary research projects, team-coaching in environmental research, consulta-
tion for designing and organising environmental research-programs, process
evaluation and participatory methods for integrating representatives from
municipalities, enterprises and private initiatives into research projects.
x
Contributors xi
xiv
Preface xv
Dierk Bauknecht
Ren Kemp
Jan-Peter Vo
Freiburg, Maastricht, Berlin, May 2005
NOTE
1. The German Research Ministry funded the workshop series within the programme on
socio-ecological research as a measure to develop the research infrastructure for sus-
tainability research. It was chaired by the ko-Institut the Institute for Applied Ecology,
as part of the project Schnittstellenentwicklung fr die Integration akademischer und
praxisbezogener Forschung im Bereich Sozial-kologie (knowledge interfaces).
Introduction
1. Sustainability and reexive
governance: introduction
Jan-Peter Vo and Ren Kemp
INTRODUCTION
3
4 Introduction
development brings complexity and uncertainty to the fore as key issues for
sustainability. Sustainability cannot be translated into a blueprint or a
dened end state from which criteria can be derived and unambiguous deci-
sions taken to get there. Instead, it should be understood as a specic kind
of problem framing that emphasises the interconnectedness of dierent
problems and scales, as well as the long-term and indirect eects of actions
that result from it. Societal discourse on sustainability has highlighted the
ambiguity of social goals, uncertainty about cause and eect relations
and the feedback that occurs between steering activities and social, tech-
nological and ecological development. Sustainability calls for new forms of
problem handling. These dier from the forms that are adequate for delim-
itable, decomposable problems that can be managed in a linear way. The
concept of sustainability has brought with it recognition of the limits of
rigid analysis and the inadequacy of policy approaches that aim at plan-
ning and achieving predetermined outcomes.
From this perspective, sustainable development is more about the organ-
isation of processes than about particular outcomes. It is about the modes
of problem treatment and the types of strategies that are applied to search
for solutions and bring about more robust paths of social and technological
development. We set out to explore these new modes of societal problem
treatment under the heading of reexive governance.
REFLEXIVE GOVERNANCE
assess them and to get competences together to act on them they require
putting aside the isolation of instrumental specialisation, widening lters
of relevance, trading o values and engaging in interaction with other spe-
cialists. In short, these problems require transgressing the cognitive, evalu-
ative and institutional boundaries, which, paradoxically, undermines the
modernist problem-solving approach. Problem solving becomes paradox-
ical in that it is oriented towards constriction and selection to reduce com-
plexity but is forced into expansion and amalgamation to contend with the
problems it generates (see Beck 1993). This is what we call the constellation
of reexive problem handling or, on the societal level, reexive governance.
Reexivity has two dierent but related meanings here that are often
confused in accounts of reexive modernisation. The rst meaning of
reexivity refers to how modernity deals with its own implications and side
eects, the mechanism by which modern societies grow in cycles of produc-
ing problems and solutions to these problems that produce new problems.
The reality of modern society is thus a result of self-confrontation. This can
be called rst-order reexivity. The second meaning of reexive moderni-
sation refers to the cognitive reconstruction of this cycle in which problem
solving through instrumental rationality generates new problems. The
impacts of technology, scientic knowledge production and the legitimacy
and eectiveness of democracy are examples of problem areas where
such reection has brought up critical reassessments of rational problem-
solving methods and led to the development of alternative methods and
processes of problem handling that are more open, experimental and learn-
ing oriented. Often these approaches aim to foster interaction between
dierent perspectives and actively explore the uncertainties, ambivalences
and control problems articulated in such a confrontation of rationalities.
Constructive technology assessment, deliberative policy making and trans-
disciplinary research are alternative concepts to rational problem solving
that all underlie concrete practices. New problem-handling paradigms and
institutional arrangements based on critical assessments of modern
problem solving and its reexivity have themselves become characteristic
features of reexive modernisation. But these phenomena are reexive in a
dierent way from that of the self-confrontation of modernisation with its
own side-eects. They represent a second-order reexivity that entails the
application of modern rational analysis not only to the self-induced prob-
lems, but also to its own working, conditions and eects. In this way, second-
order reexivity interrupts the automatism of executing problem-solving
routines. It transcends particular rationalities, and breaks the vicious circle
of rst-order reexivity. Reexive modernisation, or reexive governance,
comprises both the condition of being shaped through its own side-eects
and the transcendence of this cyclic pattern through reection of the
Sustainability and reexive governance 7
Heterogeneous interactions
The understanding of long-term transformations in socio-ecological
systems such as, for example, energy production and use, transport or agri-
culture requires knowledge about the very heterogeneous elements of these
systems. Such elements include technological artefacts and networks,
chemical substances in soil, water and the atmosphere, the organisation of
rms and markets, political institutions, scientic theories and cultural
values and attitudes. Knowledge is needed about the processes in which
they each change and about how they relate and interact with each other.
Conventional disciplinary science does not deliver this kind of knowledge
about the interlinked and complex nature of reality (Gallopn et al. 2001:
228). Instead, it concentrates on a very specic selection of elements and
interactions analytical slices of reality. In real world entanglements,
however, there is no clear boundary between these categories and the net-
works of cause and eect that cut across them. Each specialised perspec-
tive denes the systemic embedding of the particular analytical abstraction
with which it is concerned as non-existent. In specic cases, this may be
methodologically justiable because linkages have been found to be
insignicant. Parts of reality can sometimes be viewed in isolation without
losing important aspects. In most cases, however, especially in the area of
sustainability problems, linkages extend well beyond the scope of the prob-
lems as they are dened by disciplines and the cognitive models that are
used to understand them.
The knowledge restrictions of specialised perspectives relate not only to
scientic disciplines but also to the scientic method of knowledge pro-
duction in general. The full set of factors and interactions that are relevant
in real world problem settings cannot be handled through systematic mod-
elling alone. More synthetic kinds of knowledge, gained from practical
experience, are an important complementary source. Knowledge produc-
tion for sustainable development cannot therefore rely only on scientic
knowledge produced within the institutions and along the methodological
guide rails of formal science. It also needs to integrate the tacit knowledge
of societal actors. This tacit knowledge often cannot be subjected to con-
ventional methods of scientic inquiry. It can only be generated in inter-
active settings in which knowledge is co-produced by scientists and actors
from respective elds of societal practice. But also with respect to practice,
Sustainability and reexive governance 11
Uncertainty
The interdependence of social, technological and ecological elements
makes system transformation a complex and uncertain process. The overall
process, its factors and drivers, cannot be analysed by linear models of
cause and eect because feedback is pervasive. If the process of sustainable
transformation for example, of electricity provision or agriculture is
further understood as a process that takes place within a multi-level struc-
ture of nested subsystems at the local, regional and global level, the inter-
action on each level adds to the overall complexity. The result is that
socio-ecological transformation cannot be predicted. Unpredictable inter-
actions may give rise to self-stimulating processes like self-organisation, or
to destructive resonance. Examples of such contingencies include topics in
public discourse, social movements, BSE, strategic action under regulation
and stock market crashes. Thresholds for catastrophic change cannot be
dened by a single parameter but rather are driven by a conuence of many
factors, not all of which can be suciently ascertained to determine corri-
dors of safe levels of activity. Examples of this kind of diculty include
ecological pressure that causes a breakdown of ecosystem resilience, social
injustice that causes upheaval or tax increases that lead to an economic
depression. This is a fundamental constraint because of the impossibility
of measuring all incremental factors that are relevant, especially the human
factor. Non-linear system dynamics may give precisely those apparently
minor factors a signicant voice in where the system will go, as in the
buttery eect (Gleick 1998; Byrne 1998). Here an apparently small eect
tips the balance (examples can be found in Gladwell 2000).
This is why it is not possible to rely on simple models of the causes under-
lying sustainability problems. Even if complexity is excluded from cogni-
tive models, the world still remains as complex as ever and the connections
that are ignored will still be eective (Drner 1989). Inadequate problem
constructions thus return in the form of unexpected consequences when
strategies are implemented in the real world (Bhret 1990). This means that
12 Introduction
Path dependency
In addition to being unpredictable, socio-ecological developments and the
eects of human activity within them feed into a continuous process of
structural changes. These ongoing transformations are sometimes more
subtle, hidden in the background of system structures, and sometimes more
visible as in the overthrowing of established patterns. Increasingly, it is
human activity that shapes world development including its ecological
dimension even when it is not the intention to do so. The global climate
is a prominent example. Deep-rooted changes associated with modernisa-
tion cannot be attributed to particular policies or other strategies. Instead,
such changes are brought about as the aggregate and long-term eect of
unsuspected daily practices in production, consumption and political regu-
lation (Rip and Kemp 1998). Metaphorically speaking, one could say that
future socio-ecological system structures grow behind the backs of the
actors who create them.
In this continuous process of development, patterns emerge in which
social values and institutions, technology and ecological systems become
interdependent. Positive feedback may occur between developments in
technology, corporate organisation, regulation, consumption habits and
Sustainability and reexive governance 13
need to be carried out for specic assessment situations. The broad partici-
pation of aected societal actors in the process of goal formulation is
necessary, because their values and respective perception of problems con-
stitutes a basic condition of sustainable social development.
18
multiple scales dynamics and eects societal impact endogenous to are distributed
(society, of intervention transformation among actors
technology,
ecology)
Strategy Trans-disciplinary Experiments and Anticipation of Iterative participatory Interactive strategy
requirement knowledge adaptivity of long-term goal formulation development
production strategies and systemic eects
institutions of measures
Sustainability and reexive governance 19
Sustainability goals cannot be dened objectively once and for all. This
would require ascertainment of the necessary conditions for the long-term
20 Introduction
REFERENCES
Arthur, W.B. (1997), Increasing Returns and Path Dependence in the Economy, Ann
Arbor, MI: University of Michigan Press.
Bechmann, G. and Frederichs, G. (1996), Problemorientierte forschung: zwischen
politik und wissenschaft, in G. Bechmann (ed.), Praxisfelder der
Technikfolgenforschung: Konzepte, Methoden, Optionen, Frankfurt am Main/
New York: Campus, pp. 121.
Beck, U. (1993), Die Erndung des Politischen, Frankfurt am Main: Suhrkamp.
Beck, U. (1994), The reinvention of politics: towards a theory of reexive mod-
ernization, in U. Beck, A. Giddens and S. Lash (eds), Reexive Modernization,
Cambridge: Polity Press, pp. 155.
Beck, U., W. Bonss and C. Lau (2003), The theory of reexive modernization:
problematic, hypotheses and research programme, Theory, Culture & Society,
20, 133.
Bergmann, M. (2003), Indikatoren fr eine diskursive evaluation transdisziplinrer
forschung, Technikfolgenabschtzung Theorie und Praxis, 12 (1), 6575.
Bhret, C. (1990), Folgen: Entwurf fr eine aktive Politik gegen schleichende
Katastrophen, Opladen: Leske Budrich.
Byrne, D. (1998), Complexity Theory and the Social Sciences: An Introduction,
London/New York: Routledge.
Collingridge, D. (1980), The Social Control of Technology, London: Frances Pinter.
Conrad, J. (1997), Nachhaltige entwicklung: ein kologisch modernisiertes modell
der moderne?, in K.-W. Brand (ed.), Nachhaltige Entwicklung: Eine
Herausforderung an die Soziologie, Opladen: LeskeBudrich.
Dobuzinskis, L. (1992), Modernist and postmodernist metaphors of the policy
process: control and stability vs. chaos and reexive understanding, Policy
Sciences, 25, 35580.
Drner, D. (1989), Die Logik des Misslingens: Strategisches Denken in komplexen
Situationen, Reinbek bei Hamburg: Rowohlt Verlag.
Elzen, B., F. Geels, P. Hofman and K. Green (2002), Socio-technical scenarios as
a tool for transition policy: an example from the trac and transport domain,
paper presented at the workshop Transitions to Sustainability through System
Innovations, University of Twente, Enschede 29 October.
Funtowicz, S., J.R. Ravetz and M. OConnor (1998), Challenges in the use of
science for sustainable development, International Journal of Sustainable
Development, 1 (1), 99107.
Sustainability and reexive governance 27
Gallopn, G.C., S. Funtowicz, M. OConnor and J.R. Ravetz (2001), Science for the
21st century: from social contract to the scientic core, International Journal of
Social Science, 168, 21929.
Gladwell, Malcolm (2000), The Tipping Point: How Little Things Can Make a
Dierence, Boston, New York, London: Little, Brown and Company.
Gleick, J. (1998), Chaos: Making a New Science, London: Vintage.
Godet, M. (1987), Scenarios and Strategic Management, London: Butterworth.
Hirsch Hadorn, G. (2003), Unity of Knowledge in Transdisciplinary Research for
Sustainability, retrieved from http://greenplanet.eolss.net on 20.09.2003.
Jahn, T. and P. Wehling (1998), Gesellschaftliche Naturverhltnisse: Konturen
eines theoretischen Konzepts, in K.-W. Brand (ed.), Soziologie und Natur:
Theoretische Perspektiven, Opladen: LeskeBudrich, pp. 7595.
Kemp, R. (1994), Technology and the transition to environmental sustainability:
the problem of technological regime shifts, Futures, 26, 102346.
Kohler-Koch, B. and R. Eising (eds) (1999), The Transformation of Governance in
the European Union, London: Routledge.
Kooiman, J. (1993), Modern Governance: New Government-Society Interactions,
London: Sage.
Luhmann, N. (1990), kologische Kommunikation: Kann die moderne Gesellschaft
sich auf kologische Gefhrdungen einstellen?, Opladen: Westdeutscher Verlag.
Matthes, F. (2002), Nachhaltigkeit als politisches Konzept, Gaia, 11 (2), 915.
Mayntz, R. (1998), New challenges to governance theory, European University
Institute, Jean Monnet Chair Paper RSC, No. 98/50, Florence: European
University Institute
Mayntz, R. (1999), Funktionelle Teilsysteme in der Theorie sozialer
Dierenzierung, in R. Mayntz (ed.), Soziale Dynamik und politische Steuerung:
Theoretische und methodologische berlegungen, Frankfurt am Main/New York:
Campus, pp. 3869.
Nowotny, H., P. Scott and M. Gibbons (2001), Re-thinking Science: Knowledge and
the Public in an Age of Uncertainty, Cambridge: Polity Press.
Pierson, P. (2000), Increasing returns, path dependence, and the study of politics,
American Political Science Review, 94 (2), 25167.
Rhodes, R.A.W. (1996), The new governance: governing without government,
Political Studies Association, 1996 (XLIV), 65267.
Rip, A. (1998), The dancer and the dance: steering in/of science and technology,
in A. Rip (ed.), Steering and Eectiveness in a Developing Knowledge Society,
Utrecht: Uitgeverij Lemma BV, pp. 2750.
Rip, A. and R. Kemp (1998), Technological Change, in S. Rayner and E.L.
Malone (eds), Human Choice and Climate Change, Columbus, OH: Batelle Press,
pp. 32799.
Schimank, U. (1988), Gesellschaftliche teilsysteme als akteurktionen, Klner
Zeitschrift fr Soziologie und Sozialpsychologie, 40 (3), 61939.
Schimank, U. (1996), Theorien Gesellschaftlicher Dierenzierung, Opladen:
LeskeBudrich.
Schneider, V. and P. Kenis (1996), Verteilte kontrolle: institutionelle steuerung in
modernen gesellschaften, in P. Kenis and V. Schneider (eds) Organisation und
Netzwerk, FrankFurtam Main and New York: Campus, pp. 743.
Stacey, R.D. (1996), Management and Science of Complexity, Research-
Technology Management, 39 (3), 810.
Thompson Klein, J., W. Grossenbacher-Mansuy, R. Hberli, A. Bill, R.W. Scholz
28 Introduction
INTRODUCTION
31
32 Reflections on reflexive governance
For purely pragmatic and methodological reasons that is, without any
intent to hypostatize the terms in question into categories of ontological
status and signicance I make a distinction in this essay between a rst
and a second modernity. The present context does not permit a full descrip-
tion of the premises from which I proceed in making this distinction, let
alone all the elds in which and topics to which this distinction nds appli-
cation. In fact, the second modernity can be traced from the sphere of
science and technology, through that of social inequality, lifestyles and
forms of biography, the organization of work, forms of the interrelation of
capital and enterprise, up to the state and politics in both national and
transnational contexts (Beck and Bon 2001; Beck and Lau 2004). Here,
I shall consider only some elements of the theory which are important for
the concept of reexive governance.
Speaking very generally, the theory of a second modernity can be divided
into three related complexes: the theorem of the global risk society, the
theorem of forced individualization and the theorem of pluridimensional
globalization. All three theorems represent developments of the same line
of argument such that each interprets and reinforces, and is interpreted and
reinforced by, the others. Global risk society, individualization and
globalization (or, alternatively, cosmopolitization) are conceptionalized
as radicalized forms of a dynamic of modernization. As the twentieth
century gives way to the twenty-rst century, this dynamic, which is turning
back reexively in the sense of a kind of self-confrontation upon itself,
is now dissolving the familiar formulae of simple modernity.
The rst the classical or high modernity, which is specically asso-
ciated with industrial society, was characterized by a logic of organization
and action, which is only now, at a time when its practical relevance is
diminishing, becoming clearly recognizable as such. This logic involved
establishing of extremely ne divisions between categories of people, things
and activities and making distinctions between spheres of action and forms
of life such as facilitating an unambiguous, institutional ascription of com-
petence, responsibility and jurisdiction. Today, the limitations of this logic
of ne division and unambiguousness are becoming ever more evident. It
is proving increasingly dicult to justify and substantiate, and in some
areas it has completely collapsed. This logic of unambiguousness one
Politics in the global risk society 33
acquires its own independent dynamic. One might almost say that the social
congurations characteristic of the global risk society arise as a conse-
quence of the actual thought and action of human beings and institutions
coming to prevail over the assumed manifestations of industrial society
(the consensus regarding progress; the abstraction from ecological con-
sequences and perils; the optimism with respect to the limits of control and
supervision). The global risk society is no option which might have, or
might still be, chosen or rejected in a process of political debate. This soci-
etal form is rather one which arises due to the autonomous dynamic of
processes of modernization which have acquired an impetus of their own
and which are quite blind to consequences and quite deaf to warnings of
danger. These processes, indeed, tend, taken as a whole, to latently engen-
der various self-imperilments which go to delete, to transform and to politi-
cize the foundations of that rst modernity associated with industrial
society.
reective (in the narrower sense of the word). That is to say, society itself
becomes, for itself, a theme and a problem. What, at bottom, tends to
disturb here might be called the return of uncertainty into society. By this
is meant, in the rst instance, that ever more social conicts are treated
not, as hitherto, as problems of social ordering but rather as problems of
social risk. The dening quality of such risk-problems is that they are not
susceptible of univocal solutions but are rather distinguished by an essen-
tial ambivalence which may indeed be, in most cases, theoretically grasped
by recourse to calculations of probability but cannot, by such calculations,
be eliminated. It is in this inherent ambivalence that there consists the
dierence between risk-problems and problems of social ordering, the
latter being per denitionem oriented toward unambiguousness and decid-
ability. So that the return of uncertainty into society implies in the second
instance also an almost inevitable dwindling, in the face of the multiplic-
ation of ambiguities or of problems susceptible of being handled only in
terms of the logic of this and that both, of faith in the project of the tech-
nical construction and ordering of the social.
Risks not only presuppose decisions but also promote them in indivi-
dual cases as on the level of general principle. Questions of risk cannot
be translated into terms which will render them questions of social order-
ing, because these latter questions tend as it were to choke on the plural-
ism immanent in questions of risk and to metamorphose ultimately,
unavowedly and despite whatever pseudo-objective statistical faades
might be erected in the attempt to hide the fact, into questions of power
and questions of subjective moral decision. In other words, questions of
risk compel or, put more cautiously, incite to the recognition of
ambivalence (Zygmunt Bauman).
more to treat the whole world as its laboratory, thus spreading risks across
the globe.
This point about the collapse of boundaries and the new mobility of risks
has fundamental implications for the social sciences inasmuch as it helps
us to the insight that sociology (as well as historiography, political science,
jurisprudence, and so on) remain bogged down in a methodological
nationalism. What is this methodological nationalism, and why is it a
problem?
Methodological nationalism rests on the assumption that modern
society and modern politics are synonymous with a society and a politics
organized in terms of nation states. The state is understood as creator,
supervisor and guarantor of society. Societies (the number of which is
equal to the number of nation states) are understood as mere containers,
arising and subsisting in a space dened by the power of the state. This way
of thinking, which equates societies with national societies and conceives
of the former as by their nature territorially limited, is deeply ingrained in
all that the social sciences understand by concepts and perspectives, by
gathering and comparing data and conducting empirical research. It is, one
may say, constitutive of the sociological imagination (Beck 2004).
We need, indeed, to distinguish clearly between two aspects of this
problem. Where such dogmas are manifest in the words and attitudes
of individual social actors, I speak of a national point of view; where
they prove to be implicit in the third-person-language and third-person-
perspective of the social sciences, I speak of methodological nationalism.
This distinction between the perspective of subjects acting in society and
the perspective of the social scientist is an important one because there is
no logical connection between the two. The correspondence here is to be
traced back merely to the similarity in the conditions of their historical
emergence. The rise of the social sciences coincided with the rise of the
nation-state, of the system of international politics and of nationalism. It
is this specic historical involvement which alone accounts for the method-
ologically nationalist axioms upon which the social sciences proceed and
which dictate that nation, state and society constitute the natural social
and political forms of the modern world.
At some point in the not too distant past (say, around the 1960s) there
occurred a qualitative transformation in the perception of social order. This
order ceased, at this time, to be understood primarily as a conict centered
on the production and distribution of goods; from then on, the primary
issue became the contradiction between the production and distribution of
40 Reflections on reflexive governance
bads and the claims to control raised by the established institutions of the
nation states. This shift in the categories of self-perception threw modern
societies manner of organizing their own institutions and functions into a
crisis of global interdependence which has expressed itself in phenomena
of widely divergent political value and signicance climate change, global
poverty, transnational terrorism, the BSE crisis, AIDS, and so on. This
crisis of global interdependence I call the global risk society. Also thrown
into crisis by this shift are that social science and that political theory which,
combining Karl Marx with Max Weber, attempt to understand modern
society in terms of the categories capitalism and rationalization. It is this
unleashing of perils and uncertainties engendered by the civilizing process
itself and characterized above all by a global interdependence, as well as the
dominance of a public, mass-media-dominated perception of risk, which
constitutes a transition from one epoch into another.
What distinguishes new risks from old risks and dangers and why must the
dynamics of conict typical of the global risk society be understood as a
second, reexive phase of radicalized modernization? These questions I
intend to answer by means of ten theorems or characterizing clarications:
lie by their nature beyond all possible control and things which
happen in fact presently not to be under control. Dangers in the pre-
modern sense these are not, because they rest on decisions and there-
fore raise questions regarding the attribution of responsibility and the
just distribution of blame and of costs. Political counter-measures are
in every case seen to be imperative. Neither national nor international
political authorities and multi-national companies are often today
in the same case can point to the fatal uncontrollability of the
modern world to absolve themselves of the obligation to take action;
rather, they are placed by the discourse of risk conducted throughout
the global public sphere under an absolutely irrecusable pressure to
justify themselves by action. They are damned to counter-measures.
This intense expectation of counter-action alone suces to lend life
to the contrafactual hypothesis of controllability, even when all avail-
able models of response prove inadequate. Not to take action in the
face of recognized risks is politically out of the question regardless
of whether the measures taken do in fact minimize the risk, increase
it or have no eect at all.
8. The politics of risk construction and the politics of risk minimization:
here we must distinguish between two forms of risk politics, namely
(A) the politics of risk construction (the social construction of risks)
and (B) the politics of risk minimization (provision, prevention, min-
imization and appropriate cross-border intervention). (B) presup-
poses (A), while (A) helps to dene (B)s sphere of action. Both
represent forms of the denition of reality: (A) on the cognitive level,
(B) on the level of an actual moulding and remoulding of this reality
by political action; (A) is a dening characteristic of sociological
constructivism, (B) rather one of politicalscientic realism.
9. Consequences for the social sciences: the global risk society poses new
challenges not only to the politician and the man in the street but also
to the practitioner of the social sciences. Not only must these sciences
recognize that social interaction is no longer so clearly spatially and
temporally dened as it was assumed to be in the old nation-state par-
adigm (critique of methodological nationalism). Global risk society
also introduces them to a model of global socialization which is at
odds with the traditional picture of positive social integration on the
basis of shared norms and values, resting as it does upon the conict
around negative values (risks, crises, dangers of annihilation). Here,
the issue is not so much the multiplication of uncontrollable risks of
a sort apt to generate a global interdependence as the freeing of these
risks from all formerly decisive boundaries and this at once on the
spatial, temporal and social plane.
44 Reflections on reflexive governance
Conicts over civilization risks in the sense just described emerge, for
example, where opinions diverge as to how far the industrialized coun-
tries have a right to demand that developing countries protect such
important global resources as the rainforests, given that the former
countries arrogate to themselves the lions share of energy resources.
A certain reasoning sees, indeed, in precisely such dierences of
opinion a reason not to speak here of a form of global socialization.
But to reason so is to make the mistake of equating society with con-
sensus. In fact, such conicts themselves already have an integrative
function, inasmuch as they make it clear that the solutions found will
have to be cosmopolitan ones. Such solutions are hardly imaginable,
however, except on the basis of new global institutions and parame-
ters and thereby also of a closing, to some degree, of the gap between
industrial and developing worlds. Just that, indeed, is a key distin-
guishing feature of the idea of reexive governance. The long-term
consequences transgressive by their nature of all borders and bound-
aries of that constitutively unexpected to which, nonetheless, there
can be no possible response but the development of uncertain expec-
tations, can provide the spark and the fundament for transnational risk
communities Folgen-entlichkeiten, or public spheres emerging
from and sustained by the necessity to deal with commonly suered
consequences which might in their turn lead to an (indeed involun-
tary) politicization and thus, should the circumstances be right, to a
reinvention of politics on a transnational or global level.
The development referred to here that whereby the space of
everyday experience becomes a space of cosmopolitan interdepen-
dence is not to be conceived of as a kind of universal love-in. The
condition evoked is one arising and subsisting in and through the per-
ceived emergency of situations of global imperilment. The risks
which make up this imperilment give rise to a pressure for coopera-
tion which cannot be ignored. With the social construction and
general social acceptance of the idea of a cosmopolitan dimension of
imperilment, there is created, in disregard of national borders and
their attendant enmities, a space of common responsibility and
common action which, analogously to the space formed by the terri-
tory of a nation, can (although it need not) provide an impulse for
subjects without prior bond with one another to engage in common
political action. This potential is realized where the socially acknowl-
edged dimension of imperilment leads to cosmopolitan norms and
agreements in other words, to an institutionalized cosmopolitanism.
Research done to date, however, into the emergence of the supra-
and trans-national forms of organization which would correspond to
46 Reflections on reflexive governance
This return of politics after the EastWest conict and after the old certainties
of the industrial epoch compels and justies a further distinction, which runs
traversely to that above, that is, the distinction between rule-directed and rule-
altering politics. The former type can certainly be creative and nonconformist,
but it operates within the rule system of industrial and welfare state society in
the nation state (or, in our terms simple modernity). Rule-altering politics, on
the other hand, aims at a politics of politics in the sense of altering the rules
of the game themselves . . .
Even inside simple politics, the bridge game, there are a number of individual
variants of a more or less sophisticated type which one can play with various
degrees of skill and mixed success. A completely dierent situation arises,
however, if the rules of the game themselves are altered or switched. The height
of confusion is attained when one plays both at once, bridge and the game of
switching its rules. People play with swapped rule systems in order to change the
rule systems themselves. Some continue to play bridge and are outraged as
others attempt to invent and implement new displaced rules of the game during
the bridge game. We face precisely this kind of normality and absurdity every-
where today.
The game of classical industrial society, the antagonisms of labour and
capital, of left and right, the conicting interests of the groups and the political
parties, continues. At the same time, many demand and actually begin to turn
the rule system itself inside out . . . Rule-directed and rule-altering politics
overlap, mingle and interfere with one another . . . The distinction between
ocial politics and sub-politics, which is oriented to the systemic structure of
society, must therefore be contrasted with the distinction between simple
Politics in the global risk society 49
are not. And it is this which makes the counter-force represented by the
consumer a counter-force which has as yet barely begun to exert itself
so dangerous for the force of capital.
In the last analysis, however, there can be no getting around the problem of
how to redene politics specically qua science and practice centred on the
instance of the state. No question, indeed, but that the advocates and
agents of global civil society play an indispensable role in the global play of
force and counter-force, and in particular in the process of the establish-
ment of global norms and values, or, in other words, of reexive gover-
nance. We should never, however, so far over-estimate this role as to think
that the contradictions, crises and indirect consequences of our second
modernity could ever possibly be entirely, and on a global scale, civilized
away by the forces embodying this merely civil-societal form of political
commitment. Whoever thinks in this way revives and repeats the old, old
error of the man who would prefer to steer clear of politics.
In the face of this error, it must be insisted upon that the globalpolitical
meta powergame cannot be restructured as a system of reexive gover-
nance cannot, that is to say, be turned from a loselose into a winwin
situation except through the transformation of politics qua politics of the
state (along with, concomitantly, state theory and political theory). The key
question, then, is: how can and must the concept and the form of organi-
zation of the state be opened up and reconstructed in the face of the
challenges posed by the global risk society? How can a cosmopolitan self-
transformation of the state be made possible?
If we can safely dismiss, as a discredited shibboleth dear to those who
think themselves above politics, the idea that the existence of global civil
society renders the renewal of state-oriented politics superuous, the same
does not hold of the new and as yet untested idea that civil society might
itself as it were seize power. Indeed, such a symbiosis of civil society and
inter-state cooperation constitutes the very nature of reexive governance.
The key question is: how can the ideas, theories and traditions of the state
be freed from their national limitations and opened up to the challenges
posed by a cosmopolitan interdependence? In order, in this connection, to
avoid altogether entering into a pointless discussion of the falsely-stated
alternative (state-centered politics or a politics of civil society), it is neces-
sary clearly to distinguish between centeredness on the state and centred-
ness on the nation state. One proceeds rightly, indeed, in abandoning
political theorys erstwhile xation on the nation state, since this latter is
today no longer the actor in international politics but only one actor among
52 Reflections on reflexive governance
many; anybody who extends, however, his critique of the nation state
xated point of view so far as to exclude, both on the analytical and the
political plane, the very possibility of a continued eective activity and a
self-transformation of the system of state relations, is throwing out the
baby with the bath-water. Reexive governance means, then, that states too
must be understood and investigated as entities contingent and politically
transformable or, where they are not already so, must be made so. The
question thus posed is: how is the transnationalization of states to become
possible?
There is such a thing as the law of the decline of the power of the nation
state: he who plays the national card in the global meta-power-game must
lose. What is required, then, is a reversal of this perspective. That is to say,
the principle also holds that the counter-force represented by states
becomes really active and eective where these latter undergo a process of
transnationalization and cosmopolitization. Only if the states succeed in
keeping step with mobile capital and in redening and reorganizing their
power-positions and their moves in the game will it prove possible to put a
halt, internationally, to the process of the collapse of state power and
authority, and even to reverse this process.
We thus see posed a whole series of questions answers to which would
amount to a veritable reinvention of politics: who is it, in fact, who has the
authority and the right to take decisions and to create institutions such as
will be able to manage the currents of international nance? What sort of
consensus is necessary, and who must be involved, in order that an appro-
priate response emerges to the worldwide climate catastrophe? Can the
decision to struggle against AIDS (or alternatively to neglect or forego such
a struggle) really be a decision legitimately taken in the private space of an
alleged non-attributability since it is one on which turn the lives of millions
of human beings? And what sort of political agents or political institutions
might ideally be proposed as appropriate to handle this problem, on what
level and with what sort of mandate? In what manner are global, transna-
tional, national and local authorizations of decisions related to one
another and at the same time distinct and separate from one another? Who
is it in fact who imposes, over the heads of the nation states, norms and
rulings which are nonetheless binding on said nation states, and which
norms and rulings do they impose, and what legitimates their doing so?
The insight is gaining ground that new, global institutions are required
in order to address such themes as global environmental damage, weapons
control, world nancial order, currents of migration, poverty and justice
and respect for human rights; no such institution, however, nor any
measure taken by such an institution, should be permitted to overlook the
fact that these global problems have also a regional that is to say a
Politics in the global risk society 53
national and local side to them. Who, then, should control these agencies?
To whom should they be responsible? Only to the national states? To which
parliaments, which public sphere? To the United Nations? To the NGOs?
If we are to escape, in our thought and our action, the trap of national-
ity, the drawing of a distinction between sovereignty and autonomy is
essential. (Methodological) nationalism rests upon the equation of the
former with the latter. Seen from this perspective, such things as economic
dependence, cultural diversication and military, legal and technical coop-
eration between states automatically lead also to a loss of sovereignty, inas-
much as they imply a loss of autonomy. However where one takes as ones
criterion of sovereignty the power to structure social existence politically
or, in other words, the success of a state in increasing the prosperity of its
population and in bringing closer to a solution such urgent national prob-
lems as unemployment, protection of the environment, crime and social
and military security then that loss of autonomy implied by increasing
interdependence and cooperation is seen to result, in fact, in a substantial
gain in sovereignty. The capacity of governments actually to guide and steer
political developments becomes ever greater, the closer the cooperation and
integration of states with one another and the greater the political advan-
tage accruing from this closeness. In short, then: the division of sovereignty
and its reconcentration into a new nexus situated on some other plane than
that of the individual state proves not, in fact, to be tantamount to a reduc-
tion in the sovereignty of this latter but rather to its expansion.
NOTE
1. The problem described here is by no means restricted to the theory of a second moder-
nity. Similar diagnoses might be formulated on the basis of Luhmanns systems theory
and of various theories of action. These theories too work with models of reexive
processes which are not self-stabilizing but lead rather to contradictions, ruptures or
dialectical suicides (Beck and Holzer 2004) of entire systems.
REFERENCES
Bauman, Zygmunt (1995), Moderne und Ambivalenz: das Ende der Eindentigkeit,
Hamburg: Junis.
Beck, Ulrich (1993), Die Erndung des Politischen, Frankfurt a.M.: Suhrkamp.
Beck, Ulrich (1996), Wissen oder NichtWissen? Zwei Perspektiven, reexiver
Modernisierung, in Ulrich Beck, Anthony Giddens and Scott Lash (1996),
Reexive Modernisierung, Eine Kontroverse, Frankfurt a.M.: Suhrkamp,
pp. 289315. [World Risk Society, Cambridge: Polity Press]
56 Reflections on reflexive governance
Beck, Ulrich (2002), Macht und Gegenmacht im globalen Zeitalter, Frankfurt a.M.:
Suhrkamp. [Power in the Global Age, Cambridge: Polity Press, 2005]
Beck, Ulrich (2004), Der kosmopolitische Blick, Frankfurt a.M.: Suhrkamp. [The
Cosmopolitan Vision, Cambridge: Polity Press, 2005]
Beck, Ulrich and Wolfgang Bon (eds) (2001), Die Modernisierung der Moderne,
Frankfurt a.M.: Suhrkamp.
Beck, Ulrich, Wolfgang Bon and Christoph Lau (eds) (2004), Entgrenzung und
Entscheidung, Frankfurt a.M.: Suhrkamp. [Second modernity as a research
agenda: theoretical and empirical explorations in the meta-change of modern
society, British Journal of Sociology, 2005]
Beck, Ulrich and Edgar Grande (2004), Kosmopolitisches Europa, Frankfurt a.M.:
Suhrkamp. [Cosmopolitan Europe, Cambridge: Polity Press, forthcoming]
Beck, Ulrich and Boris Holzer (2004), Reexivitt und Reexion: Ulrich Beck and
Christoph Lau, Entgrenzung und Entscheidung, Frankfurt a.M.: Suhrkamp,
pp. 16593.
Beck, Ulrich and Christoph Lau (2004), Entgrenzung und Entscheidung, Frankfurt
a.M.: Suhrkamp.
Daase, Christopher (2002), Internationale Politik, in Christopher Daase, Susanne
Freske and Ingo Peters (eds) Internationale Risikopolitik, Baden-Baden: Nomos,
pp. 936.
Douglas, Mary and Aron Wildavsky (1983), Risk and Culture, Berkeley, CA:
University Press.
Giddens, Anthony (1955), Beyond Left and Right: The Future of Radical Politics,
Cambridge: Polity Press.
Wehling, Peter (2001), Jenseits des Wissens?, in Zeitschrift fr Soziologie 30,
pp. 46584.
3. Reexive modernisation
as a governance issue, or: designing
and shaping re-structuration
John Grin
INTRODUCTION
In this chapter, I will assume (cf. Grin, 2005) that the concept of reexive
modernisation, which I will more accurately circumscribe below, provides
a sensible orientation for socio-technological development in a variety of
domains as they nd themselves in late modernity. My purpose is to present
a view on reexive modernisation as a governance issue. What does it mean
to consider reexive modernisation as a governance issue? A quick and basic
answer is easy to give. Whatever denition of governance we take, it centres
on the idea of what we may loosely indicate as shaping the market and
society [and science J.G.] into a desired form. (Pierre and Peters, 2000: 1)
This basic denition implies that the two basic questions on reexive mod-
ernisation as a governance issue are: (1) how to determine the reexively
modern shape that society and the market should take? and (2) how to
eectively shape society and the market? And, indeed, it is these two ques-
tions on which this chapter will focus.
Yet, in order to deal with these questions in any fruitful way, we have to
gain a deeper understanding of what it means to consider reexive mod-
ernisation as a governance issue. As we will see, such understanding implies
a re-formulation of these questions in more specic and subtle terms. Why
is this necessary? One point here is that governance and reexive modern-
isation are referring both to concepts (and elaborations of these concepts
into theory) about phenomena, and to these phenomena themselves. A
second point is that both, as concepts, explicitly consider the institutions of
state, market, science and society and their relations (and the ways in which
they are conceived) not as givens, but as objects of more or less considerable
scrutiny and change. Thus the expression reexive modernisation as a
governance issue is essentially non-trivial.
Therefore, in the next section I will discuss the concept of governance in
57
58 Reflections on reflexive governance
Governance must be understood here in a more precise sense than the cur-
rently fashionable, casual use of the term for designating any kind of social
regulation. We will deal with the concept as one that refers to new modes of
governing, of shaping the market and society into a desired direction, that
dier from classicalmodernist government. A range of factors are men-
tioned in literature as the rationale for the shift from government to govern-
ance (for example, Rhodes, 1997; Pierre and Peters, 2000; Kooiman, 2003).
While authors dier concerning the meaning of and the mutual relations
between these factors, it is probably fair to say that most lists include: the
changing (views on) the capacity of states to shape society and the market;
the deterioration of the states self-evident authority vis--vis increasingly
fragmented societies; the relative autonomy of economic and societal devel-
opments that challenge that nation state and its relations with the market and
society; and globalisation processes that imply another challenge to a nation
states sovereignty. The quintessence of the dierence between government
and governance is, in line with this rationale, often described by saying that
practices of governance (that is to say, of shaping the market and society)
can neither be located exclusively in nor be solely directed by
politicaladministrative institutions (as the government concept presumes)
but essentially involves interactions between state, market and society.
This depiction is problematic not in the way it portrays governance, but in
the way in which it considers classicalmodernist government. Underlying
Reflexive modernisation as a governance issue 59
that conception is a presumption that goes back to the received view that,
in the course of modernity, the state, market, society and science have
become separate realms. To be sure, it is not dicult to see that functional
dierentiation between these institutional realms has indeed occurred. Yet,
if we take this received view as an adequate description of modernity, we
have to paraphrase Latour (1993) never been modern. The received view
is too simple in one crucial respect: it neglects the fact that the development
of these institutional realms in ways that vary between nation states and
societal domains can best be described as a process of co-evolution, guided
by a common orientation. That orientation may be summarised in two
dogmas, which belong to the core of the Enlightenment project. The rst one
is that it is possible to know Truth on the basis of universal knowledge,
grounded in some Archimedal point. The second dogma is that it is possible
to control reality on that basis, and that this will yield social progress by
freeing humans from fate.
These two dogmas received widespread adherence in the course of the
sixteenth and seventeenth centuries (Kumar, 1995: 7880), as a response to
what Richard Bernstein (1983) has called the Cartesian Anxiety: the quest
for certain, universal, grounded knowledge to deal with the widely felt
threat of chaos. While the notion of progress through control was, in the
path of the work of Galileo, Newton and other natural scientists, initially
primarily related to nature, it soon evolved into a worldview that also
included the idea of a controllable social reality. From the early eighteenth
century onwards, this view has guided the development of nation states.2
Views on the cosmos and on the polity melted into a Cosmopolis (Toulmin,
1990), and rational knowledge became the basis on which nation states
relied to ensure the legitimacy of their actions and the success of their
attempts to promote social progress for the people within their territories.
In addition, especially since the Industrial Revolution, it was this vision that
provided orientation to a process of ne-tuning between nation states, soci-
eties, the market and science. (Kumar, 1995: 812; Gill, 2003: 11548) As a
consequence, these dierent realms have co-evolved as parts of the same
project. In modern nation states, domains like healthcare, agriculture,
water management, transportation and so on have grown into systems of
ecient and carefully interrelated institutions that are tailored to nurture
knowledge-driven development that was supposed to and in many respects
actually did yield socialeconomic progress.
The concept of reexive modernization theory starts from the recogni-
tion that these simple modernization processes of rst modernity (Beck,
1997) have now been recognized to bring with them risks and side eects
that society does not tolerate, or no longer tolerates. This is because modern
institutions, which have emerged around this development towards progress
60 Reflections on reflexive governance
through control, have developed blind spots for risks and side eects and
lack capacities for designing and pursuing strategies which are able to deal
with them eectively. This is partly due to the fact that modernisation
processes have brought about a logic of homogenisation and virtual elim-
ination of local knowledge or, with an Aristotelean notion, of metis - the
craft to take contextual conditions into account (Scott, 1998: 302; 30941).
This disdain for and the associated undermining of metis - has reduced
societys capacity to compensate for these blind spots.
The idea behind the concept of reexive modernisation is the idea (Beck,
1992) that the very processes that were designed to and did yield progress,
also produced side eects and risks. The same institutions that nurtured
progress have developed blind spots for the negative eects that come with it.
A re-orientation of modernisation is therefore necessary towards the vision
of a radicalised modernity or second modernity (Beck, 1997): the vision
that we may and must eventually realise the demand of the Enlightenment.
The normative dimension of this concept of reexive modernisation is espe-
cially clear from the way in which Beck contrasts it with the arbitrariness
which he nds with some post-modern thinkers, and which he rejects
because, in his view, it amounts to throwing away the child (the demand of
Modernity: using rationality to improve the human condition) with the
bathwater (side-eects and risks) (ibid., p. 14). Such a re-orientation, he
further assumes, is possible and is occurring (see also Beck et al., 2003). It
presumes and brings with it3 a process of institutional transformation to
remedy the limited sight of existing institutions.
It is precisely in this respect that the concept of reexive modernisation
qualies the concept of governance. There is a tendency in literature, and
even more so in practice, to focus governance on the transformation of
the ways in which government and societal and market actors are dealing
with each other. Governance then gets elaborated through such notions
as network management (Kickert et al., 1997), public participation
(Newman, 2001) or promoting self-organisation (Rhodes, 1997). To be
sure, each of these (and other) notions provides a range of opportunities to
organise governance beyond the modernist control fashion. This is,
however, not a sucient condition for using these concepts as ingredients
for a governance concept for reexive modernisation. The reason is that,
without further measures, practices of network management, public par-
ticipation, self-organisation and so on will reect many of the assumptions
of rst modernity, and reproduce the patterns of action typical for it.
Governance practices based on these concepts may help to promote
reexive modernisation as a phenomenon if, and only if, these concepts
and practices are designed from the perspective of reexive modernisa-
tion as a concept, a Vision of bringing about a radicalised modernity
Reflexive modernisation as a governance issue 61
through not only other modes of action, but also a profound institutional
transformation.
This has two important implications. The rst is that these processes of
transformation from a particular substantial perspective (dealing with risks
and side eects in relation to progress) need in concreto be underpinned
by a substantive understanding. More precisely, they presuppose an under-
standing, for that concrete context, of the substance of the specic transfor-
mation of societal development sought, of the associated institutional
transformation and of the precise relations between them. Given that the
substance and praxis of policy analysis appears to be inuenced by its insti-
tutional embedment (Hoppe and Grin, 2000), this presents a challenge which
policy analysts normally do not have to meet, and therefore requires the
development of rather new types of policy analytical practice. We will return
to this subject in the nal section of this chapter on the basis of an example.
The second implication is that the locus of governance, the polity, cannot
be found in any of the current institutions, or in any well-dened location in
between them. Over the past decades, a rather complex institutional land-
scape has developed, which combines important modernist institutions with
a variety of institutional arrangements that have emerged more recently
(Van Tatenhove et al., 2000). Governance is thus facing what Hajer (2003)
has called an institutional void, a lack of xed and generically usable insti-
tutional arrangements in which it can eectively take place. As a conse-
quence, the polity itself has become discursive. Where we might look for
contextually suitable polities is a question we can only deal with after we have
taken up a question that needs be answered rst: (why) is it meaningful to
consider reexive modernisation as a governance issue?
practices with which they might ally to counter that threat and produce
reexive modernisation. A better understanding beyond the mere recogni-
tion of potential resistance is needed of how these practices may reach out
to each other in such a way that they may start to reinforce each other and
together contribute to a process of increasingly comprehensive transforma-
tion, both substantively and institutionally.
The second observation to be made here relates to that point. Unless the
sort of understanding just mentioned is developed, a focus on practices
that create signicant institutional change may leave us vulnerable to the
classical mistake of structuralist approaches (to which, to be sure, they do
not subscribe): a focus on structural transformation as determinant of trans-
formation of action, without considering ways in which the existing
structure may ght back. Latours attention to the role of collective inter-
pretation and his related lever metaphor are interesting and highly relevant
ingredients for an approach that does more justice to the duality of structure
(Giddens, 1984). It oers opportunities for elaborating the general prescrip-
tion that Fox and Miller (1996: 91) derive from structuration theory: to
transform normally recursive practices through discursive will formation.
Discursive will formation, though far from trivial, may help both to act
beyond existing structures and to transform these structures. This possibil-
ity is implied by structuration theory as well as all other recent social
theory. As a range of authors from very dierent backgrounds including
Law (1992), Joas (1993), Leydesdor (2001) and Archer (2003) have
observed, these theories share the idea that structures operate through an
acting agent, who in principle may exhibit reexivity as Law nicely puts it,
structure is a verb.
However, in order to provide guidance to such moments of discursive
will formation, Latours proposal for attributing to reexive modernisation
primarily a criticalreective role needs to be complemented with an elab-
oration of how this might help to transform practices and identify and
create new opportunities for these practices to connect with each other,
and contribute to a more comprehensive process of what I like to call
re-structuration: the interrelated transformation of structure and action
through structuration processes guided by the deliberate re-orientation
which Latour holds to be re-modernisations primary meaning.
These two observations imply that we may consider reexive modernisa-
tion not only as a concept but also as a phenomenon, provided that this
concept gets dened in a contextually meaningful way for particular prac-
tices; provided that it gets a chance to provide direction to these practices;
and provided that these practices are made to connect with each other in
some more or less orchestrated way, just as occurred throughout simple
modernisation processes, as illustrated below in the case of agricultural
64 Reflections on reflexive governance
product subsidies introduced early after the war) and freeing the labour force
for industry.
This change initiated a very rapid pace of rationalization (Bieleman,
2000). The primary sectors share in the labour force decreased from 19 per
cent in 1947 to 5 per cent in 1990; land use for the primary sector diminished
to some 30 per cent; and the amount of capital goods (machines; cattle;
buildings) increased by 80 per cent. A main focus was on exporting animal
produce. Domestic production of food in the Netherlands increased from
typically 1520 per cent of the domestic demand in 1945, up to typically
200300 per cent half a century later. Simultaneously, the high quality of
Dutch food specialties on the one hand, and the competitive prices of inten-
sively produced bulk goods on the other hand signicantly improved the
economic potential of the sector. Progress was based on increasing control
over nature: new generations of cows were designed to yield 1.5 per cent
more milk every year; their udders were shaped to be milked by machines;
grain was designed to be harvested more easily by giving it uniform length
and shape; diseases were controlled through vaccination of animals and the
use of pesticides in crop production.
This successful simple modernisation has been signicantly facilitated
by its embedment in smoothly functioning institutional provisions. Of par-
ticular importance were two institutional triangles. The first is the so-called
OVO triad (OVO being the Dutch acronym for research, information and
education), which generated knowledge and technology through innovative
agricultural research and disseminated it to agricultural practice through
information services to practising farmers as well as through agricultural
schools where new generations were educated. The dominant, mutually
recognised task division of farmers and researchers remained such that
researchers generated knowledge and technology, which farmers were sup-
posed to apply. By basing their work on experience with farming (many agri-
cultural researchers come from farming families) as well as imposing an
understanding of farming as a knowledge-intensive, rationalised enterprise,
the utilisation of knowledge and technology, guided by the activities of the
OVO triad, occurred relatively smoothly (Bieleman, 2000; Van der Ploeg,
1999).
The second triangle was the iron triangle of the Agricultural Ministry,
agricultural branch organisations and agricultural specialists in parliament
(Bekke et al., 1994; Wisserhof, 2000). It was based on a strong consensus
partly maintained through personal commitments and unions of dierent
kinds on the policy objectives that were already mentioned. The consen-
sus also pertained to the strategy for realising them: stimulating further
knowledge and technology development, financial measures, land redistri-
bution to enable concentration and specialisation, improvements in water
66 Reflections on reflexive governance
production, designed around animal welfare as the central concern, run into
existing food safety regulations, which, in addition to product criteria, also
include demands on the production process, which is tailored to existing
intensive agriculture. Government ocials frequently find it dicult to be
responsive to such problems, partly because national regulation is embed-
ded in European legislation, partly because exceptions are dicult to realise
within the existing legal framework, with principles like equality before the
law, firmly anchored in the Rechtsstaat.
Let us now turn to the first basic question of governance: how to determine
the reexively modern shape that society and the market should take?
Taking together the conclusions reached in sections 2 and 3, we need an
answer to the question of where political judgement must take place: in the
classical bodies of the nation state; in the variety of practices now charac-
terising many policy domains (as we have seen in the example of Dutch
agricultural policy), each of which may make claims to elaborate reexive
modernisation meaningfully; or in between these loci?
While the nation state can still claim democratic legitimacy to a larger
extent than all kinds of more or less ad hoc institutional arrangements, it is
a crucial point of departure of governance literature that it simultaneously
faces tough problems in defining what constitutes meaningful elaborations
for societal and economic actors. How does one define a direction for soci-
etal development that may expect suciently wide support if the authority
of expertise is no longer self-evident, if globalisation and individualisation
challenge existing delineations and identities and if societies are thus
increasingly normatively fragmented? And how can government judge
whether this direction will actually be realised in practices in which at least
some sub-set of society participates? The simple answer that such judgement
be left to these practices is too simple, not only because of the legitimacy
aspect, but also because, as we saw at the end of section 3, such practices
cannot be left to themselves.
In answering these questions, one fruitful point of departure is the recog-
nition that reexive modernisation should not be seen as an unambiguous
objective, but rather as an essentially contestable or open-textured concept
that may be elaborated into societal practices that represent a variety of
interpretations of the concept. Consider, for instance, the ambition of a
transition to a sustainable agriculture. If understood in the more radical
sense indicated above, sustainable development can be seen as a form of
Reflexive modernisation as a governance issue 69
what practices may and which ones may not be seen as valid elaborations
of the general ambition of reexive modernisation? What rules could guide
the relations between parliaments, such in-between fora and social prac-
tices? What are adequate heuristics for arriving at a number of elaborations
that are large enough to oer something to a significant part of normatively
fragmented societies and small enough to prevent the need for an imprac-
tical variety of R&D programmes, hallmarks with which to communicate
them to consumers, regulations to ensure food safety, machinery for the
associated agricultural practices and so on?
Interesting and promising answers to these questions may be (and are
being) given and tried in practice. Examples of the diversity of practices
can be found in the dierent programmes in which reexive modernisa-
tion of Dutch agriculture is now being waged. For the first time ever on
17 February 2004, the Agricultural Commission of the Dutch Parliament
met outside the parliamentary sessions in a joint session with farmers and
others from the rural areas in the north of the country. There is discussion
between market parties on the legitimacy and practical implications of the
existing variety of hallmarks for sustainably produced foodstus. These
exemplary practices raise new questions (including the question of how
they relate and might be related to each other) and thus need more atten-
tion than can be given here and more reected answers than can be given
now. Asking these questions are expressions of what it means to consider
governance from the concept of reexive modernisation; developing
and testing answers is part of considering reexive modernisation as a
governance issue.
Do not plan in order to organise but plan to alter the existing social
mechanisms, whether market or not, that govern x.
Show some modesty: focus on just a well-defined segment of life, spe-
cialised, even narrow, rather than vast, synoptic and broad. That is
[J.G.], it should be focused on specific practices.
Planning rarely succeeds through a big step; rather it should aim at
an endless succession of short and fairly rapid steps in a process of
trial-and-error learning or serial adjustment.
Fourth: There may be we do not yet know enough big dierences
between a succession of short rapid steps that is inuenced by a long
term perspective, and one that is not, the former probably being the
more successful form of planning and decision-making. (Lindblom,
1999: 478)
located at the same spots where we located, in the previous section, polit-
ical judgement: in the institutional arrangements of the political commu-
nity as a whole; in socio-economic-scientific practices; and in between.
And, second, this account of planning reinforces and presupposes the
notion of taking a plurality of practices as a point of departure a notion
that has appeal both from the perspective of the problem of homogenisa-
tion and elimination of practice, which we have seen (in section 2) to be
associated with simple modernisation, and from the perspective of the
approach to democratic decision-making proposed in the previous section.
the assumed law reects the usual practice that all players downstream
obtain some additional added value from the price increase of the primary
product. The anticipation that consumers will not be prepared to pay an
additional price neglects consumption patterns that have recently emerged
(grazing; increased reliance on prepared food rather than primary prod-
ucts Van Otterloo, 2000), which rely on products in which players in the
centre of the food chain put significantly more added value on primary prod-
ucts. Here again, institutional adaptations are needed to be able to go
beyond this multiplication lawand to somehow accept relatively small price
increases of the primary product in the significant value added by the
processing industry.
What do these examples teach us in terms of the first question above? They
indicate that the devil is in the detail, that seemingly self-evident assump-
tions and anticipations need to be put under critical scrutiny in moments of
discursive will formation. They also indicate that the diculties implied
may be dealt with through a combination of initial small steps beyond the
rules of existing institutions and longer term institutional transformations.
In the case of the first example, one short-term step might be to start an
experimental project in which results of a transdisciplinary study on novel
ways to maintain animal health in that project are tried out. Another simul-
taneous step might be the start of a more comprehensive programme
(involving both scientists and practitioners, contributing their metis) - for
developing the recursive control paradigm in veterinary sciences, as well as
its consequences for other areas of the animal sciences (housing, breeding,
feeding, and so on). Next steps might involve improvement and extension of
the experimental project on the basis of the experiences gained, as well as
on the insights gained from the programme on recursive control. Ultimately
in the longer term, new distinctions may emerge surrounding this pro-
gramme: between science and practice (Grin, 2005) and within and
between scientific disciplines. A similar strategy of step-by-step transfor-
mation may be followed to deal with diculties due to market features.
Against this background, let us now turn to the final two questions raised
in the introduction to this section. Reexive modernisation on a more
encompassing level may emerge from these practices, provided synergy is
realised between them. Such synergy may result from actions by planners
who make smart connections (Grin et al., 2003), but also from these prac-
tices reaching out to each other. It is in this way that what I called above
re-structuration may be induced. To elaborate this further into a gover-
nance concept, two additional insights may prove particularly helpful.
The first relates to the fact that recent social theory has shifted the ana-
lytical viewpoint from the chicken-or-egg question of structure and action
(does S determine A, or is it the other way around?) towards practices in
Reflexive modernisation as a governance issue 77
which they shape each other. This theorem of the duality of structure is of
course the core of Giddenss structuration theory; but it is also central to
for example Bourdieus (1977) notion of the habitus as well as to the way in
which Leydesdor (2001) has elaborated Luhmanns theory. Common to all
these approaches is the fact that the mutual shaping of structure and action
takes place through the work done by acting subjects structure, in a phrase
of John Law, has become a verb.
The implication of the preceding is the insight that governance should
focus on providing central actors in a variety of practices7 with the strategic
insight to induce a succession of steps in which structure and action inter-
act with each other towards a radical modernity. Meta-theoretical insights
from the theories mentioned may be enriched and made more operational
by empirical studies of earlier processes of regime change (Geels, 2002)
and by dierent bodies of middle-range theory, including methods of
regime analysis that are based on complex system theory (Rotmans et al.,
2000; 2001); and insights from the field of science and technology studies
concerning the interdynamics of niches, regime change and trends of
change in the wider environment (Rip and Kemp, 1998; Roep et al., 2003; as
well as Kemp and Loorbach, Chapter 5 in this volume, Rip, Chapter 4 in
this volume).
The second insight that may be of use here concerns the importance of
guiding visions. Relying on a wide range of empirical and theoretical liter-
ature from science and society studies, Dierkes et al. (1995) have argued that
successful processes of socio-technological change derive their success, to
a significant extent, from guiding visions. These visions have a dual func-
tion that is well expressed in the German term Leitbild: they serve as a
mental image of an attainable future shared by a collection of actors and
they guide the actions of and interactions between those actors. An attain-
able future here is one that oers a perspective of going beyond established
assumptions and anticipations. A vision with such characteristics can serve,
as the authors put it, the functional equivalent of institutional arrange-
ments, in those cases where existing arrangements are object of transfor-
mation and therefore cannot be taken as guidance. Developing the vision
of radical modernity into context-specific visions (in plural) may therefore
contribute to joint action for reexive modernisation.
These two insights may enhance each other: visions which portray a world
in which we can do better (Roep et al., 2003: 206) can help identify in ways
suggested by the above examples the concrete strategies needed on the
niche and regime level. Conversely, trends that have been discovered to yield
strategic opportunities for regime transformation may be integrated into
visions. Together, they may give esh and blood to Latours metaphorical
lever eect.
78 Reflections on reflexive governance
The examples also suggest that this is far from a trivial task. Development
of new approaches to policy analysis, which are tailored to support a variety
of actors in this task, is an additional topic for elaborating a governance
concept for reexive modernisation. Such a topic deserves much more atten-
tion than it has been given both here and in literature more generally (but see
Forester, 1999 and Fischer, 2003).
NOTES
1. In fact, the insights presented here will be further developed in the context of the Dutch
Knowledge Network on System Innovations, of which the author is co-director and in
which scientists from a wide variety of areas (complex system theory and integrated analy-
sis, the social history of technology, innovation studies and the policy sciences) cooperate
with each other and with practitioners (Rotmans et al., 2003).
2. In processes that have been far from uniform between nation states.
3. This dual expression reects the duality of structure (Giddens, 1984).
4. Countries like Germany and France had basically similar policies, and the first European
Commissioner of Agriculture was Mansholt, the first post-war Dutch minister of agri-
culture. See Ackrill (2000).
5. The risks and side-eects, together with more established criteria as economic viability,
are taken as the dimensions.
6. For similar examples in the area of health care, see Grin (2004).
7. In line with what I argued on the loci for both political judgement and planning, this
includes actors on the level of the political community as a whole; in socio-economic-
scientific practices; and in between.
REFERENCES
Ackrill, Robert (2000), The Common Agricultural Policy, Sheeld: Sheeld
Academic Press.
Archer, Margaret S. (2003), Structure, Agency and the Internal Conversation,
Cambridge: Cambridge University Press.
Beck, Ulrich (1992), Risk Society. Towards a New Modernity, London: Sage.
Beck, Ulrich (1997), The re-invention of politics. Rethinking Modernity in the Global
Social Order, Cambridge: Polity Press.
Beck, Ulrich, Wolfgang Bonns and Christoph Lau (2003), The theory of reexive
modernisation. Problematic, hypotheses and research programme, Theory,
Culture and Society, 20 (2), 133.
Bekke, Hans, Jouke de Vries and Geert Neelen (1994), De Salto Mortale van het
Ministerie van Landbouw, Natuurbeheer en Visserij. Beleid, Organisatie en
Management op een Breukvlak, Alphen aan den Rijn: Samson H.D. Tjeenk
Willink.
Bekke, Hans and Jouke de Vries (2001), De Ontpoldering van de Nederlandse
Landbouw. Het Ministerie van Landbouw, Natuurbeheer en Visserij, 19942000,
Leuven/Apeldoorn: Garant.
Benhabib, Seyla (1992), Situating the Self. Gender, Community and Post-modernity
in Contemporary Ethics, New York: Routledge.
Reflexive modernisation as a governance issue 79
Benhabib, Seyla (2002), The Claims of Culture. Equality and Diversity in the Global
Era, Princeton, NJ and Oxford: Princeton University Press.
Bernstein, Richard J. (1983), Beyond objectivism and relativism. Science,
Hermeneutics and Praxis, Philadelphia: University of Pennsylvania Press.
Bernstein, Richard J. (1993), The New Constellation. The EthicalPolitical Horizons
of Modernity/Postmodernity, Cambridge, MA: MIT Press.
Bieleman, J. (2000), Landbouw, Deel I (p. 11233) in: Techniek in Nederland in de
Twintigste Eeuw: Landbouw and Voeding, Zutphen: Walburg Pers.
Bos, Bram, Peter Groot Koerkamp and Karin Groenestein (2003), A novel
design approach for livestock housing based on recursive control with
examples to reduce environmental pollution, Livestock Production Science, 84,
15770.
Bos, Bram and John Grin (forthcoming), The Hercules eect of dealing with risks
within modern institutions: lessons from a project on sustainable husbandry,
Science, Technology and Human Values.
Bourdieu, Pierre (1977), Outline of a Theory of Practice, Cambridge: Cambridge
University Press.
Collingridge, David (1980), The Social Control of Technology, London: Macmillan.
De Wilt, J.G. (2004), Paper prepared for a meeting at the Centre for Prospective
Technology Studies, Joint Research Center at Sevilla.
Dierkes, M, U. Homann and L. Marz (1995), Visions of Technology. Social and
Institutional Factors Shaping the Development of New Technologies, Frankfurt
and New York: Campus Verlag/St. Martins Press.
Fischer, Frank (2003), Reframing Public Policy: Discursive Politics and Deliberative
Practices. Oxford: Oxford University Press.
Flyvbjerg, Bent (2001), Making Social Science Matter. Why Social Inquiry Fails and
How it can Succeed Again, Cambridge: Cambridge University Press.
Forester, John (1999), The Deliberative Practitioner, Cambridge, MA: MIT Press.
Fox, Charles J. and Hugh T. Miller (1996), Postmodern Public Administration.
Toward Discourse, London: SAGE.
Geels, Frank W. (2002), Understanding the Dynamics of Technological Transitions:
A Co-evolutionary and Socio-technical Analysis, Enschede: Twente University
Press.
Giddens, A. (1984), The Constitution of Society. Outline of the Theory of
Structuration, Cambridge: Polity Press.
Giddens, Anthony (1991), Modernity and Self-Identity. Self and Society in the Late
Modern Age, Cambridge: Polity Press.
Gill, Graeme (2003), The Nature and Development of the Modern State, Houndmills
and New York, NY: Palgrave Macmillan.
Grin, John and Henk van de Graaf (1996a), Technology assessment as learning,
Science, Technology and Human Values, 20 (1), 7299.
Grin, John and Henk van de Graaf (1996b), Implementation as communicative
action. An interpretive understanding of interactions between policy actors and
target groups, Policy Sciences, 29 (4), 291319.
Grin, John, Henk van de Graaf and Philip Vergragt (2003), Een derde generatie
milieubeleid: Een sociologisch perspectief en een beleidswetenschappelijk pro-
gramma, Beleidswetenschap, 17 (1), 5172.
Grin, John, Francisca Felix, Bram Bos and Sierk Spoelstra (2004), Practices for
reexive design: lessons from a Dutch programme on sustainable agriculture,
International Journal of Foresight and Innovation Policy, 1 (12), 14669.
80 Reflections on reflexive governance
Grin, John (2005), Knowledge society: old wine in new bottles, or a new contract
between science and society?, to be published in Joske Bunders ed. (2005):
Sharing Knowledge? Exploring the Interfaces Between Science and Society,
Amsterdam: Boom.
Hajer, Maarten A. (2003), Policy without polity: policy analysis and the institu-
tional void, Policy Sciences, 36 (2), 17595.
Haraway, Donna J. (1991), Simians, Cyborgs and Women: The Reinvention of
Nature, New York: Routledge.
Hayek, Friedrich (1960), The Constitution of Liberty, London: Routledge & Kegan
Paul.
Hennis, Marjoleine (2001), Europeanization and globalization: the missing link,
Journal of Common Market Studies, 39 (5), 82950.
Hoppe, Rob and John Grin (2000), Trac goes through the TA machine: A cul-
turalist comparison, in Norman J. Vig and Herbert Passchen (eds), Parliaments
and Technology: the Development of Technology Assessment in Europe, New York:
SUNY Press, pp. 273324.
Joas, Hans (1993), Pragmatism and Social Theory, Chicago and London: Chicago
University Press.
Kickert, Walter J.M., Erik-Hans Klijn and Joop F.M. Koppenjan (eds) (1997),
Managing Complex Networks: Strategies for the Public Sector, London: Sage.
Kooiman, Jan (2003), Governing as Governance, London: Sage.
Kumar, Krishan (1995), From Post-Industrial to Post-Modern Society. New Theories
of the Contemporary World, Oxford and Malden, MA: Blackwell.
Latour, Bruno (2003), Is re-modernization occurring and if so, how to prove it?,
Theory, Culture and Society, 20 (2), 3548.
Law, John (1992), Notes on the theory of the actor-network: ordering, strategy and
heterogeneity, Systems Practice, 5, 179393.
Leydesdor, Loet (2001), A Sociological Theory of Communication. The Self-
Organization of the Knowledge-Based Society, Universal Publishers.
Lindblom, Charles E. (1959), The science of muddling through , Public
Administration Review, 39, 7988.
Lindblom, Charles E. (1965), The Intelligence of Democracy, New York: Prentice
Hall.
Lindblom, Charles E. (1979), Still muddling, not yet through, Public Administration
Review, 59, 51726.
Lindblom, Charles E. (1999), A century of planning, in Michael Kenny and James
Meadowcroft (eds), Planning Sustainability, London and New York: Routledge,
pp. 3965.
Loeber, Anne (2004), Practical wisdom in the risk society. Methods and practice of
interpretive analysis on questions of sustainable development, Amsterdam:
University of Amsterdam (PhD thesis).
Meadowcroft, James (1999), Planning for sustainable development: what can we
learn from the critics?, in Michael Kenny and James Meadowcroft (eds),
Planning Sustainability, London and New York: Routledge.
Newman, Janet (2001), Modernising Governance: New Labour, Policy and Society,
London: Sage.
Pierre, J. and B. Guy Peters (2002), Governance, Politics and the State, Basingstoke:
Macmillan.
Priester, P.R. (2000), Landbouw, part Ib in: Techniek in Nederland in de twintigste
eeuw: Landbouw and Voeding. Zutphen: Walburg Pers, pp. 65125.
Reflexive modernisation as a governance issue 81
Rip, Arie and Ren Kemp (1998), Technological change, in Steve Rayner and
Elizabeth L. Malone (eds), Human Choice and Climate Change, Columbus, OH:
Batelle Press, pp. 32799.
Roep, D., J.D. van der Ploeg and J.S.C. Wiskerke (2003), Managing technical
intitutional design processes: some strategic lessons from environmental
co-operatives in the Netherlands, Netherlands Journal of Agrarian Studies, 51
(12), pp. 195217.
Rotmans, J., R. Kemp, M.B.A. van Asselt, F.W. Geels, G. Verbong and K. Molendijk
(2000), Transitions and Transition Management, Maastricht: ICIS.
Rotmans, J., R. Kemp and M.B.A. van Asselt (2001), More evolution than revolu-
tion: transition management in public policy, Foresight, 3 (1), pp. 1531.
Rotmans, Jan, John Grin, Johan Schot and Ruud Smits (2003), A multidisciplinary
research programme on transitions and system innovations, presented at the
Open Science Meeting of the International Human Dimensions Programme,
Montreal, 1820 October.
Scott, James (1998), Seeing like a State, New Haven, CT: Yale University Press.
Toulmin, Stephen (1990), Cosmopolis. The Hidden Agenda of Modernity, Chicago:
University of Chicago Press.
Van der Ploeg, Jan-Douwe (1999), De virtuele boer, Assen: Van Gorcum.
Van Gunsteren (1975), The Quest for Control, New York: John Wiley.
Van Otterloo, Anneke H. (ed.) (2000), Part II (voeding), in Techniek in Nederland
in de twintigste eeuw: Landbouw en voeding, Zutphen: Walberg pers.
Van Tatenhove, Jan, Bas Arts, Pieter Leroy (eds) 2000, Political Modernisation and
the Environment. The Renewal of Environmental Policy Arrangements, Dordrecht:
Kluwer Academic Publishers.
Vickers, Georey (1995 [1965]), The Art of Judgment. A Study of Policy Making
centenary edition in the Advances in Public Administration series, London: Sage.
Wildavsky, A. (1973), If planning is everything, maybe its nothing, Policy Sciences,
4, pp. 13753.
Wildavsky, A. (1979), The Art and Craft of Policy Analysis, London: Macmillan.
Wisserhof, Johan (2000), Agricultural policy making in the Netherlands: beyond
corporatist policy arrangements?, in Jan van Tatenhove, Bas Arts and Pieter
Leroy (eds), Political Modernisation and the Environment. The Renewal of
Environmental Policy Arrangements, Dordrecht: Kluwer Academic Publishers,
pp. 17598.
4. A co-evolutionary approach to
reexive governance and its ironies
Arie Rip
82
A co-evolutionary approach to reflexive governance 83
A CO-EVOLUTIONARY PERSPECTIVE
Let me start with evolutionary theories and then add mutual translations
or mutual selections which can add up to patterns.4 This route is denitely
plausible for scientic and technological change, but can be taken to apply
to social change and social order more generally, even if it may be dicult
to develop specic theories. As a perspective, it can still be used to draw
attention to certain phenomena.
There is an abundance of evolutionary theories about technological
change, and some evolutionary theories about science and its institutions.5
There are good reasons for such theories, because central to modern science
and technology is the introduction of novelties, which like biological muta-
tions will at rst be tentative and uncertain hopeful monstrosities (Mokyr
1990, Stoelhorst 1997), but can grow and modify, and become accepted.
Evolutionary theories in biology often emphasize that variation as well as
selection are blind so that there is no guarantee of the quality of the out-
comes other than that they will t the contingencies of the selection envi-
ronment. For science and technology, and for social life generally, the
selection environment is not blind, and when variations, that is, novelties, are
produced, there is anticipation on eventual selection up to attempts to
change the selection environment so as to increase the chances for the vari-
ation to survive. Van den Belt and Rip (1987) showed (using the label of
quasi-evolutionary theory) that nexuses between variation and selection
emerge (with test laboratories for new products as an example) and posi-
tioned the existence of paradigms in science and technology as a temporary,
but forceful stabilization of selection and thus continuity in variation leading
to a trajectory or path of further development.
In the co-evolution of science, technology and society, patterns at the insti-
tutional and societal level also emerge and stabilize (and open up again). For
example, there is a diuse social contract between science and society visible
in expectations about science, its mandate and division of responsibilities,
and the institutional arrangements. The phrase Science, The Endless
Frontier (after Vannevar Bushs report to the US President in 1945) is often
used to characterize the post-Second World War social contract, with its
A co-evolutionary approach to reflexive governance 85
dual character of funding open-ended basic research on the one hand that
would somehow deliver great things to society, and on the other hand, having
big public laboratories, for example, for defence, nuclear energy, new mate-
rials (Guston and Kenniston 1994). The quasi-autonomous dynamics of
science appear to be so strong that governance actors cannot do much more
than try to modulate what is going on anyway.6 In fact, during most of the
twentieth century, science policy was dened as the support of science. It is
only recently that more interventionist measures are considered and imple-
mented which then lead to concerns that emphasis on short-term relevance
could undermine the quality of science, or at the very least introduce epis-
temic drift (Elzinga 1985). Science policy actors now experience the tension
between accommodating to ongoing dynamics, and assuming a political
responsibility which implies some intervention.7
In other words, the earlier diuse social contract has started to break
down. By now, a new regime of Strategic Science, that focuses on wealth
creation and support for decision making and quality of life, is emerging.
This includes a more distributed character of knowledge production, which
is visible in the way new elds in ICT, genomics and nanotechnology are
organized (Rip 2002a).
The patterns of such regimes enable productive work and interactions,
while at the same time constraining them. This can be welcomed as well as
criticized. The evolutionary perspective does not oer immediate entrance
points for normative evaluation that is, as it were, delegated to the selection
environment but a second-order normativity can be drawn on: the import-
ance of allowing for further evolution. Specically, stabilization of patterns
and of emerging irreversibilities more generally could create a lock-in that
is dicult to escape. In the case of recent changes in and around science,
I have argued that the regime of Strategic Science might be closing in upon
itself too soon, because of strong economic and decision-making pressures
to deliver, and to exclude alternative, or just dierent approaches, like
community-oriented research and indigenous knowledges. Without neces-
sarily pronouncing on the value of these approaches per se, I argue that they
play a role in maintaining heterogeneity (Rip 2000).
Another feature where second-order normativity can be articulated is the
increasing reexivity of the co-evolution of science, technology and society.
The rise of explicit science policy making (since the Sputnik shock of 1957)
is a rst indicator. Technology assessment, debated expertise, and partici-
pation are recent phenomena, and have become accepted as a necessary
element (Rip 2002c). Such reexivity is not just a matter of discourse and
debate, even if that is part of it. It is about institutionalized feedback rela-
tions, which handle in this case interactions between science, technology
and society. In other words, the de facto governance arrangements become
86 Reflections on reflexive governance
general rule. Or better, there are lots of rules, dominant positions etc., but these
are contingent and cannot be taken for granted. Instead of steering, there is
reective (and reexive) intervention: mutual translations (thats what happens
anyway) are now seen as the basic process. (Rip 2002)
This is more than the well-known criticism of rational planning and control
approaches and reference to interactive approaches, policy networks, self-
regulation and so on. Distributed coherence is now the eect of reexive
co-evolution, and a variety of agents are involved who all contribute to
governance. Among this variety, these are reexive agents like Constructive
TA agents described in Schot and Rip (1977): ideally, all actors could
take up that role. The present book implies, already in the way the editors
and authors view their task, that there are sustainability agents who can
prot from their analysis. Lindbloms disjointed incrementalism is not
only practical advice to a governance actor with limited power and limited
insight, but also a way to advance what he calls intelligent democracy,
where all citizens would be reexive-governance agents. Lindblom advo-
cates a political position. Schot and Rip are more ironical, and recognize
the struggles between opponents of various kinds that are necessary to have
incentives to act and learn.
NON-MODERN STEERING
AND ANTICIPATION-IN-ACTION
How to develop the rst-round diagnosis further and translate it into strate-
gies for reexive governance? Basically, it must be steering from within
the co-evolutionary process (what I have called modulation), possibly
also referring to second-order normative notions of variety, reexivity
and distributed coherence. The notion of steering, with its implication of
an agent faced with an object to be steered, is of course misleading
since the steering agent is part of an evolving system including the object
and himself. To keep the tensions visible, I will use the term non-modern
steering, a contradictio in terminis, as a programmatic concept. The enlight-
ened modernist version of reexive governance then becomes a specic
version of non-modern steering, where anticipation-in-action takes the co-
evolution into account, but primarily in order to better achieve ones goals.
My approach in this section is to collect possible instances of non-
modern steering, and draw out insights in how to do steering-from-within
and what the problems and some of the paradoxes are. Let me start again
with science and science policy.
The co-evolution of science, technology and society cannot be steered in
a simplistic interventionist way other than for short periods of time and/or
A co-evolutionary approach to reflexive governance 89
in special circumstances. For science policy agents this is clear, for example,
in the way they depend on the aggregation of views and insights in the
scientic community to set priorities, even if they can make a dierence by
deciding on the funding. But even there, they are subject to credibility pres-
sures; for example, which science policy agent can nowadays refuse to
support genomics and nanotechnology? Further analysis is possible, in
particular in terms of multi-level development with mutual dependencies
between a policy and governing decision level, an intermediary level (for
example, funding agencies) and the level of research performing actors (Rip
1998). This framework can be used to understand overall developments in
research systems (Rip and Van der Meulen 1996), but also to understand
how actor strategies emerge and interact (Morris 2004).
Important for the general idea of non-modern steering is the opportunity
the framework oers to include the eect of context on policy instruments.
In contrast to a major (even if now being criticized) thrust in the policy lit-
erature, policy instruments should not be discussed in isolation. They do not
work because of their intrinsic characteristics but because of the context in
which they are applied, and in particular, the amount of repair work that is
done at other levels. This helps to articulate the global notion of steering
from-the-inside and adds the possibility that success derives from repair
work elsewhere in the overall system.
Such repair work happens all the time, and unavoidably so, but it tends
not to be recognized because of the modernist illusion. I have argued a
number of times that the apparent eectiveness of some modernist policies
(in science policy, in risk regulation) derives from the repair work done
during implementation and ongoingly in local practices. Thus, acceptance
of local repair work in order to keep things going is necessary, and one
component of non-modern steering is to ensure its quality rather than try
to control it from a distance. In other words, as Bruno Latour (1991) claims,
we have never been modern, and the hybrid monsters backgrounded by
modernist ideologies actually continue to do their work and thus ensure
that the modernist venture carries on, in spite of the ideologies.11 Non-
modern then does not negate modernism, but draws attention to its actual
practices.
There are instances of non-modern steering. Opening up learning spaces,
as emphasized in the Introduction to this book, can be seen as non-modern
steering. The orientation then is less towards solving problems (of sus-
tainability), but towards creating and maintaining spaces for working
towards solutions. This might include increasing reexivity as an institu-
tional capacity. In a co-evolutionary perspective, reexivity and attendant
learning is located in the processes and at the system level, not in the heads
of individuals.12
90 Reflections on reflexive governance
ACKNOWLEDGEMENT
Without the encouragement and comments of Jan-Peter Vo, this chapter
could not have been written. His contributions have been substantial, and
this is only partially visible in the text.
NOTES
1. I have used the term post-modernist to characterize this second approach (Rip 2002),
because of its connotation with a variety of small stories rather than one big story, as
Lyotard (1984 [1979]) characterized the contrast. See also Visscher and Rip 2003, who
identify dierent types of business consultants: enlightened modernists, ironists and
post-modernists.
2. Beck contrasts reexive with reective, the latter indicating reection on what is hap-
pening. If such reection is part of further action and interaction, it would enhance
reexivity. Jan-Peter Vo adds to this (personal communication, 2004): There is rst-
order reexivity, the self-shaping or self-constituting nature of a societal arrangement
via feedback from its own action in and on the world. But then also (in late modernity),
A co-evolutionary approach to reflexive governance 95
over actions is now seen as a complete modernist ction. (. . .) We do not run more
dangers than before, but (. . .) we are now entangled, whereas the modernist dream was
to disentangle us from the past.
12. Rip (2000d), a similar point about repertoire learning already in Rip (1986).
13. The European Commission is now establishing what they call Technology Platforms for
that same purpose.
14. An example is the way that attempts to stimulate the use of local knowledge in rural agri-
culture in KwaZulu-Natal (South Africa) get locked into supplying organic food at
premium prices (if the volume and transport problems can be solved). See Rip (2003).
15. A variety of tools and approaches to anticipate so as to improve the actors own posi-
tion are available. Learning curves, product life cycles, trajectories all presume contin-
uation of the past into the future. Actors must still decide whether to t and thus
reinforce existing dynamics, or to stretch and modify the environment so as to accom-
modate their new venture.
16. I am indebted to Jan-Peter Vo for this turn of phrase and the underlying thought.
17. Interestingly, there is now a movement towards decentred, open-ended environmental
management in which, in practice, restoration strategies do not orchestrate the return to
a specied earlier stage of an ecosystem, but engage practically with the physical envi-
ronment and learn what is possible and desirable (Castree 2001). Conversation occurs
with the ecosystem (and with relevant human actors).
18. See note 2.
19. Cf. Beck (1992), pp. 171 .
20. The title of a report of the European Environmental Agency, Late lessons from early
warning (Harremos, 2001) together with the existence of such an agency and their
writing such reports, is indicative. As the reference to early warningindicates, such antici-
patory knowledge is part of actors strategies and reactions, and its fate, i.e. uptake, is
determined by existing and emerging overlapping strategic games. An example of over-
lapping strategic games determining uptake is the story of uorochlorocarbons and the
ozone layer from the mid-1970s onwards. An early warning from Molina, uncertainty
exploited by industry to argue against regulation, the rst phase of checking and expand-
ing knowledge: while some regulatory agencies started to regulate anyway because they
wanted to protect their turfor extend their scope before other agencies step in (Rip 1992).
21. The emergence of forecasting in the 1950s and 1960s, and the way it was taken up in some
policy making is a clear example. Of course, forecasting activities (now also foresight)
can be part of anticipation-in-action.
22. For its regularity over the last four decades, Moores Law of regular increases in the
density of units on a chip (an integrated circuit) and the speed of micro-processors,
depends on the continuity of the strategic game among the chip producers and their gov-
ernment and R&D allies, using Moores Law as a reference.
23. For examples from product development see Deuten and Rip (2000), in societal debate
on acid rain, Hajer (1995). See also my analysis of social science and evidence-based
policy making (Rip 2001). The narrative approach has further implications, for example
how it can carry the past into the future.
24. Cf. Rip 2002d on Max Miller, and Rip 1986 on collective learning in controversies
shaped by forceful foci. Deuten (2003) showed that technological learning, for example
about reinforced concrete, occurred only because of demanding customers.
25. I use the common-sense version of Machiavellian here, which refers to The Prince not to
Machiavelli who wrote The Republic.
26. Donald Campbell has made a similar suggestion about the tribal norms necessary for
scientic communities in order to produce good science (Campbell 1979). I have taken
up this point in my discussion of scientic expertise, regulation and the precautionary
principle (Rip 2001a).
27. In French stylistic guidelines, the quote appears again and again, and is sometimes followed
by a reverse version: Il nest pas ncessaire dentreprendre pour esprer ni de persvrer
pout russir. (http://www.courtois.cc/citations/travail.html, visited 13 November, 2004).
A co-evolutionary approach to reflexive governance 97
28. In the same vein, the city of Rotterdam features a line from the painter and poet,
Lucebert, in big letters on top of a prominent building: Alles van waarde is weerloos
(everything of value is defenceless). See also proliferation of the sentence on buildings
and placards, Trust me. I am you. Modern societies are reexive!
REFERENCES
Beck, Ulrich (1992), Risk Society. Towards a New Modernity, London: Sage.
Beck, Ulrich, Wolfgang Boriss and Christoph Lan (2003), The theory of reexive
modernization: problematic, hypotheses and research programme, Theory,
Culture & Society, 20, 133.
Callon, Michel, John Law and Arie Rip (1986), Mapping the Dynamics of Science and
Technology. Sociology of Science in the Real World, London: Macmillan Press.
Campbell, Donald T. (1979), A tribal model of the social system vehicle carrying
scientic knowledge, Knowledge, 1 (December), 181201.
Castree, Noel (ed.) (2001), Social Nature: Theory, Practice and Politics, Oxford:
Blackwell Publishers.
Coleman, James S. (1990), Foundations of Social Theory, Cambridge, MA: Harvard
University Press.
Deuten, J. Jasper (2003), Cosmopolitanizing technology: studies of four emerging
technological regimes, PhD Thesis, University of Twente.
Deuten, J. Jasper and Arie Rip (2000), Narrative Infrastructure in Product
Creation Processes, Organization, 7, 6791.
Deuten, J. Jasper, Arie Rip and Jaap Jelsma (1997), Societal embedment and
product creation management, Technology Analysis and Strategic Management,
9, 21936.
Elzinga, Aant (1985), Research, bureaucracy and the drift of Epistemic criteria,
in Bjrn Wittrock and Aant Elzinga (eds), The University Research System. The
Public Policies of the Home of Scientists, Stockholm: Almqvist and Wiksell
International, pp. 191220.
Guston, David H. and Kenneth Kenniston (eds) (1994), The Fragile Contract.
University Science and the Federal Government, Cambridge, MA: MIT Press.
Hajer, Maarten A. (1995), The Politics of Environmental Discourse. Ecological
Modernization and the Policy Process, Oxford: Clarendon Press.
Hanf, Kenneth, and Theo A.J. Toonen (eds) (1985), Policy Implementation in
Federal and Unitary Systems, Dordrecht: Martinus Nijho Publishers.
IAF and CRIC (2002), ESRC Project on Genomics and Society. Six reports, avail-
able from http://les 1.man.ac.uk/cric/genomics
Jasano, Sheila (ed.) (2004), States of Knowledge. The co-production of science and
social order, London: Routledge.
Latour, Bruno (1991), Nous navons jamais t modernes. Essai danthropologie
symtrique, Paris: La Dcouverte.
Latour, Bruno (2003), Is re-modernization occurring and if so, how to prove it?
a commentary on Ulrich Beck, Theory, Culture and Society, 20, 3548.
Leydesdor, L. (2000), The triple helix: An evolutionary model of innovations,
Research Policy, 29, 24355.
Leydesdor, Loet (2001), A Sociological Theory of Communication. The Self-
Organisation of the Knowledge-Based Society, Universal Publishers.
98 Reflections on reflexive governance
Four analytical essays and a critical debate on the future of scholastic endeavour,
Amsterdam: Amsterdam University Press, pp. 99148.
Rip, Arie (2002b), Co-Evolution of Science, Technology and Society, expert
review for the Bundesministerium Bildung and Forschungs Frderinitiative
Politik, Wissenschaft und Gesellschaft (Science Policy Studies), managed by the
BerlinBrandenburgische Akademie der Wissenschaften, Enschede: University
of Twente, 7 June.
Rip, Arie (2002c), A co-evolutionary perspective on ELSI, CTA and other attempts
at re-contextualisation of science and technology in society, paper presented to
the meeting of the European Association for the Study of Science and
Technology, York, 31 July3 August 2002.
Rip, Arie (2002d), Systematic learning without systems? Commentary on Max
Miller, Sozialer Sinn, Heft 3, 43543.
Rip, Arie (2003), Technological innovation in context, invited keynote speech at
the meeting of the Lowlands Innovation Research Network, Louvain (Belgium),
14 January 2003.
Rip, Arie and Aard Groen, (2001), Many visible hands, in Rod Coombs, Ken
Green, Vivien Walsh and Albert Richards (eds), Technology and the Market.
Demands, Users and Innovation, Cheltenham, UK and Northampton, MA, USA:
Edward Elgar, pp. 1237.
Rip, Arie and Rene Kemp (1998), Technological change, in S. Rayner and
E.L. Malone (eds), Human Choice and Climate Change, Columbus, OH: Battelle
Press, 1998. Volume 2, Ch. 6, pp. 32799.
Rip, Arie, and Barend J.R. van der Meulen (1996), The post-modern research
system, Science and Public Policy, 23, 34352.
Schimank, Uwe (1988), Gesellschaftliche Teilsysteme als Akteurktionen, Klner
Zeitschrift fr Soziologie und Sozialpsychologie, 40, 61939.
Schimank, Uwe (1992), Spezische Interessenkonsense trotz generellen
Orientierungsdissens. Ein Integrationsmechanismus polyzentrischer
Gesellschaften, in Giegel, Hans-Joachim (Hrsg) Kommunikation und Konsens
in moderne Gesellschaften, Frankfurt: Suhrkamp.
Schn, Donald (1984), The Reective Practitioner: How Professionals Think in
Action, New York: Basic Books.
Schot, Johan and Arie Rip (1997), The past and future of constructive technology
assessment, Technological Forecasting and Social Change, 54, 25168.
Schuyt, C.J.M. (1972), Recht, orde en burgerlijke ongehoorzaamheid, Rotterdam:
Universitaire Pers Rotterdam.
Stoelhorst, J.W. (1997), In search of a dynamic theory of the rm. An evolution-
ary perspective on competition under conditions of technological change,
with an application to the semi-conductor industry, Enschede: University of
Twente.
Van den Belt, H. and A. Rip (1987), The Nelson-WinterDosi model and synthetic
dye chemistry, in W.E. Bijker, T.P. Hughes and T. Pinch (eds), The Social
Construction of Technological Systems: New Directions in the Sociology and
History of Technology, Cambridge, MA: MIT Press, 13558.
Van Gunsteren Herman R. (1976), The Quest for Control. A critique of the ratio-
nal-central-rule approach in public aairs, London and New York: John Wiley.
Verganti, Roberto (1999), Planned exibility: linking anticipation and reaction in
product development projects, Journal of Product Innovation Management, 16,
36376.
100 Reflections on reflexive governance
Visscher, Klaasjan and Arie Rip (2003), Coping with chaos in change processes,
Creativity and Innovation Management, 12 (2), (June) 1218.
Ziman, John (ed.) (2000), Technological Innovation as an Evolutionary Process,
Cambridge: Cambridge University Press.
Strategies for sustainable system
transformation
5. Transition management: a reexive
governance approach
Ren Kemp and Derk Loorbach1
INTRODUCTION
A CHANGING WORLD
Our society is always changing. Over the past decades, however, driven by
transnational trends such as internationalization, informatization and indi-
vidualization (Schnabel, 2000), the process of social change has become
103
104 Strategies for sustainable system transformation
new institutions, new technologies and new regulations within the ecologi-
cal limits dened.
Taking such a comprehensive approach towards sustainable develop-
ment requires a redenition of policy-making in this context; a more evo-
lutionary and adaptive strategy is needed that allows for self-organization
within certain limits, both ecological limits as well as social limits (often set
by government, but they can include social norms as well). The central
focus of such a strategy should be to realize long-term and large-scale inno-
vations or transitions towards more environmentally and socially benign
societal systems (such as agriculture or energy supply). Since realizing long-
term, diuse sustainability goals should be the aim of such policies, an
emphasis on interaction, experiment and learning is crucial. In this chapter
we will rst address the concept of transition as a systems way to view
societal change. Secondly, we will present the approach of transition man-
agement as a new form of governance for sustainability and will focus on
the reexive elements in the strategy.
price change, policy act or a new technology but are the result of develop-
ments in various domains which sustain each other: technology, economy,
institutions, behaviour, culture, ecology and images/paradigms (p. 20). The
process of transition is non-linear; slow change is followed by rapid change
when things reinforce each other, which again is followed by slow change in
the stabilization stage.
Although transitions are characterized by non-linear behaviour, the
process itself is a gradual one, typically spanning one or two generations
(2550 years).3 The nature and speed of change dier in each of the tran-
sition stages:
Indicator(s)
for social
development
Stabilization
Breakthrough
Take-off
Predevelopment
Time
A transition
systems that oer functional services and disservices, housing interests and
organizations with capabilities and mental models, who function in a world
of beliefs, values, capital goods, prices, settlements, lifestyles and novelties.
By denition, complex systems are highly chaotic and impossible to steer in
the sense of command-and-control because of numerous feedback loops,
inert institutions and unpredictable developments. Rather than focusing
on individual components of these systems, governance for sustainable
development should try to nd governance strategies at a systemic level
and try to change the condition of the system for the better. This idea has
been accepted by the model of transition management of Rotmans and
others (Rotmans et al., 2000; Kemp and Rotmans, 2001, 2002; Loorbach
and Rotmans, 2002). Through transition management one tries to inuence
the direction and speed of transitions by coordinating and enabling the
processes that occur at dierent levels in a more systemic and evolutionary
way, which leaves room for variation and selection mechanisms and innova-
tion. This can be done in many dierent ways, through various types of steer-
ing mechanisms, none of which incidentally can be prescribed or even
repeated in detail.
In the following pages we will give a description of our model of transi-
tion management accepting that there are other models. When we speak of
transition management we mean the ICIS-MERIT model of transition
management in whose development we were involved ourselves. It is an
attempt at goal-oriented modulation, not an attempt to achieve predened
outcomes through planning and control. Transition management works
with dynamics not against them. Ongoing developments are exploited strate-
gically. Transition management for sustainability tries to orient dynamics to
sustainability goals. The goals are chosen by society through the political
process: the systems to satisfy these goals are worked towards in an adaptive,
forward-looking manner. The goals and policies for furthering the goals are
constantly assessed and periodically adjusted in development rounds.
Policies will dier across the dierent transition phases. In early phases,
policy should be concerned with the formulation of transition goals
and engage in the formulation of sustainability visions (quality images),
which are re-assessed during later phases. The attention to innovation will
be a continued feature of all phases; it is not just something for the early
phases.
A schematic view of transition management is given in Figure 5.3.
Policy actions are evaluated against two types of criteria: (1) the imme-
diate contribution to policy goals (for example in terms of kilotons of CO2
reduction and reduced vulnerability through climate change adaptation
measures), and (2) the contribution of the policies to the overall transi-
tion process. This means that under transition management, policies have
110 Strategies for sustainable system transformation
State of
development
of solutions Sustainability
Transition management: oriented towards long-term sustainability visions
goals and visions, iterative and reflexive (bifocal)
Problem Denition
The starting point for transition management for sustainability is the per-
sistent problems of existing functional systems. Because of the system-
inherent nature of the problems they do not have a single owner. Rather
than looking for a villain to blame, one looks for a common problem
denition in which all problems are considered, not just those problems that
can be dealt with relatively easily. Transition management is targeted at
widely acknowledged problems that require a response for which no ready-
made solution is (or will be) available. Often these are not single problems
but a range of problems. For energy for example, the problems are depen-
dence on scarce (non-renewable) resources (oil, natural gas), emissions of
greenhouse gases stemming from the combustion of fossil fuels that cause
climatic change, price volatility from shortfalls in supply often as a result of
wars, and the military conict over oil resources and oil power. By develop-
ing a shared problem perception based on the input of dierent actors,8
Transition management 113
those actors involved will adjust their own problem denitions and percep-
tions because of a better understanding of the nature of the problem and
the perspectives held by other actors and accordingly their behaviour (that
is, second-order learning). This however will only come about if enough time
and energy is invested in these discussions. A problem here is that each solu-
tion to these problems has its own disadvantages. In the short term there are
all kinds of tradeos. The aim of transition management is to provide an
environment in which these trade os are made visible and can be negoti-
ated. This will be a collective task for which one needs transition goals that
reect societal aspirations.
nal image. The images should be adjusted as a result of what has been
learned by the actors in the various transition experiments. The participa-
tory transition process is thus a goal-seeking process, where the transition
images change over time (the transition goals are likely to remain the same).
This diers from so-called blueprint thinking, which operates from a xed
notion of nal goals and corresponding visions.
Transition paths are possible routes towards the nal images. The images
do not necessarily have to be consistent (only with the vision) and multiple
paths can be developed for one image (see Figure 5.4). It is important to
incorporate interim goals and objectives in the transition paths that become
more concrete the closer they are to the present. The transition paths
however also have to reect the necessary trend breaks and behavioural and
institutional changes, the uncertainties associated with the pathway and the
barriers and chances for implementation. Finally, practical experiments
(programmes) are planned, which are targeted at exploring the transition
paths derived from the analysis and developed strategy.
The vision, in combination with the images, the transition paths and the
experiments, forms the joint transition agenda at the tactical level. This is
where coalitions come together around specic options or expectations, for
example specic technologies or new institutions. For example, one could
think of developing new consumption patterns for sustainable agriculture;
a coalition to develop this idea and explore it further would include local
Basket of
images
Multiple
pathways
Knowledge Integration
per se. Transition management therefore is a network strategy that also tries
to use the networks of the participants in the transition arena to spread the
thoughts developed there. The discussions with other actors (about the
(complex) system, its problems and dynamics, but about strategies as well),
will have to be extensive and confrontational enough to lead to second-
order learning amongst the participants. In practice, this means that the par-
ticipants will reect on their own dispositions, their own practices and their
own roles within the larger context. They will supposedly take home such
new insights as well as the new ideas on cooperation, solutions, and so on.
By creating within their own organizations new arenas that address more
specic elements of the common approach and strategy, they will contribute
to realizing a structure of arenas and thus knowledge and experiences can
be shared and exchanged between these arenas. If actively pursued, such an
elaborated structure could be seen as an instrument for knowledge produc-
tion, knowledge diusion and knowledge integration (McElroy, 2003).
Too many atrocities of stupidity and immorality have been based on anticipa-
tory rationality, and too many eorts to improve human action through import-
ing technologies of decision engineering have been disappointing. (March and
Olsen, 1995: 1989)
This clearly shows the limitation of the use of anticipatory outcomes but
is probably too negative with regard to anticipation. In transition manage-
ment, experiences inform next steps more than grand visions do.
alteration of the status quo than can a mere infrequent major policy change.
(Lindblom, 1979: 520)15
Organizing a multi-
actor network
Developing
Evaluating, sustainability
monitoring visions and
and learning transition
agendas
are the result of the interaction itself and not of individual choices or
demands. By co-developing visions and agendas and collectively carrying
out practical projects and experiments, the mutual adjustment of these per-
spectives and expectations takes shape.
Biomass International;
New Gas Services;
Sustainable Industrial Production;
Toward a Sustainable Rijnmond (an industrial ecology project);
Policy Renewal.
Strategic
2020 1015% in power production 1520% in transport
goals
A. Gasification
B. Pyrolysis
Exp Transition
paths
C. Biofuels
Exp
EOS
2003
CONCLUSIONS
NOTES
systems-thinking is not a straitjacket, but rather a way to identify main drivers, be aware
of side eects and develop more integrated strategies.
3. The time span is not a dening characteristic but a result.
4. Because of changes in human activity, technologies and production processes used, the
ecological impact of the activities changes.
5. In Rotmans et al. (2000 and 2001) this phase is called the acceleration phase.
6. Transitions can be seen as system innovation at the highest level of societal systems.
These societal systems can be broken down into sub-systems (for example the regulatory,
the technological or the user sub-system) at which level system innovations take place.
Within these sub-systems, simple or singular innovations occur.
7. Other examples of system innovation are: biomass-based chemistry, multiple sustainable
land use (the integration of the agricultural function with other functions in rural areas)
and exible, modular manufactured construction (Ashford et al., 2001).
8. Discussions within the transition arena have to be based on a systems approach, which
allows for a more comprehensive and integrated analysis of the problem. The focus has
to be on the issue of what the structural origins of all the individual (symptomatic) prob-
lems of the dierent actors are.
9. See also section 7. The website www.energietransitie.nl gives an overview of the result of
discussions amongst Dutch actors related to energy, which resulted in the vision, images
and transition experiments.
10. The discussions are based on a participative systems analysis in which the dierent par-
ticipants contribute specic knowledge about specic parts of the system so that they
together develop an integrated image of the systems at hand and the main causal rela-
tions and dynamics within this system.
11. Context control may be viewed as a form of planning (see Meadowcroft 1997, p.27).
12. A nice discussion of adaptive policy with operational elements is Walker et al.s paper
(2001).
13. The criticism of anticipatory rationality should probably not be taken as criticism of
anticipation or a call for short-sightedness but as a criticism of a particular method for
dealing with the future: strategic planning. According to Club of Rome member
Mesarovic (2001), sustainability requires anticipatory democracy.
14. We take the view that path dependencies cannot altogether be prevented, each act will
inuence future acts in ways that are not entirely clear. Incrementalism, portfolio man-
agement and the stimulation by policy of robust solutions help to circumvent but not
altogether prevent the problem of suboptimal solutions. Lindblom (1997) proposes
relying on the intelligence of interaction by relying on partisan mutual adjustment.
15. Of course there is a danger of conservatism but forces of conservatism (in the form of
special interests, veto powers, and timid/unimaginative thinking) always play out them-
selves, at any time and place, as noted by Lindblom in a defence to his critics.
16. Not just incremental steps are taken. From a contemporary point of view, we have dis-
continuous policies and steps.
17. What is referred to here is what we described in section 5 as a basket of images. From all
scenario studies, it was concluded that biomass will play a signicant role in any future
energy supply system. The form in which biomass will be used however is uncertain.
Within the context of the overall vision for a sustainable energy supply, dierent images
have been developed (gasication, pyrolysis and biofuels), linked to dierent transition-
paths, which together must lead to the overall ambition.
18. Similarly, transition management could also be considered an example of a mixed scan-
ning approach, a hierarchical mode of decision-making that combines higher-order,
fundamental decision-making with lower-order incremental decisions that work out
and/or prepare for the higher order ones (Etzioni, 1986). The fundamental choices are
the long-term goals, the creation or abandoning of programmes for system innovation,
reliance on certain ways of decision-making. The fact that we can use dierent labels for
transition management shows that the ideas behind it are not new; what is new is the
operationalization of these ideas.
128 Strategies for sustainable system transformation
REFERENCES
Ashford, Nicholas, Wim Hafkamp, Frits Prakke and Philip Vergragt (2001),
Pathways to Sustainable Industrial Transformation: Cooptimising Competitiveness,
Employment and Environment, Cambridge, MA: Ashford Associates.
Beck, Ulrich (1997), The Reinvention of Politics: Rethinking Modernity in the Global
Social Order, Cambridge, UK: Polity Press.
Butter, Maurits (2002), Een handreiking voor transitiebeleid, Delft: TNO-STB.
Coenen, Frans (2002), The role of stakeholders in changing consumption and pro-
duction patterns, expert report for OECD seminar Improving Governance for
Sustainable Development, held 223 November 2001, available at http://www.
oecd.org/dataoecd/50/21/1940033.pdf.
Collingridge, David (1980), The Social Control of Technology, London: Pinter.
Dirven, Jan, Jan Rotmans and Arie-Pieter Verkaik (2002), Samenleving in
Transitie. Een vernieuwend gezichtspunt, LNV, ICIS en Innovatienetwerk
Groene Ruimte in Agrocluster, April.
Eising, Rainer, and Beate Kohler-Koch (1999), Introduction: network governance
in the European Union, in Rainer Eising and Beate Kohler-Koch (eds.) The
Transformation of EU Governance, London: Routledge.
Etzioni, Amitai (1986), Mixed scanning revisited, Public Administration Review,
January/February, 814.
Geels, F.W. (2002a), Technological transitions as evolutionary reconguration
processes: a multi-level perspective and a case-study, Research Policy, 31(8/9),
125774.
Geels, F.W. (2002b), Understanding the dynamics of technological transitions, a
co-evolutionary and socio-technical analysis, PhD thesis, Enschede: Twente
University Press.
Gibbons, M., H. Nowotny, C. Limoges, M. Trow, S. Schwartzman and P. Scott
(1994), The new production of knowledge: the Dynamics of Science and Research
in Contemporary Societies, Thousand Oaks, CA: Sage.
Grunwald, Armin (2000), Technology policy between long-term planning require-
ments and short-ranged acceptance problems. New challenges for technology
assessment, in John Grin and Armin Grunwald (eds), Vision Assessment: Shaping
Technology in the 21st Century Society. Towards a Repertoire for Technology
Assessment, Berlin and Heidelberg: Springer.
Gunderson, L.H. and B. Holling (2002), Panarchy: Understanding Transformations
in Human and Natural Systems, Washington DC: Island Press.
Kasemir, B. and M. Van Asselt (1999), Integrated assessment: multiple perspec-
tives in interaction, International Journal of Environment and Pollution, 11 (4),
40725.
Kemp, Ren, and Luc Soete (1992), The greening of technological progress: an
evolutionary perspective, Futures, 24 (5), 43757.
Kemp, Ren and Jan Rotmans (2001), The management of the co-evolution of
technical, environmental and social systems, paper for international confer-
ence Towards Environmental Innovation Systems, September 279 2001,
Garmisch Partenkirchen, Germany forthcoming in Matthias Weber and Jens
Hemmelskamp (eds) Towards Environmental Innovation Systems, Springer Verlag.
Kemp, Ren and Jan Rotmans (2002), Managing the transition to sustainable
mobility, paper for international workshop Transitions to Sustainability
through System Innovations, University of Twente, 46 July; published in Boelie
Transition management 129
Elzen, Frank Geels and Ken Green (eds), System Innovation and the Transition to
Sustainability: Theory, Evidence and Policy, Cheltenham UK and Northampton,
MA, Edward Elgar, pp. 13767.
Kemp, R. and J. Rotmans (2005), The management of the co-evolution of techni-
cal, environmental and social systems, in M. Weber and J. Hemmelskamp (eds),
Towards Environmental Innovation Systems, Heidelberg/New York: Springer
Verlag, pp. 3355.
Kohler-Koch, Beate (1999), The evolution and transformation of European
governance, in Rainer Eising and Beate Kohler-Koch (eds), The Transformation
of EU Governance, London: Routledge, pp. 1435.
Kooiman, Jan (1993), Governance and governability: using complexity, dynam-
ics and diversity, in Jan Kooiman (ed.), Modern Governance. New
GovernmentSociety Interactions, London: Sage, pp. 3548.
Latour, B. (1991), Technology is society made durable in J. Law (ed), A Sociology
of Monsters: Essays on Power, Technology and Domination, Routledge: London,
pp. 10331.
Lee, Kai N. (1993), Compass and Gyroscope. Integrating Science and Politics for the
Environment, Washington DC: Island Press.
Lindblom, Charles (1965), The Intelligence of Democracy, New York: Free Press.
Lindblom, Charles E. (1979), Still muddling, not yet through, Public
Administration Review, November/December, 51726.
Lindblom, Charles (1997), A Century of Planning, in Michael Kenny and James
Meadowcroft (eds) Planning Sustainability, London and New York: Routledge,
3965.
Loorbach, Derk, and Jan Rotmans (2002), Society in technology and technology in
society, paper presented at the international conference Engineering Education
for Sustainable Development (EESD), 24/25 October, Delft, The Netherlands.
March, J.G. and J.P. Olsen (1995), Democratic Governance, New York: The Free
Press.
Mayntz, Renate (1994), Governing failures and the problem of governability: some
comments on a theoretical paradigm, in Jan Kooiman (ed.), Modern Governance.
New GovernmentSociety Interactions, London: Sage, 920.
McElroy, M. (2003), The New Knowledge Management: complexity, learning and
sustainable innovation, Burlington: ButterworthHeinemann.
Meadowcroft, James (1997), Planning for sustainable development: what can be
learned from the critics?, in Michael Kenny and James Meadowcroft (eds)
Planning Sustainability, London and New York: Routledge, 1238.
Mesarovic, Mihaljo (2001), Evolution of sustainability concept, mimeo.
Midgley, G. (2000), Systemic Intervention: Philosophy, Methodology and Practice,
Dordrecht: Kluwer Academic Publishers.
NMP-4 (2000), Een wereld en een wil. Werken aan duurzaamheid, (A world and
a will. Working on sustainability), The Hague.
NRLO (1999), Innovating with ambition, opportunities for agribusiness, rural
areas and the shing industry, The Hague.
Rip, A. and R. Kemp (1998), Technological change, in S. Rayner and E.L. Malone
(eds), Human Choice and Climate Change, Volume 2, Columbus, OH: Battelle
Press, 32799.
Rosenberg, N. (1976), The direction of technological change: inducement mecha-
nisms and focusing devices, in Nathan Rosenberg, Perspectives on Technology,
Cambridge: Cambridge University Press.
130 Strategies for sustainable system transformation
Rotmans, J. (1998), Methods for IA: the challenges and opportunities ahead,
Environmental Modelling and Assessment, 3 (3), 15579.
Rotmans, Jan, Ren Kemp, Marjolein van Asselt, Frank Geels, Geert Verbong and
Kirsten Molendijk (2000), Transities en Transitiemanagement. De casus van een
emissiearme energievoorziening. Final report of study Transitions and transi-
tion management for the 4th National Environmental Policy Plan (NMP-4) of
the Netherlands, October 2000, Maastricht: ICIS and MERIT.
Rotmans, Jan, Ren Kemp and Marjolein van Asselt (2001), More evolution than
revolution. Transition management in public policy, Foresight 3 (1): 1531.
Scharpf, F. (1994), Community and autonomy. Multi-level policy making in the
EU, Journal of European Public Policy, 1, 21942.
Schnabel, P. (2000), SCP presentation, The Hague.
Senge, Peter (1990), The Fifth Discipline. The Art and Practices of the Learning
Organization, New York: Currency Doubleday.
Walker, Warren E., S. Adnan Rahman and Jonathan Cave (2001), Adaptive poli-
cies, policy analysis, and policy-making, European Journal of Operational
Research, 128, 28289.
WCED (1987), Our Common Future, Oxford: Oxford University Press.
Weiss, Andrew and Edward Woodhouse (1992), Reforming incrementalism: a con-
structive response to the critics, Policy Sciences, 25(3), 25573.
6. Adaptive management to restore
ecological and economic resilience
in the Tisza river basin
Jan Sendzimir, Piotr Magnuszewski,
Peter Balogh and Anna Vri
INTRODUCTION
While many of the chapters in this volume deal with the governance of
socio-technical systems to make them more sustainable, the management
of complexity has been an important issue in the area of socio-ecological
systems for quite some time. Adaptive management (AM) has been devel-
oped as a structured learning process to deal with uncertainty of socio-eco-
logical development in natural park management, sheries and so forth.
These approaches can be viewed as instances of reexive governance
since they acknowledge the impossibility of planning and control in a
complex and changing world. They anticipate unintended eects of man-
agement strategies and actually build on them in an understanding of
management as learning.
This chapter discusses the set-up of an Adaptive Management process
for renaturalisation in the Tisza River Basin (TRB) in Hungary where it
has been applied in order to deal with increasing ood risks which repre-
sent a second-order problem of modern hydro-engineering and indus-
trial agricultural practice. River basins rarely gain our attention except
in the context of great oods, but their future revolves around far more
than pulses of water. Their integrity and development also depend on
how their communities interact with ows of people, commerce, and
ideas. Sustainable progress in the evolution of these regional socio-
ecological systems (sensu, Walker et al. 2000) depends on the resilience and
dynamism of their heritage, in all its ecological, economic and social
dimensions. Over the past century, regional decline, and even collapse, of
the heritage of major river basins on every continent signal our failure to
understand or manage all the intertwined processes that drive these
complex systems.
131
132 Strategies for sustainable system transformation
To answer this challenge, it is neither science nor policy but their inte-
gration in a learning cycle along with local knowledge and practice, that will
help us adapt to the uncertainty and surprise of complex, large-scale
systems like river basins, coasts, continental plains and mountain ranges.
None of the perspectives along the way can be neglected, and small
steps seldom taken address how we steer the process of learning and man-
aging as we integrate these views, reect and experiment. Adaptive man-
agement is an on-going experiment for developing such a steering process
(Gunderson et al. 1995, Gunderson and Holling 2002, Sendzimir et al.
1999, 2003).
Adaptive management is an example of a reexive governance (RG)
approach, starting from some of the same premises and pursuing some of
the same methods as those described in the Introduction to this volume. AM
begins from the assumption that surprise is inevitable in complex evolving
systems and will confound and ultimately defeat eorts to control and
eliminate uncertainty through science, policy, or practice. If the uncertainty
inherent in natural and social evolution is compounded by inexible
attempts to control ambient variability, then management only increases
uncertainty. However, the eect is further amplied because it often emerges
after a period of initial success lulls society into complacent dependence on
short-sighted methods. Delays are potent sources of destabilizing surprise in
evolving systems (Sterman 2000). The adaptive label is partly an attempt to
recast management in the wider role as a nimble experimenter, with the
patience to consider long-term consequences. As such, management takes
place in a learning cycle as the driver and implementer of the experiments
proposed by science to probe how the world is changing and by what means
we can adapt to that change.
RG and AM jointly advocate and pursue a number of adaptive strategies
(see Table 1.1 in the Introduction). For example, both apply transdiscipli-
nary knowledge production as a means to address co-evolutionary rela-
tions between dierent dimensions (ecological, economic or social) across
multiple scales. AM invokes modelling techniques (conceptual and formal)
to integrate the insights of multiple perspectives across these scales.
Likewise, as mentioned above, AM embraces uncertainty through a cycle
that links hypothesis with policy with implementation with monitoring. In
this way we can adapt how we understand and how we react to the shifting
re-alignments of inter-relations within socio-ecological systems. The RG
goal of facing the dangers of path dependence and technological lock-in
(Arthur 1983) through anticipation of long-term systemic eects is often
addressed by looking at long-term system dynamics and sensitivities
through the use of formal models in AM exercises. AM processes are fun-
damentally iterative and participatory and have been aimed at RG strategy
The Tisza river basin 133
UKRAINE
SLOVAKIA
TISZA
Village of
Nagykr ROMANIA
T
HUNGARY i
s
z
a
R
i ROMANIA
v
e
r
SERB REPUBLIC
DANUBE
Figure 6.1 The Tisza river basin with tributaries in catchments in the
Carpathian mountain range across portions of ve dierent
national territories (Romania, the Ukraine, Slovakia, the
Federation of Serbia and Montenegro, and Hungary)
134 Strategies for sustainable system transformation
these uctuations spread like ood waves over vast areas of oodplain, sus-
pended there for considerable periods by thick, impermeable sediments.
Pre-industrial societies managed to develop cultures that could utilize and
thrive on massive, periodic ooding, by building sheries and fruit enter-
prises that made them one of the richest regions of Hungary before 1850
(Andrsfalvy 1973, Molnr 2003, Paget 1850, Tth 2002). How could such
a self-sucient region collapse to its present state of economic and social
poverty, and be threatened increasingly by the very oods on which it used
to thrive?
This chapter examines some of the sources of uncertainty from both
human and environmental sources that have combined to produce such
surprising regional collapses. It then considers adaptive management (AM)
as a reexive governance approach, which has proven useful in addressing
uncertainty in river basins and describes its implementation in the Tisza
River Basin.
Uncertainty in Nature
melt all glaciers that feed the Rhine river by 2070 (source: www.
waterandclimate.org, accessed June 2004) and consequently decrease steady
melt water inputs by 15 per cent in summer. CC impacts are expected to drive
increases in rainwater uctuations, such as a 20 per cent increase in winter
precipitation. Rising temperatures can drive higher throughput of the
hydrological cycle with higher rates of evapotranspiration (water ux
through plants and the soil back to the atmosphere) and precipitation (IPCC
2001). Higher precipitation rates will probably lead to increased spatial het-
erogeneity of precipitation, concentrating rainfall into tighter patterns of
more intense rain and thereby increasing the likelihood of ooding.
Such intensication of the volume and spatial pattern of rainfall is likely
to be amplied by human conversion of oodplain land cover from wet-
lands to agriculture or habitation. This lowers contributions of wetland
functions that buer uctuations in gases, particulates, nutrients, toxics,
water, temperature and kinetic energy (for example wave and wind). For
example, wetland productivity contributes to carbon sequestration and
evapotranspiration uxes that moderate air and water temperature vari-
ability. As a result, cooler micro-climates over river valleys and marsh plains
sustain higher rates of smaller, local rainfall events, thus buering the
basin from more intense rain upstream that generates more extreme ood
events (Pokorny et al. 1998, Ripl 1995). The combined decline of steady
base inows and increase in rain-driven input variability should increase
ood volumes and peak levels substantially. Along the Rhine river, ood
peaks should rise 20 per cent higher than previously experienced (source:
www.waterandclimate.org, accessed June 2004).
Uncertainty in Society
The sources of surprise and uncertainty in the TRB emerge not only from
the complexity of Nature but also from the ways that human intervention
has lowered the basins resilience to change. We consider here the ways that
the basins structure and function have been altered through shifts in land
use and basin morphometry (shape in three dimensions, such as elevation,
area of oodplain) by hydro-engineering.
By the nineteenth century the promise and power of the ideas of the
Enlightenment began to shake and recongure commerce and society in
Europe. As Davies (1996) notes: There is a dynamism about nineteenth-
century Europe that far exceeds anything previously known. Europe
vibrated with power as never before: with technical power, economic power,
cultural power, intercontinental power. A rising tide of energy, raw mater-
ials and people converged on growing urban agglomerations that became
regional manufacturing centres for new products and economic opportunity,
The Tisza river basin 137
which in turn drew yet more people and resources from the countryside.
Nations grew following the new, modern industrial model, swelling every
index of growth, including urban populations and the demand for food.
This demand was seen as a new opportunity by the major landholders on
the TRB. The Habsburg nobility, including the Emperor, as well as
Hungarian aristocrats and nobility, to some extent, began to convert the
agricultural land of the oodplain from a diverse polyculture of fruits, nuts
and maize to larger elds with monocultures of wheat (Kovcs 2003).
There was far more prot in providing a durable grain for the main staple
of urban populations, bread, than in handling the enormous variety of per-
ishable fruits, which were far more daunting items to store and ship. Within
a few decades Hungary would become the rst wheat-exporting nation in
Europe, earning large prots in the short run but at a far higher price in the
long run than originally imagined.
Agricultural transformation immediately raised two challenges. First,
no variety of wheat known then could withstand ooding or even high
groundwater levels for any length of time, so the Tisza had to be tamed to
keep its surges o the oodplain. Second, the twisting meanders of the
Tisza across the Great Plain had to be straightened to shorten shipping dis-
tances and make export protable. Changing the shape and content of the
elds led to the Herculean task of reshaping the entire oodplain and the
river channel itself, ocially conceived and implemented by the Austro-
Hungarian Empire as the Vsrhelyi Plan. Such radical surgery on the
landscape was an aront to tradition and sparked great controversy (see
Vay 2002), but eventually the Plan was pushed through in 1870 and the
wood and coal-red power of the nineteenth century began to sculpt the
Tisza River valley with the clean and smooth lines of an engineers ruler
(Figure 6.2).
To grow, protect and export wheat the Tisza oodplain was modied
for ood defence, and the river was straightened and deepened. Wheat
export became possible when the ow of the Tisza was concentrated in a
single, deeper channel that cut a smoothed arc through the maze of twists
and turns of a braided rivers oodplain. Overall the Tiszas length was
shortened by more than 400 km (Botri and Krolyi 1971). In addition,
the original Vsrhelyi Plan began a process that over a century eventu-
ally protected 97 per cent of the basin at risk from ooding with over
4500 km of primary and secondary dikes along the Tisza and its tribu-
taries. Overall, the dike defence system lowered the oodplain area by
more than an order of magnitude, from 38 500 km2 to 1800 km2 (Horvth
et al. 2001).
Simply mentioning that a major Tisza ood could inundate 17 per cent
of Hungary (Vradi 2001) made the threat so palpable that it locked up
138 Strategies for sustainable system transformation
a.
b.
(between 1877 and 1933) to once every three to four years (between 1934
and 1964) to almost every other year over the last decade. An engineering
race to raise the dikes started after each ood, as each subsequent ood
exceeded the previous one in height (Fetivizig 2000).
Practically in step with mounting ood statistics, regional development
has also climbed since the mid-nineteenth century, and the clash between
these two rising trends has created ever larger losses. The infrastructure of
towns and row crop farms burgeoned and spread into the ood danger
zone, the TRB oodplain, with reassurance from the apparent security of
a dike and canal ood defense system. The security promised by hydro-
engineering might hold for a decade or two, but ever-larger oods breached
these defenses, and as a result devastated homes, roads and crop elds.
Damage to built capital and commerce from one major ood event could
reach as high as approximately 25 per cent of the riverine basin GDP or
79 per cent of the national GDP (Halcrow Group 1999). The most vul-
nerable groups, such as low income farmers, are often hit hardest, deplet-
ing their scarce reserves and pushing them to the brink of bankruptcy
(Linerooth-Bayer and Vri 2003). The force of major oods perennially
rises to surpass expensive eorts to engineer, reinforce and defend the dike
system. Without a very quick, competent and heroic capability to react to
oods in crisis, the failures of hydro-engineering would be even more
apparent (ibid.).
Policy
formulation
(as test of hypothesis)
Assessment Management
action
Policy
implementation
Monitoring and
evaluation
but the following four phases oer a template of activities that integrate
research, management and practice in a cycle (see Figure 6.3).
at the expense of awareness about where the system is going. For example,
if initial policies achieve high production and prot levels, one could bank
on maximizing the prot of such success by cutting research costs, but only
if one was sure of where the system is going. The AM process strives to
avoid this pathology by broadening implementation to mean the testing
and evaluating of hypotheses (policies). This prevents the intent of policy
from being changed during implementation, and shifts the search for
eciency from cost reduction to checking whether management actions
were executed as anticipated (Gunderson, 1998). This gives implementa-
tion a disciplinary rigor of consistency in execution, because otherwise the
test of the policy becomes meaningless, and one loses the power to gain new
information about the system.
For example, in the mid-1990s when the Canadian government re-opened
the Newfoundland cod shery primarily for political reasons (unemploy-
ment), it severely endangered the scientic process of selecting criteria by
which to measure the recovery of the cod population and thereby set policy
about when to open or close the shery. The policy disconnected permission
to sh from the dynamics of the sh population, making it impossible to test
ideas based on evidence. An AM process would have posed policies as tests
of what action (or lack of action) can re-establish the resilience and carry-
ing capacity of the shery.
especially the assessment phase, modelling is not used to blend away the
messy entanglements of relations to distil them down to one rational vision
of a system. Distillation and simplication are possible, but within a far
wider context that can also generate a diversity of views at the end.
First, the modelling phase often addresses in parallel not one but many
dierent hypotheses about how a system operates to create the problem of
concern to stakeholders. Thus, stakeholders may consider not one but many
models that emerge from dierent world views, trainings and experiences.
The pressure to distil these many visions further stems from the urgency for
the group to move on from assessment and formulate policy. Thus a small
subset (perhaps one or two) of the full set of hypotheses (and perhaps
dierent models) are chosen as the basis for policy recommendations. That
subset has not been elevated to the rationalist truth but has been selected
as the most compelling sub-set of questions for the project to test. This
selection is done with full awareness of the fact that when the cycle of man-
agement implementation of policy, monitoring and evaluation again
returns to the assessment phase, other hypotheses and models may now
appear to be the most compelling. Therefore, we do not end with one model
and one explanation. Rather, the full set of models is retained while the
current favourites are tested in policy implementations. Furthermore, as
opposed to rationalist conclusions, modelling is used to dig down to a less
rational sub-terrain to reveal to stakeholders the implications of their
assumptions. The models are pictures of a world that is based on their
assumptions. Modelling exercises are opportunities for stakeholders to bore
beneath the ocial faade of conventional explanations to encounter what
their, often not very rational but profoundly powerful, sub-conscious
assumptions are. Profound questioning of ones assumptions is a major step
to honestly and earnestly examining the world through other assumptions,
through the eyes and hearts of others. As such, it oers a key to grid-lock
and confrontation that opens the door to compromise. Such compromise is
facilitated by the AM atmosphere of experimentation. If every next step is
not a proof but a test, then we are less likely to become stuck on one posi-
tion, but will continue to experiment and revise how we see the world and
how we do things.
Adaptive management is not unique. It is similar to an entire family of
approaches used to understand and manage uncertainty in ecosystems and
society (see Walker et al. (2002) for a current survey). It reects the theor-
etical and methodological training in ecology and modelling that the pro-
fessionals used as they developed AM as a framework to deal with regional
environmental crises. It represents ecologists attempts to communicate and
work with a variety of disciplines as well as the governance and business
sectors of society. It continues to evolve and mature as theory and methods
146 Strategies for sustainable system transformation
of economics and the social sciences are incorporated in its application. For
example, the study of institutions has gained equal status with that of
natural sciences and economics in the articulation and bounding of the
question at hand and in formulating hypotheses and policies (see Ostrom
(1999) and Andersson and Jansson (2004) as examples of the growing
inuence of institutional analysis in AM).
AM has not always succeeded. It has been stymied by failure to reach a
conclusion or to produce meaningful and concrete results, because partici-
pants can indulge in endless discussion and modelling if they are not vigi-
lant or are incompetently guided by facilitators. Some applications of AM
have been criticized as too theoretical and therefore impractical to act as
a framework for people without advanced education or sophisticated
means of discussing complexity. This latter point may depend on the com-
plexity of the system and the questions at hand, as well as on the skill and
patience with which AM is applied. However, AM still represents the best
opportunity to create truly multi-disciplinary, multi-sectoral cooperation
that can unite research, policy and praxis in a learning cycle.
For practitioners interested in applying AM the daunting reality is that a
scientic, transparent participatory process is not enough by itself. Success
in engaging the uncertainty of complex, evolving systems requires a vast
array of factors (Gunderson et al. 1995, Light and Blann 2000) that many
people might sweep into one broad category: luck. Experience in the Odra
river valley in Poland (Sendzimir et al. 2003) suggests that little can be
accomplished without one critical factor: a foundation of trust built over
years of fruitful collaboration between actors (government, NGOs, scien-
tists) and stakeholders. If this trust can be carefully and respectfully
expanded to embrace novel ideas and methods (such as AM, modelling and
even the scientic method) then real innovation in participatory science and
management may be possible. However, dozens of factors can potentially
derail this innovation (see Figure 1.12 in Sterman 2000). It often founders
simply because funding horizons are too short to sustain research, commu-
nity dialogue, policy implementation and monitoring long enough for real
learning to occur and to respond to long-term change.
Government, academia and NGOs in Hungary are coming to the same con-
clusion as the Dutch (Van Stokkom et al. 2002, van der Molen et al. 2002)
about ood security: hydro-engineering cannot, by itself, maintain ood
safety and stop the degradation of the biological and cultural heritage of
The Tisza river basin 147
the Tisza River Basin (TRB). The prospect of yet another round of massive
investment in dikes after failure of the defence infrastructure during the last
major ood in 2001 brought about a fundamental re-examination of alter-
natives. Several years of discussion produced a new ood protection initia-
tive in the TRB, the new Vsrhelyi Plan. This initiative is named after the
engineer who fathered the original reshaping of the Tisza in the nineteenth
century, as an ocial faade to maintain some solidarity with past engi-
neering traditions. However, for the rst time a response to the ood crisis
does not recommend raising and reinforcing dikes. Instead, the potential to
remove the dikes and use certain areas of the original oodplain is being
developed. This was never an easy consideration for water engineers in
Hungary, because dike failure leading to loss of life was the fastest way to
lose ones job and even end up in jail. But the series of dike failures over the
past century pointed only to ongoing failure as climate change and other
factors increased ood frequency and intensity.
By utilizing the oodplain for ood control, the new Vsrhelyi Plan
raises a new set of questions. Should the oodplain only be used for ood
control? What are the opportunities to re-establish the traditional roles of
ooding to store water on a dry landscape and to sustain biodiversity as
well as sh productivity? And where and how should people live and earn
a living if they must move o the ooded parts of the plain?
Beginning in 2002 a new research initiative (Kovcs 2000), sponsored
and coordinated in part by WWF Hungary, set out to answer some of these
questions in the middle reaches of the Tisza around the village of
Nagykr. It aims to explore questions related to ecology, agriculture and
sheries as part of a comprehensive assessment of the impacts of re-con-
necting the oodplain with the Tisza river channel. We now consider some
of the key elements of this programme within the context of proposing how
adaptive management could serve as a framework for linking these research
projects with management interventions and eorts to start local
oodplain-based enterprises.
The village of Nagykr is located in the middle reaches of the Tisza River
(see Figure 6.1). The village of some 10 000 people is situated on a ridge of
high ground that extends along an eastwest axis to within a few hundred
meters of the Tisza river as it swings in a wide arc from north to west. The
village ridge is surrounded by low areas that, prior to hydro-engineering,
were ooded relatively frequently (almost annually). The spatial hetero-
geneity of depressions of dierent depths in the oodplain oered a wide
variety of water storage options. Local people developed a culture that used
148 Strategies for sustainable system transformation
this dynamic water storage potential to sustain a mosaic of a rich and diverse
landscape that supported a variety of uses: orchards, grazing meadows and
sh ponds.
People today would scarcely recognize the TRB landscape and culture
that existed before the industrial revolution. Far from its current depressed
appearance, the economy of the TRB was thriving, self-sucient and rich
enough to export fruits, vegetables, timber and sh (Dr Veres Nandor,
Mayor, village of Nagykr, personal communication). Our modern sense
of diversity, tuned to the three or four varieties of any domestic fruit a
supermarket might oer, would be amazed to nd the hundreds of varieties
of fruit and nuts traditionally grown in the TRB.
Such extreme variety in agricultural produce is neither an extravagant
gesture nor some slavish adherence to tradition. This ancient tradition has
ourished for centuries in the TRB (Andrsfalvy 1973, Molnr 2003, Paget
1850, Tth 2002). It represents a strategic response by which people have
adapted to the variability in nature in every region of the globe since the
Neolithic era (McNeely and Scherr 2003). Plants and people must contend
with severe uctuations in water within the TRB, from sudden and intense
ooding that lasts for weeks over hundreds of square kilometres, to droughts
that may last for years over the entire basin. These sudden ows and slow
ebbs produced a huge variety of ooding durations over the dierent eleva-
tions on the oodplain, and the web of sluices connects the many depressions
in the landscape with the river channel, which links sh nursery areas with
the river.
A diversity of capabilities has been the traditional answer to the uncer-
tainty inherent in such extreme uctuations from wet to dry in river valleys.
Farmers along the Tisza developed fruits that were dierentially adapted
to variable durations and intensities of ooding and drought. They also
developed a local corn variety so well adapted to a short growing season
that one could plant it after the annual spring oods and still harvest a
mature plant by autumn. In comparison to the pesticides and fertilizers
required by the relatively few global varieties that consumers from the UK
to Austria recognize, these local TRB varieties have proven far more robust
against insect pests and diseases, and did not require as much pruning or
intensive care (Siposs and Kis 2002). However, continental and global var-
ieties are rapidly replacing them.
Decades of scientic research have examined the natural and social history
of the Tisza basin. This scientic legacy oers a wealth of information, but
many local people are oended by how few concrete results all this scientic
The Tisza river basin 149
The capacity to adapt and innovate is at a low tide in the TRB. Among local
stakeholders, little trust remains in central government, outside experts or
between the people themselves to solve their dilemma. This results from a
combination of prolonged and recurrent ood defence failures, degrada-
tion of natural and social heritage, economic decline and the lack of demo-
cratic traditions in a centralized political system. No sustainable future can
The Tisza river basin 151
Setting
Mapping objectives
assumptions
2 3
Problem
articulation 1 4 Finding
indicators
5
Designing
Monitoring 7
policies
evaluation
Implementation
process should oer the means for collaboration across the community to
investigate, intervene and act, and measure progress. We propose that the
AM experience of the Odra and Barycz Rivers (Sendzimir et al. 2003),
while preliminary, still suggests that local stakeholders will provide solid
experience and support if they become engaged in all phases of a trans-
parent process, especially monitoring of indicators of policy performance.
For example, an AM dialogue about the challenges of establishing a land-
scape park along the Odra River revealed the key indicator of successful
implementation as the frustration level of the foresters who were left by
default to make a similar plan work along the river some years before. As
higher governmental levels mandated a park without any extra resources
(funding, personnel or equipment) to support its establishment or execu-
tion, foresters were forced to divide their resources between normal duties
and supporting the park. Failure to work with all stakeholders to design the
park programme or to fund it eventually led to the parks collapse into
disuse as the foresters could no longer manage two jobs. Subsequently,
these same foresters came to stand against any similar new plan.
Local stakeholder experience can prove vital to experiments that explore
how to establish an economically viable culture on a re-naturalized
oodplain in the TRB. Key research tasks include: testing what varieties of
fruit and nut trees can withstand extremes of ooding and drought, what
sluice and channel morphologies establish hydroperiods that support pro-
ductive sheries, what breeds of cattle can thrive on oodplain wet meadows.
Such questions will provide stakeholders and policy makers with the most
compelling evidence of the potential oered by denaturalizing rivers. As
local stakeholders become condent, through their participation, that eco-
nomic (and hence social) development is well addressed in the AM process,
they are more likely to support the expansion of inquiry into the ecological
impacts of re-establishing river oodplain hydroperiods some of which are
complex and dicult to communicate to non-scientists. However, AM also
oers the opportunity for such questions to become more understandable
when a wider portion of the community has participated in the framing of
the questions and the eld monitoring of the experiments.
We have adapted the general adaptive management framework to suit the
needs of the project (see Figure 6.4). The modied framework consists of a
series of steps (see Magnuszewski et al. 2005 for details) that begins with a
vital step whose imposing delicacy and complexity make its proper imple-
mentation and practical description rare: forming the group of participat-
ing stakeholders and actors. This step may seem simple, either because it
seems straightforward to attempt to include every viewpoint or if one antici-
pates that one has to accept whoever is available, since few people can aord
to attend or want to participate. However, including people who refuse to
The Tisza river basin 153
154
events channel
level + productivity production
+ + + + + +
+
Silting in
River bed
side
erosion
+ channels
Normal water
+ level
Fruit tree
Groundwater diversity
level elevation +
+
Hydro-connectivity and
water storage on
floodplain
Figure 6.5 Conceptual model (mental model map) of key variables and causal loops that interact to aect Tisza
river oodplain resilience to climate related hydro-dynamic variability
The Tisza river basin 155
using them to challenge and reinforce one another. For example, we can
mesh the phases of bounding and then measuring the problem by linking the
results of the conceptual modelling phase with the process of dening indi-
cators. As conceptual modelling provides the big picture of the problem
and helps to overcome human information processing limitations, it also
makes the process of indicator selection more rigorous and leads to more
comprehensive results.
Causal loop diagramming will therefore be used as a discussion guide in
linking variables and slowly developing a graphic image of the system struc-
ture. As the web of relations takes shape, certain sections become more
understandable as identication of reinforcing and balancing feedback
loops reveals the macrostructure of the system. In this way the vast and dense
thicket of links is reduced to a smaller set of clusters of variables that tend
to interact with each other. The groups desire to focus on specic parts of
the model often generates sub-model diagrams that clarify some of the
causal details underlying the more aggregate variables and relations in the
general model.
Causal loop diagrams can been used to elicit from the participants their
underlying assumptions and mental models and to express it graphically in
the form of a map that contains key factors and processes in the region.
While a range of exercises for participating individuals and groups can be
used to dene underlying paradigms, assumptions or world-views before
scoping and modelling begins (Hare and Pahl-Wostl 2002), the give-and-
take of discussion in dening the models can also help to reveal the mental
frameworks participants rely on to judge and predict. Fundamental world
views present great opportunity and great risk. Because they lie so close to
a participants emotional core, their revelation can open profound trust and
cooperation, breaking political stalemates. On the other hand, exposure, or
even the possibility of exposure, can also provoke hostility and resistance.
Such revelations are often best left to emerge on their own from a well run
AM process that makes it natural and obvious to explore all inuences on
the problem, even how we think and construct the world in our minds.
However, some very skilled facilitators may try consciously to assess the
paradigms that participants bring to the AM process.
The mental map functions as the knowledge container, open and easily
modiable as new facts or ideas are provided or revealed during the process.
Graphic tools such as diagrams and mental maps open the discussion of
complex systems to include people who nd verbal descriptions too com-
plicated or too long and involved. Often a single map replaces pages of text
required to describe all the variables and their interactions. In every dis-
cussion a map (or maps) on the wall or in the computer will plainly show
the complex relations between nature and society in the Tisza landscape.
The Tisza river basin 157
Using the map will provide discipline for the group discussions in a
positive sense as follows: (1) dierences (and agreements) in opinions will
be articulated much more precisely; (2) gaps in understanding will be dis-
covered more eciently.
Figure 6.5 illustrates an exercise in advance modelling in preparation for
the TRB initiative. Building from the centre outward, the model evolved by
addressing key issues as clusters of interactions and described derivative or
spin-o issues by attaching new clusters to the emerging core structure. The
evolution of model structure also recapitulates the chronology of events.
As the web of relations takes shape, certain sections become more under-
standable as identication of reinforcing and balancing feedback loops
reveals the system macro-structure. The groups desire to focus on specic
parts of the model often generates sub-model diagrams that clarify some
of the causal details underlying the more aggregate variables and relations
in the general model. While model evolution is impossible to predict, we
may anticipate that a model of TRB factors and interactions will likely
examine some of the following areas. Provided that it does not create a new
set of biases for a discussion facilitator or a modeler, modelling a system in
advance of stakeholder participation can help prepare him/her by chal-
lenging assumptions and exploring dierent interpretations of variables
and model structure. Such preliminary exercises can deepen ones intuition
about how to avoid conceptual traps when exploring system structure and
dynamics.
CONCLUSIONS
No system of analysis, policy, or practice will ever eliminate surprise and
uncertainty. Innovation and novelty as well as wicked problems incessantly
emerge from evolving systems of nature and humanity, and will continue to
do so. Our understanding of complexity suggests that control and cer-
tainty are romantic notions born on the rising tide of the optimism of
the early Enlightenment. Radical and profound change from climate and
technology (genetic engineering), to name only a few sources, will certainly
surpass our current imagination of how the environment and society can be
recongured in unprecedented ways over the next few decades. Our respon-
sibility to address the impacts of evolution through new ways of learning,
managing, and discussion must engage uncertainty as a stimulus to explore
innovations and not as a basis of apprehension and apathy.
Adaptive management oers a framework to meet that challenge by
addressing, among other things, the ve strategic requirements of reexive
governance. It creates trans-disciplinary knowledge production (Strategy 1)
158 Strategies for sustainable system transformation
REFERENCES
Kovcs, Z.C. (2003), Sustainability Along the Tisza River, Oxford: Oxford Brookes
University, p. 88.
Light, S. and K. Blann (2000), Adaptive Management and the Kissimmee River
Restoration Project, Collaborative Adaptive Management Network, accession
date: 2004. Website address: http://www.iatp.org/AEAM/.
Linerooth-Bayer, J. and A. Vri (forthcoming), Floods and loss sharing: A clumsy
solution from Hungary.
Magnuszewski, P., J. Sendzimir and J. Kronenberg (2005), Conceptual modeling
for adaptive environmental assessment and management in the Barycz Valley,
Lower Silesia, Poland, International Journal of Environmental Research and
Public Health, 2(2), 194203.
McNeely, J.A. and S.J. Scherr (2003), Ecoagriculture: strategies to save the world and
save wild Biodiversity, London: Island Press.
Molnr, G. (2003), Arteri gazdalkodas Magyarorszagon, proceedings of Szovetseg
II evfolyam 2 szam, 17 January, p. 6.
Odum, H.T. (1996), Environmental Accounting: Emergy and Environmental Decision
Making, New York: John Wiley, pp. 370.
Ostrom, E. (1999), Coping with tragedies of the commons, Annual Review
Political Science, 2, 493535.
Paget, J. (1850), Hungary and Transylvania with Remarks on their Condition, Social,
Political, and Economical, New York: Arno Press.
Parmesan, C. and G. Yohe (2003), A globally coherent ngerprint of climate
change impacts across natural systems, Nature, 421, 3742.
Peterson, G., G.A. De Leo, J.J. Hellmann, M.A. Janssen, A. Kinzig, J.R. Malcolm,
K.L. OBrien, S.E. Pope, D.S. Rothman, E. Shevliakova and R.R.T. Tinch (1997),
Uncertainty, Climate Change, and Adaptive Management, Conservation
Ecology [online] 1(2), p. 4. available at http://www.consecol.org/vol1/iss 2/art4.
Pokorny, J., W. Ripl and M. Eiseltov (1998), Impacts of deforestation and
drainage on climate: landscape management and policy implications, report on
proceedings of WaterTech (conference promoting technology, science and busi-
ness in the water industry), Brisbane, Australia pp. 112.
Ripl, W. (1995), Management of water cycle and energy ow for ecosystem
control the energytransport-reaction (ETR) model, Ecological Modelling, 78,
6176.
Rittel, H. and M. Webber (1973), Dilemmas in a General Theory of Planning,
Policy Sciences, 4, 1559.
Rydzak, F., P. Magnuszewski, P. Pietruszewski, J. Sendzimir and E. Chlebus (2004),
Teaching the Dynamic Balanced Scorecard, Proceedings of the 22nd interna-
tional conference of the System Dynamics Society, 259 July 2004, Keble
College, University of Oxford, Oxford, UK.
Sendzimir, J., H. Gottgens and J. Pokorn (in review). Assessing multiple scales of
climate change impacts on northern-hemisphere freshwater wetlands, Climatic
Change.
Sendzimir, J., S. Light and K. Szymanowska (1999), Adaptive Understanding and
Management for Floods, Environments, 27(1), 11536.
Sendzimir, J., P. Magnuszewski and J. Kronenberg (2003), A scientic based frame-
work for sustainability indicators, World Wide Fund for Nature Report con-
tract number DBU AZ 18902, Berlin: WWF.
Siposs, V. and F. Kis (2002), Living with the river: LIFENature project in the Tisza
oodplain, Budapest: WWF Hungary, p. 24.
The Tisza river basin 161
INTRODUCTION
162
Sustainability foresight 163
we face dierent conditions in all three points in the case of complex prob-
lems such as the long-term transformation of utility systems.
Since the 1960s both the business and public policy sectors have systemat-
ically employed foresight methods to explore the embedding of strategies
in dynamic contexts (Godet 1987; Ringland 1998). The approach has
become popular through the Shell oil company which used it to deal with
the uncertainties of their business environment that cropped up with the oil
crises in the 1970s.
Foresight is about anticipating possible future developments in a
focal area. It diers from forecasting, however, because it recognises the
166 Strategies for sustainable system transformation
impossibility of predicting the future due to the complex dynamics that are
involved in bringing it about. Foresight conceptualises the future as open,
not determined by natural necessities, but contingent and inuenced by
human action. The future is therefore seen as malleable and apt to strate-
gic shaping, not to fatalistic adaptation. Foresight is not a process of fore-
casting the future but rather an attempt to explore the space for human
actions and interventions to shape the future. Foresight is aimed at pro-
ducing orientations rather than predictions; it provides guidance to all
actors and reduces uncertainty (Renn 2002, cited in Borup 2003, p. 3).
Foresight is not about nding out about one most probable path of devel-
opment, but rather it entails the construction of a range of dierent, equally
plausible paths of future development. Such paths are derived from the
partly contingent interaction of various factors of inuence. Foresight is
also referred to as a scenario approach to system analysis (Gallopn 2002;
Berkhout and Hertin 2002). It is qualitative and explorative and does not
aim at numerical predictions. As such it does not focus on exactitude but on
a comprehensive account of the diversity of factors from society, technol-
ogy and nature that work together in shaping transformations in the real
world. For this purpose foresight exercises seek to make use of the distribu-
ted knowledge, expectations and understanding which are contained in the
diverse perspectives of present-day actors on developments of common
concern.
By putting these expectations in the form of scenarios, they have an eect
on present-day actions and thus feed back on the development itself. The
actual results of foresight activities are therefore not the more or less prob-
able stories about alternative futures on their own, but the repercussions they
have in social interaction processes in the present (Truer et al. 2003). This
may be that expected opportunities enhance actions, which in turn support
developments that spur their actual realisation (self-fullling prophecy) or,
vice versa, that expected risks call for preventive action, which makes them
less likely to occur (self-defeating prophecy). Foresight processes thus poten-
tially shape the developments they set out to explore. As such they become
a strategic device in shaping socio-technical transformation.
How foresight, which yields alternative futures, aects present actions,
however, diers remarkably from the eect of forecasting, which yields one
most probable future. Beyond self-fullling or self-defeating eects, the
pluralistic vision which is constituted by the alternative scenarios that are
the outcome of foresight exercises has a self-reecting eect. The variety of
future developments across the scenarios calls the inevitability of each
single scenario into question and points out the indeterminacy of long-term
transformation. As such foresight can prepare decision makers for alterna-
tive courses of development and prevent premature lock-in to specic
Sustainability foresight 167
This rst phase resulted in four elaborated scenarios that represent alter-
native future structures of utility provision as well as a set of detailed descrip-
tions of highly relevant factors inuencing the transformation process. Both
developed from the interaction of heterogeneous perspectives on utility pro-
vision. By means of this procedure it is possible to overcome some limit-
ations often set by particular institutional perspectives like, for example, the
one of technology development, business or consumer protection. This
yields a trans-disciplinary and trans-professional view of the system in
which processes become central that are under everyday conditions often
externalised (such as societal acceptance for new technologies).
Another eect of the collective scenario construction is the creative
destruction of expectations and visions of future development which were
taken for granted by participants. Routine thinking about how things
unfold and what will come next could be replaced with a fan of contingent
alternatives which would each require specic strategic responses. This plu-
ralisation of the future can work as a particular kind of steering through
visions (Canzler and Dierkes 2001; Brand 2002). In this case it is not the
coordinating force of visions which become embraced as commonly held
expectations and translated into agendas (van Lente 1993; Konrad 2004),
but the ambiguity of multiple expectations that may inuence general
action orientations towards experimenting, adaptivity and cooperation.
future developments with respect to their eect on these values and a broad
range of aected actors engages in a discursive assessment of opportunities
and threats which have to be taken special care of in future transformations.5
The result of the assessment phase is the explication of threats and
opportunities of transformation from the perspective of the various actors
who are potentially aected by them. In this way critical aspects can be
identied, which form starting points for the development of adequate
strategies. Such an open-ended approach to sustainability assessment
allows for a concretisation of the abstract notion of sustainability without
passing over inherent ambiguities. It yields a map of the societal value
landscape with respect to the transformation of electricity, gas, water
and telecommunications provision. Societal goal formulation can be sup-
ported by dierentiating between facts and values and making them acces-
sible for dierentiated modes of conict resolution such as discourse about
problem framing and bargaining over distributional aspects (see Saretzki
1996).
need to be prepared for adaptation when things turn out to be dierent than
expected. This is what sustainability foresight teaches its participants as
well as the users of its products such as the scenarios, value landscape and
the integrated strategy for shaping innovation processes. Especially the
latter takes due account of uncertainty about system dynamics as well as
ambiguity of sustainability assessments and therefore follows a procedural
and experimental approach by facilitating interactive learning instead of
pushing particular best solutions, be it technologies, policy instruments or
behavioural patterns. Strategies aim to open innovation processes to inte-
grate diverse perspectives and to remain open for the revision of guiding
visions and design principles.6
Anticipation of long-term systemic eects: anticipation is the general
idea behind any kind of foresight process. The approach of sustainability
foresight in particular is to draw the system boundaries in a very broad
manner in order to become aware of distant side eects and long feedback
loops which are linked to certain strategies (for example, the four utility
sectors, action domains production, consumption and regulation, interac-
tions across the dimensions of society, technology and nature). Anticipation
is not done with impact assessment by experts but in the context of moder-
ated interaction of stakeholders from various parts of the system under
investigation. It is important to note that this approach to anticipation does
not aim to make correct projections, but to collectively explore plausible
futures by actors who themselves shape this future by their daily interac-
tions. The process may as such help to create an alignment of strategic
orientations and is therefore more about bringing about and shaping devel-
opments than predicting them.
Iterative participatory goal formulation: the discursive sustainability
assessment which serves to identify risks and opportunities of transforma-
tion acknowledges that sustainability goals must become established and
weighed against each other in broader social processes, not by scientic
experts or politicians alone. It builds on the participation of the spectrum
of dierent social actors who are aected by utility transformation in order
to lay bare the ambivalence involved in sustainability assessment and
prepare the ground for the deliberation of pragmatic judgements. Iteration
is not part of the process itself, if it is only conducted once. By proposing
the method, however, we envision that successive sustainability foresight
processes are performed in order to keep track of the changes in knowledge
and values that are part of the transformation process and adapt goals and
assessments accordingly.
Interactive strategy development: the strategy recommendations which
are elaborated in the course of the sustainability foresight process are inter-
action oriented in two ways. First, they are developed in interaction with the
Sustainability foresight 181
stakeholders who are the ones to implement them or are aected by them.
This happens in the course of group interviews linked to the in-depth analy-
sis of critical innovation processes and in the form of a strategy workshop
for drawing conclusions from these analyses. A second aspect of iterative
strategy development is linked to the kind of strategies which are developed
in the sustainability foresight process. Their procedural orientation actually
puts interactions between actors who are involved in and aected by criti-
cal innovation processes at the core of the shaping approach. Thus, strat-
egies and measures which result from sustainability foresight are indeed
aimed at moderating the self-organisation of actors who play a part in the
transformation process. Distributed capacities to inuence the course of
transformation are utilised by this means without the need to exert central
control.
orientation are therefore prone to shift problems rather than solve them.
Reexive governance arrangements, on the other hand, face limits in reach-
ing decisions which are necessary for action as long as they keep up
reexivity and do not evade to pragmatic simplications. Sustainability fore-
sight and other reexive governance arrangements therefore have to be con-
ceived as being complementary to conventional problem-solving. Their
particular value is to buer the side eects of routine problem-solving by
opening up narrow problem conceptions and recontextualising specialised
operations with the perspectives of interdependent and aected stakehold-
ers. It is in this respect that the eect of sustainability foresight should
be valued and evaluated. How the balance of reexive and positivistic
approaches to sustainable development can be evaluated and how they can
be productively combined are questions which lead us beyond the scope of
this chapter. A rst step that we have attempted to take is to contribute to a
better understanding of the specic quality of reexive governance by dis-
cussing sustainability foresight as a specic case.
ACKNOWLEDGEMENTS
We thank Dierk Bauknecht, Bas van Vliet and Ren Kemp for comments
on an earlier version and Arie Rip for the discussion of background con-
cepts which shaped our ideas beyond what is visible in the references.
NOTES
1. We appreciate funding through the programme on socio-ecological research by the German
Federal Ministry for Education and Research (www.sozial-oekologische-forschung.org)
under the project title Integrierte Mikrosysteme der Versorgung. Dynamik,
Nachhaltigkeit und Gestaltung von Transformationsprozessen in netzgebundenen
Versorgungssystemen (www.mikrosysteme.org).
2. With the concept of Constructive technology assessment, the social phenomenon of
nexus arrangements has been turned into a programmatic approach to overcome the
control dilemma in technology assessment which refers to the discrepancy between a
lack of knowledge about the eects of technological developments at an early stage and
a lack of opportunity to inuence its course at a later stage (Collingridge 1980). For a
more encompassing elaboration of the concept which has also played a role in the devel-
opment of the sustainability foresight method see Rip et al. 1995; Rip 2002; Simonis
2001.
3. The three steps are related to the distinction of system knowledge, knowledge about
goals and transformation knowledge as elements of sustainability research (see Mogalle
2001).
4. Looking at long-term developments, the uncertainty about ones own position increases.
As a result, a veil of indierence (Rawls 1999) with respect to the distribution of benets
and burdens to particular actor groups may increase the probability of future knowledge
which is less biased with respect to individual interests.
Sustainability foresight 185
REFERENCES
Arentsen, M. and R.W. Knnecke (eds) (2003), National Reforms in European Gas,
Amsterdam, Boston, Heidelberg et al.: Elsevier.
Beck, U. (1993), Die Erndung des Politischen, Frankfurt am Main: Suhrkamp.
Berkhout, F. and J. Hertin (2002), Foresight futures scenarios. Developing and
applying a participative strategic planning tool, Greener Management
International, 37 (Spring), 3752.
Borup, M. (2003), Green Technology Foresight as Instrument in Governance for
Sustainability, presented at the conference Governance for Industrial
Transformation, organised by AK Umweltpolitik und Global Change, Deutsche
Vereinigung fr Politikwissenschaft, Berlin.
Brand, K.-W. (2002), Politik der Nachhaltigkeit. Voraussetzungen, Probleme,
Chancen eine kritische Diskussion, Berlin: Edition Sigma.
Canzler, W. and M. Dierkes (2001), Informationelle Techniksteuerung: entliche
Diskurse und Leitbildentwicklungen, in G. Simonis, R. Martinsen and Saretzki
(eds), PVS Politik und Technik. Analysen zum Verhltnis von Technologischem,
politischem und staatlichen Wandel am Anfang des 21. Jahrhunderts. Sonderheft
31/2000, Wiesbaden: Westdeutscher Verlag, pp. 45775.
Collingridge, D. (1980), The Social Control of Technology, London: Frances
Pinter.
Fischer, F. (1993), Brger, Experten und Politik nach dem Nimby-Prinzip: Ein
Pldoyer fr die partizipatorische Poilcy-Analyse, in A. Hritier (ed.), PVS
Policy-Analyse Sonderheft 24/1993. Kritik und Neuorientierung, Opladen:
Westdeutscher Verlag, pp. 45170.
Gallopn, G.C. (2002), Planning for resilience: scenarios, surprises, and branch
points, in L.H. Gunderson and C.S. Holling (eds), Panarchy. Understanding
Transformations in Human and Natural Systems, Washington, DC: Island Press,
pp. 36194.
Godet, M. (1987), Scenarios and Strategic Management, London: Butterworth.
Grin, J. and A. Grunwald (2000), Vision Assessment: Shaping Technology in 21st
Century Society. Towards a Repertoire for Technology Assessment, Berlin,
Heidelberg and New York: Springer Verlag.
Grunwald, A. (ed.) (2002), Technikgestaltung fr eine nachhaltige Entwicklung. Von
der Konzeption zur Umsetzung, Berlin: Edition Sigma.
Hisschemller, M. and R. Hoppe (2001), Coping with intractable controversies: the
case for problem structuring in policy design and analysis, in M. Hisschemller
186 Strategies for sustainable system transformation
Rawls, John (1999), A Theory of Justice, (revised edn.), Cambridge, MA: Belknap
Press.
Renn, O. (2002), Foresight and multi-level governance, presented at the conference
Role of Foresight in the Selection of Research Policy Priorities, organised by
IPTS Seville.
Ringland, G. (1998), Scenario Planning: Managing for the Future, Chichester: John
Wiley.
Rip, A. (1998), The dancer and the dance: steering in/of science and technology,
in A. Rip (ed.), Steering and Eectiveness in a Developing Knowledge Society,
Utrecht: Uitgeverij Lemma BV, pp. 2750.
Rip, A. (2002), A co-evolutionary perspective on ELSI, CTA and other attempts at
re-contextualisation of science and technology in society, presented at the con-
ference Responsibility under Uncertainty, organised by European Association
for the Study of Science and Technology, York, UK.
Rip, A. and R. Kemp (1998), Technological change, in S. Rayner and E.L. Malone
(eds), Human Choice and Climate Change, Columbus, OH: Batelle Press,
pp. 32799.
Rip, A., T.J. Misa and J.P. Schot (eds) (1995), Managing Technology in Society. The
Approach of Constructive Technology Assessment, London: Pinter.
Rip, A. and J.P., Schot, (1999), Anticipating on contextualization loci for
inuencing the dynamics of technological development, in D. Sauer and C. Lang
(eds), Paradoxien der Innovation. Perspektiven sozialwissenschaftlicher
Innovationsforschung, Frankfurt and New York: Campus, pp. 12948.
Saretzki, T. (1996), Wie unterscheiden sich Argumentieren und Verhandeln?, in
Prittwitz, V.v. (ed.), Verhandeln und Argumentieren, Opladen: Leske und Budrich,
pp. 1939.
Schneider, V. (2001), Die Transformation der Telekommunikation Vom
Staatsmonopol zum globalen Markt (18002000), Schriften des MPIfG,
Frankfurt: Campus.
Simonis, G. (2001), Die TA-Landschaft in Deutschland Potenziale reexiver
Techniksteuerung, in G. Simonis, R. Martinsen and T. Saretzki (eds), PVS
Politik und Technik. Analysen zum Verhltnis von Technologischem, politischem
und staatlichen Wandel am Anfang des 21 Jahrhunderts. Sonderheft 31/2000,
Wiesbaden: Westdeutscher Verlag, pp. 42556.
Stirling, A. and P.v. Zwanenberg (2002), Precaution in the European Union: from
principle to process, paper presented at the EASST 2002 Conference on
Responsibility under Uncertainty, York: 31 July3 August.
Summerton, J. (ed.) (1992), Changing Large Technical Systems, Boulder, CO:
Westview.
Truer, B., A. Metzner and R. Hoogma (2003), The coupling of viewing and doing.
Strategic niche management and the electrication of individual transport,
Greener Management International 37, 11124.
van den Belt, H. and A. Rip (1987), The Nelson-WinterDosi model and syn-
thetic dye industry, in W.E. Bijker, T.P. Hughes and T.J. Pinch (eds), The Social
Construction of Technological Systems, Cambridge, MA: MIT Press,
pp. 13558.
van Lente, H. (1993), Promising technology, the dynamics of expectations in tech-
nological development, PhD Thesis, Enschede: University of Twente.
Vo, J.-P. (2000), Institutionelle Arrangements zwischen Zukunfts- und
Gegenwartsfhigkeit: Verfahren der Netzregelung im liberalisierten deutschen
188 Strategies for sustainable system transformation
INTRODUCTION
189
190 Strategies for sustainable system transformation
can play a key role not only in both respects, but also by dening incen-
tives and incentive structures that are compatible with the visions and
agendas, for instance in R&D policy by striking a balance between
generic support measures for R&D and other, thematically oriented
RTD programmes.
Co-ordination: orientation by visions is a powerful means of sup-
porting processes of co-ordination between dierent actors, but in
particular with respect to horizontal policy matters such as sustain-
ability, the coordination between dierent policy areas and levels
needs to be reinforced, for instance by joint action plans and har-
monised regulations. Similarly, the coordination between policy
actors and other stakeholders requires active measures, which in
most cases will rely on some kind of participation process in the early
phases of the policy cycle.
Experimentation: underinvestment in research and development
under market conditions has been one of the main arguments for
legitimising RTD policy measures since its early days. This argument
holds even more in the face of interactive, interdisciplinary and trans-
disciplinary research practices that require the involvement of a
wider range of disciplines, stakeholders and practitioners. Enabling
bottom-up experimentation (and subsequent learning about the
experiences with and the desirability of innovations) is thus another
important role for research and technology policy.
Adaptation: in an increasingly globalised economy and society,
the ability to adapt to new emerging challenges (for example, envi-
ronmental, migratory) or unpredictable events becomes an essential
asset. It implies that simple goal orientation is not sucient any more
in view of the fact that the evolution of societies depends to a large
extent on decisions made elsewhere. Experimentation and variety-
creation are prerequisites for maintaining adaptability, but adapta-
tion also requires being able to change structural characteristics of
multi-level systems. However, adaptability needs to be balanced with
stabilising characteristics such as goal orientation.
First-order and second-order learning: learning is crucial for being able
to strike the right balances, for instance between adaptation and goal
orientation, or between generic and directed research policy measures.
In the case of sustainability, both rst-order learning about past expe-
riences with policy measures (for example, by means of evaluations)
and second-order learning (that is, reexivity) about the cross-cutting
governance implications of experiences in dierent specialised policy
areas are needed (c.f. Vo and Kemp, in the Introduction to this
volume). Second-order learning implies the ability to change the role
Foresight and adaptive planning 193
of politics and policy (and thus also of other actors) in the course of the
innovation process and the system transformation.
Participation: beyond being a means to enhance the legitimacy of
policy decisions, participation also enhances the integration of tech-
nological, economic, political and institutional sources of knowl-
edge. This is particularly important in the case of sustainability where
underlying conicts about values and risk perception related to inno-
vation and new technologies make it essential to involve a wide range
of stakeholders in policy decisions about these new technologies.
Ensuring the participation of an appropriate range of actors and
stakeholders in network-based processes of preparing policy deci-
sions is thus another functional requirement of sustainability-
oriented innovation systems of which government should take care.
Adaptation over time, that is, over the policy and the innovation
cycles: policy roles may also change over time. As shown in Table 8.1
for the case of transport technology policy, the appropriate role for
EU policy changes in the course of the innovation process, and key
policy functions may move down to lower policy levels or altogether
out of the policy realm. Moreover, time is also a critical factor from
the perspective of the complex innovation systems because of the
necessity to exploit time windows of opportunity. Similar arguments
can be made with respect to the policy cycle and the resulting changes
in policy roles over time.
196
Decline
Replacement
Notes:
role has conflict with innovation phase
role does not match innovation phase
combination of role and innovation phase is possible
role ts phase
role is required in this phase
Over the last ten to fteen years, forward-looking approaches under the
headline of foresight have acquired a prominent role in policy-making, in
particular not only in science and technology policy (Gavigan 2002, Tbke
et al. 2001), but also in relation to sustainability and other long-term, risk-
prone issues. In the early years of foresight, mainly Delphi methods were
applied to assess the development potential and the competitive positions
in dierent areas of science and technology. In recent years, participatory
and panel-based approaches have become more common, not the least to
better capture the social and economic aspects related to future changes in
science and technology (see Georghiou 2001). Generally speaking, fore-
sight processes represent a mechanism with which to deal systematically
with future risks, opportunities and options by drawing on a broad range
of future expectations and by involving an equally broad range of actors in
a participatory process. It is applied with respect to emerging developments
in science and technology or socio-economic developments likely to exert
an impact and pull on scientictechnological research and development.
In some countries, foresight has become a widely used tool to support S&T
policy-making, leading to the establishment of a forward-looking political
and governance culture (Gavigan 2001).
There are several dierent types of foresight, each aiming at dierent
purposes. A rst basic distinction refers to the exploratory or normative
198 Strategies for sustainable system transformation
First of all, it has moved away from a pure focus on science and tech-
nology that dominated the scene until the early to mid-1990s to an
incorporation of rst market and then also increasingly social con-
siderations. Broadening the scope of foresight in this way can be
interpreted as a reection of the abandonment of linear models of
technological change and the adoption of a systemic understanding
of socio-technical change.
Secondly, foresight has become an increasingly participatory activity.
Initially, foresight activities were mainly based on S&T expert
opinion, but in line with broadening the scope of foresight, the notion
of expert has undergone a re-denition. With respect to participation,
one can observe similar developments in technology assessment
where the growing prominence of social, economic, environmental
and ethical concerns related to scientic and technological develop-
ments has led to a strengthening of participatory and constructive
approaches (Decker and Ladikas 2004, Joss and Bellucci 2002). Here,
the aim is to negotiate consensus on risks and opportunities or at least
achieve transparency about conicting viewpoints to contribute to a
normative debate on desirable future development paths.
Thirdly, we can see today a strong emphasis on and belief in the con-
tribution of foresight activities to shaping rather than predicting the
future. The initial Delphi exercises were still strongly inuenced by the
Foresight and adaptive planning 199
linear idea that the consensus achieved in Delphi could serve as a fore-
cast and thus as a foundation for taking preparatory actions. Today,
by bringing together in a foresight process not only experts, but in par-
ticular also decision-makers from research, industry, policy-making
and society, a shared understanding of current problems, goals and
emerging development options is expected to emerge among those
actors that have an important role to play in shaping the future. This
converging understanding of the issues at play is expected to con-
tribute to improving implicitly the coherence of the distributed deci-
sions of these actors, in line with the shared mental framework
developed. In other words, the future is shaped by aligning expect-
ations and thus creating a self-fullling prophecy. These so-called
process outputs are in the meantime often regarded as more impor-
tant than the actual substantive outputs like reports and websites.5
Both the rst and the third shortcoming could be addressed by building
on insights from strategic planning. This is in particular the case as regards
the dilemma between the need to keep future options open in order to be
able to adapt swiftly to changing circumstances on the one hand (adaptive
strategies) and the willingness to actively shape the future on the other
(proactive strategies).
In this sense, foresight and adaptive planning are instruments that facil-
itate exceeding rst-order learning about the impacts of individual instru-
ments (for example by means of conventional evaluation) and enable
second-order learning.
Over the past twenty years, one could observe a major change in
policy-making as well as in policy research, emphasising the need
for more strategically oriented inputs as a foundation for policy.
Several different approaches that had been employed thus far in a
rather isolated way have been brought together under the heading
strategic policy intelligence (Tbke et al. 2001).They comprise, for
instance, forecasting, foresight, technology assessment, scenario
development and other forward-looking approaches. In particular
scenario techniques have proven a powerful tool to integrate
elements from the aforementioned approaches into consistent
visions of and pathways towards the future. Also methodologies to
enhance policy learning from past experiences fall under the head-
line of strategic policy intelligence, for instance by way of system-
atically evaluating policy programmes, initiatives and institutions or
by comparing national practices (benchmarking). Strategic policy
intelligence is thus expected to support policy-making over the
entire policy cycle from agenda-setting to policy learning.
The strategic turn in policy-making implies a major shift in terms
of governance, and it raises new requirements with respect to the
strategic competencies of policy-makers vis--vis other actors in
the system. First of all, we are increasingly confronted with a situ-
ation, in which the necessary knowledge, competencies and intel-
ligence is distributed across society (Kuhlmann 2001), thus calling
for a higher degree of participation in policy decision in order to
draw on this distributed intelligence. Therefore, strategic policy
intelligence methods tend to employ interactive methods such as
expert panels, scenario workshops, citizen conferences and so on,
in order to complement conventional analytical methods. In other
words, strategic policy intelligence also needs to devise ways of
drawing on these distributed competencies.
Secondly, there is a clear need to devise systematic methods to
exploit prospective findings for the purposes of policy strategy
development. Although strategic policy intelligence is increasingly
Foresight and adaptive planning 207
between factors that are likely to be stable and others that are still open to
evolve in qualitatively new ways. Especially the latter ones are useful to
dierentiate one scenario from another. Such trends and trend breaks can
be of a technological as well as of a socio-economic nature.
Trends and trend breaks are the basic material from which to develop
scenarios. Here, a variety of methods is available, ranging from bottom-up
methods of constructing and clustering storylines into scenarios to top-
down methods that concentrate on selected dominant factors of inuence,
possibly supported by cross-impact analysis, Delphi methods or simula-
tions to strengthen the credibility of results. However, most of these
methods rely in the end on expert or stakeholder judgement, thus reecting
these actors implicit theories about impact chains.
What is essential for these scenarios is that they are plausible, multi-
faceted and challenging. They must be regarded as tools to stimulate think-
ing about potential future developments rather than as predictions of the
future. Moreover, the distinction between endogenous and exogenous
factors of inuence helps identify where the opportunities and limitations
for action are for the main actors addressed, that is, for instance a national
government. Trends and trend breaks also point implicitly to important
decision options of the dierent actors, that is, the strategic moves they can
make. Games (Planspiele) are a tool by which these implicit strategic
options can be made explicit.
When the basic scenarios are dened, they represent dierent frame-
works that still leave much room for further specication. In fact, they
dene a corridor of actions that needs to be rened and related to the focal
issue under investigation. As guidance for how to describe such framework
scenarios, the main types of factors of inuence can be taken as a starting
point. Typically, categories such as technologies, producers, users, policy
and so on, are dened by the participants, that is, categories that can easily
be related to actors strategies.
The normative dimension is addressed in the subsequent stage. At this
point, goals and values of the dierent actors come into play (see also
Box 2). A possible starting point is therefore a debate on visions related to
the focal issue of the exercise. This is useful in order to clarify shared (or
diverging) policy and/or societal goals, ambitions and underlying values of
the actors and stakeholders involved. A second element is related to the
potential future risks and opportunities tied to the focal issue of the
exercise.
Both elements serve in the end to agree on the dimensions along which the
exploratory scenarios are to be assessed. In many current cases, sustainable
development is taken as a guiding vision that needs to be operationalised
in terms of dierent assessment dimensions, for instance along the lines of
210 Strategies for sustainable system transformation
and policies that have been realised. The options delivered by the scenarios
have also been assessed with respect to our focal issue.
From todays perspective, portfolio analysis then looks across dierent
scenarios to assess and select those technology options and correspond-
ing policies that promise to be either robust or adaptive (or both). In other
words, robust options are fairly easy to identify, because they are posi-
tively assessed in all or most scenarios. Adaptive options have been
identied as part of a normative stage of scenario development when pos-
sibilities are sought to move the basic scenario in a more desirable direc-
tion. They are thus either crucial for avoiding major negative impacts or
for exploiting specic opportunities in a single scenario. These kinds of
insights should then serve as input for todays policy-makers to prioritize,
for instance, emerging technologies and design policies correspondingly.
Foresight and adaptive planning 213
Suggesting policy options and portfolios is just an input to actual policy design
and implementation. In other words, so far we have been mainly discussing the
early phases of the policy process. The implementation of policies, as well as
the learning processes that take place between design and implementation,
represent the wider framework in which adaptive foresight processes are situ-
ated. Moreover, emphasis has been on the strategy level, where the basic
214 Strategies for sustainable system transformation
orientations and guidelines for policy are dened to trigger and frame more
specic initiatives. In practice, however, policy design and implementation rely
on concrete actions and initiatives, often even at local level.
If the principles behind adaptive foresight are to be eective, they need to
be closely tied not only to policy design, but also to policy implementation
and learning, at the strategic as well as at the local level. More specically,
the experiences made in the course of local implementation need to be mon-
itored and fed back to strategy development.
In other words, adaptive foresight should be interpreted as part of a
broader continuous learning process that comprises the implementation and
evaluation of specic policy measures as well as monitoring relevant devel-
opments in policy at large.
Strategy development, policy design, implementation and learning
should thus not be understood as distinctly separate phases but rather as a
continuous process of mutual adjustment. This adjustment refers to goals
and objectives, to the identication of new socio-technical options, to the
growing knowledge and understanding of their impacts, to the design of
new types of policy options and to their integration into portfolios.
Within such a comprehensive setting, the impact of guiding policy strate-
gies should nevertheless not be underestimated, in particular because
public policy strategies full an orienting function for many private actors
as well, and in the best case play an implicit coordinating function in their
decision-making.
One of the main diculties of this continuous process consists of the fact
that all actors involved can recur to strategic and adaptive (and thus inter-
dependent) behaviour. This is why issues of policy coordination both
between dierent policy areas and between public and private actors have
started to play such an important role in policy-making.
In practice, processes of scenario developments and portfolio analysis will
nevertheless hardly be conducted on a continuous basis but rather be
repeated every few years, for instance in line with an update of the overall
technology and innovation policy strategy. The practical tools and methods
are available based on many years of experience with foresight, adaptive
planning, evaluation and monitoring; what is still missing is the integration
of these methods in a continuous and long-term strategy development
process.
NOTES
1. The author would also like to thank Dierk Bauknecht and Jan-Peter Vo for construc-
tive comments on draft versions of this contribution.
2. For more details see the article by Jacobsson and Johnson (2000) on functional innov-
ation systems. More specically, the conceptual framework that underlies the arguments
raised here is elaborated in Weber (2005) under the heading complex innovation
systems. It species the interactions between policy-making processes on the one hand
and innovation processes on the other, as well as the resulting dynamics.
3. See Tbke et al. (2001) for a review.
4. See also Renn (2002) who argues with respect to foresight processes that mainly episte-
mological discourses, that is, debates about potential opportunities and risks based on
a common vision and the likely developments that we can foresee without relying on pure
speculation, should be conducted at the EU level, whereas debates about assessments of
trade-os and values require stronger participation from the bottom up and taking local
specicities into consideration. However, we should also be aware that this argument
may have to be dierentiated with respect to dierent problem areas, because in a highly
internationalised or Europeanised eld (such as climate change or aeronautics, genetic
technology), the appropriate level for assessment and value discourses may well be at
these higher levels.
5. Obviously, there are also certain types of foresight exercises that have a less pro-active
intention by concentrating only on the identication of future challenges and issues
rather than aiming at solutions.
6. The term real options is used to stress their dierence from nancial options in which
portfolio management has become a standard practice.
7. The concept of real options is usually applied with respect to a specic actor or addressee
who has the choice between dierent options. It is important to be very clear about the
addressee of an adaptive planning activity and about the scope of his or her decision
options because this determines what is to be regarded as an endogenous or an exogen-
ous variable, that is, as a factor he/she can or cannot inuence. In most policy-oriented
exercises, this addressee can be society at large, but in more operational terms it is usually
a public agent like a ministry or a regional government.
8. Priority-setting in research and technology policy has become a major issue in most
European countries, both due to concerns about global competitiveness and in response
to the realisation of the European Research Area. Transparent processes, heuristics and
rationales for setting priorities are still widely absent from the policy debates, but build-
ing blocks are currently being developed (Weber et al. 2004a).
9. See also Elzen et al. (2005) on socio-technical scenarios and their application to the evo-
Foresight and adaptive planning 219
lution of dierent large technical systems. However, they mainly applied the scenario
method for purely experimental purposes and did not link them up with actual policy-
making processes.
10. The boundaries between the inside and the outside of a system do not need to be stable
but can change in the course of a scenario, thus also inuencing the range and the scope
of options for the actors considered.
REFERENCES
nology policy for sustainability, in J.N. Rosenau, E.U. von Weizscker and U.
Petschow (eds), Governance and Sustainability. Exploring the roadmap to sustain-
ability after Johannesburg, Sheeld: Greenleaf (forthcoming).
Weber, K.M., H. Gassler, W. Polt, B. Dachs and G. Streicher (2004a), Anstze und
Befunde zur Schwerpunktsetzung in der sterreichischen Forschungs- und
Technologiepolitik, Wirtschaftspolitische Bltter, 51 (3), 40518.
Weber, K.M., K. Kubeczko, K.-H. Leitner, K. Whitelegg, P. Spth, H. Rohracher
and I. Oehme (2004b), Transition zu nachhaltigen Produktionssystemen. 2.
Zwischenbericht, Seibersdorf/Graz: ARC systems research/IFZ.
Weber, K.M., R. Hoogma, B. Lane and J. Schot (1999), Experimenting with
Sustainable Transport Technologies: A Workbook for Strategic Niche
Management, Enschede/Sevilla: University of Twente/IPTS.
Weber, K.M., A. Geyer, D. Schartinger and P. Wagner (2002), Zukunft der
Mobilitt in sterreich. Konsequenzen fr die Technologiepolitik, Jahresbericht
2001, Forschungsbericht ARC-S-0185, December.
Whitelegg, K. (2005), Patchwork policy making linking innovation and transport
policy in Austria, in OECD: Governance of Innovation Systems Vol. 3: Case
Studies in Cross-sectoral Policy, Paris: OECD (forthcoming).
Knowledge production and assessment
9. Precaution, foresight
and sustainability: reflection
and reflexivity in the governance
of science and technology
Andy Stirling
225
226 Knowledge production and assessment
might be raised on this somewhat broader canvas. Certainly, where such the-
matic labels encourage the objectication and compartmentalization of
dierent governance domains, or where they are appropriated instrument-
ally to legitimate particular favoured interventions, then they may actually
militate against reflexivity in wider governance processes. I return to this
theme in the next section. For the moment, it suces to note that this kind
of compartmentalization (Weale, 2000) and instrumental legitimation
(Stirling, 2004) are all too often a feature of the relationship between sus-
tainability and wider areas of governance.
It is with this type of question about the precise nature of the relationship
between reflexivity and sustainability in governance discourses, with which
this chapter might most usefully begin. A key initial point in this regard con-
cerns the proliferating understandings of exactly what is meant by the term
reflexivity in various contexts bearing on this theme (Gouldner, 1970;
Giddens, 1976; Steier, 1991; Bourdieu and Wacquant, 1992; Alvesson and
Skoldberg, 2000; Woolgar, 1988). In particular, questions arise about dis-
tinctions between normative and descriptive usages, and about the role of
intentionality (Lash, 2001). For purposes of clarity in the present account,
it may therefore be useful to begin with a clear working distinction between
three key concepts on which this analysis will rest: unreflectiveness, reflec-
tion and reflexivity. This is not intended as an eort at general synthesis,
but simply as a consistent and transparent point of departure for the present
discussion. Accordingly, each term will be addressed in relation to modes of
representation, understanding and intervention in the governance of
science and technology. Schematic pictures of key elements in these working
denitions are provided in Figure 9.1.
Unreflectiveness is, in these terms, a governance situation in which repre-
sentations, understandings and interventions are eectively restricted to
whatever are held to be the most obvious, operational, or instrumentally per-
tinent attributes of the object under attention. Here (and in Figure 9.1), the
subject is the governance system itself. The object, in this context, might
be a resource allocation across contending scientic research programmes,
a choice among alternative technological trajectories or a decision concern-
ing wider regulatory or technology policy commitments. In the context of
discussion over sustainability, these might hypothetically be exemplied by
the way in which the governance system relates to the development of a
family of novel chemicals. Here, representations take the form of dierent
forms and styles of regulatory appraisal including risk assessment,
costbenet analysis, expert advisory committees, stakeholder negotiations
and wider political discourses over the nature and scale of contending inter-
ests. Options for governance intervention might include: mandatory
requirements on production processes, emission controls, product-specic
Precaution, foresight and sustainability 227
Object (scientific or
technological
commitment)
Representation Intervention
Subject
(system of
governance)
228
Reflection: deep, serious consideration extends to all salient aspects of the object of attention
(full representation, aspiration to objective synoptic basis and focus for intervention),
e.g. lets take account of all possible consequences before using the chemicals.
Object (scientific or
technological
commitment) Representation
Subject
(system of
Intervention governance)
Reflexivity: Attention simultaneously encompasses and helps constitute both subject and object (recursive
mutual contingency of subjective representations and interventions),
e.g. the consequences depend on our point of view and our expectations of use.
Multiple representations
Object (scientific or
technological
commitment)
229
Subject
(system of
governance)
Co-constituted interventions
Figure 9.1 Working denitions for unreflectiveness, reflection and reflexivity for the present chapter
230 Knowledge production and assessment
Uncertainty Ignorance
e.g. many carcinogens; e.g. BSE;
floods under climate change; CFCs and the ozone hole;
Problematic corporate share holder value endocrine disruption
Uncertainty Ignorance
use: level and burden of proof; use: transdisciplinary engagement;
decision heuristics; research and monitoring;
uncertainty factors; horizon scanning;
Problematic sensitivity analysis flexibility, diversity, resilience
As shown in the right hand column of Figures 9.2 and 9.3, the accepted
twofold formal denition of risk also implies, beyond uncertainty, two
further equally coherent and complementary, but still more seriously
neglected states of knowledge. These are the conditions of ambiguity and
ignorance. Ambiguity describes a state of knowledge in which the
problem lies not in the basis for determining the likelihood of each of a
variety of dierent outcomes, but in coming to a common understanding
of how to select, partition, characterize, prioritize, bound or interpret the
meanings of these outcomes (Wynne, 1992, 2002; Rosenberg, 1996;
Stirling, 1998). Ambiguity is thus about the framing of the appraisal
process the questions that are posed, the perspectives that are engaged,
the methods that are adopted, the assumptions that are made and the mode
of representing any ndings in wider governance discourses (Wynne, 1992,
2002; Stirling, 2004).
Although often intimately intertwined with uncertainty, the condition
of ambiguity is thus conceptually quite distinct. Examples of this kind of
dilemma might be seen in the denition of harm in the regulation of
242 Knowledge production and assessment
Edwards, 1986; Rosa, 1998), ignorance represents our uncertainty about our
uncertainty (Cyranski, 1986). It is an acknowledgement of the importance
of the element of surprise (Brooks, 1986; Rosenberg, 1996). This emerges
not just from the actuality of unexpected events, but from their very possi-
bility (Dosi and Egidi, 1987). It is a predicament that intensies directly in
relation to the social and political stakes that bear on a particular decision
(Funtowicz and Ravetz, 1990). It emerges especially in complex and dynamic
environments in which social agents and their cognitive and institutional
commitments may recursively influence supposedly exogenous events
(Dosi and Egidi, 1987). Indeed, it is due to this reflexive relationship between
environmental learning and social commitments, that Wynne has seminally
emphasized that ignorance entails even more intractable forms of what he
terms indeterminacy (Wynne, 1992, 2000).2
The relevance of the predicament of ignorance to the governance of sus-
tainability is obvious. Many of the most pressing specic challenges that
have arisen in this eld were, at their inception, not so much matters of
inaccurate attributions of probability to anticipated outcomes, as they were
of intrinsically unexpected outcomes. This was the case, for instance, with
recognition of halogenated hydrocarbons as key agents in stratospheric
ozone depletion (Farman, 2001), endocrine disruption as a novel mech-
anism of harm in chemical regulation (Thornton, 2000) and the resilience
to processing and interspecies transmissibility of certain spongiform
encephalopathies (van Zwanenberg and Millstone, 2001).
Of course, the specic locus of ignorance within the governance process
can vary from case to case and over time. Ignorance can be a property of a
particular institutional context for decision-making as well as a pervasive
societal condition (Stirling, 2003; EEA, 2001). The former is addressed by
processes of communication, engagement and organizational learning. The
latter is mitigated by processes for baseline monitoring, scientic research
and wider social learning. Wherever they apply, such procedures might
together be referred to as precautionary approaches to the appraisal of
scientic, technological and wider policy choices in sustainable governance.
Some examples are indicated in Figure 9.3, alongside the parallel responses
in appraisal to risk, uncertainty and ambiguity. The implications will be
further explored in the following sections.
Before turning in more detail to these precautionary approaches,
however, it is worth considering the complex way in which considerations
of reflection and reflexivity emerge from the contrasting approaches sum-
marised in Figure 9.3. It is not simply the case that those methods that are
restricted to application under risk are necessarily uniformly less reflective
or reflexive than those that are more appropriate under uncertainty, ambi-
guity or ignorance. Under each idealized state of incertitude, the array of
Precaution, foresight and sustainability 245
methods may be both applicable and useful. Beyond this, there is nothing
in this analysis that denies the essential role that science plays in the wider
appraisal of uncertainty, ambiguity and ignorance. The key point is not
that risk science is always problematic. It is that science oers a necessary,
rather than sucient, basis for the governance of sustainability.
The question then arises as to exactly how we should think about the role
and value of science in the reflexive governance of sustainability? This
raises a series of deep epistemic and ontological issues concerning the mul-
tiple natures (respectively) of knowing and being (Leach et al., 2005).
Without presuming to attempt a synthesis of the rich, complex and open-
ended character of these discourses, Figure 9.4 provides a schematic snap-
shot of some key implications of the present discussion of reflection and
reflexivity. For this purpose, the role of science is conceived as a means to
determine and represent the truth value associated with dierent possible
ontologies (ways of being in the world) and epistemologies (ways of
knowing the world).3 The initial point that immediately arises is that even
the slightest element of reflection or reflexivity refutes what might be
termed the nave realist position, to the eect that science provides precise
representations of whatever might be held to be either epistemic or onto-
logical truth.
Yet the invocation of reflexivity raises its own challenges. Discussion can
quickly descend into animated fears over the spectre of a caricature relativist
conclusion that anything goes(Feyerabend, 1975, 1978). As a paradigmatic
example of reflexivity, the adoption of a social constructivist principle of
symmetryover contending representations of scientic knowledge provides
a highly rigorous framework for analytical understandings of the social
dynamics behind these representations (Barnes and Edge, 1982; Woolgar,
1988). But this does not necessarily mean that this can be applied equally
robustly as a normative principle in governance, for the purpose of arbitrat-
ing between contending scientic and other types of knowledge. Indeed,
such an extension from analytic to normative usage would risk being unre-
flective in a similar fashion to the over-application of risk assessment. Just
as risk assessment fails to address diering grounds for condence in the
framing of analysis, so would unqualied normative use of a principle of
symmetry fail to address the manifest social fact of the existence of
diering degrees of plausibility and (self-recognised) self-consistency
attached to dierent knowledge claims about sustainability. For instance,
such caricature relativism would not only fail to adjudicate the contending
positions on global climate change. It would compel equal attention to any
representations of the science no matter how ill-conceived, inconsistent
or fanciful. At the extreme, it would neglect any particular role for specialist
expertise, empirical grounding, analytical rigour, or technical discipline.
Precaution, foresight and sustainability 247
Key
Space of epistemic/ontological
possibilities
High reflectiveness
Figure 9.4 Implications of reflection and reflexivity for the role of science
in sustainable governance
This would eectively deny the relevance of error and leave no operational
means by which governance discourses might address crucial issues of
quality or strategic bias in the wider conditioning of knowledge by circum-
stance and power.
In the mainstream governance of sustainability, the conventional resolu-
tion to the dilemma represented schematically in the lower half of Figure 9.4,
lies in recourse to various forms of what might be summarized as Popperian
fallibilist perspectives (Popper, 1963). 4 Here, it is acknowledged that the
more bullish realist claims sometimes made on the part of science are nave
and potentially misleading as a basis for policy making. In particular, falli-
bilism is more reflective than nave realism about the nature of uncertainty
248 Knowledge production and assessment
and error. It recognizes that science is only rarely able to justify the provision
of unitary, precisely formulated prescriptive representations for instance
concerning sustainability. Accordingly, this mainstream fallibilist perspec-
tive holds that science should be viewed rather as oering only approximate
representations. Yet the persistent adherence in fallibilism to a generally
unitary prescriptive understanding of the normative role of science in gov-
ernance still justies a continued (if more modest) use of the language of
science based and sound scientic decisions (Byrd and Cothern, 2000) in
sustainability discourses.
This is where the importance of the implications of reflection and reflex-
ivity begin to become clear (as shown in the top right hand corner of
Figure 9.4). For there are other possible understandings of the role of science
in the governance of sustainability, that lie beyond the conventional
dichotomy between fallibilism and relativism as alternatives to realism.
Indeed, there seems no reason in principle why we may not combine the high
level of reflection shown by fallibilism to uncertainty and error with the deep
degree of reflexivity shown by relativism towards ambiguity. This fourth type
of understanding over the role of science in governance might be described
as a kind of grounded perspectivism5 (Hales and Welshon, 2000). It is
grounded because it includes a role for criteria of self-consistency, societal
robustness and analytic or empirical quality. It is perspectivist because it
acknowledges that the latitude for divergent framings of such consistency,
robustness or quality in knowledge extends beyond the monocentric
approximations of fallibilism.
In other words, under grounded perspectivism, it is acknowledged (with
fallibilism) to be possible to discriminate between dierent representations
of the science on the basis of their plausibility or self consistency under
any particular set of framing conditions. Accordingly, unlike caricature rel-
ativism, there is reflection over uncertainty and the possibilities of bias or
error. Yet grounded perspectivism also displays reflexivity over the plural-
ity of possible socially contingent and institutionally conditioned framings
of science, each of which will yield dierent criteria of plausibility and self-
consistency. It thereby refrains from the unreflexive assumption that there
will always be a unitary (if approximate) coherent scientically founded
basis for governance commitments. Instead, grounded perspectivism enter-
tains the possibility that multiple, equally scientically valid, representa-
tions may prescribe radically dierent governance interventions.
The practical implications of this grounded perspectivist position for the
governance of sustainability are obvious. Even the most authoritative
expertise, the most detailed bodies of evidence and the most rigorous modes
of analysis typically admit a number of equally valid but mutually inconsis-
tent interpretations. Yet, just because a number of representations may be
Precaution, foresight and sustainability 249
equally true, does not mean that many are not false (Rosa, 1998; Stirling,
2003). The key challenge is thus presented as one of combining reflection
over uncertainty and error with reflexivity over ambiguity and ignorance.
Rather than aspiring to derive a single uniquely science based representa-
tion, this requires reflexive procedures for exploring and arbitrating among
a limited number of equally self-consistent, plausible and scientically
founded but mutually incoherent alternatives. By denition, this is not a
matter on which science alone can play a lead role. In order truly to realize
the implications of both reflection and reflexivity in governance of sustain-
ability, the bottom line is an emphatic paraphrase of the Churchillian injunc-
tion that science should be on tap, not on top (Lindsay, 1995).
It was argued in the last section that twin imperatives for reflection and
reflexivity compel more broad-based and pluralistic understandings of the
role of science in the governance of sustainability. Conventional reduc-
tive aggregative risk-based techniques are of only limited applicability. A
variety of alternative but neglected responses to uncertainty, ambiguity and
ignorance were identied, which oered varying degrees of reflectiveness
and reflexivity. This raises the obvious question as to how we might articu-
late these disparate approaches alongside risk-based techniques in a coher-
ent operational fashion, and how we might in practice apply them as part of
more reflective and reflexive institutions for the governance of sustainability.
Perhaps the most useful and prominent arena for critical discussion on
this theme, lies in heated policy discourses over the precautionary principle.
Arising repeatedly in dierent guises since the 1972 Stockholm Environment
Conference (ORiordan and Cameron, 1994), the precautionary principle
found its rst coherent formal shape in the Vorsorgeprinzip in German envi-
ronmental policy (Boehmer-Christiansen, 1994). Since then, initially
through the campaigning and lobbying eorts of international environ-
mental organisations, precaution has moved from the eld of marine pollu-
tion (Hey, 1992) into diverse areas such as climate change, biodiversity,
genetic modication, chemicals regulation, food safety, public health and
trade policy (Sand, 2000). As a result, the precautionary principle has
become a potent and pervasive element in contemporary risk and environ-
ment policy (ORiordan and Cameron, 1994; Fisher and Harding, 1999;
Raensberger and Tickner, 1999; ORiordan and Jordan, 2001). Precaution
forms a key intrinsic element of sustainability the classic and most globally
influential exposition being Principle 15 of the 1992 Rio Declaration on
Environment and Development. Here, the crucial operational passage holds
250 Knowledge production and assessment
This chapter began by noting the fact that prevailing discourses of techno-
logical progress suppress a general politics of technology. Given this back-
ground, the institutional integration of precaution and foresight discourses
constitutes a key element in the development of more reflective and reflexive
governance for sustainability. In short, precautionary foresightinvolves the
adoption of more long-term, holistic, integrated and inclusive social
processes for the exercise of explicit and deliberate social choice among
contending scientic and technological trajectories (Stirling, 2003; 2004).
In particular, this entails a series of concrete elements that are well
documented and increasingly the subject of parallel methodological and
institutional experimentation in the individual elds of precaution and fore-
sight. In drawing some nal practical conclusions for policy making, it is
interesting to note how closely these specic emerging and converging fea-
tures of precautionary foresight relate to each of the ve adequate strate-
gies for reflexive governance that are developed in the Introduction of this
book by Vo and Kemp. To illustrate this fact, each strategy may be taken
in turn.
Beyond the integration of precaution and foresight, both as elds and cul-
tures of interest, there are three somewhat more specic ways in which we
may envisage more integrated frameworks of knowledge production for the
sustainable governance of science and technology. The rst is that attention
should extend across an array of contending technology and policy options,
rather than restricting attention to the opportunities, competitiveness,
acceptability, tolerability or safety of a single option taken in isolation
(Johnston et al., 1998; OBrien, 2000). This resists outcomes under which
the governance process is driven in a path-dependent fashion by those inter-
ests and perspectives that happen to be most privileged in the framing of the
initial problem formulation, research question or regulatory issue.
The second aspect of additional reflection, concerns the extension of the
scope of social appraisal to address production systems taken as a whole
(including full resource chains and life cycles), rather than a single product,
process or technology viewed simply in terms of the use phase (Jackson
and Taylor, 1992; MacGarvin, 1995; Tickner, 1999). This holds out a
signicant role for more broad-based analytical tools, such as life cycle
assessment, and energy inputoutput analysis as well as more wide-ranging
discursive processes.
The third area for integration involves attending to the pros (benets,
justications and purposes) as much as the cons (costs, risks and wider
impacts) (Jackson and Taylor, 1992; MacGarvin, 1995). It is interesting
that in dierent ways this resonates with all sides in a typical technol-
ogy governance controversy. Proponents are often concerned by what they
hold to be a dearth of attention to the claimed benets of their favoured
technologies. For their part, sceptics are typically frustrated by the lack of
scrutiny of these claims, when compared with the exhaustive regulatory
demands to substantiate any suspected disadvantages. It is a rare opportu-
nity indeed to meet both concerns simultaneously.
goal formulation. The key common challenge faced here lies in nding
practical ways to articulate complex forms of integrated transdisciplinary
appraisal with deeper and more inclusive forms of stakeholder engage-
ment and citizen deliberation this time in strategy implementation. For
its part, the theme of interactive assessment of contending strategic
options is a well-established and highly topical governance discourse in its
own right (Grin et al., 1997; de Marchi et al., 1998). Divergent rationales
for interactive engagement here extend well beyond precaution and fore-
sight to include general discourses on democratic governance. Even with
respect to highly specic operational matters of strategy development,
they raise a range of profound normative, substantive and instrumental
issues.
Under many normative democratic points of view, for instance, more
inclusive forms of interactive engagement are quite simply the right thing
to do, irrespective of the outcomes. Under more instrumental under-
standings, interactive engagement is justied on the grounds that it may
help secure greater condence or trust in existing institutions and proced-
ures (Fiorino, 1989; NRC, 1996) or support for particular decisions (Vo
and Kemp, this volume) or social intelligence for the purpose of manag-
ing residual adverse reactions (Grove-White et al., 2000). Here, the aim may
be to provide legitimation for particular outcomes. The specic argument
for interactive engagement highlighted in the precaution and foresight dis-
courses, however, invokes a third rationale, which we may call substantive.
Here, the aim is one of establishing a broader knowledge base and more
eective social learning in order to achieve better outcomes in the imple-
mentation of strategy (Stirling, 2005).
It is important to distinguish between these three (normative, instrumen-
tal and substantive) rationales and orientations for interactive engagement
in strategy development. This is so, not least, because each approach embod-
ies signicantly dierent forms of reflexivity with respect to the prevailing
distribution of political, institutional and economic power (Stirling, 2004).
This is relevant, for instance, with respect to Vo and Kemps (this volume)
identication of the importance of the alignment of such interaction
towards a single collective strategic goal. This kind of hegemonic ambition
is consistent with an instrumental, but less with normative or substantive
approaches to inclusive engagement. Returning to issues raised at the begin-
ning of this chapter in relation to legitimation discourses over sustainabil-
ity, the presumption of an orientation towards alignment may foster a
certain vulnerability to the justicatory exercise of power (Collingridge,
1980; Wynne, 2002).
It is perhaps in the achievement of greater reflexivity over the role of
power in the framing of interaction in strategy development, that the
Precaution, foresight and sustainability 259
ACKNOWLEDGEMENTS
Thanks are due to Ren Kemp and Jan-Peter Vo for the initial stimulus,
subsequent patience and invaluable advice, to Brian Wynne for longstand-
ing inspiration and to Adrian Smith, Erik Millstone and Melissa Leach for
some extremely helpful comments. This chapter builds on a number of
earlier discussions, especially Stirling (2003) a great debt is also owed to
commentators on these preceding incarnations. The usual disclaimer
applies to the more embarrassing passages.
262 Knowledge production and assessment
NOTES
1. This may represent one specic element in Becks observation (Chapter 2, this volume) of
a current meta-transformation of the political.
2. The relationship between Wynnes (1992) seminal concept of indeterminacy and that of
ambiguity (Stirling, 1999) is briefly discussed in Stirling (2003). Indeterminacy can be
interpreted relatively unreflexively as a physical feature of quantum or nonlinear systems
(Faber and Proops, 1994; Ruelle, 1991), separable from subjective social processes
(Harremoes, 2000). As such, unlike uncertainty, ambiguity and ignorance, it suggests
an objective property of an observed system, rather than a state of knowledge that is
co-constituted by the subjective conditions of the observer. In any event, Wynne himself
also later uses the present term ambiguity in an essentially similar sense (2002).
3. I am extremely grateful to Melissa Leach for reminding me of the potential for treating
contending ontologies, as well as epistemologies, in these terms (see Leach et al., 2004).
This is reflected in Figure 9.4. But with gratitude for enlightening conversations with
Erik Millstone I believe that the basic point here applies in principle to a variety of
understandings of the relationship between ontology and epistemology (see Quine, 1961;
Wittgenstein, 1953).
4. I also owe to Erik Millstone a valuable clarication over the correct terminology on this
point.
5. I use the present term in a sense which broadly resonates at a societal level with the more
individualistic Nietzschean concept of weak perspectivism (Nietzsche, 1968; Hales and
Welshon, 2000). This stands independently of any subsequent political baggage (Ortega
Y Gasset, 1980), and usefully transcends an alternative term like epistemological plural-
ism, in suggesting an engagement not only with epistemic issues, but also with ontology
and power.
6. This normative call for enhanced balance between processes of closing down and
opening up in the social appraisal of scientic understandings and technological poten-
tialities displays interesting resonances with Becks programme of cosmopolitan interde-
pendence (Chapter 2 this volume).
REFERENCES
Adam, B., U. Beck and J. van Loon (2000), The Risk Society and Beyond: critical
issues for social theory, London: Sage.
Alvesson, M. and K. Skoldberg (2000), Reflexive Methodology, London: Sage.
Amendola, A., S. Contini and I. Ziomas (1992), Uncertainties in chemical risk
assessment: results of a European benchmark exercise, Journal of Hazardous
Materials, 29, 34763.
Arrow, K. (1963), Social Choice and Individual Values, New Haven, CT: Yale
University Press.
Arthur, W. (1989), Competing technologies, increasing returns, and lock-in by his-
torical events, Economic Journal, 99.
Barnes, B. and D. Edge (1982), Science in Context: readings in the sociology of
science, Milton Keynes: Open University Press.
Beck, U. (1992), Risk Society: towards a new modernity, London: Sage.
Beck, U. (1994), The reinvention of politics: towards a theory of reflexive mod-
ernization, in U. Beck, A. Giddens and S. Lash (eds), Reflexive Modernisation:
politics, tradition and aesthetics in the modern social order, Cambridge: Polity
Press.
Precaution, foresight and sustainability 263
Beck, U., A. Giddens and S. Lash (1994), Reflexive Modernisation: politics, tradi-
tion and aesthetics in the modern social order, Cambridge: Polity Press.
Berkhout, F., J. Hertin and A. Jordan (2001), Socio-economic futures in climate
change impact assessment: using scenarios as learning machines, Tyndall
Centre Working Paper No. 3, Norwich: Tyndall Centre for Climate Change
Research.
Berkhout, F., A. Smith and A. Stirling (2004), Socio-technological regimes and
transition contexts in B. Elzen, F. Geels and K. Green (eds), System Innovation
and the Transition to Sustainability: theory, evidence and policy, Cheltenham, UK
and Northampton, MA, USA: Edward Elgar.
Berlinski, D. (1976), On Systems Analysis: An essay concerning the limitations of
same mathematical models in the social, political, biological sciences, Cambridge:
MIT Press.
Bernstein, P. (1996), Against the Gods: the remarkable story of risk, London:
Wiley.
Bezembinder, T. (1989), Social choice theory and practice, in C. Vlek and
D. Cvetkovitch, Social Decision Methodology for Technical Projects, Dordrecht:
Kluwer.
Bijker, W. (1995), Of Bicycles, Bakelite and Bulbs: toward a theory of sociotechnical
change, Cambridge, MA: MIT Press.
Blair, T. (2000), T. Blair, speech delivered by the UK Prime Minister to The
European Bioscience Conference, London, Monday, 20 November, available at:
http://www.monsanto.co.uk/news/ukshowlib.phtml?uid4104.
Boehmer-Christiansen, S. (1994), The precautionary principle in Germany:
enabling government, in T. ORiordan and J. Cameron (eds), Interpreting the
Precautionary Principle, London: Cameron May.
Bohmann, J. (1996), Public Deliberation: pluralism, complexity and democracy,
Cambridge, MA: MIT Press.
Bonner, J. (1986), Politics, Economics and Welfare: an elementary introduction to
social choice, Brighton: Harvester Press.
Bourdieu, P. and L. Wacquant (1992), An Invitation to Reflexive Sociology,
Cambridge: Polity Press.
Boyce, A. (1996), Environment Technology: preserving the legacy, Chichester: Wiley.
Brealey, R. and S. Myers (1988), Principles of Corporate Finance, 3rd edn, New
York: McGraw Hill.
Brooks, H. (1986), The typology of surprises in technology, institutions and devel-
opment, in W.C. Clark and R.E. Munn (eds) (1986), Sustainable Development of
the Biosphere, Cambridge: Cambridge University Press.
Brown, G. (2004), speech delivered by the UK Chancellor to UK Government
Conference on Advancing Enterprise, London, 26 January, available at:
http://www.hm-treasury.gov.uk/newsroom_and_speeches/speeches/
chancellorexchequer/speech_chex_260104.cfm.
Byrd, D. and C. Cothern (2000), Introduction to Risk Analysis: a systematic
approach to science-based decision making, Government Institutes, Rockville
Maryland.
Callon, M., J. Law and A. Rip (1986), Mapping the Dynamics of Science and
Technology: sociology of science in the real world, Basingstoke: Macmillan.
CEC, (2000), Towards a European Research Area, Communication from the
Commission to the Council, COM(2000)6, Brussels: Commission of the
European Communities, January.
264 Knowledge production and assessment
EEA (2000), Chemicals in the European Environment: Low Doses, High Stakes?,
Copenhagen: European Environment Agency.
EEA (D. Gee, P. Harremoes, J. Keys, M. MacGarvin, A. Stirling, S. Vaz and
B. Wynne 2001), Late Lesson from Early Warnings: the precautionary principle
18982000, Copenhagen: European Environment Agency.
European Science and Technology Observatory (1999), On science and pre-
caution in The Management of technological risk, volume I: synthesis study
report to the EU Forward Studies Unit by European Science and Technology
Observatory (ESTO), IPTS, Sevilla, EUR19056 EN, available at: ftp://ftp.jrc.es/
pub/EURdoc/eur 19056IIen.pdf.
Europta, L. Kluver et al. (2002), European participatory technology assessment:
participatory methods in technology assessment and technology decision-
making, Danish Board of Technology, Copenhagen, available at: http://www.
tekno.dk/pdf/projekter/europta_Report.pdf.
Faber, M. and J. Proops (1994), Evolution, Time, Production and the Environment,
Berlin: Springer.
Farber, S. (1995), Economic resilience and economic policy, Ecological Economics,
15, 1057.
Farman, J. (2001), Halocarbons, the ozone layer and the precautionary principle,
in EEA, 2001.
Feyerabend, P. (1975), Against Method, London: Verso.
Feyerabend, P. (1978), Science in a Free Society, London: Verso.
Fiorino, D. (1989), Environmental risk and democratic process: a critical review,
Columbia Journal of Environmental Law, 14, 501.
Fischer, F. (1990), Technocracy and the Politics of Expertise, Newbury Park, CA:
Sage.
Fischo, Baruch (1995), Risk Perception and Communication Unplugged: Twenty
Years of Progress, Risk Analysis, 15 (2), 13745.
Fisher, E. and R. Harding (eds) (1999), Perspectives on the Precautionary Principle,
Sydney: Federation Press.
Ford, J. (1983), Choice, Expectation and Uncertainty: an appraisal of G.L.S. Shackles
Theory, Totowa, NJ: Barnes and Noble.
Forster, M. (1999), How do simple rules t to reality in a complex world?, Minds
and Machines, 9, 54364.
Funtowicz, S. and J. Ravetz, (1989), Managing the Uncertainties of Statistical
Information, in J. Brown (ed.), Environmental Threats, London: Pinter,
pp. 95117.
Funtowicz, S. and J. Ravetz (1990), S. Funtowicz, J. Ravetz, Uncertainty and Quality
in Science for Policy, Amsterdam: Kluwer.
Giddens, A. (1976), The New Rules of Sociological Method, London: Hutchinson.
Giddens, A. (1990), The Consequences of Modernity, Stanford, CA: Stanford
University Press.
Gouldner, A. (1970), The Coming Crisis of Western Sociology, London: Heinemann.
Grabher, G. and D. Stark (1997), Organizing diversity: evolutionary theory,
network analysis and postsocialism, Regional Studies, 31 (5), 53344.
Graedel, T. and B. Allenby (2002), Industrial Ecology, 2nd edn, New Jersey:
Prentice Hall.
Grin J., H. van de Graaf, and R. Hoppe (1997), Technology Assessment through
Interaction: a guide, The Hague: Rathenau Institute.
Grove-White, R., P. Macnaghten, S. Mayer and B. Wynne (1997), Uncertain world.
266 Knowledge production and assessment
Genetically modied organisms, food and public attitudes in Britain, Centre for
the Study of Environmental Change, Lancaster University.
Grove-White, R., P. Macnaghten and B. Wynne (2000), Wising up: the public and
new technologies, Centre for the Study of Environmental Change, Lancaster
University.
Grupp, H. and Linstone H. (1999), National technology foresight activities around
the globe: resurrection and new paradigms, Technological Forecasting and Social
Change, 60, 8594.
Hacking, I. (1975), The Emergence of Probability: a philosophical study of early
ideas about probability induction and statistical inference, Cambridge: Cambridge
University Press.
Hanley, N. and C. Spash (1993), CostBenet Analysis and the Environment,
Cheltenham, UK and Northampton, MA, USA: Edward Elgar.
Hales, S. and R. Welshon (2000), Nietzsches Perspectivism, Illinois University
Press.
Hansen, A. and C. Clausen (2000), From participative TA to TA as participant
in the social shaping of technology, TA Datanebank Nachrichten, 3 (9),
October.
Harremoes, P. (2000), Methods for integrated assessment, paper to International
Workshop on Information for Sustainable Water Management, Nunspeet, The
Netherlands, 2528 September.
Hayek, F. von (1978), New Studies in Philosophy, Politics, Economics and the History
of Ideas, Chicago University Press.
Hey, E. (1992), The Precautionary Principle and the LDC, Erasmus University.
HMT (2004), Science and innovation investment framework, 20042014, London:
HM Treasury, July, available at: http://www.hm-treasury.gov.uk/spending_review/
spend_sr 04/associated_documents/spending_sr 04_science.cfm.
Hogwood, B. and L. Gunn (1984), Policy Analysis for the Real World, Oxford:
Oxford University Press.
Holling, C. (1994), Simplifying the complex: the paradigms of ecological function
and structure, Futures, 26 (6), 598609.
Hughes, T. (1983): Networks of Power: electrication in western society 18801930,
Baltimore, MD: Johns Hopkins University Press.
Hunt, J. (1994), The social construction of precaution, in T. ORiordan and
J. Cameron Interpreting the Precautionary Principle, London: Earthscan.
Irvine, J. and B. Martin (1984), Foresight in Science: Picking the Winners, London:
Pinter.
Irwin, A. (1995), Citizen Science: A Study of People, Expertise and Sustainable
Development, London: Routledge.
Jackson, T. and P. Taylor (1992), The precautionary principle and the prevention
of marine pollution, Chemistry and Ecology, 7, 12334.
Jaeger, C., O. Renn, E. Rosa and T. Webler (2001), Risk: uncertainty and rational
action, London: Earthscan.
Janssen, R. (1994), Multiobjective Decision Support for Environmental Management,
Dordrecht: Kluwer.
Jasano, S. (1990), The Fifth Branch: Science Advisers as Policymakers, Cambridge
MA: Harvard University Press.
Jaulin, L., M. Kieer, O. Didrit and . Walter (2001), Applied Interval Analysis,
London: Springer Verlag.
Jaynes, E. (1986), Bayesian methods: general background, in L. Justice (ed),
Precaution, foresight and sustainability 267
INTRODUCTION
The more accepted the concept of sustainability becomes, the more obvious
are the shortcomings of current forms of policy making and knowledge
production. The departmentalisation of policy making and the isolation of
research disciplines cannot provide the integrated solutions needed to
pursue sustainable development. Implementing sustainable development
depends on the integration of dierent areas of policy making and
conicting goals with research and practical experience. However, attempts
at integrating and co-ordinating policy areas, research disciplines and praxis
have been increasing over recent years. Not only are there more mechanisms
and initiatives to increase the coherence and co-ordination of policy elds
to support sustainable development, but also in research policy the design
of programmes has begun to reect the need for greater integration.
Research activities have always been valued for their contribution to fur-
thering the aims of sustainable development both through activities on
understanding the natural environment and through analysing the impact
human activities can have. Such research activities have led to the develop-
ment of a wealth of instruments for dening what sustainable development
means in concrete areas and measuring the current distance from it. More
recently, however, the demands on the research system have been changing
and expanding. Research is no longer seen as the mere provider of infor-
mation on the basis of which policy decisions can be taken and targets
determined. Research is now expected to be a more active player in facili-
tating and dening change. It should play a role in moving from the provi-
sion of information to the development of integrated solutions. Research
on goals and indicators for sustainable development can be brought
together with the problem of nding context-specic solutions. In this way
abstract goals can be confronted with looking at the expectations, values
and behavioural patterns of the actors involved.
273
274 Knowledge production and assessment
Over the past two decades most European countries have designed and
established research programmes that specically focus on sustainable
development. Putting sustainability as the main goal of the programmes
The (re)search for solutions 275
funds and more generic funds. However, this research has shown that in
many cases programmes are important ways for the policy-making process
to introduce and direct new areas of research.
The aim of this book is to analyse to what extent existing policies already
contain elements of reexive governance. Such an analysis can be con-
ducted by assessing the ve strategy implications outlined: transdiscipli-
nary knowledge production, adaptivity of strategies and institutions,
anticipation of long-term eects, participatory goal formulation and inter-
active strategy development. These ve elements can be applied to the way
in which research programmes are designed and established. However,
before the programmes can be analysed it is necessary to develop a frame-
work to assess the programmes.
The (re)search for solutions 277
There is no such thing as a denitive role for research that supports sus-
tainable development. Dening the role research should play in the tran-
sition process takes place separately in each country on the national level.
The reason for starting with this meta-level approach to analysing the
programmes is that it provides an important contextual background for
understanding the design and development of the individual programmes.
Analysing the denition sets the framework within which the programmes
are established and implemented. Dening the role that research can play
278 Knowledge production and assessment
Following the denition of the role that research can play for sustainable
development, comes the need to create specic thematic programme lines. In
many cases, the overlying denition given to the research activities plays a
large role in creating the thematic content of the programmes by inuencing
the methods used to dene the content. The methods can range from linking
the thematic content to national sustainable development plans to using
methods such as foresight and backcasting to identify long term goals and
measures to achieve them. However, although the programmes use dierent
methods to dene the content, there are similarities in the approaches taken,
as programmes that support sustainable development tend to take a broader
and systemic approach to dening the thematic content of the programmes.
The focus is on systems or on chains and thus allows a problem to be exam-
ined and analysed from a holistic point of view and to understand the causes
and not just the eects of the problem. A regional focus (either rural or
urban) to such programmes is a common way of dening such a system.
Others approaches to dene the borders of the system under analysis include
innovation processes, consumption and production processes and health
issues. The six most common programme categories that can be identied are:
The above list gives an idea of the systems approach that research pro-
grammes in support of sustainable development take. However, although
many of the overall thematic topics appear to be similar, it would be mis-
leading to believe that the implementation and the detailed denition of the
programmes within the specic national contexts is the same in each cate-
gory. There are considerable thematic dierences between the programmes
in the dierent countries as the thematic approach is often determined by
national cultural specicities and problems.
The (re)search for solutions 279
Selecting Projects
Once the thematic content of the programmes is in place, the next level to
be dened is the selection of individual projects within the programmes.
This concerns the issue of how to ensure that the individual projects
comply with the overall thematic and programme goals and to guarantee
that the project level also fulls criteria related to sustainable development.
Many programmes have developed methodologies for implementing
research for sustainable development and have dened sets of criteria (see
below) for ensuring that each individual project contributes to further the
aims of sustainable development. This has been important to ensure
quality assurance in projects that do not adhere to formal current research
methods and practices.
Since the rst sustainable development programmes were established in
the Netherlands, the body of literature on designing research for sustainable
development has been growing. There is considerable consensus that
research for sustainable development diers from more conventional types
of programmes and that the design of such research activities needs to adapt
to integrate both inter-disciplinary and trans-disciplinary research activities.
Using conventional research organisation methods in such circumstances
would not achieve the desired results. Some programmes have put the
process design element rst, ahead of thematic goals, claiming that achiev-
ing the right process will have a signicant eect on the thematic content.
Some of the criteria used for selecting projects are now relatively well
understood whereas the inclusion of others in the design of research pro-
grammes has not yet been fully operationalised. In many cases this is due
to the fact that these concepts are not easily introduced, but entail a com-
plete restructuring of the research process. The most commonly used
elements for designing programmes include interdisciplinarity, stakeholder
involvement, inclusion of dierent time frames and geographical areas and
an emphasis on the dissemination of the results.
Inter-disciplinarity, or the integration of dierent academic disciplines in
a research project is often dened as a prerequisite for problem-solving
research activities and the majority of the SD targeted programmes dene
the inclusion of dierent disciplines as a core element of the research activ-
ities. Equally, stakeholder involvement (trans-disciplinary research) or the
involvement of non-academic stakeholders has been dened as one of the
main criteria for selecting projects for sustainable development research.
Stakeholders not only have a dierent understanding of the problem, they
are also able to implement change. Other criteria for selecting are not as
well understood. These include the inclusion of dierent time frames, geo-
graphical areas and dissemination despite being considered important.
280 Knowledge production and assessment
Learning
Country Programme
Austria Austrian Landscape Research
(individual Austrian Programme on Technology for Sustainable
programmes) Development
PFEIL 05 Programme for Research and Development in
Agriculture, Forestry, Environment and Water
Management
Belgium Scientic Support Plan for a Sustainable Development Policy 1
(umbrella (Sustainable management of the North Sea, Global
programmes Change and Sustainable Development, Antarctica 4,
and sub- Sustainable Mobility, Norms for Food Products, Telsat 4,
programmes) Levers for a Sustainable Development Policy and
Supporting actions)
Scientic Support Plan for a Sustainable Development Policy 2
(Sustainable Modes of Production and Consumption,
Global Change, Eco Systems and Bio-diversity,
Supporting Actions and Mixed Actions)
Scientic Support to an Integration of Notions of Quality and
Security of the Production Environments, Processes and
Goods in a Context of Sustainable Development
Germany Research on the Environment
(umbrella Research on Sustainable Economic Management,
programme Regional Sustainability, Research on Global Change,
and sub- Socio-Ecological Research)
programmes)
The Netherlands Economy, Ecology and Technology (EET)
(umbrella Dutch Initiative for Sustainable Development (NIDO)
programme HABIFORM (Expertise Network Multiple Use of Space)
with struc-
tured and
coordinated
individual
programmes)
282 Knowledge production and assessment
Country Programme
Sweden Urban and Regional Planing
(individual Infrasystems for Sustainable Cities
programmes) The Sustainable City
Economics for Sustainable Development
Sustainable Forestry in Southern Sweden
Sustainable Food Production
Sustainable Coastal Zone
Sustainable Management of the Mountain Region
Paths to Sustainable Development Behaviour, Organisations,
Structures (Ways Ahead)
Innovation Systems Supporting a Sustainable Growth
UK3 Environmental Strategy Research Programme
(individual Towards a Sustainable Urban Environment
programmes) EPSRC Infrastructure and Environment Programme
Environment Agency Sustainable Development R&D
Programme
Sustainable Development Commission
Sustainable Technologies Initiative LINK Programme
the policy-making process. The rst support plan was conceived in 1996 and
moved into its second phase in 2000. The plan is closely linked to the Belgian
Federal Plan for Sustainable Development. Although the aim of the pro-
gramme is to support an integrated approach of both policy areas and
scientic disciplines, it does not display the same level of methodological
development that forms the basis of the German and Dutch programmes.
The nal two national programmes described in this section, Sweden and
the UK, have had signicantly higher barriers to overcome in their respect-
ive research systems. This has inuenced their ability to dene research for
sustainable development. The funding structures of both countries are
fragmented and consist of individual research councils who bear the main
responsibility for dening research activities in their own elds. The
co-ordination barriers to dening research policy for sustainable develop-
ment are considerably higher than in the three countries described previ-
ously. In Sweden and in the UK there are many more smaller and less
co-ordinated research programmes that thematically t into the competen-
cies of one of the funding bodies. This does not mean that the individual
research activities are not interesting and well developed, but that the
approach taken towards research for sustainable development is neither
as methodologically nor as thematically developed as in the Dutch or
German case.
The UK has taken several steps in recent years to rectify the lack of
co-ordination and strategy regarding research policy for sustainable devel-
opment. It has established a research network for sustainable development.
The network aims to become a clearing house for research that addresses
sustainable development. Research for sustainable development is dened
as research that combines environmental research with either the economic
or social dimension, or the one-plus-one criterion (Eames and Policy
Studies Institute 2001).
The type of research and the role of research for sustainable development
identied by national research policy dene to a certain extent the frame-
work for the thematic content. This can be seen in the case of the Dutch
programme described above and its focus on innovation processes.
However, the focus in the Netherlands is equally formed by the specic
Dutch geographical, economic, social and environmental situation. The
thematic focus is closely linked to immediate environmental problems.
The Netherlands has a particular geographical situation that consists of a
high water table and slow drainage. On the other hand, the country gener-
ates high levels of waste due to its processing activities mainly from its
286 Knowledge production and assessment
to address the dierent levels. The structured nature of the activities in the
Netherlands entails good coverage of issues and also consists of formal
reviews that are able to identify important gaps, also through foresight
studies. This is by far the most sophisticated method for dening short,
medium- and long-term goals in research programmes.
interest groups from the region. The goal of the projects is to combine
theory with practice and to initiate social change through developing new
models and options for acting. Experiences so far have highlighted the
diculties of integrating the social dimension. Addressing all three dimen-
sions of sustainable development together can cause conicts. However,
valuable lessons have been learned concerning the involvement of non-
scientic actors. Such actors have to be included in the project at an early
stage and kept involved throughout. They should not just be handed the
nished project.
Another example is the funding measure, Framework for Innovations
Towards a Sustainable Economic Behaviour. This sub-programme aims to
expand on the achievements made in technical eciency by including the
social dimension. It looks at three central questions: the role of the State,
the development of indicators for sustainable development, and the inte-
gration of political institutions such as the Federal Environment Oce, the
Ministry of the Environment and the Ministry of Trade and Commerce
into the research process, as these institutions are responsible for the frame-
work conditions for innovation. One of the interesting aspects of the Socio-
Ecological Research Programme is the continuous development of the
research methodology. An example is the project Evalunet, an infrastruc-
ture project, which is aimed at developing assessment methods and crite-
ria for trans-disciplinary research.
The Belgian approach is dierent again as its current Scientic Support
Plan for a Sustainable Development Policy (SPSD II) consists of two parts
that are linked by separate projects. The rst part has a greater social
science focus on issues such as energy, agro-food and transport, and the
second part has a greater environmental focus on climate and ecosystems.
These two parts do not attempt to integrate disciplines. This is left up to the
mixed actions that aim to draw on the results of both parts. The selection
criteria for these projects are based on the precautionary principle, vertical
and horizontal policy integration and social equity to name just a few
(Verbeiren 2002). The Belgium programme is very focused on the require-
ments of the policy process. It does not include actors other than those
from the policy-making process.
Developing selection and assessment criteria that are sensitive to
research for sustainable development is also important. The German
Socio-Ecological Programme has taken steps in this direction and assesses
a projects ability to formulate the problem eld clearly. Projects are fur-
thermore assessed as to their relevance to three cross-cutting central
topics: theoretical development of methods and instruments, the relation-
ship between theory and practice, and the relationship between gender and
environment.
The (re)search for solutions 289
Learning
NOTES
1. This chapter is based on the paper Organising Research for Sustainable Development: An
assessment of National Research Programmes rst presented at the Conference on the
Human Dimensions of Global Environmental Change Knowledge for the Sustainability
Transition: The Challenge for Social Science, 67 December 2002 in Berlin.
2. The paper is based on the results of the study Identifying and assessing national research
activities on sustainable development set up through the ESTO network in February
2001. The study identied and assessed the national SD research programmes of seven
European countries: Austria, Belgium, Germany, The Netherlands, Portugal, Sweden and
the UK. One hundred and two programmes were assessed using both a set of thematic
and process-orientated criteria.
3. To be able to compare the programmes across the seven countries, not all UK programmes
identied in the country report as addressing SD have been included in this table.
4. NIDO Nederlands Initiatief voor Duurzame Ontwickkeling (Dutch Initiative for
Sustainable Development).
5. EET Economics, Ecology and Technology Programme.
6. ICES-KIS is the Inter-ministerial working group that advises on strengthening the know-
ledge infrastructure. It advises on the spending of a budget drawn from natural gas rev-
enues. This budget funds the programmes HABIFORM Expertise Network Multiple
Use of Space, CONNEKT transport congestion programme, NIDO Economics,
Ecology and Technology Programme and SKB Soil Quality Management.
7. It should be noted that Belgium (together with the Netherlands) is one of the only coun-
tries that specically links its sustainable development programme to its federal sustain-
able development plan. In other countries the responsibilities are divided among dierent
ministries or research councils.
The (re)search for solutions 293
REFERENCES
INTRODUCTION
FROM INTERDISCIPLINARITY TO
TRANSDISCIPLINARITY
294
Integrating perspectives in the practice of transdisciplinary research 295
CHALLENGES IN TRANSDISCIPLINARY
RESEARCH
Focusing on Diversity
Steering Flexibility
The introductory chapter to this book refers to the need for new strategies
of coping with the complexity of interlinked social, technological and eco-
logical development claiming that reexive governance modes should be
based on processes for continued learning and modulation of ongoing
developments, rather than on complete knowledge and maximization of
control (Vo and Kemp, this volume). Could this reexive approach for
dealing with complexity possibly be ecient not only for organizing tran-
sition processes in society but equally productive within the process of
transdisciplinary knowledge production and value balancing itself ?
Frequent painful experiences with unpleasant instabilities of agreements
and a general lack of control and predictability in transdisciplinary
research have revealed the striking similarities between steering problems
in society and steering problems in complex research settings.
The mix of disciplinary and professional project partners in transdisci-
plinary teams is not only a strategy to maximize the knowledge base of a
project but also a strategy to reproduce the heterogeneity of competing
real world interests and rationales. It consequently results in a multitude
of uncertainties and unforeseeable team dynamics. Far from fullling
classic standards of traditional disciplinary research, such transdisci-
plinary projects seem to systematically undergo radical relaunches con-
cerning research objectives, methodological strategies, time planning and
product development. In consequence, the projects are often confronted
with reproaches that concern insucient quality management (muddling
through) and reduced scientic consistency of the results due to the mul-
titude of adaptations in research plans.
That is the reason why there is in addition to knowledge integration a
second key challenge in transdisciplinary research: the implementation of
exible but nonetheless stable steering processes and decision models that
derive eciency not from xed targets and criteria but from uttermost
Integrating perspectives in the practice of transdisciplinary research 299
In watching and analysing the special team dynamics, which are being set
o by priorily focusing dierences within a team instead of common
denominators, a surprising impression could be experienced throughout
two long-term monitoring studies for an extensive Austrian research
programme on sustainable regional development.2 Within the monitoring
study experiences of researchers have been analysed with dierent team
models, working methods and integration strategies implemented in
11 interdisciplinary and transdisciplinary projects of this large research
programme. These teams had up to 20 members, including a multitude of
disciplines and many non-scientic partners.
It was one of the most interesting and useful outcomes of this study, that
team performance is not necessarily combined with a high degree of accor-
dance amongst team members but that it rather seems closely connected
to the conscious handling of team heterogeneity and to a very open and
down-to-earth analysis of divergences. A most striking result of the moni-
toring study was the observation that a number of projects that had been
merged by the programme management and that consequently passed
through an extremely dicult start phase of team building, performed very
well in the end: these merged teams soon enjoyed and cultivated the very
direct style of discussion they had developed during the hot kick-o periods
of the projects, and they felt more satised with their scientic results than
other teams, that had started o in harmony and that later on had great
diculties in establishing a culture of open confrontation of contradictory
views. What was it, that enabled these merged projects to perform success-
fully after having passed an extremely dicult start, marked by experiences
of open disunity and heavy conicts? These teams had done something in
the beginning, that is unusual amongst new teams that generally want to
300 Knowledge production and assessment
four programmes (Loibl 2003). The three-level steering approach has been
derived by integrating the ndings of the preceding Austrian monitoring
study and the following international survey. It has been successfully tested
in dierent contexts like project meetings, team coaching and quality
circles. Its objective is to connect project development and team building by
training group capacities in reecting dierent functional rationales, as well
as dierent individual relevance lters and success criteria. The steering
model is based on the experience that a fruitful solution to the problems
caused by team diversity is to use a self-reexive learning approach by fre-
quently and systematically changing between three dimensions:
1. The rst dimension reects the real-world problems upon which the
research project focuses: transdisciplinarity strives to produce innova-
tive insights into the dynamics behind the examined problems and to
create sustainable solutions by better integrating and balancing the
diversity of societal interests, institutional structures and sector poli-
cies (external heterogeneity).
2. The second dimension constitutes the operative working process
within the research project: it is this second analytical dimension that
reects the plurality of disciplinary knowledge, of practical expertise
and of working methods, forming the joint knowledge base of trans-
disciplinary project teams: the diversity of various techniques of
problem analyses and data interpretation and the dierent cultures of
communication and project management (internal heterogeneity).
3. It is the third dimension, which represents the meta-level of building a
common understanding for patterns and functionality of heterogene-
ity itself, which stimulates the crucial process of knowledge-integration
within transdisciplinary teams. For the societal project of sustain-
ability, it is of increased importance to discover the dynamics behind
the formation of dierent value systems and relevance perceptions in
society. Moreover, it is just as important for the success of heteroge-
neous teams to nd out which relevance criteria are of higher import-
ance to specic team members than to others and to understand why
it is these very criteria and not others. Relevance scales and success crit-
eria employed within the research team are obviously not strictly iden-
tical to those of the actors in the problem eld, but they will
correspond if the transdisciplinary team is well composed. The third
dimension of analysis focuses on eects of societal and institutional
plurality within the microsystems of transdisciplinary research teams.
A frequent problem that occurs when teams want to work with this three-
dimensional approach is that the very special competencies which are
Integrating perspectives in the practice of transdisciplinary research 303
task forces. The dominant orthodox strategies and majority positions within
their background systems get out of sight or are considered outdated at best.
This might work very well until the reform-oriented results begin to cause
irritation amongst the more conservative colleagues or superiors of the
department and at this moment many transdisciplinary projects encounter
heavy turbulence sometimes so heavy that the projects have to undergo
complete relaunches or to live with the fact that their results will have very
little impact because of system resistance. In order to optimize the output
(products) as well as the outcome (policy implementation) of transdiscipli-
nary sustainability research, it is of predominant importance to understand
the various dynamics of system stabilization and reformation.
continuing as planned would be higher than the risk of change for the nal
success of the project. But in order to take such decisions after having
reected on interim ndings and team experiences, it is necessary to
monitor constantly the progress of a project and to oblige explicitly all
team members to keep an eye on the question of whether the research path,
the instruments and products still correspond to the increased and inte-
grated knowledge base the project has reached at that time.
The possible range of research questions and corresponding team com-
positions in sustainability research make it extremely dicult to dene a
set of general success criteria of project progress to be observed for self-
monitoring. The eort of discussing and agreeing on a set of contextual-
ized success criteria for the progress of the specic project is considered a
crucial element for reexive project steering, because it is mostly through
this eort that tacit knowledge bases of the team members can be made
explicit and available for the entire team.
Constantly monitoring and analysing the performance of social systems
(be this a research team, an enterprise, a community or society as a whole)
and systematically opening up the complexity of competing but inter-
linked interests is essential for promoting sustainable development. But
in order to manage transdisciplinary research (and reexive governance)
successfully in practice, a complementary challenge of closing down again
this process of unfolding the multiplicity of heterogeneous interests and
action strategies must be mastered (Vo, Kemp and Bauknecht, Chapter
16, this volume). After the phase widening up the analytical approach to a
maximum, it is necessary to narrow the focus again and to reduce the com-
plexity of perceptible system dynamics for developing a manageable port-
folio of research products and action strategies. In order to master this
challenge, a reliable model of nal decision-making must be formally
agreed in the beginning of a transdisciplinary research project. It is indis-
pensable though, to integrate strong participatory elements in this decision
model because disciplinary as well as functional dierentiation needs the
element of participation for reintegrating partial problem perceptions and
identifying externalities and blind spots. Team members should not only
participate in designing the research plan and in dening the rules for team
co-operation, they should in particular participate in decisions on method-
ological question, criteria weighting and interpretating results.
In order to produce coherent project results and not just a sample of multi-
disciplinary problem views and solution approaches, transdisciplinary
teams need to integrate working steps and interim ndings throughout
Integrating perspectives in the practice of transdisciplinary research 307
CONCLUSIONS
NOTES
REFERENCES
INTRODUCTION
Policies for science and technology must always be a mixture of realism and
idealism. (Freeman, 1991)
313
314 Development of technology and policy
SNM has been promoted since the late 1990s (Schot et al., 1994; Kemp
et al., 1998; Weber et al., 1999; Kemp et al., 2001; Hoogma et al., 2002).
Work has been conducted in several areas, including transport, energy and
wastewater. Many of these are ex post analyses in which an SNM interpret-
ation is made for the success or otherwise of experiments in sustainability
(success being niche growth, branching and inuence upon the incumbent
technology system). SNM is, however, proposed as a forward-looking
policy approach. As such, it has been discussed in a Working Paper by the
Science and Technology Foresight Oce of the European Commission
(von Schomberg, 2002). Most advanced, however, are Dutch government
attempts to incorporate a version of it into its technology policy (Hoogma
et al., 2002; Rotmans, 2003; VROM, 2003; Ministry of Economic Aairs,
undated).
316 Development of technology and policy
Origins
Objectives
Model of change
The introduction of novelty has been studied in great detail. However, the
adoption of novelty is decisive for society, not its introduction. Adoption is an
active process, and has elements of innovation itself. Individual behaviour,
organisations, and society have to rearrange themselves to adopt, and adapt to,
innovation. In this sense, the introduction of a new technology is an unstruc-
tured social experiment. (Rip and Kemp, 1998: 338)
the Americas in the nineteenth and early twentieth century) that reinforced
demand for regular passenger shipping, and sustained steam technology
diusion. The second set of lessons is consequently more outward looking:
which types of institutional reforms can help niche growth? SNM advo-
cates describe the purpose of this set of lessons as:
From the perspective of regime management, the Danish policy is very inter-
esting. It conrms our model of technological transitions about the importance
of the coincidence of successful niche policies against the backdrop of chang-
ing regimes. It also shows the importance of learning, the creation of new actor
networks, and changes in the institutional framework. More importantly, it
demonstrates some of the advantages of a exible, sequential policy aimed at
modulating the dynamics of socio-technical change into socially benecial
directions and using windows of opportunity within the evolving dominant
regime. (Kemp et al., 2001: 287)
It is also a case whose roots lie in the AT movement. Here is a direct link
between the two approaches. What SNM considers an exemplar is consid-
ered a success story in grassroots innovation (Douthwaite, 2002). It was AT
idealists that were opposed to nuclear power who seeded the billion-euro
export industry of today.
Origins
The AT movement was the R&D department for Utopia. It combined the
reality of environmental degradation with the idealism of the new left and
counter-culture (Veldman, 1994). Activists were interested in technologies
that would serve a society radically dierent to industrial capitalism
(Dickson, 1974). They demanded a transformation of technology systems
(and society) into forms that did not threaten ecological catastrophe, and
which were much more convivial in use. AT would not be as alienating or
soul-destroying to work and live with, compared to the mass production
and consumption oered by the large corporations. AT was utopian in
the sense that widespread expansion of the niches they created would be
virtually impossible within the existing structure of society (Dickson,
1974: 99). Might the same be said of SNM?
Today, apart from a visitor centre in Wales (the Centre for Alternative
Technology attracts around 60 000 visitors each year), the label AT is
rarely used in the UK. The Alternative Technology Group at the Open
University which was home for some AT movement intellectuals
became the Energy and Environment Research Unit in 1986. What was the
Urban Centre for Alternative Technology in Bristol is now the Centre for
Sustainable Energy, with over 30 sta promoting community energy pro-
jects nationally. Nevertheless, alternative niches and ideas continue to be
(re)created at the grassroots level. Beneath the slow evolution of govern-
ment and corporate policies for sustainable technology, networks of grass-
roots innovators continue to create alternatives to the mainstream. Niches
exist in community-supported organic food schemes, eco-housing and
community energy projects. Not all participants will have heard of AT, nor
SNM for that matter, and yet, in their own way, these grassroots innovators
reclaim an approach last attempted by AT.
Niche-based approaches to sustainable development 323
Objectives
Model of Change
A further debate within the movement was over the purposes of AT niches
in relationship to social change. Some activists foresaw the widespread
diusion of AT, but recognised that this would not be possible without a
radical transformation of society. Others were primarily interested in devel-
oping the ecologically-harmonious technologies that would facilitate their
retreat from the system into relatively self-sucient communities (usually
in rural locations). The idyll created by the latter might or might not be
intended as a beacon for others to follow, if they so wished (Rivers, 1975).
An accommodation was reached between the social revolutionaries and
the back-to-the-land folk, because they could meet around the locus of AT
niches. For the rst group, niches pregured the technology systems that
Niche-based approaches to sustainable development 325
So far we have introduced both SNM and AT. SNM was introduced as a
candidate in governance innovation for dealing with reexive governance
in technology issues. The objective now is to try and draw some lessons
through comparative analysis. There is a gap of 30 years between AT and
Niche-based approaches to sustainable development 327
SNM. The gap is felt in the language of each approach and the ideals
associated with their experiments. So what can we learn from a compari-
son, and with what implications for future research? Table 12.1 compares
the approaches in summary form. This section discusses the issues that a
comparison with AT raises for SNM.
Niche Purposes
Widening Participation
with business and government. SNM calls for government and business to
engage more with civil society in innovation. Whilst widening participation
brings in wider knowledge, it also opens up the challenge of reaching con-
sensus between diverse groups, social values and priorities. More radical
niches will experience a greater gulf between advocates and mainstream par-
ticipants, presenting a challenge for reexive governance strategies.
Beyond creating knowledge about alternatives, what are the precise mech-
anisms available for niche-based change? In SNM, niche growth generates
transformations. Given its evolutionary orientation, it is unsurprising that
growth is ensured through: (1) a degree of niche compatibility with the
dimensions of the incumbent regime; (2) niche performance should be
robust, and (3) that the niche shows development potential, for example
through extension to new applications and, especially, markets (Weber
et al., 1999). Thus the idealists who might initiate a niche need to be joined
by more entrepreneurial systems builders. The managerial challenge is to
carry niche development through to the stage when it can attract the big
capital needed for the niche to become a commercial prospect. The paradox
is that the good compatibility criteria can imply that niches at radical odds
with the incumbent regime will need to oer considerable positive returns
if they are to attract investment. The alternative technologies developed in
GLEB networks struggled because, besides a hostile political climate, the
social needs they served lacked purchasing power in markets (Mole and
Elliott, 1987). In a sense, the AT movements attachment to higher level
political change recognised that their niches would never attract such
capital unless investment incentives (and institutions) were altered radically
through public policy.
There is another mechanism by which niche initiatives might inuence
mainstream change (Geels, 2002). Some aspect of the niche activity may
Niche-based approaches to sustainable development 331
modication? In other words, unless the mainstream buys into the values
embodied in the niche, then the scaling-up of the niche will be distorted
perversely or simply not happen because adoption demands too much
change to the mainstream. On the other hand, if socio-technical practices
can easily be stripped of values, and adapted to suit incumbent interests
and contexts, then the niche socio-technical conguration may well diuse
much more easily, but with key sustainability features becoming lost in the
translation. Niche inuence is wider, but less profound.
CONCLUSIONS
AT was rooted in civil society, and was inclined toward political lobbying
as well as grassroots initiative. The AT experience illustrates how innovative
ideas can emerge through informed dissent beyond conventional business
contexts. SNM is oriented more toward policy-makers and technology
producers, and suggests how they may engage with change by including
leading users and other participants. In practice, the ability to create niches
with transformation potential is distributed, unevenly, across a variety of
social actors. It makes sense to include technology producers, users, regu-
lators and others with the necessary skills, money, tools, knowledge, and
power to force legitimate change.
However, as AT debates illustrate, there exists exibility over how a
sustainability problem gets interpreted. Diverse solutions are favoured
dierently by dierent actors, depending upon their ideas, values and expe-
rience. Integrated knowledge production will have to navigate the (some-
times conicting) social values attached to dierent knowledge claims and
priorities. In short, how can world views be bridged and the energies and
capabilities of political activists and business investors harnessed more
eectively? Though both elements are important for progressive change, it
may simply be unrealistic to expect them to co-ordinate neatly in the
systematic way anticipated by some governance innovations.
Niche-based approaches to sustainable development 333
and lever in the resources to create their ideal niches. Governance strategies
committed to sustainable development might consider how better to
support this innovative zeal. This chapter has tried to argue such support
must not merely be quantitative. There is a qualitative dimension in the way
support transmits niche lessons to the mainstream and facilitates action for
their incorporation. Indeed, new interest in niche-based approaches rekin-
dles a challenging research and policy agenda.
ACKNOWLEDGMENTS
This chapter draws upon research funded by the UK Economic and
Social Research Council (ESRC) under their Sustainable Technologies
Programme. I am grateful to all those people who have given freely of
their time, conversation and insights during the course of the research. Of
course, the views expressed here remain my responsibility.
NOTES
1. http://www.blueprint-network.net/.
2. This chapter is based on an earlier version published as Smith, A. (2004) Alternative
technology niches and sustainable development, Innovation, Management, Policy and
Practice, 6(2), pp. 22035, in A. Griths (ed.), Corporate Sustainability: Governance,
Innovation Strategy, Development and Methods (ISBN 0975043625), eContent
Management, Maleny, Queensland. I am grateful to the publishers for permission to
develop it into this chapter.
3. Grassroots innovators develop bottom-up solutions to public problems. They tend to
come from outside mainstream rms and operate instead in civil society arenas. They are
often idealists, and experiment with social innovation as much as technological hard-
ware. Indeed, limited access to nancial and other resources can force interest to focus
upon the software issues of social organisation and new rules.
4. Data supplied by the Danish Wind Industry Association.
5. The focus now is limited to sustainable energy issues, but it retains AT sensibilities with
a concern for community action. See http://eeru.open.ac.uk/natta/welcome.html.
6. Amory Lovins, a high-prole soft energy advocate, worked as Energy Campaigner for
FoE in the 1970s.
7. Which later aliated to the Labour Party.
8. The Conservative government introduced legislation restricting trades union activity;
local government spending was brought under tighter central control; and GLEB dis-
appeared with the abolition of the Greater London Council in 1986.
9. Community groups running home insulation schemes, for example, have evolved into
regional dissemination centres, such as the Centre for Sustainable Energy mentioned
earlier.
10. Such as the thousands of students who studied AT as part of courses at the Open
University.
11. Of course, uncoordinated socio-economic trends and changes might also manifest
tension in particular socio-technical regimes, such as demographic changes and water
demand. Niches may still oer solutions.
Niche-based approaches to sustainable development 335
REFERENCES
Meadows, D.H., D.L. Meadows, J. Randers and W.W. Behrens III (1972), The
Limits to Growth, New York: Universe Books.
Ministry of Economic Aairs (undated), Energy transition: impulse for sustain-
ability and innovation, Ministry of Economic Aairs, The Hague.
Mole, V. and D. Elliott (1987), Enterprising innovation: an alternative approach,
London: Frances Pinter.
Nelson, R. and S.G. Winter (1982), An Evolutionary Theory of Economic Change,
Cambridge, MA: Harvard University Press.
Ornetzeder, M. (2001), Old technology and social innovations: inside the Austrian
success story on solar water heaters, Technology Analysis and Strategic
Management, 13 (1), 26978.
Rip, A. and R. Kemp (1998), Technological change, in S. Rayner and E.L. Malone
(eds), Human Choices and Climate Change, Volume 2 Resources and Technology,
Columbus, OH: Battelle.
Rivers, P. (1975), The Survivalists, London: Methuen.
Rotmans, J. (2003), Transitiemanagement: Sleutel voor een Duurzame Samenleving,
Assen: Koninklijke Van Gorcum.
Schot, J., R. Hoogma and B. Elzen (1994), Strategies for shifting technological
systems: the case of the automobile system, Futures, 26: 106076.
Schot, J. (1998), The usefulness of evolutionary models for explaining innovation.
The case of the Netherlands in the nineteenth century, History and Technology,
14, 173200.
Schot, J. and A. Rip (1997), The past and future of Constructive Technology
Assessment, Technological Forecasting and Social Change, 54, 25168.
Schumacher, E.F. (1973), Small is Beautiful, London: Blond and Briggs.
Soil Association (2004), Food and Farming Report 2004, Bristol: Soil Association.
Stirling, A. (1998), On the Economics of Diversity, SPRU electronic working
paper series, available at: http://www.sussex.ac.uk/spru/1-6-1-2-1.html.
Veldman, M. (1994), Fantasy, the Bomb and the Greening of Britain, Romantic
Protest, 19451980, Cambridge: Cambridge University Press.
Von Schomberg, R. (2002), The objective of sustainable development: are we
coming closer?, working paper from Oce of Science and Technology Foresight,
DG RTD, European Commission, Brussels.
VROM, Ministry of Housing, Spatial Planning and the Environment (2003),
Transitions progress report: making strides towards sustainability, VROM, The
Hague.
Wainwright, H. and D. Elliott (1982), The Lucas Plan. A New Trades Unionism in
the Making?, London: Allison and Busby.
Waks, L.J. (1993), STS as an academic eld and a social movement, Technology in
Society, 15, 399408.
Weber, M., R. Hoogma, B. Lane and J. Schot (1999), Experimenting with
Sustainable Transport Innovations: a Workbook for Strategic Niche Management,
Enschede: University of Twente Press.
Willoughby, K.W. (1990), Technology Choice. A Critique of the Appropriate
Technology Movement, London: ITDG.
Winner, L. (1979), The political philosophy of alternative technology: historical
roots and present prospects, Technology in Society, 1, 7586.
13. The sustainable transformation
of sanitation
Bas van Vliet
INTRODUCTION
337
338 Development of technology and policy
personal drinking water use (Vewin, 2004). Apart from wasting high
quality water, the dilution of human waste by huge amounts of water
makes waste water treatment highly inecient in terms of capacity use,
energy consumption and overall treatment performance. Instead of being
recovered, the nutrients from human waste are ultimately landlled, incin-
erated or discharged to open waters.
An increasing number of environmental scientists and engineers (GTZ,
2003; Lens et al., 2001) argue that all of these valuable ows could and
should be brought back into a closed-loop system, which recovers the nutri-
ents from human waste and minimises the use of energy, water and mater-
ial resources. Closed-loop systems dier radically from contemporary
waterworks and sewer systems in both technical and social respects. To
implement closed-loop systems, a strategy is needed which incorporates
both technical and social changes within the practices of sanitation. Rather
than an expert-driven approach, which leads to environmentally optimal
technical systems, there is a need to widen the closed-loop approach to a
more reexive strategy including participatory goal denition by a wide
variety of actors rather than only sanitation experts.
This chapter aims to draft the key elements of a reexive strategy of
socio-technical change in waste water systems. Firstly I elaborate in
Section 2 on the kind of socio-technical changes that are (implicitly) pro-
posed in closed-loop systems to solve the second-order sustainability prob-
lems of sewer systems. Section 3 presents two cases of experimentation with
such systems in the Netherlands. The cases are selected on the premises that
they dier in particular in terms of reexive governance. This is followed
by an evaluation of the cases (Section 4) based on the ve process require-
ments for reexive sustainable governance as presented in the Introduction
of this volume. The conclusion addresses the lessons learned and the key
elements of a strategy of sustainable transformation of sanitation as well
as a critical assessment of its implications.
Multi-level Change
Closing the loop means that the very principles of current water works
should be altered, and that large-scale physical infrastructures of water
works should be abolished and replaced over time. Proponents of such
closed-loop or eco-sanitation systems aim to introduce radically dierent
house-on-site and community-on-site systems to treat human waste. These
systems minimise water use to transport faeces and urine so as to obtain
The sustainable transformation of sanitation 339
LANDSCAPE LEVEL
The landscape metaphor refers to the structural character of its inuence
on technological development: technological trajectories are guided by
the gradients in the landscape. Changes at this level only occur gradually
and very slowly. An example of such a gradient is the robustness of socio-
cultural notions and practices that have been developed around the avail-
ability of the sewer system (and the connected devices like water closets).
At the beginning of the twentieth century, water closets became highly
desired artefacts to have installed in modern homes. Today, in many coun-
tries and areas were there is no sewer system, the water closet is still a land-
mark of a modern lifestyle. Moreover, most people who have one would
not consider the use of water closets and sewer systems as particularly
problematic, on the contrary: they are quite happy with it. The principle
of ush and forget is especially appealing when it comes to the handling
of human waste. An attempt to change the system radically will probably
render harsh public opposition, especially when such a system would
require extra eort with, or would cause a (more frequent) sight or smell
of human waste.
REGIME LEVEL
The regime level refers to the cognitive, social and technical rule-sets that
are embodied in practices, artefacts and organisations. The rules are carried
out by a heterogeneous network of producers, providers, users, researchers,
governmental and non-governmental bodies.
The new eco-sanitation techniques require a dierent regime as com-
pared to the systems in place. Sewer systems and other water works are
mostly centrally managed and controlled by state-owned organisations.
It has taken a whole century to construct not only a huge network of
pipes, pumps and treatment facilities, but also to set up a legislative and
organisational framework to govern the entire system. Eco-sanitation
The sustainable transformation of sanitation 341
SOCIO-TECHNICAL VARIABLES IN
ECO-SANITATION SYSTEMS
Large-scaled system
Central organisation Low involvement of end
users
A
Combined water
Separated water flows
flows
B C
High-involvement of Decentralised organisation
end users Small-scaled system
takers in selling the apartments, they were the rst to object to experiments
that may aect either the price or the attractiveness of the apartment to the
potential buyer, or both. The municipality, although committed to the exe-
cution of the project, felt responsible for the well-being of its (future) citi-
zens, as well as for the proper management of the whole system.
At the start of the project, many aspects of the experiment with vacuum
toilets and anaerobe treatment systems were still unresolved from seem-
ingly trivial issues like the shape and colour of toilets to more crucial
aspects of management and transfer of technology after the experiments
came to an end. The initiators (a group of environmental engineers) tried
to resolve these problems during a number of project meetings as well as
bi-lateral meetings; however, there was no satisfactory result. Both the
municipality and the project developers decided in 2003 to withdraw from
the project and another building site had to be found to implement the
proposed technology.
In hindsight, the following can be said about why the project failed in
achieving its goal, which was the testing of eco-sanitation techniques in a
residential area. A rst lesson that can be drawn is that apart from a com-
mitment of the municipality and a subsidy for the installations, one has
to deal with many more conditions under which such anomaly in the
current system can be developed. Municipalities and water boards are
legally responsible for the proper handling of human waste and they want
solid guarantees that alternative systems will work as well as, or even
better than, the systems they are used to working with. Project developers
and real estate agencies are even more reluctant to accept the risks of
technical or managerial failure. As long as there are as many uncertainties
as in the current project, the willingness of all these actors to co-operate
is limited.
Furthermore, commitment of institutional partners was not adequately
secured before the project started. Before the subsidy was granted, there
was a commitment of the local city alderman to join the proposed project,
as it would very much add to the municipalitys green image. Once the
project really started, however, the municipality was represented by the civil
servants of Public Works, who did not show an equally green commitment
as their political boss. Furthermore, there was a major disagreement about
the character of the group of end users. The initiators of the project
claimed that there are enough potential buyers in Wageningen who are
committed to the environment to join such an environmental innovation
project voluntarily. The builders (project developers) disputed this and
claimed that any alteration in the standard sanitary system will be consid-
ered a hazard, even to green consumers.
346 Development of technology and policy
The two cases presented above only show a small segment of the variety
of todays innovations and experimentation in eco-sanitation. Moreover,
they were not deliberately set up to test a special mode of governance,
rather to test technologies and new arrangements for sustainable living.
However, the analysis of the two cases can provide a starting point to draft
reexive governance strategies to transform sanitation systems into a more
348 Development of technology and policy
applied in dierent social contexts and adaptive enough to cater for chang-
ing insights during the course of transition. For sure, this is easier said than
done. In most cases of eco-sanitation, especially in developed countries, not
only is a large sewage infrastructure very close by, but so are the institutions
supporting it. The two cases discussed here showed that in the rst place
one should massage existing institutions, rather than invent new ones.
Only when traditional parties like water boards, municipalities and utilities
are willing to cooperate, may one invent new forms of management that
will t the new socio-technical arrangements. An example of such new
management might be the management of reed-bed lters by working
groups of inhabitants.
The main cause of failure in the rst case was that institutional changes
seemed to be considered of minor importance compared to the technical
installations and were to be dealt with only after the start of the whole
project. The second case demonstrated new forms of management and col-
laboration (between municipality, future residents, architects, water boards
and urban planners) that were conducive to the success of the technical
implementation of eco-sanitation.
The case of Culemborg showed that the projects aim was not just to
experiment with new technologies, but also to reach sustainability in
housing and living. The whole area was planned to full the explicit overall
vision of sustainable living as articulated in various workshops by the
350 Development of technology and policy
Culemborg and many other cases. The blurring boundaries between provi-
sion and consumption and the coming into place of new intermediaries
(community associations for maintaining systems, caretakers, control
systems) should go hand in hand with new, interactive strategy develop-
ment of which the Culemborg case is an example. Top-down regulation and
planning that is so characteristic of sewage systems should be replaced by
a process of trial and error, citizen participation, discussions with experts,
requests for subsidies, competition for eco-city awards and so forth. The
case also showed that it is very dicult to develop niche projects like this
systematically: the local geographical and institutional circumstances seem
to require specic local approaches. However, there are still some general
lessons to learn from the processes of both the Culemborg and Wageningen
cases in terms of strategy development. These will be addressed in the fol-
lowing, concluding section.
CONCLUSION
Closing loops take a long time, and eco-sanitation is still at a very early
stage. Likewise, the suggested building blocks of an implementation strat-
egy move a few rst steps away from straightforward technological experi-
menting and towards more reexive governance of a transformation
to sustainable sanitation. Obviously, the proposed building blocks of a
reexive strategy require a lot more of stakeholders time and eort than
a straightforward technological approach. But if it makes the dierence
between eventual failure and success, it is worth the eort.
NOTES
1. The project was funded by the Dutch Ministries of Economic Aairs and Environment
and Spatial Planning and executed by a consortium of environmental engineers, munici-
palities, utilities and housing developers to develop, study and implement several tech-
niques of on-site anaerobic waste water treatment in the built environment. The aim of
the project is to enhance the diusion of eco-sanitation within the Netherlands, by setting
up niche projects in which researchers, municipalities, consumers and other actors can
learn about the new technology.
2. Article 2, Algemene bouwvoorschriften voor zuiveringssystemen en inltratievoorzienin-
gen (General building regulation for treatment systems and ltration utilities), Chapter 2
of Uitvoeringsregeling Lozingenbesluit Bodembescherming (Tweede Kamer der Staten
Generaal, 18-12-1997, source: www.overheid.nl)
3. The Dutch abbreviation stands for Ecological Centre for Education, Extension and
Advice.
REFERENCES
GTZ (2003), Ecosan closing the loop, Proceedings of the 2nd international sym-
posium on ecological sanitation, 711 April, Lbeck.
Kaptein, M. (1999), Eco Housing Project EVA Lanxmeer (Culemborg),
Sustainability: a matter of people, presentation text, 26 October.
Kemp, R., J. Schot and R. Hoogma (1998), Regime shifts through processes of
niche formation: the approach of strategic niche management, Technology
Analysis and Strategic Niche Management, 7, 13968.
Lens, P.N.L., G. Zeeman and G. Lettinga (2001), Decentralised Sanitation and
Reuse Concepts, systems and implementation, London: IWA Publishing
(Integrated Environmental Technology Series).
354 Development of technology and policy
Rotmans, J., R. Kemp and M. van Asselt (2001), More evolution than revolution:
transition management in public, Policy Foresight 3 (1), 1531.
Schumacher, E.F. (1975), Small is Beautiful, a study of economics as if people mat-
tered, London: Blond Briggs.
Shove, E. (2003), Comfort, Cleanliness and Convenience: The Social Organization of
Normality, Oxford: Berg Publishers.
Siemensma, M. (2000), De ontwikkeling van decentrale sanitatie en hergebruik in
een woonwijk, [The development of decentralised sanitation and reuse in a res-
idential area], MSc thesis, Wageningen: Wageningen University.
Southerton, D., H. Chappells and B. van Vliet (eds) (2004), Sustainable
Consumption: the Implications of Changing Infrastructures of Provision,
Cheltenham UK and Northampton, MA, USA: Edward Elgar.
Van Vliet, B. (2002), Greening the Grid, The Ecological Modernisation of Network-
bound Systems, Wageningen: Wageningen University (PhD Thesis).
Van Vliet, B., H. Chappells and E. Shove (2005), Infrastructures of Consumption.
Environmental Innovation in the Utility Industries, London: Earthscan.
Vewin (2004), Huishoudelijk waterverbruik www.vewin.nl.
Werner, C., P.A. Fall, J. Schlick and H.-P. Mang (2003), Reasons for and principles
of ecological sanitation, paper presented at Ecosan Closing the Loop,
Proceedings of the 2nd International Symposium on Ecological Sanitation, 711
April, Lbeck, pp. 2330.
Zeeman, G. and G. Lettinga (1999), The role of anaerobic digestion of domestic
sewage in closing the water and nutrient cycle at community level, Water,
Science & Technology, 39, 18794.
14. The transition towards sustainable
production systems in Austria:
a reexive exercise?
Philipp Spth, Harald Rohracher, K. Matthias
Weber and Ines Oehme
SUMMARY
355
356 Development of technology and policy
process so far has only been loosely tied to policy-making, thus predomi-
nantly staying at an experimental level and not signifying a general shift
of Austrian R&D policy towards reexive governance. Although the
specic process described here has to be seen as a rather isolated experiment
within the Austrian context happening in the shadow of the governmen-
tal R&D policy rather than driving it some conclusions about the applic-
ability and pitfalls of the concept of reexive governance can yet be drawn.
We discuss prerequisites of a participative assessment of socio-technical
pathways especially with regard to political mandate and constraints for the
delegation of such socio-technological decision-making processes to non-
governmental actors.
INTRODUCTION
by this programme since December 1995, has been to develop options for
a sustainable development of Austrian landscapes and regions.
At about the same time, the National Environmental Plan (NUP) BMU:
Bundesministerium fr Umwelt (Detter 1995) was formulated, which obvi-
ously focused mainly on the environmental dimension of sustainability.
It was followed by the Federal Sustainability Strategy in 2002, which was a
reaction to changing requirements at the European and international levels.
In the late 1990s, a major research and technology initiative was started at
the Federal Ministry for Transport, Innovation and Technology (BMVIT),
which aimed at the development of technologies for sustainable develop-
ment (Nachhaltig Wirtschaften). For the purpose of this chapter, one of its
sub-programmes entitled Factory of Tomorrow4 is of major interest. This
programme focuses on the advancement of research and technologies that
are supposed to deliver a double dividend; that is, they are supposed to
enhance the competitiveness of Austrian industry while at the same time
contribute to the establishment of a more sustainable production system.
In general, the lack of horizontal cooperation and the division of labour
between ministries in matters of research, technology and innovation
policy implies that major pre-conditions for reexive governance could not
be met. With BMVIT focusing on technologically-oriented R&D and tech-
nology policy initiatives, the changes to the broader policy framework and
further preconditions for a structured transition towards sustainability
(stakeholder inclusion, social and economic dimension of sustainability)
can be achieved only with diculty. Consequently, the programme
Nachhaltig Wirtschaften mainly focuses on environmental technologies and
demonstration projects that could serve as a visible orientation and model
for adopters (beacons as they are called in the programme). There is not
much connection to the general sustainability research of the Ministry of
Science and Education (for example, focusing on ecosystems, sustainable
lifestyles, transdisciplinary research or social dialogues on sustainability),
nor is the programme closely integrated in initiatives of the responsible
departments for environment or energy.5
Leading industrial and academic researchers from dierent elds and
institutions of science and engineering relevant to the scope of the pro-
gramme were invited to contribute to the designing of the programme. As
a consequence, the programme acquired a strong technological orientation,
whereas wider contextual issues relevant to sustainable development were
addressed in the aforementioned cultural landscape programme hosted by
the Federal Ministry for Science, Education and Culture.
Nevertheless, the Factory of Tomorrow programme aims to support
the structural shift towards eco-ecient management and a sustain-
able production system through research, technological development,
The transition towards sustainable production systems in Austria 361
The actual calls for project proposals, which identied specic problems to
be solved,6 put this rather broad framework into more concrete terms. With
regards to any criteria for sustainability, the way is clear for applicants and
the jury to operationalise sustainable production and decide on what is
promising in terms of sustainability. Furthermore the overall funding
volume is comparatively small (about 10.54 million for the rst three calls
20013).7 Of this budget, production represents a very broad research
domain. As a result, the 88 projects that have been funded so far are rather
diverse and only loosely integrated into joint research agendas.
These pitfalls are part of the reason why the programme funded an
accompanying strategic research project, which tried to embed these tech-
nological projects in broader strategies of a transition towards sustainable
development and to foster interaction and coherence. This accompanying
project is seen as an experiment for further initiatives to integrate and
implement research carried out in the programme. The establishment of a
special advisory group, which includes representatives from dierent
funding bodies, ministries and the Austrian Research Council, expresses
the interest of these research funding organisations to nd new ways to link
research with implementation strategies and longer term planning. The
strategic accompanying project set out to demonstrate the applicability of
transition management approaches for R&D programmes by focusing on
362 Development of technology and policy
Bioreneries in Austria
While some research eorts were made in the late 1980s with regard to the
use of green biomass for ethanol production and use of grass bres for
paper production, research with respect to the concept of green biorenery
was only intensied in the late 1990s (Narodoslawsky 1999).
Bioreneries were promoted in Austria to begin with because of their
assumed positive impact on regional development by (1) stabilising the
current use of green land, thus keeping the landscapes open and attractive
and (2) direct impact on the economic structure of rural regions. Kromus
et al. (2004) go on:
The decentralized nature of the raw material generation for the Green Biorenery
calls for decentralising . . . This will generate jobs in rural areas thus considerably
reinforcing the economic structure of these regions as well as raising the skills of
the workforce in these regions as operating this technology needs up-to-date tech-
nical knowledge especially in the eld of chemical engineering.10
The exploitation of grassland might have intriguing side eects like the conser-
vation of cultural landscapes and the improvement of the stay option of
farmers. Therefore interest is focused on the type of green bioreneries, using
grass from meadows as feedstock . . . In contrast to the centralized Biorenery
concepts in Europe, the Austrian Green Biorenery focuses on a decentralized
system based on grass-silage . . . with lactic acid and amino acids (proteins) as
key products from silage. (Kromus et al. 2004)
Recent years have indeed seen extensive research activities and attempts
to set up pilot studies, which demonstrate the potential of green bioreneries
in the region of southern Styria. Before now, industry has not been inten-
sively involved in the funding of these research activities, though. Research
has thus been limited to a number of projects within the Factory of
Tomorrow programme, complemented with equity-funded activities by
several research institutions. Yet a national network of researchers and insti-
tutions interested in biorenery-related R&D has been set up and coordi-
nates research on an informal level.
The aforementioned hopes in bioreneries, which were presented mainly
by a small group of researchers, are to a large extent based on assumptions
which have not yet been claried. With respect to R&D policies the gov-
ernment agencies nd themselves in a (common) state of ambiguity. The
technologies and concepts of bioreneries might yield a high potential to
solve several problems in one go (sustainable production, income gener-
ation, landscape preservation). At the same time, it has not been assessed
at this early stage of technology development how the innovation would
impact on the broader socio-technical context, and under which conditions
this impact would be positive. So how can policy deal with promising tech-
nologies and concepts like these in an adaptive and reexive manner and
take into account the broader systemic requirements of transition
processes towards sustainable socio-technical systems?
Selection and
specification Framework scenario 2 Image of the future 2
of transition
fields
Portfolio-based
policy strategy
scenario development process (see above). The status quo and current
developments within these elds have then been explored by literature
and Internet research and 1520 in-depth interviews per transition
eld. It is essential to have a clear understanding of the transition elds
and how they can be systematically represented in terms of the three
levels of analysis: (1) broad normative goal or need area, (2) transition
eld, and (3) individual technological experiment.
2. Scenario development: the following scenario development process
aimed to develop rst a set of dierent possible scenarios basically in
an explorative way by asking what is likely to happen in the future
rather than what actors want to happen. On a practical level, the par-
ticipants of a rst workshop collected important factors of impact on
the future of, for example, bioreneries. These impact factors were then
combined in plausible impact constellations and eects called story-
lines. One such storyline for example read: The industrialisation of
milk production in Austria is likely to further reduce the use of green
land for grazing, resulting in the succession of woodlands. Another
read: High political support for biogas production and respective
nancial incentives are likely to narrow the biomass feedstock left for
bioreneries. These storylines were then clustered in consistent fami-
lies, thus building the foundation of three dierent scenarios. This
methodology could be called inductive or bottom-up scenario devel-
opment as opposed to the top-down deduction of scenarios from a
variation of two or three main factors. The latter approach was used to
develop the well-known Shell scenarios on the future of energy con-
sumption and others (Van der Heijden 1996). The three scenarios of
two to three pages each were further elaborated and completed by the
project team after the rst workshop.14 It was important to ensure that
each of the scenarios could not be simply classied as desirable or not
desirable; rather each of them should demonstrate its positive and
negative aspects in order to provide a multi-faceted, plausible and at
the same time challenging image of the future. The results were sent to
the participants, who discussed and further specied them at the begin-
ning of the second workshop.
3. Sustainability assessment of scenarios: the scenarios diered of course
in terms of their overall sustainability orientation. This was intended
and is useful with regard to the deduction of policy perspectives,
because it promotes further discussion on how to move towards sus-
tainability even under detrimental conditions. The following normative
phase15 started with the selection of the most important criteria for
an assessment of the sustainability potentials of the scenarios.16 These
criteria were negotiated among the participants17 on the basis of a
The transition towards sustainable production systems in Austria 371
suggest that the scenario process could have served much more as a catalyst
if the mandate and the project budget had been larger.18
The actors involved in the scenario workshops modied their views on the
subject during the exercise, for example, with regard to the probabilities of
certain framework conditions, research priorities and strategies for the pro-
motion of bioreneries. The scenario process stimulated new debates about
long-term perspectives for the transition eld of bioreneries. By giving the
participants an opportunity to reect upon the contribution and conditions
required for success of their projects, the process facilitated second-order
learning eects. A growing awareness for critical framework conditions and
necessities for a successful development of bioreneries in Austria could be
observed. The participants stated that it was the rst opportunity for them
to develop such detailed images of the future and assess strategic action
with regard to these scenarios. Although it was a rather homogenous group
of people, the scenario process was a novel opportunity to exchange views
that had not been exchanged before.
As discussed above, the scenario-building process was only partly suc-
cessful in enabling interaction between researchers and other actors outside
the R&D programme, as only few corporate actors and actors from other
policy elds participated in the workshops.
We can now come back to the two guiding questions of this chapter:
clear how dicult it is to mobilise these actors towards the joint eorts of
strategy development and implementation (that is, how demanding it is in
terms of time, money and legitimacy). The experiences lead us to conclude
that R&D policy can only exert a major impact on long-term socio-techni-
cal transitions towards sustainable production if it is implemented in the
context of a broad participatory process. Compared to these requirements,
the resources for the experimental project on bioreneries denitely were
not sucient to trigger anything like a transition process. Further and more
substantive steps would be needed to initiate joint and coherent eorts of
the dierent parties involved.
CONCLUSIONS
ACKNOWLEDGEMENT
NOTES
REFERENCES
Bundesministerium fr Umwelt (1995), Nationaler Umweltplan, Vienna: BMU.
Coenen, Reinhard (2002), Sustainable development new challenges for technol-
ogy assessment, in Andrew Jamison and Harald Rohracher (eds), Technology
Studies and Sustainable Development, Munich, Vienna: Prol Verlag GmbH,
pp. 12743.
Danner, H., B. Mazingaidzo, B. Molzbichler, M. Neureiter and R. Braun (1999),
Austrian and European markets for products from a green biorenery, in pro-
ceedings of the 2nd International Symposium The Green Biorenery, Feldbach,
Austria.
Elzen, Boelie, Frank Geels, Peter Hofman and Ken Green (2002), Socio-technical
scenarios as a tool for transition policy an example from the trac and trans-
port domain, in proceedings of the workshop Transitions to Sustainability
through Systems Innovation, Enschede: University of Twente.
Geels, Frank W. (2002), Understanding the dynamics of technological transitions,
PhD thesis, Enschede: University of Twente.
Gleich, A. v., R. Haum and U. Petschow (2004), Guiding principles for sustain-
ability, kologisches Wirtschaften, (5), 2930.
Grunwald, Armin, R., Coenen, J., Nitsch, A. Sydow and P. Wiedemann (2001),
Forschungswerkstatt Nachhaltigkeit. Wege zur Diagnose und Therapie von
Nachhaltigkeitsdeziten, Berlin: Sigma.
Industriewissenschaftliches Institut (2001), Stoiche Nutzung Nachwachsender
Rohstoe in sterreich Marktanalyse und Handlungsmanahmen, Studie im
Auftrag des Bundesministeriums fr Verkehr, Innovation und Technologie,
Vienna: BMVIT.
Kamm, B. and M. Kamm (2004), Biorenery systems, Chemical and Biochemical
Engineering Quarterly, 18 (1), 16.
Katter, R., H. Mackwitz, Michael Narodoslawsky, K. Payer, W. Stadlbauer, Brigitte
Wei, Robert Wimmer and M. Wrgetter (1999), Nachwachsende Rohstoe:
Ergebnisse aus der Vorbereitungsphase fr das Impulsprogramm Nachhaltig
Wirtschaften, Studie im Auftrag des Ministeriums fr Wissenschaft und Verkehr,
Graz.
Kromus, S., B. Wachter, W. Koschuh, M. Mandl, C. Krotscheck and
M. Narodoslawsky (2004), The green biorenery Austria development of an
integrated system for green biomass utilization, Chemical and Biochemical
Engineering Quarterly, 18 (1), 712.
Kromus, Stefan (1999), Elaboration of a decentralized green biorenery for the
Austrian region of Feldbach: a sustainable concept?, Master of Science Thesis,
Wageningen: Wageningen Agricultural University (WAU).
382 Development of technology and policy
INTRODUCTION
383
384 Development of technology and policy
Uncertainty
A core element of the agrobiodiversity problem is uncertainty and lack of
reliable knowledge.
The rst diculty is to measure the exact extent of the loss. As a con-
comitant of the diversity of analytical levels of analysis (ecosystem, species,
genetic diversity) and of scientic approaches (see above), there is a lack of
widely accepted indicators.15 Also, data availability varies depending on
which level of genetic diversity is to be described. Diculties arise even at
the apparently unproblematic level of recording the numbers of species,
races and varieties.16 Assessments of the various components are con-
ducted separately and without an integrated evaluation of agricultural bio-
diversity as a whole. These limits to scientic analysis imply uncertainty
about the full scope of agrobiodiversity loss.17
The second dimension of uncertainty relates to the consequences of
agrobiodiversity loss and to risk assessment. In major ignorance as to the
value of agrobiodiversity (and of biodiversity in general), and partly even
as to its functions,18 interactions and causal relations, it is unclear what
specic impact the decline of agroecosystem, species and genetic diversity
will have (CBD 2000a: 92). This applies to ecological impacts as well as to
economic costs and socio-cultural consequences. Plausible evaluations of
the risks and dangers can be made (see above). It is for example undis-
putable that agrobiodiversity loss diminishes options for future breeding
and agriculture. However, there is no way to take informed decisions on
which parts of agrobiodiversity, which races, populations, lines and vari-
eties, eventually genotypes, genes and alleles are worth maintaining and
which are not. This decision depends on normative evaluations as well as
on information about future needs that cannot be known at present: about
tomorrows diseases, nutrition needs, consumer demands, possible envir-
onmental events and agricultural development paths, and so on. The uncer-
tainty inherent in the agrobiodiversity problem suggests the use of the
precautionary principle as a reexive governance strategy.
A third dimension of uncertainty relates to the eects of political inter-
ventions. The complexity of ecological, economic and social eects makes it
dicult to determine what side eects agrobiodiversity conservation and uti-
lization strategies might cause. On the one hand, agrobiodiversity strategies
390 Development of technology and policy
can have unintended impacts in the social and economic realm, and they can
even cause negative feedback within agrobiodiversity itself.19 On the other
hand, non-agrobiodiversity policies have ambiguous eects on agrobiodi-
versity. One case in point is trade liberalisation. A reduction of trade-distort-
ing domestic support (subsidies) might lead not only to agricultural output
substitution, but also to input substitution and to output price impacts. It is
very hard to assess which impact will predominate, to disentangle short-term
from long-terms eects, to assess their direction and relative magnitude and
therefore also their eects on agrobiodiversity (CBD 2003a). For instance, the
reduction of domestic support might lead to a decline in the application of
fertilizers and other agricultural chemicals and thus indirectly promote the
use of more crop varieties as a means to reduce risk of pests. It might also have
a countervailing impact, because production pressures lead to technological
changes that induce further specialisation like mono cropping.
Path dependencies
Path dependencies in the eld of agrobiodiversity loss exist both in biolog-
ical terms (relating to the problem itself) and in economic and political
terms (relating to the societal capacity of reacting to the problem).
First, agrobiodiversity loss is irreversible. Once a species, race, line,
variety etc. has died out, certain paths abruptly end. In other words, the
plants and animals can no longer be used, either in breeding, agriculture,
in landscape maintenance, or for hobby or any other purpose. A process of
path-closing also takes place when a breed or variety is exclusively pre-
served ex-situ in gene banks. Important knowledge about the animals/
plants characteristics, utilisation, about the genetic materials traits and
substances of content, are lost, rendering the further utilisation of the
animals and plants unlikely.
In economic terms, innovation dynamics aect the maintenance of agro-
biodiversity. There is an incentive for breeding companies to work with
those genetic resources that have already been successfully developed into
high-performing products by previous breeders or with which they them-
selves have experience. By means of their investment, breeders design a
path. To the extent that knowledge is accumulated and breeding progress
advances, the productivity and economic attractiveness of other varieties,
lines, races or populations that are not developed, decreases (Wol and
Petschow 2004). Thus, the potential for the adaptation of crops and live-
stock (and implicitly of mankind) to future requirements is lost.
Finally, political path dependencies prevent a transformation of unsus-
tainable farming practices and breeding strategies: For decades, the high-
productivity, low-sustainability practices have been subsidised. The
subsidies and the respective agricultural alignment have been pushed
The transformation of agriculture 391
Approach
Conservation of Use of agrobiodiversity Development of agrobiodiversity
agrobiodiversity (A) (B) (C.1, C.2)
Object Protection of existing plant and animal Development of Protection of
genetic resources new varieties and existing plant and
lines/individuals animal genetic
resources
Development of
393
new varieties and
lines/individuals
Development of
more diverse
agroecosystems
Implementation Ex situ conservation On-farm use Technology-oriented Sustainable
transformation of transformation of
breeding: Introduction breeding28
of new traits through within agricultural
genetic engineering production chain
Table 15.1 (continued)
Approach
Conservation of Use of agrobiodiversity Development of agrobiodiversity
agrobiodiversity (A) (B) (C.1, C.2)
394
(e.g. organic farming), development,
niche markets processing,
distribution,
marketing,
consumption,
awareness raising
Scope and thrust Segmental Segmental Segmental Holistic
Static Dynamic Dynamic Dynamic
Risk orientation Insurance Precaution Risk Precaution
kinds of logic and possess unequal strength to assert themselves in the policy
process.
The multi-level structure of national, EU and international policies
further complicates things for governance that favours agrobiodiversity.
Some examples: eorts of the German redgreen government to introduce
a harmonised budget line on genetic resource grants were thwarted by con-
servative State (Lnder) governments. A facilitated registration procedure
of conservation varieties cannot be introduced by EU Member States as
long as the Commission delays the creation of an important implementa-
tion guideline. Multilateral provisions on plant genetic resources for food
and agriculture could only be introduced after seven years of tough nego-
tiations under the auspices of the FAO. Conicts of interest, as well as
inter- and intra-institutional friction and negative coordination practices,
are crucial reasons for such multi-level complications.
Similarly, in the social realm, it is dierent groups like environmentalists,
seed networks, hobby gardeners, or NorthSouth activists whose activities
touch on the issue of agrobiodiversity, but who do not necessarily have the
explicit goal of sustaining it. It will therefore be hard to mobilise them for
collective action.
Farmers rights
The concept of Farmers Rights was rst developed in the conict between
developing and developed countries on the question of access to genetic
The transformation of agriculture 403
of their own budgets, decisions had less explosive force. Their implemen-
tation is subject to a budget salvo.
CONCLUSIONS
NOTES
9. Allele are dierent versions of the same gene that cause dierent specications of a trait.
10. Article 5 (iii) and Article 8, UPOV Convention (UPOV 1991).
11. Such as disease and pest resistance, adaptation to harsh conditions and poor quality feed.
12. Famous examples of economic disasters spring from genetic monoculture such as the
nineteenth-century Irish potato famine and the US pest Corn Leaf Blight in 1969.
13. Food security means a state when all people, at all times, have physical and economic
access to sucient, safe and nutritious food to meet their dietary needs and food prefer-
ences for an active and healthy life (FAO 2004b).
14. In the recent past, distributive conicts have intensied between these approaches as
research grants and resources for professorships are being channelled into molecular
biology and genetic engineering, while subjects like taxonomy and ethnobotanics are
being neglected (BMVEL 2002: 11; Swiss Biodiversity Forum 2002: 10). Severe losses of
basic knowledge impend, threatening not only the monitoring of taxa but also prevent-
ing research and development of management strategies.
15. For example, with the help of methods and indicators of molecular biology, genetic
variability become visible in cases where other approaches identify a high degree of
uniformity.
16. For instance, the taxonomic classication of genetically heterogeneous land races is
somewhat arbitrary.
17. For an impression of the opposing assessments compare, for example, Smale (1997),
ASSINSEL (1996) against, for example, Thrupp (1997), Vellv (1993), Fowler and
Mooney (1990).
18. Such as the ecosystem services provided by agricultural biodiversity.
19. An example: state aid for the cultivation of old cereals (Emmer, Einkorn) in Austria led
to unfair competition with non-subsidised Emmer and Einkorn in Germany, thus ham-
pering their locally adapted cultivation and breeding in the neighbouring country.
20. Deutscher Bauernverband (DBV).
21. The Common Agricultural Policy (CAP) was established in the European Economic
Community as early as 1957.
22. However, in the meantime the EP has a say in some specic agrobiodiversity matters as
they fall under the heading of health and consumer politics with an EP co-decision pro-
cedure. In the 1990s, the EP successfully pushed for the rst major EU initiative on agro-
biodiversity, Directive 1467/94/EC on genetic resources in food and agriculture.
23. Interestingly enough, agrobiodiversity may count both as man-made and as natural
capital. In the logic of the sustainability sciences and their capital forms typology,
however, it is more to the point to interpret agrobiodiversity as natural capital.
24. Conservation outside natural habitats or agricultural use, that is, in zoos, botanical
gardens and above all in gene banks and by cryo-conservation.
25. The actual cultivation and husbanding, respectively, of crops and livestock.
26. In this case, the conict of means overlaps with a conict of values.
27. This applies especially to the use strategy and the more holistic of the two development
strategies.
28. As opposed to the technology-oriented transformation of breeding, the envisaged
changes in breeding aim at widening the breeding targets, and at reducing the universal
spreading of genotypes/genetic traits, etc.
29. The FAO on the international level is an exception.
30. For an account of the conventional, predominant structures of agrobiodiversity gover-
nance, see Wol (2004b), IW et al. (2004b), Raustiala and Victor (2004), Kameri-
Mbote and Cullet (1999), Bragdon and Downes (1998).
31. CGIAR is the Consultative Group on International Agricultural Research, an associa-
tion of public and private members supporting 16 research centres worldwide, founded
in 1971. Within the reseach process, farmers have been involved in various stages,
ranging from the initial documentation of genetic diversity and associated indigenous
knowledge (for example, Sandoval-Nazarea 1994) to the identication of methods to
assist the continued selection and maintenance of local cultivars (for example, Godbole
2000).
The transformation of agriculture 411
32. The delimitation between participatory research (knowledge production) and participa-
tory breeding (strategy implementation) is in practice not always easy to draw.
33. The background of the project commissioned by the German Ministry of Science is a
highly innovative ministry programme on socio-ecological research. The programme is,
however, under political pressure to stand its ground. As elsewhere, transdisciplinary
research is rather the exception than the rule in the German research scene.
34. The loss at dierent levels of agrobiodiversity was empirically shown for wheat, chicken
and pigs in Germany. Breeding-related, economic, legal and political driving-forces as
well as the dominant societal discourses were identied. See www.agrobiodiversitaet.net.
35. The CBD is an international treaty on the protection and sustainable use of biodiversity
and on the fair and equitable sharing of benets arising out of the utilisation of genetic
resources. It was adopted in 1992 at the Rio summit and has 188 parties to date.
36. An ecosystem as understood in the CBD means a dynamic complex of plant, animal
and micro-organism communities and their non-living environment interacting as a
functional unit (Article 2, CBD). Unlike the understanding of habitats, this denition
does not specify any particular spatial unit or scale.
37. A forerunner for this perspective may be seen in the literature on agroecology (Altieri
1987). With its broad perspective on optimising not the production of commodities but
the agroecosystem as a whole it shifted the emphasis in agricultural research away from
disciplinary approaches towards multidisciplinary research on complex interactions
among and between people, crops, soil and livestock.
38. Only by this can the risks linked to agrobiodiversity loss be avoided and the ecological
and agricultural benets of agrobiodiversity realised. Also, by using agrobiodiversity,
on-farm adaptivity of breeding and farming strategies to changing environmental
requirements is in-built.
39. Breeding.
40. Processing, retailing, and marketing.
41. The dierent functions of biosphere reserves are reected in dierent protection zones:
core areas are devoted to long-term protection; in buer zones, only activities compati-
ble with the conservation objectives can take place; in an outer transition area sustainable
resource management practices are promoted and developed. Stakeholder participation
in designing and carrying out the reserves functions shall be provided for (UNESCO
1996).
42. However, the nancial basis of the projects is notoriously unstable.
43. Essential steps of adaptive management are the assessment of a problem, the design,
implementation and monitoring of management, its evaluation and adjustment.
44. Despite the value of these principles for agrobiodiversity maintenance, the ecosystem
approach within the CBD has not been applied in an integrated way to issues of agri-
cultural biodiversity so far (Dec. VII/11, Annex II). While this is at least partly put down
to the fact that agriculture is practised largely on lands under private ownership (ibid.),
it might also mirror the comparatively low perception of agricultural biodiversity within
the CBD.
45. For example, the State of the Worlds Plant Genetic Resources Report (FAO 1996) or the
FAO Domestic Animal Diversity Information System.
46. For example, the measurement of eective population sizes in livestock populations in
Germany (BMVEL 2003).
47. In the context of the (non-binding) International Undertaking on Plant Genetic
Resources for Food and Agriculture (IU), the FAO Conference in 1989 dened Farmers
Rights as rights arising from the past, present and future contributions of farmers in
conserving, improving, and making available plant genetic resources, particularly those
in the centres of origin/diversity (Res. 5/89). As pledge to the recognition of Breeders
Rights the FAO member states recognised Farmers Rights. However, a substantive
denition of those rights was not made. It was envisaged that the realisation of Farmers
Rights should ensure a ow of benets from the use of plant genetic resources, to
farmers and their communities. FAO Resolution 3/91 therefore laid down that Farmers
Rights will be implemented through international funding on plant genetic resources,
412 Development of technology and policy
which will support plant genetic conservation and utilization programmes, particularly,
but not exclusively, in the developing countries. More radical proposals strived for
developing Farmers Rights into some form of property rights equivalent to Plant
Breeders Rights. Both Agenda 21 (1992) and the Global Plan of Action for the
Conservation and Sustainable Utilization of Plant Genetic Resources for Food and
Agriculture (1996) mention the objective to realise Farmers Rights.
48. International Treaty on Plant Genetic Resources for Food and Agriculture.
49. Article 9.2, ITPGR.
50. Another rationale of PVS and PBB is the experience in developing countries that new
varieties from national or international breeding programmes meet with low acceptance
by farmers.
51. Hybrid breeds, and in future potentially GURTs (Genetic Use Restriction Tech-
nologies), too.
52. Seeds are distributed on a loan basis and recovered after harvest.
53. Though the introduction of high-yielding varieties and commercial crops has aected
the availability of the seeds of traditional cultivars in developing countries.
54. See Table 1.1 in Vo and Kemp Chapter 1, this volume: while the middle row provides
the analytical framework (specic problem features), the bottom row establishes pre-
scriptive strategy requirements.
55. For example, development of breeding technologies, division of labour and scientication
of agriculture, emergence of the environmental and animal welfare movements, change
of the farmers role in modern societies, seed regulation, nutrition-consciousness,
farmers self-determination vs. dependence on multinational agribusiness, NorthSouth
distribution of biological diversity and of the technology to make use of it, the specic
contribution of women to agrobiodiversity maintenance, international trade politics, the
value but lacking remuneration of traditional knowledge, risk awareness in society,
scientic discourses and curricula, rural development and global food security.
56. For example, reducing agrobiodiversity loss to a breeding problem.
57. This could be the case for example when participatory national policies on the transfor-
mation of agriculture have negative impacts on the global South.
REFERENCES
Almekinders, C., L. Fresco and P. Struik (1995), The need to study variation in
agroecosystems, Netherlands Journal of Agricultural Science, (43), 12742.
Altieri, M. (1987), Agroecology. The Scientic Basis of Alternative Agriculture,
Boulder, CO: Westview Press.
Anderegg, Ralf (1999), Grundzge der Agrarpolitik, Munchen: Oldenbourg.
ASSINSEL (1996), ASSINSEL position on maintenance of and access to plant
genetic resources for food and agriculture (PGRFA), adopted in May, Geneva.
ASSINSEL (1997), Feeding the 8 billion and preserving the planet, Geneva.
BMVEL (Bundesministerium fr Verbraucherschutz, Ernhrung und
Landwirtschaft) (2003), Nationales Fachprogramm Tiergenetische Ressourcen,
Bonn.
BMVEL (Bundesministerium fr Verbraucherschutz, Ernhrung und
Landwirtschaft) (2002), Nationales Fachprogramm zur Erhaltung und nach-
haltigen Nutzung panzengenetischer Ressourcen landwirtschaftlicher und
gartenbaulicher Kulturpanzen, Bonn.
Bragdon, S. and Downes, D. (1998), Recent policy trends and developments related
to the conservation, use and development of genetic resources, Issues in Genetic
Resources 7, 1998.
The transformation of agriculture 413
King, A. (2000), Tools for Participatory Research on Crop and Tree Diversity, available
at http://www.ipgri.cgiar.org/themes/human/Articles/Particip-research2.doc.
Khler-Rollefson, I. and J. Wanyama, (eds) (2003), The Karen Commitment, pro-
ceedings of a Conference of Indigenous Livestock Breeding Communities on
Animal Genetic Resources, Karen, Kenya, 2730 October 2003.
Mller, K.J. (1999), On-farm improvement and original seeds in Germany, in T.
Gass, L. Frese, F. Begemann and E. Lipmann (eds), Implementation of the
Global Plan of Action in Europe Conservation and Sustainable Utilization of
Plant Genetic Resources for Food and Agriculture, proceedings of the European
Symposium, 30 June 3 July 1998, Braunschweig, Germany: Rome.
Pearce, D.W. and G. Atkinson (1995), Measuring sustainable development, in
D.W. Bromley (ed.), The Handbook of Environmental Economics, Oxford: Blackwell.
Perlman, Dan L. and Glenn Adelson (1997), Biodiversity: Exploring Values and
Priorities in Conservation, Oxford: Blackwell Science.
Ploeg, Jan Douwe van der and Henk Renting (2002), Environmental co-operatives
reconnect farming, ecology and society, in Bertus Haverkort, Katrien vant
Hooft and Wim Hiemstra (eds), Ancient Roots, New Shoots: Endogenous
Development in Practice, London: Zed Books pp. 2227.
Raustiala, K. and D. Victor (2004), The Regime Complex for Plant Genetic
Resources., International Organisation, 58 (2).
Rengalakshmi, R., D. Dhanapal, Oliver King and T. Boopathy (2003),
Institutionalizing traditional seed exchange networks through community seed
bands in Kollihills, India, In CIP-UPWARD (ed.), Conservation and Sustainable
Use of Agricultural Biodiversity. A Sourcebook, Vol. II, Los Banos, Laguna, The
Philippines, pp. 3028.
Sandoval-Nazarea, V. (1994), Memory banking protocol: A guide for document-
ing indigenous knowledge associated with traditional crop varieties, Los Banos,
Laguna: UPWARD, 1994 Training Document Series: 2.
Settle, W. (2003), Case Study No. 11. Ecosystem management in agriculture.
Principles and application of the ecosystem approach, in FAO: Biodiversity
and the Ecosystem Approach in Agriculture, Forestry and Fisheries.
Proceedings of a satellite event on the occasion of the Ninth Regular Session
of the Commission on Genetic Resources for Food and Agriculture, Rome,
1213 October 2002.
Smale, M. (1997), The Green Revolution and wheat genetic diversity: some
unfounded assumptions, in World Development, 25 (8).
Soriano Niebla, J.J. (2002), quoted in N. El-Hage Scialabba, C. Grandi and
C. Henatsch, Case study no. 4: Organic agriculture and genetic resources for
food and agriculture, in FAO: Biodiversity and the Ecosystem Approach in
Agriculture, Forestry and Fisheries, Proceedings of a satellite event on the occa-
sion of the Ninth Regular Session of the Commission on Genetic Resources for
Food and Agriculture, Rome, 1213 October 2002.
Sthapit, B., A. Subedi, S. Gyawali, D. Jarvis and M. Upadhaya (2003), In situ con-
servation of agricultural biodiversity through participatory plant breeding in
Nepal, in CIP-UPWARD (ed.), Conservation and Sustianable Use of
Agricultural Biodiversity. A Sourcebook, Vol. II, Los Banos, Laguna, The
Philippines, pp. 31121.
Swiss Biodiversity Forum (2002), Visions in biodiversity research. Towards a new
integrative biodiversity science, availale at http://diversitas.mirror.ac.in/news/
visions_pdf.
416 Development of technology and policy
INTRODUCTION
This concluding chapter is more than a summary of the arguments pre-
sented in the preceding chapters. One could say that we take a reexive
approach to the theme of the book. We reconsider the ideas from the
Introduction in the light of the ideas it has prompted in the chapters.
In this way, we can rework a concept of reexive governance that incor-
porates feedback from theoretical as well as more practical areas of appli-
cation. This feedback promotes learning with respect to the concept of
reexive governance. Experiences that are gathered on the basis of
reexive governance reproduce and modify the conceptual framework and
shape further experience they are an example of conceptual structura-
tion, to paraphrase Giddens (1984/1986). This concluding chapter can
thus be seen as a view on an emerging path of thinking and practice in
societal governance and problem solving.1
We proceed by rst discussing the relationship between sustainable devel-
opment and reexive governance in more depth. Here, the initial hypothe-
sis from the Introduction becomes substantiated by evidence from the
chapters. Sustainable development serves as a label under which a funda-
mental transformation of governance, in the context of reexive modern-
isation, is politically negotiated. Seen in this light, sustainable development
is indeed more than an empty phrase; it is both a symptom and a catalyst
of what Beck (1994) describes as reexive modernisation.
A second point is a more explicit concern for the quality of the outcome
of processes of reexive governance. Does reexive governance actually
produce better results? This question refers to the need for criteria of pro-
cedural quality, since it is not possible to arrive at a solid denition of the
right outcome of problem handling for sustainable development. Such cri-
teria can support an assessment of reexive governance without getting
419
420 Conclusions
trapped in the temptation to predene the results for learning processes and
thereby negate the very strength of the approach.
A third point is that we add to the question of the location of reexive
governance: in which types of interaction and at what level of social
organisation does reexive governance take place? In this respect, we
introduce dierent levels as a conceptual extension to reexive govern-
ance. This view acknowledges that the levels at which problems are
addressed and the interplay of governance processes across dierent
levels are an important dimension. In this volume, both Beck and Wol
show this aspect with respect to transnational governance as a response to
the limits of political organisation in nation states. Other chapters in
this volume, like those of Loibl and Whitelegg that discuss knowledge
production, show how reexive governance in research plays a role on
both a macro-level of programme management and a micro-level of
project management.
As a fourth point, we add a fundamental qualication to the concept of
reexive governance by introducing the ecacy paradox. This concept
refers to the contradicting requirements of opening up and closing down in
social problem-solving processes (see also Stirling 2005). On the one hand,
problem-oriented interactions need to be opened to take account of the
interaction of diverse factors, values and interests. This is necessary to
produce robust knowledge and strategies. On the other hand, selection of
relevant factors, decisions about ambiguous evaluations and convergence
of interests are necessary to take decisions and act. The strategy elements
of reexive governance, as presented in the Introduction, address the need
to open up various specialised kinds of problem solving to allow for inte-
grated assessment and coordinated strategies. The ecacy paradox draws
attention to the fact that eective governance requires these strategy ele-
ments to be complemented with appropriate strategies to reduce complex-
ity and achieve stable strategies. The proposed way to deal with these
paradoxical requirements is to combine opening up with closing down,
for example, by organising problem-handling processes in sequences of
opening up and closing down (compare the discussion of exploration and
exploitation in March 1991).
At the conclusion of this nal chapter, we summarise the concept of
reexive governance and formulate our position on its overall potential for
furthering the societal search for sustainable development. As a last step
towards the unfolding of the concept of reexive governance, we outline an
agenda for further research and practical experimentation.
A view on an emerging path 421
and press for their ideas and perceptions of problems in public dis-
course. Participatory knowledge production and strategy develop-
ment and implementation are based on insight into social pluralism
and distributed intelligence an insight that relates fundamentally to
the ideal of democracy.
sixth strategy element of reexive governance that covers all three dimen-
sions of problem solving: problem analysis, goal formulation, and strategy
development and implementation.
The chapters in this volume have uncovered a number of insights as to
the level of social organisation at which reexive governance is taking place
and the creation of problem-specic institutional settings. The chapter by
Kemp and Loorbach provides one example of tailor-made problem-
solving spaces, namely the transition arena, which they call a new institu-
tion for interaction and an open and dynamic network in which dierent
perspectives, dierent expectations and dierent agendas are confronted,
discussed and aligned where possible. Interestingly, the transition arena is
very uid, changing its size, task and participant prole throughout the
transition management process and thereby creating a congruency between
the shape of the transition arena and the problem on its agenda. Kemp and
Loorbach also introduce a concept that has been referred to in several other
chapters. They dierentiate between three levels of socio-technical systems:
macro-landscape, meso-regimes and micro-niches. Reexive governance
can in principle be located on all of these levels. Smith describes two
approaches that have chosen the niche level as the appropriate place to
foster system change. Other approaches, such as sustainability foresight,
suggested by Vo et al., highlight the need to coordinate niche activities and
developments on the regime level.
Looking at the chapters by Loibl and Whitelegg, we nd another
example of how reexive governance can be placed on dierent levels and
how these interact. While Loibl analyses reexive governance within
research projects, Whitelegg looks at the reexive governance of research
programmes. The latter includes both the governance of the programme
itself for example, the learning/adaptability of programmes or participa-
tion of stakeholders to dene priority areas and the promotion of
reexive governance within research projects. The chapter by Loibl also
points us to the fractal and nested nature of reexive governance that
operate at dierent levels. This chapter focuses on one of the ve reexive
governance strategies set out in the Introduction, namely, integrated know-
ledge production. Yet while exploring the practice of transdisciplinary
knowledge production as an example of a reexive strategy element in soci-
etal governance, it turns out that other reexive strategy elements are also
at work in the governance of the research process itself. Those elements are
needed to deal with complexity, heterogeneity and distributed resources in
transdisciplinary research processes.
While it is an important insight that reexive governance can and must be
developed on dierent levels, it is mainly the chapters in the section on
strategies for sustainable system transformation that emphasise interactions
A view on an emerging path 429
Having arrived at the end of the book, perhaps it is good to state what we
hope to have achieved. First and foremost, we hope we have generated an
interest in the very idea of reexive governance, realising that this is only a
rst step. Second, we hope we have shown that reexive governance repre-
sents a radical innovation with respect to dominant modernist regimes of
governance and that it needs to be taken up by theorists. And third, we hope
we have shown that reexive governance is for real that it already exists
in various forms.
In the Introduction we introduced ve strategies which can be derived
from the reexive governance perspective (integrated knowledge produc-
tion, experimentation and adaptivity of strategies, anticipation of long-
term eects, interactive participatory goal formulation and interactive
strategy implementation). In this concluding chapter we added the con-
gruency of governance and problem space as a sixth strategy element. We
suggest that these six strategies are central elements of a conceptual reper-
toire which can further the development of practices of reexive govern-
ance. Their application injects second-order reexivity into governance
processes, leading actors to reconsider their embedding in wider system
contexts and review the problem denitions, goals, options and strategies
coming out of it. In this way, governance gets prepared to deal with the
rst-order reexivity of modernisation, the spiralling up of problems and
problem solving as a result of unexpected side-eects.
The dierent chapters have demonstrated that existing governance
systems already include elements of reexivity that go beyond the con-
frontation of social groupings with unintended consequences. There are
indeed many instances, in diverse areas of practice, of new governance
approaches based on the reection and anticipation of unintended conse-
quences, in which the handling of uncertainty, ambivalence and distributed
control plays a central role. In the terminology proposed at the beginning
of this chapter, one could say that there is broad evidence for the emergence
of second-order reexivity on top of the rst-order reexivity of societal
development. As a fourth and nal point, we hope that we have been able
to show how these quite fundamental changes in society are linked to the
concept of sustainable development, which plays an important role as a cat-
alyst of social discourse and change.
436 Conclusions
NOTES
governance, it does not connect easily to the occupation of modern development with
itself, which appears in the repairing of the undesired side-eects of its own working. This
aspect is strong in the concept of reexive modernisation. Environmental protection and
technology assessment are examples of societal governance that is oriented towards its
own results without concern for the link between objective problems and subjective
approaches to problem solving. This material reexivity of governance can be observed
even when it is not cognitively reconstructed by the actors who conduct environmental
protection or technology assessment. As for the concept of reexive governance, we
further use a notion of reexivity that includes the unintentional and even unreected
self-confrontation of social action. To avoid confusion, however, it is advisable to
introduce a clear dierentiation between unintended reexivity as a condition of gover-
nance being confronted with side-eects and its cognitive reection and correspond-
ing adaptation of problem-handling practices as new governance approaches that cope
with side-eects by incorporating uncertainty, ambiguity and distributed control. The rst
form of unintended reexivity can be labelled rst-order reexivity; the second form of
reected reexivity can be labelled second-order reexivity.
REFERENCES
439
440 Index
risk 1415, 20, 24, 91, 226, 231, 238, Berlinski, D. 238
24042, 2456, 250, 252, 260, Bernstein, P. 237
389 Bernstein, R. 59, 69, 70
technology 85, 206, 227, 237, 238, Bezembinder, T. 242
242, 252, 254, 328, 371, 425 Biedermeier 49
ASSINSEL 392 Bieleman, J. 64, 65, 66
Association for Biodynamic Vegetable Bijker, W. 234, 236, 329
Plant Breeding 405 biological diversity
Atkinson, G. 391 Convention 25, 384, 38990, 3979,
Atlantic Conveyor 135 401, 406
Austria see also agrobiodiversity
bioreneries 355, 3578, 36279 biomass 1078, 115, 123, 124, 355, 357,
Factory of Tomorrow 212, 355, 357, 362, 364, 370, 372
35962, 365, 373, 3749 biopolymers 212
freight transport 210 bioreneries (Austria) 355, 3578,
Institute for Technology Assessment 36279
377 biosphere reserves 400, 403
Landscape Research 281, 286, black water 346, 348
35960 Blair, T. 233
Research Council 361, 368 Blanc, G. 322
research and development 35862, Blann, K. 144, 146, 149
3749 blind spots 60, 301, 306, 388
SD programmes 281, 286 Blueprint network 314
sustainable production 25, 35579 blueprint thinking 114, 118
transdisciplinary research 24, 295, Boehmer-Christiansen, S. 249
299, 3012 Bohmann, J. 227
autonomy 53, 55, 58, 86 Bhret, C. 5, 11
Bon, W. 32
back-to-the-land activists 3245 Bonner, J. 242
Balanced Scorecard approach 155 Borup, M. 166
Balzer, I. 190, 284 Bos, B. 67, 75
Bamm, A. 295 Botri, I. 137
Barnes, B. 246 bottom-up approach 320, 328, 333,
Bauman, Z. 38 369, 370, 378, 408
Bayesian methods 238, 240, 241 bounded rationality 72, 73
beacon principle 360, 377 Bourdieu, P. 77, 226
Bechmann, G. 11 Boyce, A. 234
Beck, U. 4, 6, 7, 323, 36, 39, 41, 44, Boyle, G. 325
46, 489, 53, 5962, 69, 83, 86, 93, Bradshaw, T. 257
126, 168, 225, 230, 237, 419, 421, Brand, K-W. 168, 174
427, 430 Brealey, R. 239
Bekke, H. 65, 67 breeding 25, 4037
Belgium 281, 2845, 2889, 291 Brezhnev doctrine 49
Bell, S. 155 Broecker, W.S. 135
Bellucci, S. 190, 198 Brookeld, H. 385
benchmarking 206 Brooks, H. 244
Benhabib, S. 70 Brouwer, H. 124
Bergmann, M. 11 Brown, G. 233
Berkhout, F. 166, 243, 253, 313, 317, Brundtland Commission 3
333 Bundeswehr 49
Index 441
European Union 16, 545, 67, 155, Farmers Rights 25, 384, 4024, 406
195, 196, 363, 391, 399 Federal Plan for Sustainable
CAP 66, 67, 383, 391, 407 Development (Belgium) 281, 285,
Europta, L. 237, 253, 254 286
EVA Lanxmeer, Culemborg 3467, feedback 4, 1113, 19, 85, 105, 125,
34952 339, 383, 389, 419, 422
evaluation 12, 1415, 425 loops 91, 109, 1567, 180, 308, 399
adaptive management 141, 1434 Fetivizig 139
transition management 11516, Feyerabend, P. 246
12022 bre composites 212
Evalunet 288 Fiorino, D. 258
expectations rst-order learning 1923, 205
of bioreneries 3624 rst-order reexivity 67, 422, 435
foresight and 194, 197, 199200 rst modernity 32, 34, 36, 59, 60, 61,
sustainability foresight and 169, 172, 64
174, 177 Fischer, F. 78, 169, 243
experiential learning 121 Fischer, M.M. 208
experimentation 12, 18, 93, 158, 333, Fisher, E. 249, 251, 255
422 Flyvbjerg, B. 70
niche-based approaches 245, Folgen-entlichkeiten 45
31334 Folger, T. 243
opening up/closing down 4335 Ford, J. 243
sanitation systems 25, 33753 foresight
scenario process 3767 adaptive 23, 189218
system innovation 10711, 11516, participative/reexive 3678
11819, 121, 1916, 2078, precautionary 24, 253, 25461
31321, 426 recent developments/decits 197200
see also adaptivity reexivity and 24953
exploratory scenarios 19 shaping transformation 1658
adaptive planning 200204 sustainability 23, 16284
methodology 20514 Forester, J. 78
sustainability foresight 166, 16971, Forster, M. 240, 255
1734, 17780 Fox, C.J. 63
externalities 5, 306 framing 90, 246, 248, 261
see also side-eects Frederichs, G. 11, 276
Eyzaguirre, P. 404 Freeman, C. 313
freight transport (Austria) 210
Faaland, J. 3212 Friends of the Earth 325
Faber, M. 243 Frhlich, J. 208
Factory of the Future 371, 374, 375, fundamental uncertainty 134, 135
377 Funtowicz, S. 9, 189, 238, 244
Factory of Tomorrow 212, 355, 357,
35962, 365, 3739 Gallopin, G.C. 10, 166
fallibilism 2478, 250, 260 Gavigan, J.P. 190, 1978
FAO 384, 3857, 392, 396, 398400, Geels, F.W. 77, 108, 31618, 33031
404 genetic engineering 33, 392
Farber, S. 257 genetic erosion 385, 388, 399
Farman, J. 244 genetic resources 385, 392, 393, 3957,
Farmer Field Schools 25, 401, 407 4025, 407
Farmers Association 391 Georghiou, L. 197
444 Index